The African Programme for Onchocerciasis Control (APOC) at mid-point: history, achievements and future challenges

THE ArRrcAN pRoGRA 4ME FoR oNcHocERcrASrs coNTRot (tooc) AT MID-POINT: HrsToRY/ ACHTEVEMENTS AM FUrlrRE CHATTENGES Matfltsct'tyts chltdlor yilltultnl ns n xyylcnnt nrtlu xnnls of rroyul ncfunc eiltmnsttolog on/anoc 29.tt,2rnt l. F, PREFACE APOC AT MID-POINT : SO FAR SO GOOD PREFACE APOC AT MID-POINT: SO FAR SO GOOD The African Programme for Onchocerciasis Control (APOC) was launched in December 1995 on the tidal wave of the resounding success of the 2l-year old Onchocerciasis Control Programme in West Africa (OCP). Six years later and now at the mid-point of its pre- determined existence it is time to take stock and plan for the second half. This special Supplement contains a set of articles that focus on some key areas of the activities of APOC in the first phase. Each article makes a critical appraisal of the major achievements and shortcomings of the Programme from the start of operations in 1996 and identifies the main challenges for Phase 2. A succinct account of the state of affairs at the birth of APOC would help to put the achievements and the challenges in better perspective. The ultimate goal of APOC is "to eliminate onchocerciasis as a disease of public health importance and an important constraint to socio-economic development throughout Africa". The prescribed strategy by which this goal is to be attained is "the establishment of a self-sustainable ivermectin treatment programme" in the high-risk zones of all the endemic countries outside the OCP area. Where feasible, control would be effected by local vector eradication. The five main pillars around which the success of APOC was predicated are: 1. The empowerrnent of the endemic communities within the context of Primary Health Care (PHC). 2. A unique global partnership between the private and public sectors including the affected communities themselves, forming a formidable alliance in pursuit of a common objective. 3. The development and opportune arrival on the scene, through research, of innovative tools and strategies that make the proposed control operations both feasible and cost-effective. 4. The unprecedented commitment of a Pharmaceutical Company - Merck & Co. Inc., to donate I through the Mectizan@ Donation Programme (MDP) the ivermectin (Mectizan@, for as long as needed to treat onchocerciasis, and 5. The vast experience of OCP and the synergy derivable from the physical proximity and temporal overlap of the two programmes. There were major concerns also. Notable among these was the probability of 'donor- fatigue', which has largely receded in the face of the impressive Economic Rate of Return (ERR) and general satisfaction with APOC's progress over the years. There were also some doubts about securing appropriate receptivity and commitment of the participating countries to the APOC initiative given the increasing and conflicting demands of emerging diseases on national budgets. A third and daunting concern was related to the scientifically derived estimation that ivermectin would have to be consumed by at least 65 o/o of the target populations of all the hyper- and meso-endemic areas at least once every year for upwards of 20 years if the objective of eliminating onchocerciasis from the continent is to be achieved. The progress made towards overcoming these concerns are contained in the accompanying papers. There have been many changes since APOC commenced operations. The pre-APOC estimate of number of people a risk and to be treated has more than doubled as a result of the application of REMO/GIS. The number of people treated every year with ivermectin, using APOC's Community-directed treatment (ComDT) approach has risen sharply from 8 million in 1996, to over 20 million in 2000. Substantial capacity building has taken place on all fronts, including diverse training programmes that range in content from administrative competence to skills needed for field operations. Devolution of operational research to endemic country scientists and the promotion and empowerment of local non-governmental organizations (NGDOs) by the International NGDOs are crucial investments towards sustainability of the programme that augur well for other future health and development programmes. The prospects of effectively integrating Community directed treatment with 2 Ivermectin (CDTI) into the health system, and its potential as an entry point for other programmes are very bright. The potential impact of APOC on the disease spectrum and the health services of participating countries are enorrnous. With the empowernent of the endemic communities within the context of PHC the health care stakes and bar of performance have been raised to a level that must be sustained in Phase 2 of APOC. There are many challenges to be faced in APOC Phase 2. Some of these have been identified and discussed in the seven articles of this supplement. Most of these understandably have to do with sustainability. Suffice it to state that if these challenges are to be successfully met then the partnership that has brought APOC to this commendable stage would have to remain committed. The Management of APOC and its support systems, like the Technical Consultative Committee (TCC), need to be adequately strengthened so as to be able to cope with the ever expanding scope of activities. This is more so now that OCP which has provided a substantial amount of administrative and infrastructural back up to APOC has only a few more months before winding up. As has been observed in all previously successful public health disease control programmes, notably the global smallpox eradication programme and OCP, contemporaneous scientific research is a sine qua non for ultimate success. APOC's main investment in operational research is channeled through WHO/TDR where the search for a macrofilaricide that is suitable for mass treatment and a new tool for rapid monitoring of treatment continues. In this regard it is nice to end on a cheerful note. As we were about to go to press, report of a TDR-APOC supported multi-centre study has indicated that a rapid epidemiological method for identiffing areas and levels of Loa loa endemicity has been developed that would provide an important tool for dealing with one of the major challenges that is being caried over from APOC phase I to phase 2, namely the issue of severe adverse events (SAE) following treatment with ivermectin in some areas where onchocerciasis and loiasis coexist. 3 According to the Report of the External Mid-Term Evaluation, "APOC has made significant and satisfactory progress towards meeting its objectives". This Supplement tells part of the story on how it has done so, how far it has gone and how much farther it must go before the curtain is drawn on its operations. Dr. Azodoga SEKETELI Director, African Programme for Onchocerciasis Control (APOC) 4 PARTNERSHIP AND PROMISE: EVOTUTION OF THE AFRICAN RIVERBTINDNESS CAMPAIGNS 1Partnership and Promise: Evolution of the African Riverblindness Campaigns AUTHORS: B. BENroN,' J. BuMp,'''' B. Lrcse,3Atto A. SEKETELI,a ADDRESSES OF AUTHORS: Onchocerciasis Coordination Unit, The World Bank, 1818 H. Street NW, Washington, DC 20433, U.S.A. Institute of the History of Medicine, The Johns Hopkins University, 1900 E. Monument Street, Baltimore, Maryland 21205, U.S.A. Human Development Network, Africa Region, The World Bank, 1818 H. Street NW, Washington, DC 20433, U.S.A. African Programme for Onchocerciasis Control, 01 B.P. 549, Ouagadougou, Burkina Faso Address for Correspondence Dr J. Bump Institute of the History of Medicine, The Johns Hopkins University, 1900 E. Monument Street, Baltimore, Maryland 21205, U.S.A E- mail: jbump@worldbank.org Short running title: Partnership and Promise of APOC ) 3. 4. ABSTRACT The article describes the evolution of the partnership between various health and developmental agencies that has sustained the campaign against riverblindness in Africa. The international community was oblivious to the devastating public health and socio-economic consequences of onchocerciasis until towards the end of the 1960s and the beginning of the 1970s when a UNDP- supported Mission to West Africa and a visit to the sub-region by the President of the World Bank culminated, in 1974, in the inauguration of the Onchocerciasis Control Programme in West Africa (OCP). OCP was a landmark event for the Bank as it represented its first ever direct investment in a public health initiative. The resounding success of the OCP is a testimony to the power of the partnership which, with the advent of the Mectizan Donation Program, was emboldened to extend the scope of its activities to encompass the remaining endemic areas of Africa outside OCP. The paper discusses the progress that has been made in consolidating the partnership and the prospects of adapting the various strategies of the African Programme for Onchocerciasis Control (APOC), such as the community-directed treatment (CDTI) concept so as to entrench an integrated approach that couples strong regional coordination with empowerment of local communities to address many other health problems. INTRODUCTION The control of West Africa's riverblindness is a glowing, but unlikely, success story in international public health. Despite its crippling effects, this once-widespread disease remained virtually undetected during the colonial period. Lost among the numerous and varied tropical diseases, onchocerciasis ranged unchecked throughout much of the 20h century as well, attracting international interest only in the last 50 years. Running since 1974, the Onchocerciasis Control Programme in West Africa (OCP) has eliminated riverblindness (onchocerciasis) as a public health problem in the l0 countries where it operates. Inaugurated in 1995, the African Programme for Onchocerciasis Control (APOC) extended operations to include the remaining 19 infested countries on the African continent. Given the low international profile of the disease and the technical difficulty of its control, it is particularly impressive that OCP and APOC have been successful. This article outlines how onchocerciasis became visible to two previously independent communities-international health experts and international development professionals-and follows the ongoing results of their collaboration, the OCP and APOC programmes. Onchocerciasis is a debilitating welter of skin disease, blindness, and itching to its sufferers (Buck, 1974). These symptoms are the body's immunologic response to millions of tiny worms (microfilaria), spewed forth by adult worms (macrofilaria), which live coiled in the human host for 10-14 years. If removed and unwound, these adult worms routinely measure 60 or more centimeters long (Malatt and Taylor, 1992). Conveyed in a juvenile phase via unlucky bites of the aptly named blackfly Simulium damnosum, onchocerciasis was, until recently, endemic to West Africa and even now remains a serious public health problem in the majority of countries on the continent. Commonly called riverblindness after its geographic locus and most visible symptom, those infected are cursed with interminable itching, thickening and depigmentation of the skin, and in an average l}Yo of cases, permanent blindness. In some heavily infected villages of pre-OCP West Africa, this figure was several times higher- blindness among adults often ran above 30% (Waddy, 1969). When highly prevalent, the disease eventually forces communities away from the fertile river valleys as village blindness rates reach devastating proportions. With too few able-bodied people left to tend fields, food shortages and economic collapse evict residents. Moving to hardscrabble highlands offers some ) respite from further infection, but is not without its own problems-poor soils and little water cripple farming efforts. Since 1974, OCP has sought to end this scourge. Despite riverblindness' wide-ranging consequences, it was an unlikely target of the international health community. Except to the affected communities, onchocerciasis was invisible until recently. [t went unnoticed by colonial administrations even as they fought schistosomiasis, sleeping sickness, syphilis, malaria, and a range of other labor-related diseases. Its transmission cycle was not described until 1926 (Blacklock, I926a, 1926b). It was not definitively linked to blindness until almost 1950 (Waddy, 1949). Riverblindness is also an exclusively rural disease, affecting only the poorest and most remote communities-populations with the fewest resources and the least access to health services. Further, the disease is technically difficult to control because of the worm's long lifespan, the prevalence of infected populations, the vector's very long flight range, and the absence of suitable pharmaceutical options before 1987. These factors combine to suggest that onchocerciasis could have easily gone unchecked, ravaging locally, unnoticed internationally. All the more curious then, that the international health community should find one of its greatest successes in onchocerciasis control. Rising awareness among international health professionals Africans were all too familiar with onchocerciasis, but full knowledge of its characteristics and symptoms eluded western medics until after World War II. During the colonial period, "craw- craw," as it was then known, was identified as a skin disease. Reaffirming European notions of disease and place, craw-craw went unnoticed as one among countless tropical afflictions. A more precise understanding began to emerge in 1875, when Surgeon John O'Neill of the H.M.S. Decoy identified craw-craw as a filarial disease, that is, one caused by a parasitic worm. Moored off Cape Coast Castle, Ghana, O'Neill borrowed six patients from Addah Fort Hospital, noting "[craw-craw's] intractability, contagiousness, and irritating nature so aroused my attention that I was induced to bestow much time on its microscopic examination, and succeeded at length in discovering a filaria..." (O'Neill, 1875, p.265). The filaria itself was later described by Leuckart in 1893, who named it "Onchocerca volvulus." The next breakthrough waited more than 30 years. In a pair of 1926 publications, D. B. Blacklock of the Sir Alfred Lewis Jones Research Laboratory in Freetown, Sierra Leone cleverly unraveled the parasite's transmission and development cycle (Blacklock,1926a, D). Blacklock 4 traced the parasite through the blackfly vector-Sizulium damnosum-discovering that, as with malaria, the onchocerciasis parasite undergoes an intermediate maturation in the insect. These discoveries of riverblindness' cause and transmission cycle were followed by a deepening understanding of the disease's effects. Though West African blindness rates were long known as among the world's highest, onchocerciasis-historically viewed as a skin disease-has only recently received it's share of the blame. In its stead, the role of other perennial culprits was magnified-Vitamin A deficiency and trachoma, to name two. A tentative link between blindness and onchocerciasis had been made as early as 1916 by Robles (Luna, 1918), working in Guatemala, where the disease had been inadvertently imported by slave traders (Nelson, 1991), and by Rodhain in 1920, working in the Belgian Congo. Community observations by Hissette in 1931 were the first indication of a major public health problem-studies in the Sankuru river area of Congo revealed among the population 20o/o blindness and 50% "ocular complications." Finally, a definitive study by Waddy in L949 conclusively linked onchocerciasis and Africa's rampant blindness. Present-day understandings of onchocerciasis consequently emphasize blindness, in addition to the long-identified dermatologic symptoms. As the full effects of the scourge became known, onchocerciasis attracted more and more attention from European and American specialists; particularly well represented were researchers from ORSTOMT (assigned to OCCGE2), reflecting longstanding French colonial involvement in the region. Many of ORSTOM's studies later proved crucial to OCP (Le Berre, 1966; Duke, 1990) Armed with knowledge of the disease and its effects, public health workers struggled with a limited arsenal. Chemotherapeutic options suramin, a macrofilaricide, (Van Hoof, 1947) and diethylcarbamazine (DEC), a microfilaricide, (Hewitt, 1947) both proved unworkable outside the hospital because of severe side effects. A surgical option, mass nodulectomy, proved ineffective in trials (WHO 1987). Vector control remained a possibility, but struck many as unworkable in light of the environmental problems of DDT and the unsuccessful Global Malaria Eradication Programme of the 1950s and 1960s. Large-scale international attention first came to the disease in Africa in a 1968 technical conference in Tunisia (WHO 1969), initiated largely by M6dicin-G6n6ral Pierre Richet, Dr. ' Office de la Recherche Scientifique et Technique d'Outre-Mer (ORSTOM), now named " Institut de Recherche pour le Developpement (IRD)". 2 Organisation de Coordination et de Coop6ration pour la Lutte contre les Grandes Endemies. 5 Ansari, WHO's Chief of Parasitic Diseases (Duke, 1990), and an important paper later published by B. B. Waddy (1969). Arranged by USAID, OCCGE, and WHO, delegates convened to discuss the technical feasibility of controlling onchocerciasis in West Africa, where its effects were most severe. Conference attendees concluded that given sufficient resources, control might be possible, but would take two decades or more. Four years later, a UNDP-supported Preparatory Assistance to Governments (PAG) mission left for West Africa to plan a long-term control strategy. The development assistance community becomes involved Over the course of the 20'h century, onchocerciasis worsened in West Africa. Growing populations, forced migration, climatic changes, and colonial ambivalence all exacerbated an upswing in the centuries-old tide of infection (Hunter, 1966). On the one hand, the rising disease toll multiplied the humanitarian and economic impact of onchocerciasis. On the other hand, an unusually harsh series of droughts intensified the need for agricultural development- particularly in the oncho-infested river valleys, with their better soils and ready water sources. The confluence of these two pressures exacted its cost in health and productivity, but also forged the alliance that would eventually succeed against the widening scourge. By 1972, the international health community was already mobilizing to fight the disease. Because of the drought and famine, the international development community was also focused on the fate of West Africa. Accordingly, and at the same time as the PAG mission, World Bank President Robert McNamara went to West Africa, where he saw the broken villages and fallow fields of the endemic zones during a flight to Ouagadougou, Upper Volta (now Burkina Faso), a then- common feafure of the West-African savanna. From the air he saw curious clusters of circular forms-the surviving skeletons of once-occupied homes. Upon landing in Ouagadougou, McNamara, struck by the devastation of the disease, quickly arranged a special trip to Bobo Dioulasso, Burkina Faso's second largest city, where he would meet PAG mission leader B. B. Waddy and French scientists, Rene Le Berre, an entomologist, and Jean-Jacques Picq, a microbiologist in the French military (World Bank, 1972). Deeply concerned, McNamara made his own investigation, discussing onchocerciasis at length with the scientists in Africa. Upon his return to Washington he continued his inquiry, making use of his extensive connections, including researchers at the Wilmer Eye Institute at the Johns Hopkins University. McNamara became convinced that control of the disease was technically possible, given enough time and money-projected at 20 years and $120 million 6 1972 dollars (WHO, 1,973). Armed with personal conviction, McNamara threw his own weight and that of the institution he led behind the budding onchocerciasis control efforts. However, in managerial terms, the World Bank was not equipped to direct a public health campaign. In fact, at that time, the World Bank had never even made a loan for a health programme. Financing OCP would be the Bank's first initiative in the sector, a striking departure from the early '70s norm of large dams and massive infrastructure projects. Meanwhile, WHO already had scientists at work, and UNDP and FAO were developing follow- up agricultural plans. McNamara saw his comparative advantage-the World Bank could use its reputation and leverage with donors to fund the control programme and manage the finances. He insisted on unrestricted contributions made as grants only, and prepared donors for a lengthy campaign. The Onchocerciasis Control Programme in West Africa Through a combination of persistence, dedication, and happenstance, the Onchocerciasis Control Programme evolved from an ambitious plan to a sterling example of disease control. At the nexus of health and development, riverblindness served as a rallying point for many previously distant groups. The launching of OCP in 1974 formalized an unprecedented collaboration between seven host countries, four international organizations, and nine donors. Sponsoring agencies included the World Health Organization as the executor, the World Bank as the fiscal agent, and the United Nations Development Programme and the Food and Agriculture Organization sharing various development planning responsibilities. This partnership joined intemational health professionals with their counterparts from the international assistance community. The confluence of these two groups reflected growing appreciation for the importance of health to development. When Programme operations began in 1974, the blackflies that transmit riverblindness were so numerous that protection against them in most rural areas was impossible. These swarns of flies were kept infective by a large parasite reservoir in the human population. In the mid-1970s, entomologists recorded fly biting rates in the thousands per person, per day in some heavily infected areas (Walsh, 1977). In these and other so-called " oncho zones," riverblindness reached devastating proportions, often infecting 90%o or more of the population (Crisp, 1956; Brown, 1962). Without a drug treatment, the only effective approach to onchocerciasis control was to intemrpt parasite transmission by substantially reducing the density of infected flies. The flies could be targeted at the larval stage because their breeding 7 sites were confined to areas of fast-flowing water, such as rapids and dam spillways. Larvicide spraying could therefore be concentrated on these locations, killing the disease-transmitting flies before maturity. Strict environmental monitoring and follow-up protective measures ensured the long-term health of non-target fish, flora, and fauna. Until the late 1980s, by necessity, OCP was based exclusively on this strategy of vector control. Options expanded in the 1980s when Merck's Mectizan@ (ivermectin, MSD) was shown to be effective against the microfilariae that manifest riverblindness (Aziz et al., 1982a,1982b; Sutherland and Campbell 1990), without the harmful side effects of DEC (Greene et al., 1985; Larviere et al., 1985; Awadzi et al., 1986). Because ivermectin does not kill adult worrns, which continue to produce damaging microfilariae, keeping the symptoms at bay therefore requires regular doses for the remainder of the adult worm's life, up to 14 years. The advantages of ivermectin are twofold. First, it relieves the symptoms and prevents further damage. Second, because it kills the juvenile worrns, blackflies are less likely to ingest (and later transmit) the parasites even if they do bite a victim while he or she still harbors productive adult worms. This second benefit, slowing transmission, is particularly helpful when beginning control efforts in new areas because it initially reduces the microfilarial load in the population faster than larvicide spraying. In the original OCP area, control through larviciding had been mostly achieved by the time ivermectin became available, but the drug proved a key advantage in heavily infected areas and added a therapeutic dimension to the programme, providing relief for victims and arresting its progress. Additionally, World Bank staff, in particular were intrigued by the expansion possibilities held by the new drug. Following French regulatory approval in 1987 (Brown and Neu, 1990), OCP conducted extensive field trials to assess ivermectin's public heath potential (Remme et al, 1990; Dadzie et al. 1991). It was already known as a potent drug for individuals, but could it intemrpt regional transmission? What coverage would be necessary to protect communities? Underwriting these investigations was Merck's pledge to donate ivermectin in whatever quantity needed and for as long as necessary. Community trials were supplemented with extensive modeling using ONCHOSIM, a transmission simulator developed with OCP data by researchers at the University of Rotterdam (Plaisier, 1990). These intensive epidemiological studies indicated that with long-term ivermectin coverage alone-no larvicide spraying-it might be possible to control riverblindness. The African Programme for Onchocerciasis Control 8 IEmboldened by their success in West Africa and empowered by Merck's donation, the riverblindness partnership embarked on a broader mission-defeating the disease throughout the continent. Building on the knowledge and experience gained in OCP, the sponsoring agencies in 1995 launched a second programme to combat the rest of Africa's riverblindness, the African Programme for Onchocerciasis Control (APOC). APOC emphasizes the ivermectin strategy studied initially under OCP-using long-term distribution of the drug to eliminate sickness and slow transmission by reducing the parasite reservoir in humans. Larviciding is used only peripherally because APOC's vast area makes spraying too costly, and thick forests covering the area's principal rivers render aerial delivery ineffective, as well. APOC is made possible by Merck's continued donation of unlimited supplies of Mectizan. This generosity and the drug's effectiveness have facilitated the expansion of riverblindness control to the remaining infested areas of Africa. Riverblindness control based on ivermectin presents some advantages over larviciding alone, such as immediate relief for victims, but also raises new challenges, the largest of which involves sustaining a drug coverage threshold long enough to intemrpt transmission. With rural diseases such as riverblindness, people who most need the drugs are often the hardest to serve, living beyond the reach of national health services. And riverblindness victims need doses every six to twelve months as long as they harbor even one adult worm. Further, communities must take the drug for two decades or more to have any definitive impact on transmission. For this reason, APOC's Community-Directed Treatment initiative (Com-DT) has developed extensive networks of Community Drug Distributors (CDDs), appointed by their peers to work with APOC's NGDO (non-governmental development organizations) partners to distribute Mectizan on a sustainable basis and to share knowledge of the disease. APOC's reliance on the regular and continued administration of the drug makes this distribution crucial to success. WHO's expert epidemiologists estimate that communities at risk will need to take ivermectin for approximately 20 continuous years to eliminate the disease as a public health problem. During this period, these scientists calculate, APOC must ensure at least 650/o treatment coverage throughout the affected communities. In 2000, members of the Community-Directed Treatment network achieved an average ivermectin coverage of 74%o in the targeted communities (WHO 2001). These distribution efforts were augmented by a strong commitment to capacity-building: in the same year APOC trained or re-trained more than 77,000 people. The 22 million people now reached annually by APOC represent less than half of those who will be targeted as APOC expands over the next few years. 9 OCP's 27 years of operations form the knowledge base on which APOC is founded, though on the surface it would appear that the two control strategies are unrelated. In fact, APOC's approach represents a natural evolution from OCP, incorporating new treatment strategies but relying heavily on OCP's technology and fine-tuned database of riverblindness epidemiology, which has been painstakingly constructed over nearly three decades. The involvement of communities is also a natural extension of the devolution integral to OCP all along. While the first staff members in Ouagadougou were primarily Westerners, the Programme set out quickly to hire and train Africans. Since 1980, for instance, all Directors have been Africans. Overall, OCP has funded several hundred graduate degrees, all awarded to Africans who were working against riverblindness (Samba, 1994). By 1984, ten years into the prograrnme, 96oh of the staff were African. Now Africans comprise more than 99o/o of the personnel. In addition to devolution, OCP made groundbreaking contributions in other areas as well. From the earliest days, OCP has embodied a culture of operations research, which has rescued the Programme more than once. Around 1980, for instance, the blackflies began developing resistance to the main larvicide (Kurtak, 1990). Thanks to several years of ongoing research, OCP scientists were able to develop and test alternatives. Using some of these new options in a rotation overcame the resistance in the fly population, restoring the effectiveness of vector control. Another major difficulty came to light when OCP entomologists discovered that the fly's effective flight range was 400 kilometers or more (WHO, 1987)-nearly 10 times original estimates (WHO, 1973). The problem appeared as infected flies were discovered re-invading previously controlled areas (Garms et al., 1979). Ongoing epidemiological and entomological research revealed the problem while superb management and redoubled donor support allowed OCP to expand into the nearby zones where new flies had been discovered (Philippon et al., 1990). Similarly, APOC's operations were underwritten by extensive preparatory research including epidemiological mapping (De Sole er al., 1990), studying distribution methods (TDR, 1996), examining the sustainability of the Com-DT approach (WHO, 2000), and investigating community compliance and the effects of ivermectin on skin disease (Brieger et al., 1998). Future Prospects APOC is fundamentally about riverblindness, but the ivermectin distribution network it has created stands ready to serve as an essential vehicle for addressing other widespread health problems. As Merck has done with Mectizan, several drug companies are prepared to donate medicines that can be delivered via APOC at minimal cost to the poorest communities. Among 10 the simplest and most effective would be vitamin A capsules, given free by Hoffman-LaRoche, which would help prevent blindness in children and improve general health in the rest of the population. Aside from the health benefits, providing additional drugs would strengthen the position and effectiveness of community Mectizan distributors. Great care must be taken not to overwhelm this nascent distribution network, but in the case of vitamin A, little training would be needed, and everyone can take it. Further building local capacity by distributing vitamin A would pave the way for making available several other donated drugs in the communities that have been the hardest to reach. The full value of APOC's distribution network lies in its potential. In it, the international community has a ready pathway to deliver medicines to those who need it most. Drugs already available at no cost include the following, all of which have the same dosing schedule as Mectizan: Vitamin A to prevent malnutrition, blindness, death Azythromycin to cure trachoma, now the leading cause of blindness in Africa Ivermectin and albendazole to stop transmission of lymphatic filariasis (elephantiasis), one of Africa's leading causes of long-term disability Despite recent, sharp reductions in price, AIDS medications are still quite expensive, but are among the many health interventions that could conceivably be delivered through this network in the future. Once fully developed, there would be virnrally no limit to the materials that could be distributed, including health information, condoms, multivitamins, and vaccines. There is the very real possibility of knocking out not just one health problem, but several major diseases-enough to dramatically improve public health overall, instead of just making opportunities for diseases now of secondary importance. The Community Directed Treatment (Com-DT) initiative stands as a fitting coda to the legacy established by OCP. What began as a top-down, vertical disease programme has evolved into a bottom-up, integrated approach that couples strong regional coordination with the empowerment of local communities to address not only onchocerciasis but, potentially, many other health problems as well. For much of its history, OCP has been conducted with helicopters, spraying against a formidable, but solitary foe. APOC has extended this success with the help of Mectizan to include local communities not as passive beneficiaries, but as powerful agents impacting some of their own health outcomes. This grassroots involvement, if fully strengthened, would complete OCP's transformation from an external programme to a fully African-owned and managed region-wide health system. 1l The success of OCP testifies to the power of this partnership, in which sponsors have concentrated on their respective strengths and the donor community has lent unwavering support for nearly three decades-far longer than for any other operational development programme. Given the technical hurdles, this commitment has been essential to success and remains one of OCP's finest hallmarks. No combination of shorter programmes could have achieved these results nor evolved horizontally, affording potentially much wider protection of the public health. The longstanding collaboration and goodwill developed on all sides of this partnership have generated synergies greater than those imagined, and all at remarkably low cost. Coverage has cost, at every point in both programrnes, far less than one dollar per year, per person protected (Benton and Skinner, 1990; Benton, 1998). OCP's efforts have led to widespread acclaim among public health practitioners, but perhaps more importantly, have led to broad-based support from the communities protected. This gtassroots credibility is the single most important factor in APOC's success at Mectizan distribution, and portends a transformation in the ability to deliver a wide range of medications in the future. Development assistance prograrnmes have long sought comprehensive ways of addressing widespread public health problems while ensuring sustainability, local involvement, and community empowerment. OCP and APOC have achieved these goals by virfue of nearly 30 years' dynamic partnership with increasingly active host countries and their constituent populations, proficient intemational agencies, and unswervingly dedicated donors. ACKNOWLEDGEMENTS We appreciate the valuable help given by Prof. O. O. Kale in the preparation of this paper. The assistance of Marline Alexis, Joyce Musya-Mpangu and Olympia Gjino of the Onchocerciasis Coordination unit of the World Bank is gratefully acknowledged. REFER.ENCES AWADZI, K., DADZIE, K. Y., SCHULZ-KEY, H., GILLES, H.M., FULFORD, A. J, & AZ\Z,M. e. (1986). The Chemotherapy of onchocerciasis XI: A double-blind comparative study of ivermectin, diethylcarbamazine and placebo in human onchocerciasis in Northem Ghana, Annals of Tropical Medicine and P aras itologl, 80, 433-442. t2 AZtz, M. A., DIALLo, s., DIop, I.M., LARIVIERE, M., & poRtl, vt. (1982a). Efficacy and tolerance of ivermectin in human onchocerciasis, Lancet ii, 17l-L73. AZ\Z,M. A., DIALLO, S., LARIVIERE, M., DIOP, I. M., PORTA, M., & GAXOTTE, P. (1982b).Ivermectin in Onchocerciasis, Lancet ii, 1,456-1,457 . BENToN, a. (1998). Economic impact of onchocerciasis control through the African Programme for Onchocerciasis Control: an overview, Annals of Tropical Medicine and Parasitologt, 92, (Suppl.), S33-S39. BENToN, B. & sKr.{NER, E.D. (1990). Cost-benefits of onchocerciasis control, Acta Leidensia,59, 4054rr. BLACKLoCK, D. B. (1926a). The development of Onchocerca volvulus in Simulium damnosum, Annals of Tropical Medicine and Parasitologt,20,148. BLACKLOCK, D. B. (1926b). The further development of Onchocerca volvulus LBucrenr in Simulium damnosum THeos. Annals of Tropical Medicine and Parasitologt,20r203-218. BRIEGER, W. R., AWEDOBA, A. K., ENEANYA, C. L, HAGAN, M., OKELLO, D., OGBUAGU, K. F., osANGA, o. o. & ovucA, E. B. L. (1998). The effect of ivermectin on onchocercal skin disease and severe, itching: results of a multi-centre trial. Tropical Medicine and International Health, 3, 95 1-961. BRowN, A. w. A. (1962). A survey of Simulium control in Africa, Bulletin of the World Health Organisation, 27, 51Ll-527 . BRowN, K. R. & NEU, D. c. (1990). Ivermectin-clinical trial and treatment schedules in onchocerciasis, Acta Leidensta, 59, 169-17 5. BUCK, e. a. (1974). Onchocerciasis: symptomatologt, pathologt, diagnosls. Geneva: WHO. czuSP, G. (1956). Simulium and Onchocerciasis in the Northern Territories of the Gold Coast. London: H. K. Lewis. 13 DADZIE, K. Y., REMME, J., DE SOLE, G., BOATIN, B., ALLEY, E. S., BA, O. & SAMBA, E. M. (1991). Onchocerciasis Control by large-scale ivermectin treatment, Lancet,337, 1358-9. DE SOLE, G., GIESE, J., KEITA, F. M. & REMME, J. (1990). Detailed epidemiological mapping of three onchocerciasis foci in West Africa, Acta Tropica 48,203-213. DUKE, B. o. L. (1990). Human Onchocerciasis-An overview of the disease, Acta Leidensia,59, 9-24. GARMS, R., WALSH, J. F., & DAVIES, t. a. (1979). Studies on the reinvasion of the Onchocerciasis Control Programme in the Volta River basin by Simulium damnosum s.l. with emphasis on the South-Western areas." Tropenmedizin und Parasttologie, 30, 345-3 62. GREENE, B. M., TAYLOR, H.R., CUpp, E.W., MURpHy, R.p., WHITE, A.T., AZIZ, M.A., SCHULZ-KEY,H., D'ANNA, s.A., NEWLAND, H.s., GoLDSCHMIDT, L.p., AUER, c., HANSoN, A.p., FREEMAN, s.v., REBER, E.w., & wILLIAMS, r.N. (1985). Comparison of ivermectin and diethylcarbamazine in the treatment of onchocerciasis, New England Journal of Medicine,313, 133-138. HEWITT, R. I., KUSHNER, S., STEWART, H. W., WHITE, 8., WALLACE, W. S., & SusseRow, V. (1947). Experimental Chemotherapy of Filariasislll,Journal of Laboratory and Cltnical Medicine, 32, t3r4-1329. HISSETTE, J. (1931). Sur I'existence d'affections oculaires importantes d'origine fil arienne dans certains territoires du Congo, Annales de la SociLte Belge de Mddecine Tropical, 11,45-46. HooF, L. vAN (1947). Sur la chimioth6rapie de I'onchocercose (note pr6liminaire) Annales de la Sociritd Belge de Mddecine Tropical, 27,173-77. HUNTER, J.M. (1966). River Blindness in Nangodi, Northern Ghana: a hypothesis of cyclical advance and retreat, Geographical Review, 56, 398-416. LARIVIERE, M., VINGTAN, P., AZIZ, M., BEAUVAIS, B., WEIMANN, D., DEROUIN, F., GTNOUX, J., SCHULTZ-KEY, H., GAXoTTE, p., BASSET, D., & SARFATT, C. (1985). Double-blind study of t4 ivermectin and diethylcarbamazine in African onchocerciasis patients with ocular involvement, Lancet, ii, 17 4-17 7 . LE BERRE, R. (1966). Contribution ir l'6tude biologique et 6cologique de Simulium damnosum Theobald 1903 (Diptera, Simuliidae). Paris : Office de la Recherche Scientifique et Technique Outre-Mer. LUNA, P. (1918). Disturbances of vision in patients harboring certain filarial tumors (Onchocerca). Amertcan Journal of Ophthalmolog/, (Suppl. 3) 1,122-125. KURTAK, D. (1990). Maintenance of effective control of Simulium damnosum in the face of insecticide resistance. Acta Leidensia, 59, 9 5-l 12. MALATT, A. 8., & u. n. TAYLOR (1992). Onchocerciasis, Infectious Disease Clinics of North America, (Special issue on Ocular Infections, M.Barza and J. Baum, eds.) 6, 963-977. NELSON, c. s. (1991). Human onchocerciasis: notes on the history, the parasite, and the life cycle, Annals of Tropical Medicine and Parasitologt,85,83-95. o'NEILL, J. (1875). On the presence of a filaria in'Craw-Craw'. Lancet, i, 265-266. PHILIPPON, 8., REMME, J. H., WALSH, J. F., GUILLET, P., & ZERBO, D.G. (1990). ENIOMOIOgiCAI results of vector control in the Onchocerciasis Control Programme, Acta Leidensia, 59, 79-94. PLAISIER, A. P., OORTMARSSEN, G. J. VAN, HABBEMA, J. D. F., REMME, J., & ALLEY, E. S. (1990). ONCHOSIM: a model and computer simulation program for the transmission and control of onchocerciasis. Computer Methods and Programs in Biomedicine, 31,43-56. REMME, J., DE SOLE, G., DADZIE, K. y., E. S. ALLEY, E. S., BAKER, R. H. A., HABBEMA, J. D. F., PLAISIER, A. P., OORTMARSSEN, c. J. vAN, & SAMBA, e. rra. (1990). Large scale ivermectin distribution and its epidemiological consequences, Acta Leidensia, 59, 177-191. 15 RoDHATN, J. (1920). Observations diverses concernant Onchocerca volvulus, Bulletin de la Socititti Pathologie Exotique, Paris, 13, 848-858. sAMBA, e. u. (1994).The Onchocerciasis Control Programme in West Africa. Geneva: WHO. 5STHERLAND, r. H, & ceNapeBLL, w. c. (1990). Development, pharmacokinetics and mode of action of ivermectin, Acta Leidensia 59(l-2),161-168. wADDy, B. B. (1949). "Onchocerciasis and blindness in the Northern Territories of the Gold Coast", unpublished report, 1949. Cited in K. David Patterson, "Riverblindness in Northem Ghana, 1900-1950," pp. 88-117 in Gerald W. Hartwig and K. David Patterson, eds. Disease tn African History, (Durham: Duke, 1978). wALSH, t. r. (1977) On the biology and control of vectors of human onchocerciasis-with special reference to West Africa. Paper presented to WACP-Symposium on Onchocerciasis, Accra 6-8 July 1977. woRLD BANK (1972). Unpublished archival document. Robert S. McNamara Collection, Scrapbook: Minutes-President's Council, Index Yol.2 of 2. woRLD HEALTH oRGANrsArroN (1969). Joint USAID/OCCGE/WHO techntcal meeting on the feasibility of onchocerciasis control. Tunis, l-8 July 1968. (WHO-ONCHO/69.75) woRLD HEALTH oRGANISATToN (1973). PAG Mission: Controle de l'onchocercose dans la rdgion du bassin de la Volta: rapport de la mission d'assistance preparatoire aux gouvernements de C6te d'Ivoire, Dahomey, Ghana, Haute-Volta, Mali, Niger, Togo. Gendve: WHO. woRLD HEALTH oRGANTsATIoN (1987). WHO Expert Committee on Onchocerciasis, Third Report. Geneva: WHO. woRLD HEALTH oRGANrsATroN (1996). Report of a Multi-Country Study on Community Directed Trea tment with lvermeclin. TDR/AFR/RP/96. 1 . Geneva: WHO. 16 woRLD HEALTH oRGANISATIoN (2000). Implementation and Sustainability of Community Directed Treatment of Onchocerciasis with lvermectin: Report of a Multi-Country Study. TDRflDE/RP/CDTV00. 1. Geneva: WHO. woRLD HEALTH ORGANISATION (2001). African Programme for Onchocerciasis Control Programme Document for Phase II (2002-2007) and the Phasing-Out Period (2008-2010). Forthcoming. 17 THE ACHIEVEMENTS AND CHALLENGES OF THE AFRICAN PROGRA MME FOR ONCHOCERCIASTS coNrRoL (apoc) THB ACHIEVEMENTS AND CHALLENGES OF THE AFRICAN PROGRAMME FOR ONCHOCERCIASIS CONTROL (APOC) AUTHORS A. sprBrell,t G. ADEoyE,2 A. EYAMBA,3 E. I.NORU*,4 p. DRAMEH,' u. v. AMAZIGo,' M. NoMA,' F. AGBoroN,l y. AHoLou,'o. o. xaLp6ANDK. y.DADZTE' ADDRESSES OF AUTHORS l. African Programme for Onchocerciasis Control (APOC), Ouagadougou, Burkina Faso 2. University of Lagos, Akoka, Yaba, Lagos, Nigeria 3. The Carter Center, Global 2000 River Blindness Program, Yaounde, Cameroon 4. Department of Dermatology, College of Medicine, University of Nigeria, Enugu Campus, Enugu, Nigeria 5. World Health Organization, Geneva, Switzerland 6. Department of Preventive and Social Medicine, University College Hospital, Ibadan, Nigeria 7. 01905 Osu - Accra, Ghana Address for Coruespondence Dr. Azodoga S6k6t6ti Director, African Programme for Onchocerciasis Control (APOC), 01 B. P. 549 Ouagadougou 01 Burkina Faso Tel: (226) 34.22.77 Faxz (226) 34.48.00 E-mail: seketelia@oncho.oms.bf Short running title: Achievements of APOC ABSTRACT The Community Directed Treatment with Ivermectin (CDTI) strategy of APOC has enabled the Programme to reach, empower and bring relief to remote and underserved onchocerciasis endemic communities. With CDTI, geographical and therapeutic coverage rates have increased substantially, in most areas, to levels required to eliminate onchocerciasis as a public health problem. Over 20 million people received treatment in 2000. APOC has also used REMO-GIS effectively to provide information on the geographical distribution and prevalence of onchocerciasis, as a means of identifying CDTl-priority areas, and obtaining better estimates of the numbers of people to be treated. A unique public/private sector partnership has been at the heart of APOC's relative success. Through efficient capacity building, the Programme's operations have positively influenced and strengthened the health services of participating countries. These laudable achievements notwithstanding, APOC faces many challenges during the second phase of its operations when the full impact of the Programme is expected to be felt. Notable among these is the sustainability of CDTI, its effective integration into the health care system and exploiting its potential as an entry point for other health programmes such as the lymphatic filariasis elimination programme, which would feature on the agenda of many participating countries during APOC's Phase 2. Executing these other programmes without compromising the onchocerciasis control programme itself is a major challenge to APOC. Success in meeting these challenges will depend on the continued wholehearted commitment of all partners especially governments of participating countries. 2 INTRODUCTION The vast majority of those who suffer from and are exposed to onchocerciasis live in Africa (WHO, 1995a). The most severe consequences of the disease is blindness, which may affect one third of the adult population of the most highly affected communities and the prevention of which has been the main clinical raison d'€tre for initiating the Onchocerciasis Control Programme in West Africa (OCP) in 1974 (Tsalikis, 1993; Benton et al., 2001). However in the last decade important pioneering studies, sponsored by the UNDPAVoTId Bank/WHO Special Programme for Research and Training in Tropical Diseases (TDR), have shown that onchocercal skin disease (OSD) is associated with a greater degree of morbidity than was hitherto appreciated. These studies demonstrated that severe OSD and intolerable itching cause a lot of suffering to millions of people particularly in the forest zone where the blinding form of the disease is less prevalent. (Amazigo and Obikeze, 1991, Amazigo, 1993, WHO, 1995b,). It not only causes psychosocial problems, ostracism and stigma (Okello et al., 1995; Owga et al., 1995, WHO, l995b,Brieger et al., 1998b) but OSD also has a demonstrably negative socio-economic impact on farmers' productivity, breastfeeding and school attendance. (Amazigo 1994; Kim et a|.,1997 ; Oladepo el al., 1997; Benton, 1998 ; Vlassoff et a1.,2000). The relative contributions of both major clinical patterns, ocular and dermal, to the burden of onchocercal disease and their socio-economic consequences are considerable. (Kale, 1998). The principal strategy of OCP from its inception was vector control. However the advent of ivermectin (Mectizan@) and its donation by Merck & Co. in 1987, free of charge for as long as needed, provided a second string to OCP's operational bow. The first extensive field studies on the suitability of the drug for use on a mass scale were conducted by the OCP. The favourable results of these studies led to OCP adopting mass distribution of Mectizan@ as an adjunct to vector control. It even used the drug alone in some areas (Awadzi et al., 1985; Dadzie et al., 1987, Remme et al., 1989; Dadzie et al., 1990; Remme et al., 1990; Dadzie et al., l99l; 3 Whitworth et al., l99l;1992 Guillet et. al., 1995). The application of the two strategies by OCP has led to the virrual elimination of onchocerciasis as a public health problem and an obstacle to socio-economic development in the 1l countries of OCP leading to the recognition of OCP as one of the most successful prograflrmes in the history of development assistance (Kim and Benton, 1995). The initial efforts at mass distribution of ivermectin outside OCP were made by non- govemmental development organizations (NGDOs) a few years prior to the establishment of the African Programme for Onchocerciasis Control (APOC). The first of the NGDO-facilitated distribution progralnmes in Africa that came to be referred to as the Ivermectin Distribution Programme (IDP) (Duke and Dadzie 1993), was established Nigeria in 1989. Many of the pioneering NGDOs were already well known through their activities in prevention of blindness and were already with the WHO Prevention of Blindness Programme (PBL). At that time, outside of the OCP area, authorities in most countries where the disease was endemic did not consider the disease as a public health problem and there were therefore no programmes or structures in place for its control. The NGDOs soon recognised the need to coordinate their separate and independent efforts if they were to achieve their common goal. Therefore in 199? they came together to form the NGDO Coordination Group for Ivermectin Distribution. By 1995, it had become clear that members of the Coordination Group needed considerably more resources than they could generated on their own if there was to be any significant expansion in the scope of their activities. Furthermore the various NGDOs and programmes used mobile teams and the clinic-based and Community-Based Treatment with Ivermectin (CBTI) methods in the distribution of ivermectin. These methods were clearly not appropriate or cost-effective for large-scale sustainable distribution of the drug. In 1995 the Task Force on Onchocerciasis Operational Research (OOR) of the TDR, in collaboration with the OCP, addressed the problem in a multi-country research study. The result was the development of the method of distribution 4 that has become known as the Community-Directed Treatment with Ivermectin (CDTI) (WHO,l996a). Prior to this, and in response to the need of the disparate control programmes of the various NGDOs for a rapid, reliable and cost-effective method of identifying communities endemic for onchocerciasis to be targeted for treatment, the OOR developed the Rapid Epidemiological Mapping of Onchocerciasis (REMO), (Ngoumou and Walsh, 1993; Ngoumou et al., 1994l' WHO, 1995c). The underlying concepts and design of REMO drew heavily on the experience in OCP. The World Bank provided financial support in the development of REMO, which has since become a key tool in the control armamentarium of APOC. This then was the scenario on the onchocerciasis front in Africa when APOC, inaugurated in December 1995, commenced operations in 1996 with a mandate "to build on the success of OCP and establish sustainable control (of onchocerciasis) in the remaining 19 countries in Africa where the disease was still a public health problem". (WHO,l996b). The entire Programme was manifestly predicated on the window of opportunity provided by the Mectizan Donation Programme (MDP). Thus the control strategy prescribed for APOC was "community-based treatment prografirmes with the drug ivermectin, supplemented with vector eradication in a few isolated foci". This paper is an overview of the achievements of APOC in the first (1996-2001) of two planned phases of its operations, and a consideration of the challenges for the second phase (2002-2007). Six companion articles in this supplement review in greater detail some key aspects of the activities of APOC. The Structure of APOC Governance: APOC has built a strong and effective partnership that unites the member countries their beneficiary communities, NGDOs, multilateral agencies, bilateral donors and the private 5 sector, Merck & Co. Inc. and the scientific community. The organisational framework is designed to reflect this partnership. (WHO,l996b). All partners are represented on the Joint Action Forum (JAF), the main governing body of APOC which has been meeting, annually as scheduled, to review and approve the proposed plans of action and budgets, assess global financing requirements and take decisions with regard to overall progralnme policies. The members of the Committee of Sponsoring Agencies (CSA) are the same as for OCP viz. FAO, UNDP, WHO and the World Bank. The CSA organizes medium and long term planning of APOC and OCP activities, approves National Plans and Project Proposals and takes interim decisions on behalf of JAF. Programme Management: The headquarters of APOC is in Ouagadougou where a small core staff looks after the day-to-day affairs of APOC. This core staff, in conjunction with the NGDO liaison office in Geneva, has, in spite of a heavy burden of work, maintained a high standard of operations. APOC Management has faithfully implemented programme policy, facilitated and coordinated the activities of the National Onchocerciasis Task Forces and their partner NGDOs and provided oversight on financial management. It has also assisted in the training of National and NGDO staff and facilitated the monitoring of projects. APOC is supported by a Technical Consultative Committee (TCC), made up of 10 members. The TCC has successfully reviewed and recommended for approval 63 applications for APOC funding of CDTI projects, including their national support systems in addition to four vector elimination projects. In the first phase of its operation APOC has facilitated the formation of a National Onchocerciasis Task Force (NOTF) in every participating country. Made up of officials of the Ministries of Health (MoH) and Programme Managers as well as all partner NGDOs operating in the country, NOTFs have made considerable improvement in their performance over the years. 6 ACHIEVEMENTS The major achievements of APOC include consolidation of the unique and effective partnership around which it was built, the effective and efficient delivery of ivermectin to an ever increasing population in need of treatment, the impact of its activities on the health and socio-economic status of the endemic communities and the strengthening of the existing health systems. Partnership The APOC partnership, which is the hub of its existence, has been considerably fortified and consolidated during the first phase of APOC. The partnership exists and operates at different levels, and at each level the different partners interact in different combinations. The establishment and successful sustenance of such a diverse web of interactions between the partners is deemed a major programme achievement. The spirit and letter of the Memorandum of Understanding (MOU) signed by all participating countries are now better kept than at the beginning of APOC. All partners are pulling their weight, not least the beneficiary endemic communities whose increasing active involvement in the programme is the bedrock of the partnership. Benton et al., (2001) provide a more robust review of this aspect of APOC's first phase. APOC STRATEGY 1. Community-directed treatment with ivermectin (CDTI) In the CDTI approach, endemic communities are empowered to take full responsibility for the drug delivery process, decide how and when and by whom treatment should be administered, and oversee its implementation and follow-up. The successful and effective implementation of CDTI is one of APOC's most impressive achievements. The transformation, with the full cooperation of the NGDOs that had hitherto used the CBIT mode of distribution for their pioneering IDPs before the advent of APOC, to the CDTI approach, has been accompanied by higher treatment coverage rates. Areas of civil unrest and conflict have also benefited, and the cost of operations 7 per person treated continues to fall with time. When these achievements are coupled with the enhancement of ownership of the distributive system in the communities, a potent indicator of sustainability, APOC is clearly on the way to fulfilling the most cardinal objectives of its mandate viz. the establishment of "a regional programme that emphasizes sustainability, community involvement and partnership with NGDOs with the ultimate aim of eliminating onchocerciasis as a disease of public health and socio-economic importance throughout Africa" (WHO, 1996b). CDTI is without doubt one of the most revolutionary and visibly successful health development programmes to be implemented at the community level in recent times. An in-depth review of the CDTI strategy, which continues to be fine-tuned (WHO , 2000e) and the outstanding challenges is given in a companion paper (Amazigo et al.,200la). 2. REMO/GIS Another of the creditable accomplishments of APOC has been its effective use of the REMO/GIS at both the country level and centrally for planning and management of onchocerciasis control. This has provided a cost-effective and rapid process of targeting CDTI. REMO/GIS (WHO, 1998a) has been completed in full in 12 countries and partially in five others. Only two countries, Angola and Burundi, remain to start mapping of onchocerciasis. A full account of the use of REMO as a sine qua non tool by APOC, from the time APOC took delivery of it from TDR in 1996, is given in a companion paper (Noma et a1.,2001). The figure shows the distribution, at the end of the year 2000, of APOC-approved CDTI and country headquarter support projects and the vector elimination programmes. Vpcron ERADTcATIoN In keeping with its clearly defined policy and mandate on vector control, APOC has instituted vector eradication activities in three sites: Itwara in Uganda, Tukuyu in Tanzania and the island of Bioko in Equatorial Guinea. It has also commissioned a feasibility study in Mpamba-Nkusi in Uganda. In all the project sites baseline data have been collected, training of local staff 8 Tconducted (with OCP assistance), relevant equipment obtained and good progress is being made on a broad front. APOC took over the responsibility of bringing a long-standing control project in Itwara focus to its final conclusion. No Srzalium neavei larvae have been found in the focus since September 1998, and ground larviciding treatment has been discontinued since October 2000. The experience built up in the Itwara main focus has been of benefit in the extension of activities in the Aswa and Mpamba-Nkusi foci where feasibility study indicates that ground larviciding is a practical and effective strategy for vector elimination. One of the most important and durable achievements and legacies of APOC with respect to vector elimination activities is the training of local health workers in the multiple tasks associated with vector control. This is a form of capacity building that would certainly yield fruits in future for other comparable disease control prograrnmes. Irrrpacr or MBcrrzAN DrsrRrBUTroN One of the responsibilities of APOC is to determine the long-term social, clinical and epidemiological impact of CDTI. In order to discharge this responsibility effectively and as objectively as possible it was considered necessary to measure the impact of ivermectin treatment on transmission, clinical disease (both skin and ocular), quality of life as well as the socio- economic life of the people. To this end a multi-disciplinary team, made up of biomedical and social scientists was constituted. It was charged with the responsibility of measuring and documenting the epidemiological, social and economic impact of APOC operations. A separate alrangement was made for the assessment of the long-term effect of CDTI on transmission. Baseline data consisting of social, demographic, entomological, dermatological and ophthalmologic parameters were collected and analysed between 1998 and 1999. Standard indicators for the long-term assessment of CDTI in each of the participating countries have been identified. The main indicators for the evaluation were the incidence of onchocercal ocular lesions, including the incidence of onchocercal blindness and visual impairment, as well as the 9 prevalence of onchocercal reactive skin lesions, the incidence of onchocercal depigmentation and the prevalence of troublesome itching due to onchocerciasis. Baseline entomological data are also being assembled. The results of the studies have established baseline pattems and prevalence of different manifestations of skin and eye diseases due to onchocerciasis as well as entomological indices for 12 sites. (WHO, 2000f, g). The benefits that have accrued from the activities undertaken under the aegis of APOC with respect to the mass distribution of Mectizan is discussed in this paper under three broad headings: direct health benefits, positive impact on the health system in the endemic countries and socio-economic benefits. DIRECT HEALTH BENEFITS Within a relatively short period of five years APOC has successfully expanded and accelerated treatment with Mectizan in participating countries thereby relieving skin manifestations and intolerable itching in severely infected individuals, and preventing an estimated incidence of 40,000 cases of blindness per year (WHO, 2001). Over 67 million Mectizan tablets, with an estimated value of US$201,000,000, were distributed to 20 million people in the year 2000. In addition the direct involvement in and assumption of responsibility for conducting CDTI activities by the communities has resulted in an enhanced health consciousness in the field with potential 'openings' for instituting similar community-directed approaches to the control of other health problems. Based on several studies over the years, a number of specific health benefits are expected to accrue as the CDTI Programme progresses. These include substantial (80%) reduction in new cases of optic nerve disease, a sizeable (45%) reduction in the incidence of further visual field deterioration in individuals with optic nerve atrophy, and lowering of the incidence of punctate keratitis and uveitis. (Dadzie et al., 1990; Abiose et al., 1993; Umeh, 1,997). Considerable benefits are also expected by way of reduction in the burden of disease due to onchocercal skin lesions, with up to 50o/o decline in the prevalence of severe itching and improvement in reactive skin lesion. (Burham, 1995; Brieger et al., I998a; Ogbuagu and Eneanya l0 1998; Brieger et al., 2001). There are also a number of incidental benefits, related to the broad spectrum anthelminthic and antiscabetic activity of the drug (Njoo et al., 1993; Ottesen and Campbell 1994; Nnoruka and Agu,200l). The impact of mass treatment with ivermectin on the transmission of onchocerciasis has been comprehensively reviewed. (Boatin et al., 1998). Iuptcr orv rnn Hottrru Svsreu APOC through its CDTI strategy reaches many neglected communities at "the end of the road" that have limited or no access to orthodox health services. The involvement of peripheral and district health services in supporting the implementation of the programme has resulted in the strengthening of the health services structure. Health services staff have become increasingly more intimately involved in CDTI planning and implementation. The direct participation of the national health authorities in the work of NOTFs has also helped to strengthen collaboration at the national level among national and international partners in support of health development beyond onchocerciasis control (WHO, 2001). Capacity Building and Strengthening the Health Servtce APOC has helped the MoHs to build administrative and technical capacity in health delivery in 14 countries. The technical assistance has, to a large extent, been provided through the use of specially trained national staff thus contributing to capacity building at all levels, from the community to the national. At community level, the number of communities receiving health education has tremendously increased from 18,616 as at year 1 (WHO, 1996b) to over 49,000 communities in 2000 (WHO, 1997; 1998b). A number of innovative materials using local languages, traditional media and other communication channels which are appropriate for rural African societies have been developed by APOC (WHO, 2001). Training as part of capacity building and strengthening of the Health System has been one of APOC's most visible and enduring achievements. In this the input of the NGDO partners has 1t been outstanding. Training is seen as one of the fundamental pillars for the sustainability of the onchocerciasis control programme in the post-APOC era. It is the seed of the tree from which many future control programmes would reap health nourishing and sustaining fruits. In Sudan and some other APOC countries CDTI is serving as an entry point for basic health services. Table I shows a summary of the number of people trained by APOC in specific areas of capacity building. This aspect of APOC's work and the challenges ahead are discussed by Homeida et al., (2001). In addition to these, transportation facilities made available through APOC have supported other programmes, and reduced the notorious isolation of peripheral health workers from the grassroots. Health care relationships with communities have thus been enhanced. Socut .txo EcoNo*ttc BoNnrtrs Most of the social and economic benefits that are expected from the operations of APOC will begin to become manifest during the second phase. These are comparable to those that have been so meticulously documented in respect of OCP (Kim and Benton 1995; Kim et a1.,1997; Benton, le98). OrnnnBoNorns The sustainability of CDTI depends on the extent to which communities assume responsibility for the organization of the activities and take necessary decisions. APOC's success in involving thousands of communities in an aspect of their own health care (CDTI) has been most encouraging. As at the end of year 2000, over 49,000 communities in 14 countries were actively involved in planning and managing the distribution of ivermectin through CDTI (WHO, 1997; 1998b; 1999). Furthermore, from 1999 APOC transformed from an exclusively independent to a participatory mode of monitoring and evaluation of the CDTI projects. A method for community self-monitoring is currently being developed. It is expected to reinforce a sense of ownership in t2 the endemic communities. Three companion papers in this supplement discuss various aspects of this topic in greater detail. (Amazigo et al.,200lb; Clemmons et a1.,200L; Homeida et a\.,2001). CDTI INTEGRATION AND AS ENTRY POINT TO PHC Integration of CDTI with PHC is a declared policy of APOC. It is seen as a prerequisite for the sustenance of CDTI in the post-APOC era. It is, however, recognized that the status and viability of the PHC system vary significantly from country to country and from district to district within the same country. The pace and effectiveness of integration would therefore differ from place to place accordingly (WHO, 2001). There is good indication of commitment at the country level to the inclusion of CDTI into National Health Plans. The Ministries of Health (MoHs) have allocated staff and resources, such as offices and logistics, to programme activities. The NOTFs have substantiated the concept of a National Planning and Executing agency within the Ministry, which also includes outside partners such as NGDOs and technical Experts (WHO, 2000c). One of the cardinal features of CDTI is its potential value as an entry point for a variety of other community-based health programmes (CBHPs). During the 1998/99 period, Cameroon, Central African Republic (CAR) and Sudan were among the countries that used CDTI strategy as entry point for other services (WHO, 1999). [n Cameroon, the CDTI project of the South West Province initiated special training for literate CDDs to assist the Health Care services with the identification of cases of cataract in target communities. In CAR, the method has led to the introduction of a more comprehensive primary care service in the Bocaranga district in the North- western region. In Sudan, other forms of health services were made available to communities embroiled in civil unrest through CDTI activities such as the screening organized by NOTF of 1409 persons for non-onchocerciasis blindness. Those screened were provided with medical care and one hundred and thirty one of them were surgically treated. Homeida et al., (2001) discuss this topic extensively in a companion paper in this supplement. l3 GEOGRAPHICAL AND THERAPEUTIC COVERAGE. Evidence has accumulated to show that the amount of energy, commitment and resources generated by APOC are resulting in considerable successes in respect of access to treatment in target communities (WHO, 2000c). The two basic process indicators are geographical and therapeutic coverage. With regards to the geographical coverage, there are 53 on-going CDTI projects in 14 countries. In 1999, 19,288 (71.9%) of the 23,071target communities were treated in eight countries (Cameroon, CAR, Chad, Malawi, Nigeria, Sudan, Taruania, and Uganda). (WHO, 2000a). All target communities received treatment in Tanzania and despite the prevailing civil war and the absence of dependable administrative structure for health care, corlmendable coverage rates (50% of the target population) were recorded during 1998199 period in Sudan where APOC partners provided treatment to 198 communities in Southern Sudan (WHO, 2000b). In 2000, 34,440 communities were actively involved in planning and managing the distribution of ivermectin. This represents a ten-fold increase in geographical coverage between 1995 and 2000. (WHO, (2000a). As regards therapeutic coverage, the target is to treat at least 65%o of the total population annually for several years. The number of people now receiving treatment every year,20 million, is more than double the 8 million figure recorded when APOC started its activities. Coverage in individual countries has increased considerably since 1998 and most projects now record treatment coverage in excess of 650/o. Experience during Phase I has shown that expansion of ivermectin distribution can be achieved at a reduced cost per head and with a satisfactory level of coverage. According to APOC's estimates, 59 million persons will need to be treated every year by 2007 . OPERATTONAL RESEARCH A fruitful partnership between APOC, OCP and TDR has, through operational research, enabled APOC to introduce improvements in its operations. The REMO/GIS technique, developed by the t4 TDR, has been a crucial tool in developing and implementing the APOC strategy. It continues to be refined. (Noma et al., 2001). Continued research has identified ways of sustaining CDTI (WHO, 2000e; Amazigo et al.,200lb) and enhancing the role of women in it (Clemmons et al., 2001). Multi-centric studies in endemic countries are on going on the development of appropriate and effective advocacy packages that would increase the receptivity and hence the commitment of policy makers to the Programme. So also are studies on rapid epidemiological methods of assessing the distribution and prevalence of Loa /oa infection. CTTALLENGES OF APOC. Strengthening of the partnership. Maintaining and shengthening the partnership that forms the bulwark of APOC operations is one of the most important challenges facing APOC during Phase 2. The efficiency and effectiveness of its operations would continue to form the critical parameters by which APOC would be judged and the best way to ensure the continued solidarity of the alliance of partners. In return APOC could look forward to all partners pulling their weight and responding promptly to the needs of the Programme as soon as those needs arise. Benton et al., (2001) have examined, in depth, the issues related to the APOC partnership. Sustainability of ivermectin treatment. The greatest challenge facing APOC is the sustainability of CDTI after the cessation of the Programme. The reports of independent monitors indicate, among other things that an aspect of the Programme that is somewhat weak is the nature and level of commitment to CDTI of authorities at different levels of operation in the some of the countries. (Amazig o et al., 2001a; b) This deficiency is most glaring with respect to budgetary provision and release of funds for projects which impact directly on treatment coverage rates. Other shortcomings include inadequacy of frontline health facilities (FLHFs), health centers, dispensaries, and health posts, and a decline in the quality of services, in terms of the maintenance, availability of essential 15 equipment, etc. When all these are taken together the need to strengthen the peripheral health system and integrate CDTI into the health service, as means of sustaining CDTI, constitute a major challenge to APOC (WHO, 2000d). Female Participation. The involvement of women in CDTI programmes varies considerably from country to country and from region to region. In general, the level of involvement of women mirrors the status of women and the cultural context in particular communities. Reports have indicated that increased participation of women in all aspects of CDTI is associated with better performance, however, scientific documentation is important for policy decisions. How this gender issue relates to CDTI and constitutes a challenge to APOC has been elaborated upon in extenso by Clemmons et al., (2001). Cost Recovery The issue of the role of cost recovery in a disease control programme like that of APOC remains controversial. Cost recovery has been put forward as a means of assuring sustainability of CDTI. Some APOC countries, notably Cameroon and Chad have, in keeping with Bamako initiative, adopted cost recovery as a national policy for PHC and apply it in their CDTI projects. However the reports on the effect of cost-recovery on such issues as treatment coverage are not at all clear. Neither indeed are the consequences of its use as source and means of defraying the cost of CDTI operations and providing incentives to CDDs. (Amazigo et al., 2001a; b,' Clemmons et al., 2001). Clearly there is a lot yet to comprehend about this thorny issue and resolving it represents a pressing challenge to APOC. MaNaCBTvTENT STRUCTURE AND DECENTRALIZATION The organizational structure of APOC has proved to be appropriate for the needs of the Programme management during Phase 1. However the Report of the External mid-term evaluation while commending the performance of APOC Management during Phase 1, identified l6 several areas that need readjustment in the way the Programme is managed for it to be able to cope successfully with and meet the challenges of Phase 2. Notable among these are staffing, planning procedures, the role of TCC, and effecting transfer of some of the functions to the national programmes (WHO, 2000c). The process of decentralisation is a challenge that APOC must continue to implement in Phase 2. CDD motivation. There is hardly any issue surrounding the operation of CDTI that is as controversial as that of the motivation of CDDs. Two companion papers in this supplement examine the information available on the subject from a number of projects in the same as well as different countries (Amazigo et al,200la; b). Clearly there is a lot more that is unknown than known about the intricacies surrounding that motivation of "volunteers" in control programmes such as APOC. That local factors are of importance cannot be gainsaid. It is doubtful however if a universal formula to the subject can be readily found. The challenge to APOC therefore is to prepare guidelines that are sufficiently flexible yet well focused for use under different circumstances within the context of CDTI. Funding of country-specific operational research on the topic may be a good starting point. IDENTIFICATION OF LOCAL NGDOS FOR CDTI PROJECTS There is a consensus among partners that the identification, promotion and empowerment of local National NGDOs as partners, will help to sustain the onchocerciasis control Programme, particular in the post-APOC era. Currently all the NGDOs operating in partnership with the APOC are, with but two exceptions, International NGDOs. The exceptions are one national NGDO in Nigeria and Liberia. These have been groomed by international NGDOs and have taken on CDTI responsibility for a few districts within project areas. The challenge in the second phase of APOC is to identify and nuture more and more competent local NGDOs who can sustain CDTI over the long run. Each international NGDO should be encouraged to adopt at least one l7 NGDO for that purpose and APOC should be prepared to provide any reasonable support needed for effecting this. CHALLENGES OF REMO AND OPERATIONAL RESEARCH The need to complete REMO in all parts of the endemic countries, and to carry out refinement as necessary is a priority for APOC. As soon as the promising preliminary findings of the multi- country study on a rapid assessment and mapping procedure for Loa loa are confirmed, its application as a control programme tool would be become an important and urgent assignment for APOC. The overlay of loiasis and onchocerciasis maps would provide the Programme with a powerful means of avoiding the Prograrnme compromising problems associated with severe adverse events (SAE) in a places where the two disease co-exist. A comprehensive discussion on REMO and the related challenges appear in this supplement (Noma et a1.,200I). Meanwhile the search for a suitable macrofilaricide that would have a tremendous impact on the Programme strategy continues in earnest. So also are studies on the nature and implications of the failure of a few people, in parts of OCP, who have been treated with ivermectin but have failed to respond appropriately to treatment. The issue of these non-responders has, in some quarters, been used to raise the spectre of parasite resistance to the drug. CIIALLENGES ON INFORMATION, EDUCATION AND COMMUNICATION (IEC) IEC and Advocacy are important components for the sustainability of CDTI and all programme activities. Technical assistance should be provided by APOC to projects to develop better IEC and advocacy strategies, messages and materials through a participatory approach. Special attention should be given to develop innovative materials that utilize traditional media and other communication channels that are appropriate to rural African societies (e.g. songs, theatre display, testimonials and interviews; broadcasts on radio, child to child activities, community entertainment, education festivals, etc). Communication channels should utilize local languages and special effort to target women and youth needs to be emphasized. l8 CONCLUSION There is ample evidence that APOC has recorded a series of significant achievements in the first phase of its operations. As it moves into the second and final phase some of the challenges which the Programme would have to contend with have been identified. A number of ways of dealing with some of these challenges have been suggested. It is recognized however that other challenges will arise in Phase 2, just as in Phase 1, which have not been foreseen. The ultimate achievement of the objectives and goal of APOC will depend as much on the maintenance of the sound managerial principles and performance that characterized Phase I as on the alliance of forces that bind the Programme together remaining strong, responsive and committed during Phase 2. AcxNowT,EDGEMENTS The achievements of APOC owe a lot to the participating countries, non-governmental development Organizations (NGDOs), Technical Consultative Committee (TCC), Committee of Sponsoring Agencies (CSA), donor community and WHO/TDR who have provided solid support in this unique partnership in the fight against the scourge of onchocerciasis. In particular, we are indebted to Dr Ole Christensen for his unstinting support to APOC. REFERENCES ABIOSE, A., JONES, B. R., COUSENS, S. N., MURDOCH, I., CASSELS-BROWN, A., BABALOLA, O. E., ALExANDER, N. D. E., NUHU, I., EVANS, J., IBRAHIM, u. F. & MAHMooD, e. o. (1993). Reduction in incidence of optic nerve disease with annual ivermectin to control onchocerci asis. Lancet, i, 130-134. AMAzIGo, u. o. (1993). Onchocerciasis and women's reproductive health: indigenous and biomedical concepts. Tropical Doctor, 23, 149 -15 1. l9 AMAZIGO, u. o (1994). Detrimental effects of onchocerciasis on marriage age and breasrfeeding. Tropical and Geographical Medicine, 46, 322-325 AMAZIGO, u. o. & OBIKEZE, D. s. (1991). Socto-culturalfactors associated with prevalence and intensity of onchocerciasis and onchodermatitis among adolescent girls in rural Nigeria. SER/TDR Project Report, Geneva: WHO AMAZIGO, U. V., BRIEGER, W. R., KATABARWA, M., AKOGUN, O., NTEP, M., BOATIN, B., N'DOYO, J., NOMA, M. & SEKETELI, A. (200La). The challenges of community-directed treatment with ivermectin (CDTI) of the African Programme for Onchocerciasis control (APOC). Annals of Tropical Medicine and Parasitologt, ?95, (Suppl .), ??? - ???. AMAZIGO, U. V., OBONO, O.M., DADZIE, K.y., REMME, J., JfyA, J., NDYOMUGYENYI, R., ROUNGO U, J-8., NOMA, M. & SEKETELI, A. (2001b). Monitoring community-directed treatment progralnmes for sustainability: lessons learned. Annals of Tropical Medicine and Parasitology, ?95, (Suppl .), ??? -??? AWADZI, K., DADZIE, K. y., SCHULZ-KEY, H., HADDOCK, D. R. W., GILLES, H. M. & AZV,M. A. (1985). The chemotherapy of onchocerciasis X. An assessment of four single dose regimes of Mk-933 (ivermectin) in human onchocerciasis. Annals of Annals of Tropical Medicine and Parasitologt, 79, 63-7 8. BENTON, B. (1998). Economic impact of onchocerciasis control through the African Programme for Onchocerciasis Control: an overview. Annals of Tropical Medicine and Parasitologt,g2, (Suppl.), 533-539. BeNtoN, B., BUMI, J., LIESE, B. & SEKETELI, A. (2001). PenrNeRsHrp AND eRoMISE: EVoLUTToN OF THE AFRICAN RIVERBLINDESS CAMPATGNS. Annals of Tropical Medicine and Parasitologt, ?95, (Suppl .), ???-??? 20 BOATIN, B. A., HOUGARD, J. M., ALLEY, E. S., AKPOBOUA, L. K., YAMEOGO, L., DEMBELE, N., SEKETELI, A. & DADZIE, K. Y. The impact of Mectizan on the transmission of onchocerciasis. Annals of Tropical Medtcine and Parasitologt,92, (Suppl. l), 546-60. BRIEGER, W.R., AWEDOBA, A. K., ENEANYA, C. I., HAGAN, M., OKELLO, D., OGBUAGU, K. .F., oSANGA, o. o. & ovucA, E. B. L. (1998a). The effect of ivermectin on onchocercal skin disease and severe, itching: results of a multi-centre trial. Tropical Medicine and International Health, 3, 95 1 - 96L. BRIEGER, w.R., oSHINAME, F. o. & oSoSANYA, o. o. (1998b). Stigma associated with onchocercal skin disease among those affected near the Ofiki and Oyan Rivers in western Nigeria. Social Science and Medtcine, 47, 84 I -853. BRIEGER, w. R., KALE, o. o. & oSoSANYA, o. o. (2001). Development of reactive onchocercal skin lesions during a placebo-controlled trial with ivermectin among persons without lesions at baseline. Tropical Doctor,3l, 96-98. BURHAM, c. (1995) Ivermectin treatment of onchocercal skin lesions: observations from a placebo controlled double blind trial in Malawi. American Journal of Tropical Medicine and Hygiene,52,270-276 CLEMMONS,L., AMAZIGO, U. V., BISSECK, A-C., MYSUYA-MPANJU, J., NOMA, M., OYENE, U., EKPO, u., KATENGA, S. & SEKETELI, e. (2001). Gender issues in community-directed treatment with ivermectin CDTD of the African Programme for Onbchocerciasis control (APOC). Annals of Tropical Medicine and Parasitology, ?95, (Suppl . ), ???-??? DADZTE, K. y., BIRD, A. c., AWADZLK., SCHULZ-KEY, H., GILLES, H. M. & AZlz, u. n. (1987). Ocular findings in a double-blind study of ivermectin versus diethylcarbamazine versus placebo in the treatment of onchocerciasis. Journal of Ophthalmologlt,Tl,TS-85 DADZTE, K. Y., REMME, J., ALLEY, E. s. & DE soLE, c. (1990). Changes in ocular onchocerciasis four and twelve months after community-based treatment with ivermectin in a holoendemic 21 onchocerciasis focus. Transactions of the Royal Society of Tropical Medicine and Hygiene, 84, 103-108. DADZIE, K., REMME, J & DE SOLE, c. (1991). Changes in ocular onchocerciasis after two rounds of community-based ivermectin treatment in a holoendemic onchocerciasis focus. Transacttons of the Royal Society of Tropical Medicine and Hygiene,85,267-271 DUKE, B. o. L. &otoztr., K. Y. (1993). Procedural manualfor lvermectin Distribution Programmes. wHo/PBLlg3. 3 5, 33 -37 Geneva: WHO. GUILLET, P., SEKETELI, A., ALLEY, E.S., AGOUA, A., BOATIN, B.A., BISSAN, Y., AKPOBOUA, L. K. B., QUILLEVERE, D. & SAMBA, E. M. (1995). Impact of combined large-scale ivermectin distribution and vector control on transmission of Onchocerca volwlus in The Niger Basin, Guinea. Bulletin of the World Health Organisation,73,199-205. HOMEIDA, M., BRAIDE, E., ELHASSAN, E., AMAZIGO, U. V., LIESE,8., BENTON, 8., NOMA, M., ETvA'ALE,D.,DADZTE, K. y., KALE, o.o.& sBrBrBlq a. (2001). Community-directed treatment with ivermectin (CDTI) strategy of the African Programme for Onchocerciasis Control (APOC) and its potential for providing additional health services to the poorest populations. Annals of Tropical Medicine and Parasitology, ?95, (Suppl.),???-??? KALE, o. o. (1998). Onchocerciasis: The Burden of disease. Annals of Tropical Medicine and Parasitology, 92, (Suppl.l), S101-Sl 15. KIM, A. & sENroN, n. (1995). Cost-benefit analysis of the Onchocerciasis Control Program (OCP). World Bank Technical Paper, 282, Series on River Blindness Control in West Africa. KIM, A., TANDON, A., HAILU, A., BIRRIE, H., BERHE, N., AGA, A., MENGISTU, G., ALI, A., BALCHA, F., cEBRE-MrcHAEL, T., BTzNNEH, A.& GEMETCHU,t. (1997). Health and Labour Productivity: the Economic Impact of Onchocercal Skin Disease (OSD). Policy Research Working Paper No. 1836. Washington, DC: World Bank. 22 NGouMou, P. & wALSH, J. F. (1993). A manualfor Rapid Epidemiological Mapping of Oncho cerciostu. TDR/TDE/ONCH O 193.4. Geneva: WHO. NGouMou, P., wALSH, J. F. & MACE, J. M. (1994). A rapid mapping technique for the prevalence and distribution of onchocerciasis: a Cameroon case study. Annals of Tropical Medicine and Parasitologt, 88, 463-47 4. NJOO, F. L., BELLING, G. A. C., OOSTING, J., VEETER, J. C. M., STILMA, J. S. & KIJLSTRA, e. (1993). Concurrent parasitic infections in onchocerciasis and occurrence of adverse reactions after ivermectin treatment. American Journal of Tropical Medicine and Hygiene, 48,652-657. NNoRUKA, E. N. & AGU, C. E. (2001). Successful treatment of scabies with oral ivermectin in Nigeria. Tropical Doctor, 31, 15-18. NOMA, M, NWOKE, B. E. B., NUTALL, I., TAMBALA, p., ENYONG, p., NAMSENMO, A., REMME, J., AMAZIGo, u. v., KALE, o. o. & sprcrplI, e. (2001). Rapid Epidemiological Mapping of Onchocerciasis (REMO): Its application by the African Programme for Onchocerciasis Control (APOC). Annals of Tropical Medicine and Parasitology, ?95, (Suppl.),???-??? OGBUAGU, K. F. & ENEANYA, c. I. (1998). The effect of ivermectin on onchocercal skin disease: clinical findings. Annals of Tropical Medicine and Parasitology, g2, (Suppl. 1), S139-145. oK-ELLo, D. o., ovucA, E. B. & ocwer- oK_ENG, J. w. (1995). Dermatological problems of onchocerciasis in Nebbi District, Uganda. East African Medical Journal, 72,295-298 oLADEPO, O., BzuEGER, W. R., OTUSANYA, S., KALE, o. o., OFFIONG, S. & TITILOYE, M. (1997). Farm land size and onchocerciasis, status of peasant farmers in South-western Nigeria. Tropical Medicine and International Health. 2, 334-340. ONCHOCERCIASIS CONTROL PROGRAMME. (1973). Onchocerciasis control tn the Volta River Basin Area. Report of the Preparatory Assistance Mission (PAG) to the governments of Dahomey, Ghana,Ivory Coast, Mali, Niger, Togo And Upper Volta. Geneva, OCP, 73,1: 23 orrESEN, E. & CAMPBELL, w. C. (1994). Ivermectin in human medicine. Journal of Antimicrobial Chemotherapy, 34, 19 5 -203. ovucA, E. 8., OKELO, D. O., OGWAL OKENG, J. W., ORWORTHO, N. & OPOKA, F. O. (1995). Social and psychological aspects of onchocercal skin disease in Nebbi district of Uganda. Easl African Medical Journal, 72,, 449 -453. REMME, J., BAKER, R. H. A., DE SOLE, G., DADZIE, K. y., ADAMS, M. A., ALLEY, E. S., AVISSEY, S. K. & wnLSH, J.F. (1989) A community trial of ivermectin in the onchocerciasis focus of Asubende, Ghana. 1. Effects on the microfilarial reservoir and the transmission of Onchocerca volvulus. Tropical Medictne and Parasitolog/, 40, 367 -37 4. REMME, J., DE SOLE, G., DADZIE, K. Y. ALLEY, E. S., BAKER, R. H., HABBEMA, J. D., PLAISIER, A. p. vAN OORTMARSSEN, c. J. & SAMBA, E. M. (1990). Large scale ivermectin distribution and its epidemiological consequences. Acta Leidensia, 59, 177 -191. TSALIKIS, G. (1993) The Onchocerciasis Control Programme (OCP) in West Africa: A review of progress. Health Policy and Planning, 8, 349-359. UMEH, R. E. (1997). Use of computerised visual function tests in a community-based study of onchocerciasis in Nigeria. Nigerian Journal of Ophthalmology, 5,41-45. VLASSOFF, c., WEISS, M., OVUGA, E. 8., ENEANYA, C., NWEL, p. T. BABALOLA, S. S., AWEDOBA, A. K., THEoPHILUS, B., coFIE, P. & SHATABI, P. (2000). Gender and the stigma of onchocercal skin disease in Africa. Social Science and Medicine, 50, L353-1354 WHITWORTH, J. A. G., GILBERT, C. E., MABEY, D. M., MAUDE, G. H., MORGAN, D. & taYLOR, O. W. (1991). Effects of repeated doses of ivermectin on ocular onchocerciasis: community-based trial in Sierra Leone. Lancet,338, I 100-1103. woRLD HEALTH ORGANISATION (1995a). Onchocerciasis and its control: Report of a WHO Expert Committee on Onchocerciasis Control. wgo Technical Report Series, 852. Geneva: WHO. 24 woRLD HEALTH oRGANISATIoN (1995b). The Pan-African Study Group on Onchocercal Skin Disease. Report of a multi-country study. ron/ern/npl95.l. Geneva: WHO. woRLD HEALTH ORGANISATIoN (1995c). Supplemental guidelines for Rapid Epidemiological Mapping of Onchocerciasis (REMO). TDR/TDF/oNCHo/95.1. Geneva: WHO. WORLD HEALTH ORGANISATIoN (19964). Community-Directed Treatment with lvermectin. Report of a multi-country study. TDR/AFR/RI/96.1. Geneva: WHO. woRLD HEALTH ORGANISATION (1996b) African Programme for Onchocerciasis Control (APOC) Programme Document for Phase I. Ouagadougou: APOC. woRLD HEALTH ORGANISATION (1997). Progress Report of the World Health Organisation. Jaf-Fac. 3.2. Ouagadougou: APOC. woRLD HEALTH ORGANISATIoN (19984). Guidelines for analysis of REMO data using GIS. TDR/TDF/coMDT 98.3. Geneva: WHO. WORLD HEALTH ORGANISATTON (1998b). Progress Report of the llorld Health Organisation. Jaf-Fac. 4.2. Ouagadougou: APOC. woRLD HEALTH ORGANISATTON (1999). Progress Report of the World Health Organisation .Jaf-Fac. 5.2. Ouagadougou: APOC woRLD HEALTH ORGANISATIoN (2000a). Progress Report of the World Health Organisation. . Jaf-Fac. 6.2. Ouagadougou APOC. woRLD HEALTH oRGANISATToN (2000b). Report on Partners' Meeting. Programme Implementation 2 . Ouagadougou: APOC. WORLD HEALTH ORGANISATTON (2000c). External mid-term evaluation; African Programmefor Onchocerciasis Control. wuo/epoc/cs/00.3. Ouagadougou: APOC. woRLD HEALTH ORGANISATIoN (2000d). Final Report on the sustainability of community- directed treatment of onchocerciasis with ivermectin.waolxpoc/cs/OO.1. Ouagadougou: APOC. 25 woRLD HEALTH ORGANISATToN (2000e). Implementation and sustainability of Community- Directed Treatment of onchocerciasis with ivermectin: report of a multi-country study. rDR/rDE/Rp/cor/00. 1. Geneva: WHO. woRLD HEALTH oRGANISATIoN (2000r. Partners' meeting: monitoring CDTI projects. (U npub I is hed D o cu m en t) Ouagadougou : APOC woRLD HEALTH oRGANISATIoN (20009). Partners' meeting report: Long-term impact assessment of APOC operations. (Unpublished Document). Ouagadougou: APOC. woRLD HEALTH oRGANISATIoN (2001). Programme Document for Phase 2. Technical Consultative Committee (TCC) Review. (Unpublished Document). Otagadougou: APOC. 26 Table I Number of people trained through APOC in specific areas of capacity building Number of oeoole trained bv APOC Area of Capacity Building Financial management of Trust fund Vector elimination activities Independent monitoring Health Care personnel t999 43 33 68 7,570 2000 55 27 60 13,456 Total 98 60 128 2,1026 Total 89 6 40,203 77,554 117,757 21 Fig: REMO-GIS MAP FOR APOC PROJECT SITE Chad CDTI: 1 CAR CDTI: 2 HQ:1 Nigeria CDTI:26 HQ:1 Sudan CDTI: 2 HQ: I Liberia CDTI: 1 Ethiopia CDTI: 1 HQ: 1 Cameroon CDTI: 8 HQ: 1 Uganda CDTI:4 Vector: 2 Eq. Guinea CDTI: 1 Vector: I Tanzania CDTI:4 Vector: 1 HQ: 1 Gabon CDTI: 1 Malawi CDTI: 1Congo CDTI: 1 D.R. Congo CDTI: 1 HQ: 1 28 )\ -rl -'.<( RAPID EPIDEMIOLOGICAL MAPPING OF oNCHOCERCTASTS (nrUO), rTS APqLTCATTON By THE AFRICAN PROGRAMME FOR ONCHOCERCIASIS coNrRor (npoc) IRAPID EPIDEMIOLOGICAL MAPPING OF ONCHOCE,RCIASIS (REMO): ITS APPLICATION BY THE AFRICAN PROGRAMME FOR ONCHOCERCTASTS CONTROL (APOC) AUTHORS M. Noue,'B. E B. Nworp,2 I. NuteLL,3 P. A. TaNaseLA,o P. ENvoNc,sA .NerrrsnNuo,6 J.Rnuur,3 U. V. Arrrazrco,rO. O. Kar,B'nNoA. Sersrplrl ADDRESSES OF AUTHORS African Programme for Onchocerciasis Control (APOC) Department of Parasitology & Entomology, School of Biological Sciences, Imo State University, Owerri, Nigeria World Health Or ganrzation, Geneva, Sw itzerland National Onchocerciasis Task Force (NOTF) Secretariat, Blantyre, Malawi Tropical Medicine Research Station, Kumba, Cameroon Ministdre de la Sant6 et de la Population, Bangui, R6publique Centrafricaine Department of Preventive and Social Medicine, University College Hospital, Ibadan, Nigeria * Addres s for Co rrespo n dence Dr. Mounkaila Noma African Programme for Onchocerciasis Control (APOC), 0l B. P.549 Ouagadougou 01 Burkina Faso Tel: (226) 34.22.77 Fax: (226) 34.48.00 E-mail: nomam@oncho.oms.bf 1. 2. 3. 4. 5. 6. 7. Short running title: Applications of REMO by APOC 2ABSTRACT One of the fundamental challenges that the African Programme for Onchocerciasis Control (APOC) has had to face is how to identify the endemic communities where its mass ivermectin treatment operations are to be carried out in conformity with its stated objective of targeting the most highly endemic, affected and at risk populations. This it has done by adopting the technique of Rapid Epidemiological Mapping of Onchocerciasis (REMO) that provides data on the distribution and prevalence of onchocerciasis. Integration of REMO data into the Geographical Information System (GIS) enables delineation of zones of different levels of endemicity which is an important step in the planning process for onchocerciasis control. Zones are included in or excluded from the APOC-funded programme of Community-Directed Treatment with Ivermectin (CDTI) depending on whether or not their levels of onchocercal endemicity reach the threshold set by APOC. This paper describes the application of the REMO/GIS technique by APOC in its operations and identifies the challenges that are outstanding. 3INTRODUCTION Following the successful introduction of mass treatment with ivermectin (Mectizan@) by the Onchocerciasis Control Programme in West Africa (OCP) as an adjunct to vector control in the programme area, (Alley et al., 1994), and based on the pledge by Merck & Co. to donate Mectizan@, (ivermectin, MSD) free of charge, for as long as needed, some countries in collaboration with some United Nations agencies and non-governmental development organizations (NGDOs) outside the OCP area initiated the ivermectin distribution prograrnme (IDP). The methods used, under the IDP initiative, for the selection of communities for ivermectin treatment in the pre-APOC days have been described (Taylor et al., 1992, WHO, 1992). In order to strengthen, consolidate and promote their efforts towards achievement of their common goal of disease control, the NGDOs participating in the IDP joined forces to establish a coordinating structure in 1992. At the country level, NGDO coalitions were created, first in Nigeria followed by Uganda. The methods used for the distribution of the drug were mainly those of mobile teams and programme-owned community based drug delivery systems. There was in most cases little or no community input into planning for or involvement or active participation in the distribution process. By 1995, it had become clear that the coordinating groups had reached the upper limit of what could be achieved by the unsustainable IDP methods being used and the resources they were able to generate. The inauguration of the African Programme for Onchocerciasis Control (APOC) in 1995 opened a new chapter and opportunities in the control of this disease in Africa. The objective of APOC is to establish, within a period of 12 to I 5 years, effective and self-sustainable community-directed ivermectin treatment throughout the endemic areas in the geographic scope of the Programme. It also aims to eradicate the vector, wherever possible, by using environmentally safe methods, in selected and isolated foci (WHO, 1996). 4From the inception, the philosophy of the APOC has been to use the limited funds available to it to provide ivermectin treatment to the people suffering the most from the disease and at the highest risk of developing severe complications (WHO, 1995a; 1996). It became imperative therefore to prioritize allocation of resources according to needs (Benton, 1998). To achieve this there was a need to devise better, relatively more rapid and reliable, yet cost-effective methods of obtaining relevant information than hitherto. Better information was required on the geographical distribution of the disease, as well as the identification of high- risk communities where the burden and public health importance of the disease are highest. PROCESS OF DETERMINING THE DISTRIBUTION OF ONCHOCERCIASIS AND IDENTIFYING CDTI PRIORITY AREAS The disease and its vector Human onchocerciasis or river blindness is a disease resulting from infection by a parasitic filarial worrn, Onchocerca volvulus. The adult worms live mainly in subcutaneous tissue usually aggregated and embedded in fibrous tissue known as onchocercal nodules. The female worrns produce millions of microscopic embryos called microfilariae (mf). The parasite (mf) is transmitted from one person to another by female black flies belonging to the genus Simulium during blood meals. The vectors breed in fast-flowing, relatively unpolluted and well oxygenated rivers and streams. In addition the vector has an effective flight range, when seeking blood meal, which is unlikely to exceed 15 km. (Le Berre, 1966; Phillipon, 1978; Quill6v6r6, 1979). These vector ecology and behaviour characteristics determine the distribution and intensity of transmission of onchocerciasis (WHO, 1,976; De Sole e/ al.,1,99la,b). Thus the most severely affected communities are almost invariably located within 10 km of a vector-breeding site. The design and application of the method of rapid epidemiological mapping of onchocerciasis (REMO) is based on and makes use of these facts. REMO surveys methods To collect data on onchocerciasis that is relevant for its operations, APOC relies on the REMO. REMO was developed under the auspices of the UNDP/World Bank/WHO Special 5Programme for Research and Training in Tropical Disease (TDR) and field tested in Cameroon. (Ngoumou et a\.,1994). It is a simple, non-invasive, and practicable process that is easy to apply over a wide-range of bio-ecological zones, with no socio-cultural or religious restrictions. REMO consists of three stages: 1. the division of the country into bioclimatic/biogeographical zones with reasonable degree of uniformity with regard to the potential for onchocerciasis: 2. the selection of communities to be surveyed per zone in order to determine whether onchocerciasis is present or not and, if present, to give an indication of distribution and severity. (Selection is optimally biased towards communities at highest risk); 3. and a measure of the level of endemicity of onchocerciasis in the selected communities using a rapid epidemiological assessment (REA) method, based on nodule prevalence in a sample of adults. The basic processes of REMO have been amply described (Ngoumou and Walsh, 1993, Nwoke, 1993, WHO, 1995b). Planning and Execution of REMO surveysfor APOC The National Onchocerciasis Task Force (NOTF) of each APOC country is responsible for submitting to APOC Management for consideration, a proposal for REMO in the country. The NOTF prepares a budget for the REMO exercise and assembles a national mapping team, which is usually composed of experienced geographers, entomologists, epidemiologists, nurses and physicians. APOC Management reviews the project proposal, including the budget submitted to it by the NOTF for the REMO surveys. It thereafter appoints a team of external experts whose mission is to train the national team members on REMO methods. The panel of experts assists the national team to plan the field surveys, within the revised budget, and participates in field activities. 6ClassiJication of a country into REMO zorres The division of a country into REMO zones is performedin2 to 4 hours by the mapping group consisting of nationals and the external experts. The division of a country into different bio- geographical zones is based on the density, spatial distribution and host seeking behavior and the physiology of the Simuliid vector. Where the initial biogeographical division is too large to provide a desirable degree of uniformity, it may be necessary to subdivide. Selection of the communities to be surveyed The objectives at this stage of the mapping exercise are to select a limited number of villages to be surveyed which will show whether onchocerciasis is present or not and if present, give a rough indication of distribution and severity. Two types of communities are selected, high risk and secondary communities. The high-risk communities are those situated in the areas most likely to have the worst disease profile. These are communities in the immediate vicinity of major potential vector breeding sites, as revealed on available topographical maps. Three to four experts make the selection of high risk communities based on the following criteria: o the selected community should be located close to river banks, preference being given to communities close to rapids; o the selected community should be a "first line" community (i.e. villages without other human settlements between them and the river/rapids), with preference being given to isolated communities; o atleast one high risk community should be selected in each distinct stretch of river; . on the main river, at least one high risk community should be selected every 30-50 km, o at least one high risk community to be selected in the valley of every major tributary. It is recommended to choose an alternative village for each of the high-risk villages selected. The altematives should have similar characteristics to the original choices and be likely to suffer 7the same challenge of onchocerciasis transmission. An altemative village is surveyed if the original choice cannot be located or reached or has an uncooperative population. For each high risk village or its alternative, the mapping group should choose a related secondary village, which should be located at least 10 km farther way from the likely main source of vectors. The aim of selecting secondary villages is to obtain some indication of the distribution and overall severity of the disease over a wider geographical area. Survey of a secondary village will only be undertaken if at least some of the high-risk communities have proved to be meso- or hyper-endemic for onchocerciasis. REA methods in REMO REA surveys should be undertaken in the dry months of the year when farming activities are less and villages are most accessible (Ngoumou and Walsh, 1993; WHO, 1995b). In each community, a random sample of 50 males aged 20 years and over are to be examined for the presence of onchocercal nodules. All persons in the sample should be engaged in rural occupations and'should have been resident in the community for at least l0 years. Males are preferred because they are more likely to be heavily infected than females and they are more amenable to physical examination by palpation. If women desire to be included in the examination, there is no objection to doing so provided this is without prejudice to the standard requirement of examining 50 adult males. The results of the examination for the two genders should be reported separately. In villages where there are less than 50 adult males available for examination the remaining number should be completed by sampling of consenting adult females. Validation of REA Validation of REA results should be undertaken within l-2 months of the completion of the REA, by a group of independent experts, for a sample of villages where the original REA was conducted (WHO, 1995b). The validation villages should be selected by a stratified random sampling method that takes account of different ecological zonal patterns and levels of L- 8reported onchocercal endemicity. Validation inspires an appropriate degree of confidence in the reliability and quality of the REMO data that APOC demands for operational purposes. Geographical Information System Analysis Data collected through rapid assessment surveys are then integrated in geographical information system (GIS) for analysis and interpretation. Once the data have been plotted on the map (Fig.1), the epidemiological patterns would be interpreted. This phase is executed by the same REMO teams that did the original sampling of the survey villages. It is important to remember that the survey data do not represent a random spatial sample, but that villages were selected in a purposefully biased manner using the available information about the river basins. The results of REA surveys are presented as pie charts on a map. (Fig.l). The analysis and interpretation should be done by river basin and river section. Thereafter it is necessary to extrapolate from these results and to classiff the endemicity levels of different areas. The ultimate aim of the analysis is to classiff subsections of the river basins into ivermectin treatment strategies. (Ngoumou and Walsh, 1993; WHO,I995b,1998) Application of REMO/GIS resultsfor CDTI Strategt On the basis of the Guidelines for Analysis of REMO Data Using GIS (WHO, 1998), the teams involved in the REMO exercises at country level have worked together to identi$r three different operational areas based on treatment shategies: 1. Definite CDTI area,2. No CDTI area and 3. Possible CDTI area. A Definite CDTI area is one for which the REMO results show clearly that onchocerciasis is highly endemic (with nodule prevalence greater than20o/o, the APOC threshold for CDTI) and the disease constitutes a significant public health problem throughout the area. No CDTI areas are made up of villages where there is very low or no transmission of O. volvulus. In a No CDTI area, onchocerciasis prevalence, expressed in nodules rate, ranges from 0 to lgoh.In such villages, onchocerciasis is not considered as a being of sufficient public health 9problem to warrant CDTI. Clinic-based ivermectin treatment may be provided in areas where there is transmission but endemicity is low. A Possible CDTI area is one where the epidemiological paffem presented by the REMO results is not sufficiently clear and therefore requires refinement of the map of the distribution of onchocerciasis through additional REA surveys. The need or otherwise for CDTI would depend on the outcome of the refinement. In REMO/GIS mapping to delineate the CDTI areas, a standard colour scheme has been adopted and accepted by the parhrers involved in onchocerciasis control (Fig. 2). It is as follows: Rled for Definite CDTI areas, i.e. priority areas where ivermectin mass treatment is indicated, Green forNo CDTI areas where mass treatment with ivermectin is not indicated or allowed and Yellow for Possible CDTI areas awaiting further refinement of REMO endemicity pattern to determine whether CDTI is or is not indicated. STATUS OF REMO IN APOC COUNTRIES As of July 2001, REMO has been completed in 12 APOC countries and partially completed in five others. In the remaining two, Angola and Burundi, REMO has not been conducted mainly because of social instability. (Table 1). The results of REMO show that in four countries, Gabon, Kenya, Mozambique and Rwanda, the levels of onchocerciasis endemicity do not reach the threshold set by APOC for CDTI. REMO/GIS results reveal a vast belt of hyper- and meso-endemic onchocerciasis stretching from South Eastern Nigeria through Cameroon, Chad, CAR, DRC, Ethiopia, Kenya, Malawi, Sudan, T arzania and Uganda. APPLICATION OF RAPID EPIDEMIOLOGICAL MAPPING BY APOC REMO/GIS has been extensively and intensively used by NOTFs, with the support of APOC, as a tool to draw onchocerciasis epidemiological maps in endemic African countries for the implementation of CDTI control strategy. The results of REMO are a prerequisite for consideration of any national plan and CDTI project proposal for APOC funding. Thus 10 APOC uses REMO as a tool for operational planning and targeting of interventions and the allocating of resources to communities in most need. REMO/GIS not only allows for visualization of priority areas for ivermectin mass distribution but also an estimation of the number of persons to be treated. At country level, REMO helps to establish a partnership between Ministries of Health (MoHs) and NGDOs for the development of project proposals for control of onchocerciasis. APOC on its part uses REMO results to explain the epidemiological distribution of the disease and to mobilize and allocate funds for the disease control. From 1996 to 2001, REMO results have been used as a fundamental basis for approval of APOC funding for 60 CDTI projects. One of the major benefits of the application of REMO by APOC is its use in the estimation of the size of target populations. REMO/GIS is used as a tool to overlay the layer of Definite CDTI areas on the layer of rural population per administrative area in order to calculate the population of CDTI zones. Taking into account the exclusion from the treatment population of children below 5 years of age, a ratio of 0.84 is applied. If all the areas studied during REMO field exercise have available population data, the estimation of the population to be treated could be directly calculated on the basis of the above 'formula'. However, this has usually not been the case for most of the areas. For each country, available data from the last census is compiled and made available in the GIS format. (Tobler et a\.,1995). The figure from the smallest census unit is used, wherever available, for the population estimates. The estimate of the numbers of people that require treatment for onchocerciasis in Nigeria was originally less than 10 million. This figure has since been revised upwards to 22,260,000 based on country-wide REMO results. Given the focal transmission pattern of onchocerciasis, the population estimates are based on the rural population, which is endemic for the disease. From a study conducted by the World Resources Institute (WRD in collaboration with the Club du Sahel in 1995 (World Resource lnstitute, 1995), it is possible to obtain rural population estimates by administrative T 11 areas for some of the APOC countries. Elsewhere no detailed information is available and therefore, an estimation of the percentage of rural population is applied to the whole country. For example, analysis of the results of REMO/GIS in Cameroon showed that 5025 villages are in the Definite CDTI zone and another 403 villages are located in the Possible CDTI (CDTI likely) zone. The REA results gave an average of 700 persons per village. On the assumption that CDTI would be required in at least 50Yo of the villages classified as CDTI likely, the estimated total target population for CDTI would be: (5025 X 700) + (403 X 700 X 0.5; :3,658,550 persons. Table 2 shows that 57,650,000 persons are estimated to be eligible for yearly ivermectin treatment in 13 APOC countries. This is twice the population originally estimated in the APOC Phase I document (WHO, 1996). These estimates will continue to be revised as REMO is completed in each country. Monitoring and evaluation of projects' performance are key elements of the APOC agenda. They enable the Programme to initiate mid-course corrections in activities as may be necessary. In order to demonstrate and measure the impact of CDTI projects, APOC has anticipated that NOTFs would need a simple system that would help them to produce routine customized maps with information on the location of health care facilities. This implies the decentralization of the use of REMO results in implementing CDTI projects. APOC in collaboration with WHO/FIealtMap Unit is therefore currently testing a customized module based on the Healthmapper for monitoring and evaluation of CDTI activities in APOC countries. HealthMapper, a user-friendly software, is a database management and mapping system, developed by the WHO/HealthMap Unit for public health planning and decision-making at all levels. It simplifies the process of data collection, storage, retrieval, management and analysis. HealthMapper manages REMO data and simplifies the use of GIS and mapping as well as data from other sources. It provides a user-friendly interface for spatial and statistical analysis (maps, tables and charts) of public health data. People at risk of onchocerciasis live in geographically isolated communities. They are usually neglected, have poor access to information and health L2 care and are served by poor transport and education infrastructure. A fundamental step for APOC in reaching such remote populations in order to sensitize them for proper and appropriate responses to their health problem is to put them on maps. REMO data and other health information integrated in the HealtMapper by NOTFs are seen by APOC as being critical for project implementation and setting up an integrated monitoring system of health events at community level. CHALLENGES In the application of REMO by APOC, field experience has shown that there are some specific challenges that need to be addressed in the next phase of the control progftlrnme. Challenges relsted to the vectors The experiences in APOC have shown that REMO is of immense value in all areas where the vectors of onchocerciasis are members of the Simultum damnosum complex. However, in some parts of Central and East Africa such as Tanzania where non-man-biting vector forms occur, the use of topographical maps alone is inadequate. This is because the examination of maps alone does not provide evidence on whether the members of S. damnosum complex likely to be present are potential vectors or not. In such a situation the standard REMO method needs modifications. Accurate and reliable information on disease transmission can only be obtained by conducting field surveys on the spot with the use of the Global Positioning System (GPS). The same considerations apply in areas, such as parts of the Democratic Republic of Congo and Uganda, where the main vectors are members of the .S. neavei group and where transmission tends to be very focal. To address these challenges, APOC called on seasoned and experienced expertise capable of designing a rational network of survey points on the spot. Equipped with GPS, the experts have conducted fruitful surveys. Challenges related to REA in REMO Prior to REMO the process for selection of communities for treatment of onchocerciasis with ivermectin under the IDP was more elaborate and tedious (Taylor et al., 1992). The REA 13 method of REMO. which is based on examination for palpable nodules is not only more rapid but more user- and client-friendly. In some cases however, such as in Sudan (Schwartz, et al., 1983, Baraka, et ol., 1995), many onchocerciasis affected persons present with a generalized and symmetrical dermatitis located on legs, arms and trunk. In this clinical form of onchocercal disease called Sowda, the subjects do not often have palpable nodules. Diagnosis is usually made on the basis of asymmetrical pruritic, papular dermatitis, a positive Mazzotti reaction, and histopathologic changes characteristic of the Sowda type of onchocercal dermatitis. The nodule prevalence rate is negligible vis-d-vis the true burden of onchocercal skin disease. Thus the classical REA method of REMO based on nodule prevalence is not appropriate under those circumstances. There is a need therefore to develop for such areas, new methods by which to define CDTI endemicity threshold comparable to the current one that is based on nodule prevalence. A first step is to determine the relationship between the classical levels of onchocercal endemicity and the prevalence of Sowda. A related challenge is to devise reasonably objective guidelines by which NOTFs can determine the limits of CDTI activities within an administrative unit when the REMO boundary traverses such a unit. A satisfactory way of balancing the expediency of socio- political considerations against the financial implications of treating all or only the REMO enclosed part of the administrative unit is needed. Challenges related to severe adverse reactions Severe adverse reactions (SARs) have been reported, following treatment with ivermectin in some areas where onchocerciasis and Loa loa infection (loiasis) co-exist, almost exclusively in central and western Cameroon. (Chippaux et al., 1996; Burham,1997; Boussinesq et al., 1998) The most serious of these reactions manifest as central nervous system disorders which may lead to coma and death. Since there are many areas within the ambit of APOC operations where the two diseases co-exist, such as Congo, DRC, Gabon, Nigeria and Sudan t4 there is a need to develop a rapid epidemiological assessment (REA) method to identify all CDTI areas where there is co-existing loiasis. At present the Programme concentrates on conducting REA systematically in all villages in CDTI designated areas where loiasis is suspected to occur, prior to mass distribution of ivermectin. All villages where onchocerciasis endemicity is hypo-endemic or absent are excluded from treatment. If the community is hyper- or meso-endemic and has never been treated or is located in an area with less than 60oh therapeutic coverage or with less than 2 treatment rounds, mass treatment with ivermectin should proceed over a fixed period of time with medical supervision from 3 to 5 days after treatment and supervision by community distributors (CDDs) from the second to the eighth days. Training and awareness of CDDs and health personnel at all level in such areas is enhanced for early detection of cases of severe adverse central nervous system reactions and prompt referral to appropriate district/regional hospitals where standardized emergency support protocols are available. CONCLUSIONS REMO/GIS has been used by APOC to identify accurately, quickly and economically communities suffering most from onchocerciasis. REMO/GIS results have become a standard tool that enables APOC participating countries to prepare national onchocerciasis control plans, to elaborate and launch CDTI projects for the control of the disease. The need to use REMO/GIS as a tool for monitoring/evaluation of the projects is foreseen to allow held implementers to follow up the performance of the key elementary units of APOC Projects which are the affected communities. ACKNOWLEDGEMENTS We are grateful to Dr Yankum Dadzie for his contributions to the use of the Rapid Epidemiological Mapping of Onchocerciasis (REMO) tool by APOC. Data for REMO exercises were collected in collaboration with members of the National Onchocerciasis Task Forces (NOTFs) and with logistic support of Ministries of Health in 17 countries; to these 15 individuals, too numerous to mention here, we are grateful. This study would not have been possible without the participation of hundreds of endemic communities; we are indebted to them. REFERENCES ALLEY, E. S., PLAISIER, A. P., BOATIN, B. A., DADZIE, K. Y., REMME, J., ZERBO, G. & SAMBA, E. M. (1994). The impact of five years of annual ivermectin treatment on skin microfilarial loads in the onchocerciasis focus of Asubende, Ghana. Transactions of the Royal Society of Tropical Medicine and Hygiene, 88, 581-584. BARAKA, O. 2., MAHMOUD, B. M; ALI, M. M., EL SHEIKH, E. A., HOMEIDA, M. M., MACKENZIE, C. D. & wLLLAMS, J. F. (1995). Ivermectin treatment in severe asymmetric reactive onchodermatitis (sowda) in Sudan. Transactions of the Royal Society of Troptcal Medicine and Hygiene, 89, 312-315. BENToN, n. (1998). Economic impact of onchocerciasis conhol through the African Programme for Onchocerciasis Control : an overview. Annals of Troptcal Medicine and Parasitologt,92 (Suppl.l), 533-S39. BOUSSINESQ, M., GARDON, J., GARDON-WENDEL, N., KAMGNO, J., NGOUMOU, P. & CrurrlUX, l. r.(1998). Three probable cases ofZoa loa encephalopathy following ivermectin treatment for onchocerciasis. American Journal of Tropical Medicine and Hygtene,58,46l-469. BURNHAM, c. u. (1997). Ivermectin where Loa loa is endemic. Lancet,350,2-3. CHIPPAUX, J. P., BOUSSINESQ, M., GARDON, J., GARDON-WENDEL, N. & ERNOULD, J. C. (1996). Severe adverse reaction risk during mass treatment with ivermectin in loiasis endemic areas. Parasitologt Today, 12, 448-450. DE soLE, G., GIESE, J., KEITU, F. M. & REMME, t. (l99la). Detailed epidemiological mapping of three onchocerciasis foci in West Africa., Acta Tropica, 48,203-213. 16 DE SOLE, G., BAKER, R. H. A., DADZIE, K. Y., GUILLET, P., KEITA, F. M. & REMME, J. (1991b). Onchocerciasis distribution and severity in five West African countries. Bulletin of the World H ealth Organization, 69, 689-698. LE BERRE, R. (1966). Contribution ir I 6tude biologique et 6cologique de Simulium damnosum Thoebald,l903 (Diptera :Simuliidae). M6m. ORSTOM. No 17 :L-204. NGouMou, p. & wALSH, J. F. (1993). A manual for Rapid Epidemiological Mapping of Onchocercr'asr's. TDR/TDE/ONCH O 193 .4. Geneva: WHO. NGouMou, p., wALSH, J. F. & MACE, J. M. (1994) A rapid mapping technique for the prevalence and distribution of onchocerciasis: a Cameroon case study. Annals of Tropical Medicine and P aras itolo gt, 88, 463 47 4. NwoKE, B. E. B. (1993). Rapid assessment of onchocerciasis endemicity in Nigeria. Nigerian Journal of Biotechnologt, 11, 213-216. pHrt-rppoN, B. (1978). L'onchocercose humaine en Afrique de I'Quest. Mem. Initiation ORSTOM, Paris, No 38, 193 pp. eurLLEvERE, D. (1979). Contribution d I dtude des caracteristiques, taxonomiques, biologiquee et vectrices des membres du complexe Simulium damnosum presents en Cote d'Ivoire. Travaux et documents de I'ORSTOM, 109, 1-304. scHwARTZ, D.A., BRANDLING-BENNETT, A. D., FIGUEROA, H., CONNOR, D. H. & GIBSON, O. W. (1983). Sowda-type onchocerciasis in Guatemala. Acta Tropica,40, 383-9. TAyLoR, H. R., DUKE, B. o. L. & tvtuNoz, B. c. (1992). The selection of communities for treatment of onchocerciasis with ivermectin. Tropical Medicine and Parasitolology, 43, 267-670. TOBLER, w. DEICHMANN, U., GOTTSGEN, J & MALOY, K. (1995). The Global Demographic Project. Universtty of Califtrnia, Santa Barbara,l995 Technical Report.95-96. woRLD HEALTH oRGANTsATIoN (1976). Epidemiology of onchocerciasis. Report of a WHO Expert Committee. Technical Report Series No. 597, Geneva: WHO. l7 woRLD HEALTH oRGANISATIoN (1995a). Onchocerciasis and its control: Report of a WHO Expert Committee on Onchocerciasts Control. Technical Report Series No. 852. Geneva: wHo. woRLD HEALTH oRGANISATION (1995b). Supplemental Guidelines for Rapid Epidemiological Mapping of Onchocercias is (RE MO). TDR/TDF/ONCHO/95. I . Geneva: WHO. woRLD HEALTH ORGANISATION (1966). African Programme for Onchocerciasis Control, Programme Documentfor Phase I (Unpublished document). woRLD HEALTH ORGANISATIoN (1992). Methods for community diagnosis of onchocerciasis to guide ivermectin-based control in Africa. TDR/TDE/oNCHo|92.2. Geneva: WHO. woRLD HEALTH oRGANISATIoN (1998) Guidelines for analysis of REMO data using. GIS. TDR/TDF/co MT /98.3, Geneva: WHO. r8 TABLE 1 Status of Rapid Epidemiological Mapping of Onchocerciasis (REMO) in 19 APOC countries as of July 2001 APOC countries Status of REMO REMO completed REMO partially completed or to be refined No REMO conducted at all Angola + Burundi + Cameroon + Central African Repub. (CAR) + Chad + Congo Republic + Democr. Rep. of Congo (DRC) + Equatorial Guinea + Ethiopia + Gabon + Kenya + Liberia + Malawi + Mozambique I I Nigeria + Rwanda + Uganda + United Republic of Tanzania + Sudan + Total t2 5 2 l9 TasI,e 2 Estimated number of persons to be treated in 13 APOC countries based on REMO/GIS results Country REMO/GIS estimates of the number of persons to be treated Cameroon 3 670 000 Central African Repub. (CAR) 700 000 Chad 1 270 000 Congo Republic 200 000 Democr. Rep.of Congo (DRC) 1,792 0000 Equatorial Guinea 240 000 Ethiopia 3 100 000 Liberia 2 360 000 Malawi l 400 000 Nigeria 22 260 000 Uganda I 530 000 United Republic of Tanzania 1 310 000 Sudan 1 690 000 Total 57,650,000 20 Fig. I Rapid epidemiological mapping of onchocerciasis results of Liberia 2t Ftc. 2 Rapid Epidemiological Results and CDTI areas in APOC Countries THE CHAITENGES OF COMMUNITY-DIRECTED TREATMENT WITH IVERMECTIN (coTil oF THE AFRICAN PROGRAIVI ME FOR ONCHOCERCIASIS coNrRor (apoc) THE CHALLENGES OF COMMUNITY.DIRECTED TREATMENT WTTH IVERMECTIN (CDTI) OF THB AFRICAN PROGRAMME FOR ONCHOCERCTASTS CONTROL (APOC) AUTHORS U. V. Arr,razrco,'w. R. BRIEGER,2 M. KaresaRwA,3 o. ArocuN, tM. NrEp,'B. BoATIN,U J. N'Dovo,'M. Noun,' A. SexttpLI.' ADDRESSES OF AUTHORS: African Programme for Onchocerciasis Control (APOC), Ouagadougou, Burkina Faso African Regional Health Education Centre (ARHEC), College of Medicine, University College Hospital (UCH), Ibadan, Nigeria The Carter Center, Global2000 River Blindness Program, Kampala, Uganda Federal University of Technology, Yola, Nigeria Groupe de Travail de Lutte contre l'Onchocercose (GTNO), Ministdre de la Sante publique, Cameroun Onchocerciasis Control Programme in West Africa (OCP), Ouagadougou, Burkina Faso Programme de Lutte contre l'Onchocercose et la C6cit6, Ministdre de la Sant6 de la Population, Bangui, R6publique Centrafricaine Dr. Uche V. Amazigo African Programme for Onchocerciasis Control (APOC),01 B. P.549 Ouagadougou 01, Burkina Faso Tetz (226) 34.29.59; (226) 34.29.s3 Faxz (226) 34.28.75 E-mail: amazigouv@oncho.omsbf Short running title: Challenges of Community-Directed Treatment 1. 2. aJ. 4. 5. 6. 7. Address for Correspondence ABSTRACT The African Programme for Onchocerciasis Control (APOC) adopted the community-directed treatment with ivermectin (CDTI) approach as its principal strategy for the control of onchocerciasis in the 19 countries of Africa that fall within the ambit of its mandate. Halfway through its 12 years mandate, APOC has gathered enough information on the main challenges to guide the activities of the second phase. An analysis of reports and other documents emanating from consultants, scientists, monitors, national and project level implementers indicated three broad categories of challenges: managerial, technical and socio-political. This review identified the most pertinent concerns under these three categories that APOC must address during the next phase to enhance the prospects of establishing sustainable ivermectin distribution systems. The major challenges include, 1) maintaining timely drug collection mechanisms, 2) integrating CDTI with existing primary health care services, 3) strengthening local health infrastructure, 4) achieving and maintaining an optimal treatment coverage, 5) establishing a process of community self-monitoring, 6) designing and implementing operations research locally, 7) ensuring adequacy of community-directed distributors, 8) increasing the involvement of local NGDOs in the Programme,9) achieving financial sustainability, l0) implementing equitable cost-recovery systems and 1l) engaging in effective advocacy. The implications of the challenges and suggestions about how they are being, or could be, addressed are also highlighted in this brief review which should be of value to other Programmes and Agencies that may be contemplating the adoption of this unique strategy. 2 INTRODUCTION The African Programme for Onchocerciasis Control (APOC) was launched in 1995 as a unique global partnership embracing the 19 African countries with endemic onchocerciasis outside the Onchocerciasis Control Programme in West Africa (OCP), donors, foundations, non- governmental development organisations (NGDOs), UN agencies, Merck & Co. Inc. the manufacturers of the drug Mectizan@ (ivermectin MSD), and the affected communities of the participating countries. (Etya'ale, 1998). Merck & Co. through the Mectizan Donation Programme (MDP), undertook to provide, and has been providing, Mectizan free of charge for as long as needed, to all onchocerciasis endemic countries throughout the world, including the 19 APOC countries.. For any given control programme or project this is a minimum of l5 to 25 years, a duration based on the life span of the adiltOnchocercavolvulus parasite. The extent of the population and geographic coverage with ivermectin treatment is a relevant factor in the calculation of need. (Boatin et a|.,1998). The objective of APOC is to establish a sustainable, community-directed ivermectin delivery system serving over 50 million people living in all the endemic communities in the 19 countries with the goal of eliminating onchocerciasis from Africa. A large, multi-country study conducted in 1994-95 showed that community directed treatment with ivermectin (CDTI) was a feasible and effective way to ensure that Mectizan reached endemic communities (WHO, 1996a; Akogun et a1.,2001). The study found that significantly higher average village population coverage figures of 68.60/o were achieved when CDTI was used compared to 62.2Yo where programme staff themselves planned and organised distribution. This difference in coverage is important because of the need to maintain a minimum annual population treatment coverage level of 65%o for at least l5 years in order to effectively control the disease (Plaisier et a1.,1990; Boatin et a1.,1998; Miri, 1998). Thus, APOC adopted CDTI as the basis of its control strategy. Since then potential effectiveness of the community- 3 directed treatment (CDT) approach for the control of lymphatic filariasis (LF) in Africa, has since been established (WHO,2000a; Gyapong et a|.,2001). The global LF elimination Programme has adopted the CDTI approach as well. The issue of community involvement in disease control programmes is not unique or peculiar to CDTI. For example, a role. for community treatment and health education in schistosomiasis control has been proposed based on research in Ghana (Aryeetey et aI.,1999; Wagatsuma et a|.,1999). Participatory community distribution techniques have been seen as being of special value to reach nomadic pastoralists with vitamin A and iron supplementation (Ettyang, 1999). Kidane and Morrow (2000) reported that mother coordinators in Tigray, Ethiopia, could be successfully trained to teach other mothers prompt recognition and treatment of malaria, resulting in a significant reduction of child deaths. The unique feature of CDTI is that the community is expected to take the leading role in planning, implementing and assessing the drug delivery project, and that this is being done simultaneously in almost 50,000 communities in Africa. CDTI is based on the principle of community participation. The key element of CDTI, as spelled out in guidelines published by APOC, is decision making by villagers themselves in selecting their own distributors (CDDs) and planning the period, dates, locations and modes of village level distribution (WHO, 1998) The advantages and benefits of changing from community-based distribution (CBIT) systems to CDTI have been expounded as sustainability and community ownership and empowerrnent as well as cost-savings to the health departments (Amazigo et a|.,1998). CDTI encourages and enables communities to organizetheir ivermectin distribution in line with their own cultural norms (Akogun et a1.,2001) and organizational structures (Katabarwa et al., 2000b). CDTI has also been seen as a stimulus for developing primary health care in remote and previously unreached rural villages (Hopkins, 1998). Recent reports have emphasised that the transition from CBIT programmes, to CDTI is a major challenge to health systems as middle level health staff often lack the management skills 4 needed to foster community participation (Miri, 1998). Community participation, including selection of their own CDDs, undergirds the basic steps in starting CDTI and is also associated with higher treatment coverage (Katabarwa et aL.,2000a). APOC has been in operation for just over five years and is funding 53 five-year CDTI projects in 14 of the 19 participating countries in Africa. [n the process of planning for the second and final phase, APOC has commissioned a review to examine the lessons learned during the first phase and the challenges of sustaining CDTI beyond the five years of APOC funding. The findings of the review are presented in this paper. MATERIALS AND METHODS Over 40 documents compiled by APOC on CDTI project planning, implementation and evaluation were reviewed. There were three broad categories of documents. Group one consisted of general overviews and assessments produced by consultants and staff of APOC for review by the Technical Consultative Committee (TCC) and other constituent bodies of APOC. The second group of documents included the projects proposals and their technical and financial reports. These contained detailed information on intended activities and budgets as well as reports on completed activities, expenditures and problems encountered. The third set of documents consisted of reports from independent scientists who led monitoring teams to projects one year after the start of implementation. These teams developed and used a series of qualitative and quantitative instruments that sought information about CDTI implementation and the potentials for its sustainability from community members, CDDs, community leaders, health staff, programme managers and political leaders. The first set of four documents (WHO, 2000b, c, d, e) provided the basis for an analytical framework to address the challenges (Appendix A). Using this framework, consultants reviewed all three groups of documents at APOC headquarters to extract and analyse the data and draw appropriate conclusions. The exercise also involved making suggestions on how best to meet the challenges in APOC s second phase. Major challenges were grouped under three categories, 5 managerial, technical, and social-political. FINDINGS AND IMPLICATIONS Monitoring reports tended to focus more on the district or local government area (LGA) level and ignore the commitment of StateA.lational Ministries of Health (MoHs). This was however a weakness of the monitoring tools. Annual and semi-annual technical and financial reports from the projects to APOC are sometimes late and are inconsistent in the format of the presentation and what is highlighted. For example, treatment coverage might be presented in detailed tables in annual reports, but only mentioned perfunctorily in the text in the following year interim reports. Although the foregoing are limitations to across project comparisons, they do not delimit the discernment of general trends in CDTI implementation. The findings of the document review are grouped under the three major categories of challenges, managerial, technical and social-political, mentioned above. Managerial Challenges Developing timely drug collection mechanisms In the context of CDTI, the community is primarily responsible for collecting its own supply of ivermectin from the nearest health facility. Reports from 14 countries indicated that 85% of communities collected their ivermectin from designated central storage points, but monitoring reports showed that about one-third of communities surveyed, health workers make ivermectin intended for distribution available to CDDs during training or re-training ahead of the scheduled distribution period. Although this ad hoc arrangement relieves the community of responsibility of choosing when to collect its own ivermectin stock, it overcomes potentially serious logistical problems and saves the community the additional cost of transportation to the central collection point. It also reduces reliance on health staff who may or may not be available at the agreed collection point on the date and at the time preferred for collection by the community. Because of the natural decay of village volunteer knowledge over time, especially among non-literate ones (Ryan et al., 1990-91), re-training should be an annual process, thus offering continual access to 6 the drug collection by villagers. Although, from the foregoing it would appear that drug collection mechanisms are not a major challenge, there are several important factors to consider. Monitoring reports documented only the first year of CDTI, and except where internal monitoring by NOTFs is undertaken annually, there is no way of knowing whether these collection systems can be maintained. The use of training as an opportunity for drug collection assumes that availability of supplies will remain synchronized with training and that training will continue to take place regularly. The question of whether communities will organize for their own drug collection for several years if drugs are not made available at training remains unanswered. Ensuring regular collection of the drug irrespective of whether it is supplied during training or not is a challenge. A number of other difficulties that pose challenges to be faced in APOC Phase 2 were also identified. These are related to delays in the process of transferring ivermectin through the various levels of the programme; transport-related problems; the issue of distance and the absence of contact health workers at material times. More than one in four communities experience delays in drug transfer. Health workers in Nigeria and Central African Republic (CAR) explain that often there is no regular means of transport from the district to the central stores at state/national level and that they often use their own transportation and funds to collect the drugs. The distance of the health facility nearest to the community may lie much farther than the recommended 5 km (WHO, 1996; 1998). Infrequency or nonexistence of transportation, either public or private, is further compounded when communities eventually reach the nearest facility or central store, only for the person in charge of drug supplies to be absent. Therefore APOC should encourage and facilitate the establishment of a flexible but effective process for timely collection of drugs by the communities during Phase 2. In Uganda, where health facilities arefar, clusters of villages come together and choose alternatives such as schools or other community institutions where supplies might be stored safely and conveniently. l Integrating with the Health Services The integration of CDTI into the existing health system was a foundation goal of APOC for assuring sustainability. A major challenge is to define standards of what constitutes integration, recognising that integration is a two-way process in which CDTI benefits from the resources of the health service and the health service takes advantage of the community orientation and procedures of CDTI. CDTI represents a unique approach to community health by stressing a working partnership between health staff and the community, with the community taking a strong leadership role. In Sudan where health care service is inadequate, CDTI has been used to disseminate health information and to serve as a vehicle for communities to convey their health needs to govemment and other agencies. The Liberian CDTI proposal, envisions that, CDTI will assrsl in cap acity butl di ng for future co mmunity p arttcip ation proj ects. Integration of a disease-specific control programme into the primary health care services can be problematic if the system into which a programme is being integrated is weak. For example, coverage dropped, supplies were delayed and wastage increased in a community distribution programme for iodized oil capsules when it was integrated into the PHC system in Tanzania (Peterson et al., 1999). Inadequate system resources and social mobilization skills contributed to the problem. The challenge is to devise means by which CDTI would not only be successful even in the presence of a weak PHC but also act as a catalyst for integration of other programmes. As observed during monitoring in Sudan, The community s expectations ofvery basic services in addition to CDTI cannot be ignored without raising serious ethical problems. A major challenge has been the lack of clear indicators for measuring progress toward integration. Several potential indicators of integration have been found in various APOC reports and documents, and are outlined here as a basis for developing a set of standards or guidelines. For example, it was documented that many district CDTI staff are also responsible for other programmes at the local govemment area level (Katabarwa and Mutabazi,1999). This may also be a reflection of staffing realities where there are not enough personnel to assign separately to 8 programmes such as malaria control, guinea worrn eradication and AIDS education. The challenge of integration here is one of balance to remove any spatial or temporal overlaps. This could be achieved through proper scheduling and adequate management training and supervision. Just as district health staff have multiple functions, so do some CDDs. Most CDDs in Cameroon, Nigeria , Uganda and Sudan, have been reported to be involved in various other health and development activities. (Homeida et a|.,2001). CDDs believe that their training in CDTI helped enabled them to perform other tasks more effectively, and community leaders value having trained people provide a variety of services that the community needs. Resource sharing is another potential integration indicator that has received little attention in APOC reports. In Nasarawa State, Nigeria, LGA onchocerciasis staff had access to and were using motorcycles from the general pool of the PHC Department. Fuel and lubricants were adequately supplied for these motorcycles and maintenance was carried out. This contrasts to the experience in Oyo State, Nigeria, where independent monitors reported that, The available vehicles and bicycles are not enough and this adversely affects supervision The same problem was seen in the CDTI of Kano State, Nigeria. It should be noted however that sharing only works when the basic system has adequate and accessible resources. {nother indicator of integration is financial.. The 1999 annual report from Uganda on the Phase I project explained how financial support for CDTI could be integrated into district level budgeting: The Ministry of Health (MoH) recognizes onchocerciasis control as a priority for action, and the programme has been specified in the guidelines on the conditional grant for Primary Health Care (PHC) to the districts. This will enqble the program to benefit from the grant. " Unfortunately, in Kabale District of Uganda, the monitoring team found that none of the district or sub-county institutions had made plans to fund CDTI activities currently, nor were arrangements being made to continue the programme after donor support ceases. The challenge therefore, is working within the different levels of government from national to district and even 9 subdistrict to ensure that commitment to planning and budgeting is institutionalized. Strengthening Internal Infrastructure The main drawback to integrating CDTI with the health system has been identified as the weak structures in existing health delivery systems. Efforts must be made to strengthen front line, district and national health care infrastructure and systems. Some of the identified problems at the front line and district levels reflect weakness in the overall structures that are supposed to sustain CDTI. These include: 1) inadequate health facilities, 2) inadequate trained health personnel. 3) lack of community outreach, 4) poor understanding and tolerance by staff of local cultures, 5) neglect of front line staff in the information flow concerning CDTI and other programmes, 6) declining maintenance of facilities, equipment and workers skill levels, 7) lack of a clear def,rnition within the front line health facilities of staff roles. It is important to note that 3,456 district and front line health workers have been trained on CDTI by December 2000. The APOC mid-term evaluation report (WHO, 20000 revealed that not enough was being done to develop the capacity at the country level for the work being handled by the TCC. During its second phase, APOC faces the challenge of encouraging NOTFs to strengthen local and national capacity for carrying out most of the current functions of the TCC such as reviewing proposals and evaluating reports. The human and financial resources of individual NOTFs in APOC-supported countries to take on these roles have not yet been assessed. It is not clear whether they can take on and sustain these roles. In meeting this challenge care should be taken to use existing structures rather than creating new ones, which may not survive after APOC funding ceases. Both the NOTFs and APOC should identify structures that will be emplaced for sustaining CDTI after 5 year of CDTI implementation. The district is the appropriate level for the emplacement of CDTI responsibilities, and APOC will need to ensure that the necessary capacity is in place for local integration. To this end the indicators of integration will need to be further developed and incorporated into monitoring tools. l0 Currently, in most APOC-supported countries, CDTI activities are carried out by district staff. However, planning, information systems, and budgeting for CDTI are not included in the routine activities at the district level in all projects. This results in low levels of commitment towards CDTI by district staff. The challenge for CDTI is to ensure that this is done, and that activities such as training and supervision by the district and national offices are included in monthly and annual budgets. It was also recommended that at the national and district level, the policy promoted should safeguard and build on the CDTI approach for all health prograrnme activities at the village level. An approach, which enhances community capacity to handle new or additional activities as they see fit, enhances sustainability of the control of onchocerciasis and other diseases of public health importance should be promoted. Technical Challenges Achieving and Maintaining Coverage Two important indicators of APOC achievement are the extent of geographic and therapeutic coverage. Therapeutic coverage is the proportion of the people in an area that is treated with ivermectin. Within five years APOC has been able to more than double the 8 million annual pre- APOC treatment coverage level in endemic countries. Unfortunately the change in the denominator used (eligible versus total population) over time has not been properly recorded by projects thus making it difficult to draw comparisons between projects and to accurately describe trends from pre-APOC era. Howeverwhere there has been consistency as in five LGAs in Cross River State, Nigeria, and four districts in Uganda from 1996 to 2000, a trend of increasing coverage is visible overall. It is of note that it was only in 1998, at the start of CDTI in these districts, that the minimum desirable target of 65%o was achieved on average. Figure 1 summarises the average coverage achieved in the nine districts during two years prior to the start of CDTI and for three years thereafter. But the greatest challenge is how to maintain the desirable level of treatment coverage over a long period of time when community interest might ll have waned and the immediate health benefits of treatment are less obvious. Data Management Another major challenge to NOTFs is developing and maintaining the capacity to manage and analyse their own data. As projects take more responsibility for their data, APOC will need their commitment to use the same standards of measuring indicators so that reliable and comparable data on coverage are available at all levels. This may require thatNOTF arrange local workshops where the importance of accurate and reliable data for assessing performance is emphasised. Geographic coverage is the proportion of all communities within an endemic area designated for treatment that were actually treated. Ideally the number of villages/communities should have been established by rapid epidemiological mapping ofonchocerciasis (REMO) prior to developing a CDTI proposal. Not all projects started with an adequate REMO, but once REMO has been completed, it may become evident that some communities under treatment in the early years of the project may need to be dropped. This may present a political challenge to governments. Two special challenges on geographical coverage concern areas with civil unrest or war with resulting population dislocations, and areas where Loa loa infection is co-endemic with onchocerciasis, particularly in parts of Cameroon where severe adverse events, including fatalities, have occurred (Boussinesqet al. 1998). Minor side effects of ivermectin include itching, swelling, fever and dizziness. These affect a small portion of the population and tend to disappear with successive treatments as the load of infection decreases. These can be managed at the village level with analgesics and antihistimines. Although these side effects are rated as minor in a clinical sense, villagers who miss a day or two of work due to such inconveniences may consider the socio-economic consequences as quite serious. Direct refusals have been attributed to personal experience of side effects or rumours of the effects on others. Indirect refusal may take the form of absenteeism during the period of distribution. This points to the challenge of training CDDs to manage the minor side effects, which would serve an additional purpose of increasing their primary health t2 care repertoire and enhancing the potential for integrating CDTI and PHC. But the fundamental challenge remains and is worth reiterating, that is how to maintain interest of a population, majority of who are not overtly ill, in the consumption of a drug for many years. Promoting Commitment to Self-Monitoring and Operational Research Project monitoring and operational research are two major issues that are essential comnritments fbr system-wi<le progranrnle perfornrance inrprovernent. Indepenclent monitoring reports front 1998-2000 have provided useful inflormation for reviewing prograrnme components such as CDD selection, record keeping procedures and communiry participation. Various countries where independent monitoring has been done have written to APOC explaining how they have addressed issues raised in the monitoring reports as seen in the following extracts from unpublished APOC documents: "The enthusiasm andwillingness of the health fficials to accept thefindings and to take immediate steps is for me a crowning experience---by and large APOC has managed to remove the negativefeeling about monitoring and I am convinced that thts will encourage the culture of internal monitoring of not only CDTI but other health programmes". "Field staff (health and NGDOs) are usually pleased to have someone visiting their projects and are usually delighted to show off what they have been doing and use the opportunity to seek opinions on how some of the constraints could be reduced". APOC is now faced with the challenge of assisting NOTFs to establish and strengthen national and local self-monitoring processes and integrating these into the local health system. Self-monitoring using both rapid quantitative and qualitative methods is necessary at the district level to discern and address the changing factors that affect coverage from year to year. Time or season of distribution may be a factor one year, while mode of CDD selection (or replacement) may be a factor the next. This will facilitate better planning and management of CDTI at national and district levels and will benefit other programmes in addition to CDTI. The basic operating methods of APOC are based on firm research results initiated by TDR 13 and the OCP including the CDTI process and the REMO procedures. APOC has, since its inception, provided opportunities and funds for individual CDTI projects to initiate their own operational research (OR) as a mechanism for identifying relevant local research needs and responding to them. Among pertinent APOC-funded OR are those on the role of women in purdah in the community-level programme implementation and on the factors that motivate CDDs. The ninth session of the TCC considered seven OR proposals from four countries that addressed issues such as the impact of cost recovery strategies, women participation in CDTI and the role of community participation in the development of health education materials. The remaining OR challenges include local factors that influence coverage. A few studies have implied that side effects in one treatment round may dampen compliance in the next (Chijioke, 2000; WHO, 2000f ) while others have documented visibly beneficial side effects such as worrn expulsion, increased vitality and general well being after treatment (Akogun et al, 2000). These are deemed to have stimulated attendance for treatment. Such studies should be undertaken in different cultural settings. Another factor whose effect on coverage is not well understood is cost recovery (Hopkins, 1998). Some studies have documented willingness to pay (Onwujekwe et a|.,1998; 1999; 2000). Others have clearly indicated that there is opposition to payment for treatment under the cost-recovery scheme and that this has a negative effect on coverage(Clemmonsetal.,2001). Thereisneedforfurtherwelldesignedandexecutedresearch on the subject. Other potential OR issues include the effects of delays in release of project funds, and the feasibility of integrating CDTI and PHC. Research problems identified in the field should be addressed through appropriately designed local research and the results should be applied immediately to benefit the programme. In addition to the challenges of identifying pertinent field-based problems, partners in CDTI face the challenge of ensuring that project staff possess adequate research skills and orientation. Ministry of Health (MoH) personnel are rarely trained to carry out operational research, and where such skills are available, programme staff are often too busy to use them. t4 The collaboration and partnership APOC has developed with research institutions and scientists should be encouraged at the local level. Harnessing the capacity of personnel of the Research and Planning Units of the MoHs to conceptualize, design, implement, analyse and utilise operational research is a challenge that APOC can address in its second phase. At present the TCC vets and approves OR projects. The challenge now is to devolve this responsibility to NOTFs in consultation with national research bodies. Social-Political Challenges Increasing Cultural Appropriateness of CDDs CDTI Guidelines (WHO, 1998) recommend that a community of 250 people should have at least two CDDs. An average 2.7 CDDs per village has been documented. (Amazigo,1999;Amazigo et al, 200I.). Four major challenges exist in maintaining an adequate pool of culturally appropriate CDDs over time, 1) motivation, 2) accountability, 3) literacy and 4) gender. While in some communities CDDs are satisfied with receiving only thanks and praise, based on the belief that they are simply fulfilling their normal obligations to the community (WHO, 2000f ,' Katabarwa et al., 2000b), in other projects staff and independent monitors reported demands for monetary or material incentives for CDD work (Katabarwa and Mutabazi, 1999; Katabarwa et al., 2000b). Factors apparently responsible for the demand for incentives include, 1) inadequate discussion and consensus building at village meetings on the roles, responsibilities and motivation of the CDD, 2) less favourable comparison of CDTI with other donor-supported programmes that give community based workers t-shirts, caps, first aid kits and even bicycles, and 3) a history of community based ivermectin delivery projects that gave small allowances to distributors prior to CDTI. (Amazigo et. a|.,2001). Ideally the CDD, like other voluntary community health workers, should be accountable to his/her community and not function as an extension worker of the formal health system (Nichter, 1997). Studies have shown that community satisfaction with CDDs work and coverage levels are negatively affected when either a single community leader or health worker chooses the l5 CDD (Katabarwa et a1.,20004). CDDs are more acceptable and function better when chosen at the smallest natural community unit, such as the kinship in Uganda (Katabarwa et a1.,2000b). Pressure by donor agencies and programme planners have often resulted in villagers using literacy as a requirement for selection of CDDs who can better fill out various treatment forms distributed by NOTFs. Because of their relatively higher levels of education such CDDs are prone to seeking work opportunities outside the area, and thus their attrition is also high. APOC and TDRboth addressed the issue of record keeping by CDDs with no or low level literacy skills (Oladepo et a|.,1999-2000). A challenge to APOC is to develop simple methods that can be used effectively by both literate and non-literate CDDs so that lack of literacy is not a barrier to selection as CDD. (WHO, 2000n. Minimal involvement of women in planning, decision-making and implementation of CDTI has been a concern raised in monitoring reports from different countries, which have documented that only 20o/o of CDDs are female. Local culture may inhibit women from taking on decision-making and leadership roles. In some instances, where women have been selected as CDDs, it was found that they deferred the tasks to their male children, in part because of the demands of their domestic roles (Akogun et a|.,2001). A conflict of ideals is evident. On one hand APOC wishes to promote women s empowerrnent in community health, and on the other it wants communities to make their own decisions on how to run CDTI in the context of the local culture. The value of having stakeholders meeting (SHM) in promoting, among other things, the participation of women has been demonstrated. (WHO, 2000fi. APOC is now encouraging CDTI projects to organize community feedback meetings to discuss distribution issues, and this may be a way to meet the challenge of involving women and other special groups in the community. The challenges posed in respect of gender attributes have been reviewed at length (Clemmons et aL.,2001). Ac hiev ing F i n an cial S ustain ab ility Funding of CDTI after the initial 5-year APOC grants is a major challenge on political will for I6 onchocerciasis-endemic countries. An unpublished document prepared for the May 2000 APOC Partners Meeting framed the issue thus: "Sustainability after the cessation of APOC funds ... is an issue, whtch requires great consideration by the TCC and all other bodies of APOC. This question requires to be looked into to examine (sic) the abilities of the partners (Governments and NGDOs) to sustain the project after the 5'h year of APOCfunding and to evaluate the indicators used by the TCCfor sustainability and to see if they werefeasible and have proved their value in thefirst project that completed the 5 years". When mass treatment with ivermectin began in the early 1990s in APOC countries, sourcing of funds was done on a project-by-project basis. The primary responsibility for fundraising was carried out by NGDOs, most of who were based in the North. The new alrangement of APOC to a great extent separates the technical assistance and fundraising roles so that the NGDO can concentrate more on helping the programme run smoothly, while the fiscal agency, The World Bank approaches the donor community. APOC funding is front-loaded so that basic infrastructure and management processes (e.g. training) are put in place early. The budget then decreases annually (Figure 2) until treatment per person approaches US $0.20 or less in the fifth year. By the sixth year the MoH, with possible NGDO assistance, takes over full financial responsibility. Documents were reviewed to learn whether current national, state and local health authorities CDTI projects are going to be in the position to fund their programme activities after withdrawal of APOC support. An original target of APOC was that, By the year 2008, all community-based systems will have been declared sustainable according to criteria to be defined by the TCC all APOC support will have ceased, and all residual support activities will have been integrated into the national health seryices." REFERENCE is mandatory! APOC requested State and Country projects to address the issue of financial sustainability in their initial proposals to APOC. On a positive note, there was evidence of future financial sustainability in reports about funding lor ll ivernrectin rlistribution in thc propose<l project areas prior to thc start o[ APOC. Analysis of eleven project proposals from Malawi, Nigeria, Sudan and Uganda showed the expected diminishing support from APOC over the five-year period, ranging from an average of 55% in the first year to 3lo/o in the final year. Final year budget estimates varied widely from l5o/oto 60Yoof totals expected from APOC. This reflects the differences in the relative financial status of APOC countries and that some countries are in situation of civil conflicts or war. On average the fifth year budgets also represented 42o/o of the first year budget. Mostly these conform to the TCC guideline of a maximum of US $0.20 per person treated in the final year of APOC support. On paper therefore, the projects look poised to carry on by themselves, but the question arises as to how the proportion of the fifth year budget supplied by APOC will be made up in the sixth and subsequent years. This raises the challenge of cost reduction. A study of CDTI implementation in three Ugandan Districts found that training costs varied, and that higher costs were negatively correlated with sustainability by the districts (Katabarwa and Mutabazi,1999). While the budgets generally reflected the expected trend in financing, the proposals were often weak in articulating future plans for guaranteeing sustainability after APOC funds cease. For example, the proposal for Ondo State, Nigeria in 1999, stated only that progress towards financial sustainability,{ "will be monitored through prompt payments of staff salaries, provision of logistic and infrastructural support." Others did not even address the issue. The inability to articulate clear plans for achieving financial sustainability was framed in the APOC commissioned study on sustainability thus: Clear self-sufficiency targets have not been prepared by countries and APOC. Self-sufficiency targets, as in the case of the Expanded Programme on Immunisation (EPI), should take into consideration such factors as the relative wealth of the country and population (WHO, 2000e). The Mid-term Extemal Evaluation of Phase I noted a related problem, that of Lack of standards for drawing up budgets (WHO, 20004. The evaluation team identified topping up of MoH staff salaries using project funds as l8 a serious threat to sustainability, and recommended that this practice be phased out immediately. In addition, reports documented the concern of APOC officials about inflated and unsustainable budget requests from countries. In a recent TDR study on indicators for sustainability, local government political and health officials were asked about their commitment to sustaining funding for CDTI (WHO, 2000fl. Responses were inevitably positive from the politicians, and while the heads of health departments were generally positive, they also expressed caution. It was the Onchocerciasis Coordinators themselves who were less than optimistic when they narrated actual experiences of using their own funds to collect drug supplies and supervise village level activities. It is unrealistic to expect anyone, especially politicians, to express doubts about sustainability when no figures are provided and the target is to be achieved in five years time, when most of them will not be there! (WHO, 2000e). A major challenge for the future sustainability of CDTI is to reconcile the gap between budgeting of funds and actually releasing them for spending. During monitoring of the first year of Phase I CDTI projects in Uganda it was found that in one district only 1.8% of actual operating funds came from government, while the rest were provided by the NGDO and APOC. This is not an isolated experience. Five Nigerian states, Adamawa, Yobe, Borno, Taraba and the Federal Capital Territory GCT) have released between $10,000 and $25,000 annually since inception of CDTI. A comparison with the original budget submitted by the Yobe project to APOC showed that the amount released was adequate. On the contrary, the amount released by the Adamawa project was much lower than expected. For example, the first year budget proposed by Adamawa, the MoH spending $240,040, of which $81,540 (34%) was for programmatic categories including training, travel, communication, supplies and health education. The second year programmatic categories were budgeted at $79,988 (59% of MoH tota[), and third year was $ 109,238 (63%). This implies either inadequate release of funds or an original over-estimate of need. t9 The local government (LGA) contribution to CDTI running costs is often not properly reflected in budget proposals, although a few states in Nigeria have a column in their budgets for LGA contributions. Average annual budget proportion attributed to LGA contributions was 25oZ in Cross River State, 27Yo in Taraba State and 46Yo in Kogi State. In a 1998 review of onchocerciasis control activities in Adamawa and Borno State, Nigeria, prior to implementation of CDTI, it was reported that LGAs consistently provided, in addition to regular staff, annual running cost funding between $500 and $2,500. Since then, LGA contributions have continued to cover the costs of printing registers and IEC materials and fueling vehicles for f,reldwork. A recent meeting of NOTF representatives in Ouagadougou recommended ensuring that contributions by the districts are recognized in APOC s Phase 2. The group concluded that, There is little doubt that the district level will be a prominent player. Expertise should be developed in such a way that it is the districts which assume responsibility and have the necessary capacity. Another cost component is that borne by the community. Community support is contingent on an understanding of and commitment to the principle of community ownership of the programme by all partners. The monitoring reports showed that"where community support is minimal, this reflects poor understanding by community of the role they are expected to play tn dectsion making as regards volunteering to be CDDs and /or supporting the CDD with cash or kind. Likewise in Ruvuma,Tanzania the monitors reported, They (the community) did not perceive the programme as their own, but that of the government and perhaps donors. The challenge remains for health staff to educate communities better on the true nature of community participation embodied in the CDTI process. Positive experiences of community support to CDDs in Nigeria and Tanzania (Clemmons el al., 2001), and in community fundraising for insecticide impregnated bednets (ITNs) in the Gambia offer a good examples for other programmes (Mills et al.,1994). Promoting the Involvement and Strengthening of Local NGDOs 20 Another challenge to sustaining CDTI is finding and encouraging the involvement of local NGDOs that have the capacity to implement CDTI plans and that are willing to take on projects. Local NGDOs could enhance the likelihood of long-term sustainability. They are often more flexible than government agencies and are often more familiar with local social and cultural ways of getting things done efficiently at the community level (Clark, 1995). There is agreement that local NGDOs should be promoted as partners to enhance the prospects of sustainability. Criteria for involvement of both local and international NGDOs address the characteristics of credibilify, presence of infrasffucture and staffing, financial base to support action and technical capabilities in both management and public health. There have been some positive experiences with local NGDOs. In Nigeria, the Mission to Save the Helplesss (MITOSATH), a local NGDO, was groomed by an international NGDO, Christoffel Blinden Mission (CBM), and is now subcontracted to take responsibility for CDTI in a few districts in the Taraba State. By 2003, all CBM supported LGAs in Taraba state will be devolved to MITOSATH. Sight Savers International (SS! is also nurturing a local NGDO, Christian Health Association of Liberia (CHAL) to take full responsibilities in CDTI in four counties in Liberia, due to its long history of providing both clinical and PHC services in the country. Local NGDOs have also shown promise for CDTI in conflict areas of the Sudan Clearly the challenge of involving local NGDOs is two-pronged. The first is identifying and building the capacity of local groups. The second is guaranteeing the political will within goverrrment and the NOTF to collaborate with local NGDOs. APOC should have a major role to play in establishing norns and standards for such collaboration. Cost Recovery There is no unanimity on the role of cost-recovery as a solution to financial self-sufficiency. It is practiced in respect of CDTI only in Cameroon and Chad. (Amazigo et. al.200l; Clemmons e/ al. 2001; Homeida et. al2001). The necessity for cost recovery has been linked by some to the perceived need to provide cash or material incentives to CDDs. As the study of sustainability of 2t CDTI pointed out, Decisions on user charges are essentially political (WHO, 2000e). The introduction of user fees in other places have resulted in decreased service utilization. In Ghana it was found that exemption procedures failed to work, but on the other hand, fees became a vital resource accounting for between two-thirds and four-fifths of non-salary operating expenses in government health facilities (Nyonator and Kutzin, 1999). The issue of cost recovery was raised in the original APOC programme document. Recognizing that the Bamako Initiative was being implemented as a policy in many countries, APOC noted that while cost-recovery was not inconsistent with the its objective of sustainability, however since ivermectin is donatedfree-of-charge, there can be no cost-recovery in respect of the value of the drug itself; cost-recovery can only relate to the cost of distribution. (WHO, 1996b). In subsequent years, the TCC noted that where the Bamako Initiative has been implemented, it is incumbent on State and Local Onchocerciasis Control Teams (SOCT/LOCTs) to examine the issue of cost-recovery, especially if CDTI is to be integrated with other aspects of PHC. Researchers have also asked questions about cost-recovery and willingness to pay for ivermectin distribution. Onwujekwe et al. (2000) asked villagers in three Nigerian communities how much they were willing to pay and their average responses ranged from $0.28 to $0.30. Some preferred to have the money collected in advance, while others preferred payment at the time service was received. [t was also determined that villagers were generally willing to develop a system of equity in which those unable to pay could be covered by more generous donations from the more well to do persons in the community (Onwujekwe et al., 1999). Unfortunately such studies merely document intentions, not actual expenditures. During the first independent monitoring exercise in Uganda in 1998, villages reported that they were being asked to contribute 100 Uganda Shillings per household toward the local running, which essentially consisted of help for the CDD to cover feeding and travel costs incurred in the implementation of CDTI. Some CDDs and focus group respondents believed that 22 this fee discouraged people from coming for the drug. In addition some CDDs and NGDO staff believed that the idea of fees was not truly a community decision in all places, but was often imposed by community leaders and health care personnel. Ultimately, these fees have been dropped. Cameroon and Chad on the other hand have been running cost recovery, with special village bank accounts set up in most cases. In 2000, NOTF of Cameroon in a meeting with community leaders revised allocation of cost recovery funds and increased payments to CDDs from 25 %o to 32oh. Criticism of the system stems from lack of accounting transparency in some instances. The challenge of cost-recovery is the establishing of a system with the full participation of the community and with accountability and transparency. By the way it now operates cost-recovery is unlikely to be a mechanism for guaranteeing the financial sustainability of CDTI beyond the community level. (WHO, 2000e). CDTI in ConJlict Areas CDTI was originally developed and tested in communities that are stable and that have a functional health system. The assumptions about community participation developed under these circumstances may however not apply in conflict areas. Four countries (Sudan, DRC, Congo, Uganda) receiving APOC grants for CDTI implementation are experiencing conflicts ranging from sporadic communal disturbances to prolonged wars where civil machinery, including the health service, has broken down completely. These conflicts impact on geographical coverage as access to communities becomes limited during conflicts. Other problems include 1) convincing the communities whose lives are threatened by violence, and indeed the authorities, that onchocerciasis is a priority 2) finding partners who have the capability or willingness to venture into such areas with the attendant problems of insecurity, and 3) random and unpredictable population movements. These make it difficult to determine census, trends in treatment coverage and select CDDs, among other CDTI processes. Solutions for maintaining CDTI in such situations must be locally formulated to take account of the extent of disruption of community life. In areas where the conflict is limited to a 23 few villages and does not last the whole year, meaningful flexibility in timing of ivermectin distribution may make CDTI implementation possible. Southern Sudan demonstrates one of the most protracted and formidable challenges with regard to CDTI implementation in conflict areas. Operation Lifeline Sudan (OLS), a partnership involving 25local and international NGOs, is coordinated by HealthNet International (HNI). OLS has been responsible for supervising and training of CDDs, and in the absence of a functioning health system the UN/OLS organized CDTI within its emergency-oriented and humanitarian framework, although it could not have access to all the endemic villages. 24 The meeting of NOTF representatives in February 2001 addressed the issue of CDTI in conflict areas and offered three recommendations, l) be flexible in implementing CDTI, 2) develop procedures for treatment during temporary armistices, and 3) expand into areas emerging from civil strife. The latter recognizes the real danger from land mines and artillery that threaten health workers and villagers and may make it impossible to reach some areas. The challenge for CDTI is to encourage NOTFs and MoH to develop plans for existing and potential conflicts. Advo cacy fo r S ustai na b ility Advocacy is another strategy for promoting sustainability. There are evidences of financial commitment and a positive attitude by a few Governments, while in other projects simply creating awareness of CDTI among policy makers has not resulted in any concrete support other than payment of salaries and provision of office spaces. From experiences like these APOC recommended that NOTFs need to intensify advocacy to ensure that policy makers make it mandatory for projects to be adequately budgeted within the PHC system". REFERENCE is mandatory. At country level, specifically in Cameroon, CAR, Ethiopia, Nigeria and Tanzania, the NGDO Coalitions have played a major role in advocacy. The experiences of independent monitors to Delta State, Nigeria, demonstrate the need for advocacy at the LGA level. According to an official in one of the LGAs visited, The Chairman, irrespective of his background, medical or otherwise, will not willingly support health activities, whtch are neither tangible nor visible for the people to see. Apart from embarking on projects for political points, they also prefer activities that are income earning. The CDTI programme did not appear to meet either requirement of the politicians. The process of independent monitoring also offers an opportunity for advocacy. Feedback begins tn the community when the monitoring team debriefs the community, the CDD and the health personnel, and is further enhanced by the debriefing of the State and NGDO project implementation team. Feedback is given to the NOTF and to the WR (WHO Representative) 25 before the report is sent to APOC. While it was noted that There is no evidence of APOC follow-up of the application of monitors recommendations in thefield, reports from the projects showed responses. In Sudan, for example, there was introduction ofNGDO collaboration, annual awards to CDDs to enhance motivation, internal monitoring and improved community mobilization. Nigeria reported undertaking retraining of trainers and supervisors, use of IEC materials and increasing women s participation after receiving monitoring debriefings and reports. After monitoring in Uganda, the NOTF reported providing uniform treatment registers for CDDs, retraining CDDs and health staff and producing IEC materials, among other activities. APOC has developed advocacy tools including several documentary videos and public service announcements. Since most policy makers are unfamiliar with life at the grassroots, these could be powerful advocacy tools. Evaluation of the use and effect of these tools are needed. The Filariasis Intervention Task Force of TDR, with APOC support, has an on-going multi- country study to strengthen advocacy strategies of CDTI projects. CONCLUSIONS AND IMPLICATIONS Bossart (1990) identified five indicators of sustainability in donor-supported health projects, which should 1) demonstrate effectiveness in reaching clearly defined goals, 2) integrate their activities fully into established administrative structures, 3) gain significant levels of funding from national sources during the life of the project, 4) negotiate project design with a mutually respectful process of give and take, and 5) include a strong training component. The foregoing presentation has shown CDTI projects to have succeeded quite well in two indicators, the first (coverage targets being met) and the fifth (training of community members as well as professionals). The major challenges lie in integrating with PHC, committing substantial local funding, and mutual negotiation, which in this case involves securing better involvement of the community in the planning process as well as better communication between front line health care personnel, managers and policy makers. The challenges to sustaining CDTI are akin to those that have been identified in the 26 reorientation of Cameroon s health system to PHC, where the health system, from the policy makers to the front line staff has been facing the challenge of operating within the philosophy of PHC as it relates to community involvement without the enabling legal, policy, training and management changes that would foster these. (Essomba et a|.,1993) The final challenge is documenting progress toward meeting the above outlined challenges during Phase 2 of APOC. APOC is alive to the challenges of Phase 2 of its CDTI operations, a handful of which have been identified. It could be said with a measure of justification that identifying the challenges is half the battle won. ACKNOWLEDGEMENTS The authors are grateful to Dr. Yankum Dadzie, former Director of APOC, Ms. Catherine Cross, Professor Oladele Kale and Professor David Molyneux for their contributions in producing this manuscript. We are also grateful to the members of the National Onchocerciasis Task Forces (NOTFs) of Cameroon, Sudan, Nigeria, Uganda, Chad, Central African Republic, Liberia, Ethiopia, Tanzania, Equatorial Guinea and Malawi for their support in the collection of data. We are indebted to thousands of communities and other partners of APOC for their unrelenting hard work in the delivery of ivermectin. REFERENCES AKOGUN, o.B, AKOGUN, M. K. & eUoU, Z. (2000). Community-perceived benefits of ivermectin treatment in north-eastern Nigeria. Social Science and Medicine,50, L4l5-1416. AKOGUN, O. B., AUDU,2., WEISS, M.G., ADELAKLIN, A.O., AKOH, J. I., AKOGUN, M. K., REMME, H, & KALE, o. o. (2001). Community-directed treatment of onchocerciasis with ivermectin in Takum, Nigeria. Tropical Medicine ond Internattonal Health 6,232-243. AMAZIGO, U. V., OBONO, O. M., DADZIE, K. y., REMME, J., JIYA, J., NDYOMUGYENYI, R., ROUNGOU, J-8., NOMA, M. & SEKETELI, a. (2001b). Monitoring community-directed treatment programmes for sustainability: lessons learned. Annals of Tropical Medicine and P aras ito lo gt, ?95, (Suppl.) ??? -??? 27 AMAZIGO, U., NOMA, M., BOATIN, B.A., ETYA'ALE, D.8,, SEKETELI, A. &Ol,OZte, K. y. (1998). Delivery systems and cost recovery in Mectizan treatment for onchocerciasis. Annals of Tropical Medicine and Parasitolology,92 (Suppl. l), S23-S31. AMAZIGO, u. (1999). Community selection of ivermectin distributors. Community Eye Health,12, 39-40. ARYEETEY, M. E., AHOLU, C., WAGATSUMA, y., BENTIL. G., NKRUMAH, F. K, & rcOJtUe, S. (1999). Health education and community participation in the control of urinary schistosomiasis. East African Medical Journal, 7 6, 324-329. BOATIN, B. A., HOUGARD, J. M., ALLEY, E. S., AKPOBOUA, L. K., YAMEOGO, L., DEMBELE, N., SEKETELI, A. & DADZIE, K. Y. (1998). The impact of Mectizan on the transmission of onchocerciasis. Annals of Troptcal Medicine and Parasttologt, g2, (Suppl.l), 546-560. BOSASART, r. (1990). Can they get along without us? Sustainability of donor-supported health projects in Central America and Africa. Social Science and Medicine,30r 1015-1023. BOUSSINESQ, M., GARDON, J., GARDON-WENDEL, N., KAMGNO, J., NGOUMOU, p. & CgtppnUX l.p. (1998). Three probable cases ofloa loa encephalopathy following ivermectin treatment for onchocerciasis. Annals of Tropical Medicine and Parasitologt, 58,461-469. CHIJIOKE, C. P. (2000) Factors affecting adverse event reporting during mass ivermectin treatment for onchocerciasis. Acta Tropica, 76, 169-173. CLARK, J. (1995). The State, Popular Participation, and the Voluntary Sector. World Development, 23, 593-60 1. CLEMMONS,L., AMAZIGO, U. V., BISSECK, A-C., MYSUYA-MPANJU, J., NOMA, M., OYENE, U., EKPO, U., KATENGA, S. & SEKETELI, A. (2001). Gender issues in community-directed treatment with ivermectin CDTI) of the African Programme for Onbchocerciasis Control (APOC). Annals of Tropical Medicine and Parasitology, ?95, (Suppl.) ??? -??? ESSoMBA, R. o., BRYANT, M. & sooeRr, C. (1993). The reorientation of primary health care in Cameroon: rationale, obstacles and constraints. Health Policy and Planning,8r232-239. 28 ETTYANG, G. (1999). A community-based health intervention programme in pastoral and agricultural Pokot communities in Western Kenya. Acta Paediatrica 88, (Suppl.),41-42. ETYA'ALE, D. E. (1998). Mectizan as a stimulus for development of novel partnerships: the international organization's perspective. Annals of Tropical Medicine and Parasitologt, 92, (Suppl. 1.), S73-S77. GYAPONG, M., GYAPONG, J. o. & owusu-SANAHENE, G. (2001) Community-directed treatment: the way forward to eliminating lymphatic filariasis as a public health problem in Ghana. Annals of Tropical Medicine and Parasitologt, g5rTT-86. HOMEIDA, M., BRAIDE, E., ELHASSAN, E., AMAZIGO, U. V., LIESE, B., BENTON, B., NOMA, M., ETYA'ALE,D.,DADZTE, K. y., KALE, O. O. & SEKETELI, e. (2001). community-directed treatment with ivermectin (CDTI) strategy of the African Programme for Onchocerciasis Control (APOC) and its potential for providing additional health services to the poorest populations. Annals of Tropical Medicine and Parasitologt, ?95, (Suppl. ),???-??? HOPKINS, e. o. (1998). Mectizan delivery systems and cost recovery in the Central African Republic. Annals of Tropical Medicine and Parasitologt, g2, (Suppl.1.), S97-S100. KATABARWA, M. N. & MUTABAZI, o. (1999). Community-directed, ivermectin-treatment programmes for onchocerciasis control in Uganda: the selection and validation of indicators for monitoring sustainability at the district level. Annals of Tropical Medicine and Parasitologt, 93, 653-658. KATABARWA, N. M., MUTABAZI, D. & RICHARDS, F. o. (2000a). Controlling onchocerciasis by community-directed, ivermectin-treatment programmes in Uganda: why do some communities succeed and others fail? Annals of Tropical Medicine and Parasitology, 94,343-352. KATABARWA, N, M., RICHARDS, F. O. & NDYOMUGYENYI, R. (2OOOb). IN rurAI UgANdAN communities the traditional kinship/clan system is vital to the success and sustainment of the African Programme for Onchocerciasis Control. Annals of Tropical Medicine and 29 Paras itology, 94, 485-495. KIDANE, G. & MoRRow, R. H. (2000). Teaching mothers to provide home treatment of malaria in Tigray, Ethiopia: a randomised trial. Lancet,356, 550-555. MILLS, A., FOX-RUSHBY, J., AIKINS, M., D'ALESSANDRO, U., CHAM, K. & GREENWOOD, n. (199a). Financing mechanisms for village activities in The Gambia and their implications for financing insecticide for bednet impregnation. Journal of Tropical Medicine and Hygiene, 97,325-332. MIRI, E. s. ( 1998). Problems and perspectives of managing an onchocerciasis control prograrnme: a case study from Plateau state, Nigeria. Annals of Tropical Medicine and Parasitologt, 92, (Suppl. 1.), S 121-128. NICHTER, u. (1997). Project diagnosis: participatory research as a first step toward community involvement in primary health care. In: Hahn RH (ed) Anthropology in Public Health, Bridging Differences in Culture and Society. New York: Oxford University Press, pp. 300-324. NyoNAToR , F . & KUTZIN, J. ( 1 999). Health for some? The effects of user fees in the Volta Region of Ghana. Health Policy and Planning,14,329-41. oLADEpo, o, BRTEGER, w. R. & orusaNvA, s. (1999-2000). Testing alternative record keeping formats not requiring literacy skills: onchocerciasis control at the village level. International Quarterly of Community H ealth Education, 19, 43-50. oNWUJEKWE, O. E, SHU, E. N., NWAGBO, D., AKPALA, C. O. & OTONrcWO, P. O. (1998). WillingneSS to pay for community-based ivermectin distribution: a study of three onchocerciasis- endemic communities in Nigeia. Tropical Medicine and International Health, 3, 802-808. oNwuJEKwE, o. E., sHU, E. N. & oKoNKwo, p. o. (1999). Willingness to pay for the maintenance of equity of a local ivermectin distribution scheme in Toro, northern Nigeria. Public Health,ll3, 193-194. 30 ONWUJEKWE, o. E., SHU, E. N. & oKoNKwo, p. o. (2000). Community financing of local ivermectin distribution in Nigeria: potential payment and cost-recovery outlook. Tropical Doctor, 30,,91,-94. PETERSON, S., ASSEY, V., FORSBERG, B. C., GREINER, T., KAVISHE, F. P., MDUMA, B., ROSLING, H., SANGA, A. B. & GEBRE-MEDHIN, M. (1999). Coverage and cost of iodized oil capsule distribution in Tanzania. Health Policy and Planning, 24,390-399. pLArsrER, A. p. vAN OORTMARSSEN, G. J., HABBEMA, J. D. F. & REMME, J. (1990). ONCHOSIM: a simulation model for onchocerciasis transmission and control. Acta Leidensia, 59,177- 191. RYAN, J. M., JOHN, G. C. & BRIEGER, w. R., (1990-91). Five-year knowledge retention by volunteer primary health workers in western Nigeria. International Quarterly of Community Health Education, ll, 123-133. WAGATSUMA, y., ARYEETEY, M. E., SACK, D.A., MORROW, R. H., HATZ,C.& KOJIMA, S. (1999). Resolution and resurgence of Schistosoma haematobium-induced pathology after community-based chemotherapy in Ghana, as detected by ultrasound. Journal of Infectious Dis eas es, 179, I 5 L 5 -l 522. WORLD HEALTH ORGANISATION (1996a). Community Directed Treatment with lvermectin: Report of a multi-country study. TDR/AFR/RI196.l. Geneva: WHO. WORLD HEALTH ORGANISATION (1996b). African Programme for Onchocerciasis Control (APOC): Programme Documentfor Phase 1. Ouagadougou: APOC. woRLD HEALTH ORGANISATION ( 1 998) Community-Directed Treatment with Ivermectin (CDTI) : a Practical Guidefor Trainers of Community-Directed Dtstributors. wHo/Apoc/rrac/98.5. Ouagadougou: APOC. WORLD HEALTH ORGANISATION (20004) Community Directed Treatruent of Lymphatic Filariasis in Africa TDIJIDE/Rr/cott/}}.2. Geneva: WHO. woRLD HEALTH ORGANISATIoN (2000b). Technical Consultative Committee - Effectiveness and 3l Challenges. Paper preparedfor the APOC Partners Meeting,Ouagadougou,22-27 May 2000.Ouagadougou: APOC. WORLD HEALTH ORGANISATTON (2000c) Programme Implementation - Operationalizatton of APOC CDTI Strategt, a background document preparedfor the May 2000 meeting of APOC s Technical Consultative Committee. Ouagadougou: APOC. WORLD HEALTH ORGANISATION (2000d) . Report of the Mid-term (Phase I) External Evaluation of AP OC. wHo/APoc/cs/00.3. WORLD HEALTH ORGANISATToN (2000e). Final Report on the Sustainability of Community Directed Treatment of Onc hocerciasis with lvermectin. WHO/APOC/CS/00. I . WORLD HEALTH ORGANISATION (2000, Implementation and Sustainability of Community Directed Treatment of Onchocerciasis with lvermectin: Report of a Multi-Country Study. rDR/IDE/Rp/cor/00. 1. Geneva: WHO. woRLD HEALTH ORGANISATION (2000g). Reporting wtth Pictures: Community-Directed Implementation of Health Projects. TDR/IDE/Rp/00.1. Geneva: WHO. 32 I Fig. 1: Trends in Coverage in Nine CDTI Districts of Uganda and Cross River State, Nigeria 70 60 50 EooI 330 20 10 1996 1997 1 998 Year 1999 2000 Pne-e DTI ADTA 0 JJ @a @ap ffi Fig.2: The CDTI Budgeting Process: Contribution of APOC and Others 100 80 60 40 20 Y1 Y2 Y3 Y4 Projeet Year Otkers APOC 0 Y5 Y6 34 APPENDIX A A FRAMEWORK FOR ANALYSING THE CHALLENGES TO CDTI Category Challenge Source Managerial Developing drug collection mechanisms including issues of delay and wastage 234 lntegrating with health services including use of CDTI strategy with other health issues and sharing of existing infrastructure and personnel for CDTI 123 4 Skengthening internal infrastructure at local, district and national levels to manage CDTI 24 Technical Achieving and maintaining 65% coverage within villages and reaching all villages 34 Addressing individual and collective refusals due to perceptions of adverse reactions 34 Conducting conect census and using standard formula for calculating coverage 4 Establishing system-wide commitment to and performance of self-monitoring and utilizing the results to improve programmes, especially monitoring of progress by communities themselves 234 Designing and conducting operational research that addresses real problems and applying the results in programming. 24 Social-Political lncreasing the adequacy and appropriateness of CDDs, preventing attrition 24 Determining and providing appropriate lncentives and motivation for CDDs 14 lnvolving women as CDDs and in community decision making 4 lncreasing involvement and building capacity of local NGOs 4 Achieving financial Sustainability after APOC funds and planning for this 12 Deciding on an appropriate balance between cost-recovery strategies vs guaranteeing affordability at all levels, reducing costs 134 Developing advocacy to gain continued support of political leaders 34 r. wHo (2000b) 2. WHO (2000e) 3. wHo(2000c) 4. WHO (20004 35 GENDER ISSUES IN COMMUNITY-DIRECTED TREATMENT WITH IVERMECTIN (cou) oF THE AFRICAN PROGRA MME FOR ONCHOCERCIASIS coNrRor (npoc) GENDER ISSUES IN COMMUNITY.DIRECTED TREATMENT WITH IVERMECTIN (CDTI) OF THE AFRICAN PROGRAMME FOR ONCHOCERCTASTS CONTROL (APOC) AUTHORS: L. Ct-BuuoNS,l U. v. AuazIco,2 A-C. Btssm,3 J. Mswe-MpANJU,o M. Nove,2 U. oynue,s U. Exro,6 S. KarpNcA,T AND A. SxErplr.2 ADDRESSES OF AUTHORS Department of Anthropology, University of Pennsylvania, Philadelphia, PA, USA African Programme for Onchocerciasis Control (APOC), Ouagadougou, Burkina Faso Department of Dermatology, University of Yaounde, Cameroon The World Bank, Washington D. C,20433, USA Onchocerciasis Control Programme Unit, Ministry of Health, Calabar, Nigeria Department of Biological Sciences, University of Agriculture, Abeokuta, Ogun State, Nigeria National Onchocerciasis Control Programme, Ministry of Health, Dar es Salaam, Tanzania Address for Correspondence Dr. Uche V. Amazigo African Programme for Onchocerciasis Control (APOC),01 B. P. 549 Ouagadougou 01, Burkina Faso Ter: (226) 34.29.59; (226) 34.29.s3 Faxz (226) 34.28.75 E-mail: amazrgou ho.omsbf 1 2 3 4 5 6 7 Short running title: Gender issues in CDTI of APOC ABSTRACT This paper reviews issues on compliance and participation by men and women in three of the countries (Cameroon, Nigeria and Tanzania) of the African Programme for the Control of Onchocerciasis (APOC). Project monitoring data from 109 focus group discussions (FGDs), 6069 household survey respondents and 89 interviews with ivermectin distributors were analyzed to gain insight into the attitudes and behaviors of men and women in relation to ivermectin treatment, and their participation in the programme. Although there are no statistically significant gender differences in coverage rates for ivermectin treatment, culturally prescribed gender relationships influence the ways in which men and women express and experience treatment-related behaviors. Gender roles also affect participation in the programme. Decision- making in communities on the selection of distributors tends to follow socio-cultural hierarchies based upon patriarchy and gerontocracy. Few ivermectin distributors (2I%) are women. Although they receive less support than their male counterparts, female distributors are just as willing to continue ivermectin distribution in the community and perform as well or better than men in this regard. The study concludes that the use of the terms "community-directed", "community participation", and even "compliance" obfuscates important gender differences that are inherent in the implementation of onchocerciasis control. Development of strategies that recognize these gender differences will have important implications for long-term adherence to treatment and the overall quality and sustainability of the programme. 2 INTRODUCTION Onchocerciasis is a disease caused by a nematode worm, Onchocerca volvulus, transmitted to humans by the bite of the black fly of the genus Simulium. The disease causes serious eye lesions and blindness, disfiguring skin disease, and unbearable itching. Approximately 20 million people worldwide are infected with onchocerciasis, and 100 million people, most of who live in Africa, are estimated to be at risk of contracting the disease. The drug ivermectin is effective in killing the microfilariae produced by the adult worm. Microfilariae are responsible for most of the pathology and resultant morbidity associated with the disease. Their elimination by ivermectin therefore brings welcome relief from the distressing signs and symptoms of the disease. Since 1987, Mectizan@ (ivermectin MSD) has been provided by a United-States based private pharmaceutical company, Merck & Co. Inc., free of charge to all onchocerciasis-endemic countries world wide with a pledge to continue to do so for as long as it is needed. The ultimate goal of the African Programme for Onchocerciasis Control (APOC) is to eliminate onchocerciasis as a disease of public health and socio-economic importance in endemic African countries. APOC's strategy of Community-Directed Treatment with Ivermectin (CDTI) endeavors to attain and sustain treatment coverage rates that are necessary for effective control of the disease. It is intended to achieve sustainable ivermectin distribution and treatment for upwards of 15 years, long after the cessation of direct APOC support. Endemic communities that are eligible for CDTI take responsibility for the timely procurement of adequate supplies of ivermectin from the health facility nearest to them, for storing the drug under appropriate conditions, and for treating all eligible individuals. The scale of implementation of the CDTI strategy since the inception of APOC has been impressive. In 2000 alone, more than 70 million 3 tablets of Mectizan@ were distributed, nearly 20 million people were treated, and over 49,000 endemic communities participated in CDTI activities in l4 countries (WHO ,2000a). Because of their socially and culturally constructed roles and relationships, men and women have different opportunities and constraints related to their adherence to ivermectin treatment and to their overall participation in CDTI activities. The present paper discusses findings from quantitative and qualitative data related to the implementation process and outcome of CDTI activities in three countries: Cameroon, Nigeria and Tanzania. This study demonstrates how a gender-sensitive analysis of health attitudes and behaviors can reveal useful information with important implications for onchocerciasis and other community-based disease control programs. This information would be completely hidden in a gender-blind approach. Since our approach to the analysis of "gender issues" in onchocerciasis control activities is multi-disciplinary, drawing from on the disciplines of community development, health behavior, and anthropology, we will begin with a brief overview of key concepts from some of these fields to which reference would be made later when discussing the findings of our study. REVTEW OF KEY CONCEPTS WITHIN THE CONTEXT OF CDTI Models of community participation APOC's CDTI strategy is anchored in key concepts in primary health care (PHC) and community development. APOC's Executing Agency, the World Health Organization (WHO) launched the PHC movement in the late 1970's. WHO favors using terms like "community involvement in health (CIH)" to intentionally imply a more active as opposed to passive engagement by people in promoting their own health, and defines CIH as both "a deliberate strategy" and a "basic right" (WHO, 1999). The PHC philosophy of actively involving communities is consequently a cornerstone of the CDTI strategy. 4 The concept of "community ownership" as the ultimate level of participation in the CDTI strategy has its origins in Arnstein's "ladder of citizen participation" (1969) whose eight rungs represent increasingly higher levels of community involvement in development projects. The lowest rung on the ladder represents the most passive level of participation, which Arnstein calls "manipulation" (where communities have little information, little or no authority, and are essentially coerced into participating in a development project imposed upon them by outsiders). The highest rung on the ladder represents the most active level of participation, which Arnstein calls "citizen control" (where communities have complete authority, power and influence in planning, managing and implementing development projects). Community participation is therefore not a static element; instead it is more a continuum of degrees of empowerment. Negotiating through an inevitable degree of tension or conflict as power shifts away from government health services and non-government development organizations (NGDOs) and towards endemic communities is consequently an inherent part of the CDTI process. C o mp liance vers us adh erenc e Whereas APOC's main focus is to establish sustainable systems of ivermectin distribution, the ivermectin treatment coverage rate is nevertheless an important indicator by which to measure progress, impact, and even "success" (WHO, 1996; Katabarwa et al., 2000). To eliminate onchocerciasis as a public health problem in endemic communities, ivermectin treatment must continue for at least 15 years, with a stable annual treatment coverage rate of at least 650/o of the total population. Long-term compliance with ivermectin treatment at the desirable level has therefore been a topic of considerable discussion at all levels of the programme. In medical terms, compliance is defined as "the extent to which a person's behavior (in terms of taking medications, following diets, or executing lifestyle changes) coincides with medical or health advice" (Haynes et al., 1979). In recent years, however, health behavior specialists have debated the authoritarian and dictatorial tone underlying the concept of patient 5 "compliance" (Akpala et al., 1993). Many are now promoting the concept of "adherence"; rather than "compliance" (Donovan and Blake, 1992;' Cameron, 1996; Playle and Keeley, 1998; Abraham et al., 1999; Kyngas et a1.,2000). Promoting the concept of "adherence" to treatment represents a shift in the patient- practitioner relationship, away from a "follow-the-doctor's-orders" model and towards a paradigm of partnership, with greater dialogue, information-sharing, and joint decision-making. Patients who have been empowered to make an educated, determined and free-willed decision to follow treatment are more likely to do so (Haynes et al., 1979; Donovan and Blake, 1992; Cameron, 1996). Thus, beyond being an ideological objective, the concept of "adherence" also implies better and more sustainable outcomes from treatment (Shu et a1.,2000). The "compliance" versus "adherence" debate is therefore extremely relevant to APOC, since the Programme's own CDTI strategy is based upon the same underlying goals of partnership and empowerment. Like "community-directed" treatment with ivermectin (CDTI), "adherence" to treatment represents the highest, most active level on the continuum of participation; while "compliance", like "communit5r-based" ivermectin treatment (CBIT), represents the lowest, most passive level of participation. The Health Belief Model The Health Belief Model states that people's health-related behaviors are influenced by how susceptible they believe they are to contracting an illness or its symptoms, their perceived severity of the consequences of the illness, how beneficial they believe the treatment would be, and how high they perceive the barriers of treatment or behavior change to be (Becker et al., 1977). The greater the perceived susceptibility to disease, the more likely people are to adopt behaviors to prevent their contracting the disease, and the greater the perceived benefits of a particular health behavior, the more likely people are to adopt that behavior. Similarly, the lower 6 the barriers associated with a particular health behavior are, the more likely people will be to adopt that behavior. Other authors have indirectly illustrated the validity of the Health Belief Model to disease control by documenting the strong direct associations between people's perceived susceptibility to onchocerciasis, the perceived severity of the consequences of the disease, the perceived benef,rts of the drug ivermectin, and treatment compliance (Akogun et al., 2000) and demonstrating associations between perceived susceptibility to malaria, perceived benefits of malaria prophylaxis, and adherence to prophylaxis regimens (Abraham et al., 1999). Promoting gender-sensitive health research According to Courtenay (2000), gender is more easily understood as a verb than as a noun, for it is a dynamic interaction between people whose roles change over their life spans and their life circumstances. Quoting other social scientists, Courtenay notes that "gender is something that one does, and does recurrently in interaction with others" Gender bias, particularly through the exclusive study of males as the standard for behaviors, knowledge, values and perspectives, and health status has been pointed out and criticized as a frequent shortcoming in anthropology (Muckhopadhyay and Higgins, 1988) as well as in health research (Puentes-Markides, 1996; Vlassoff and Manderson, 1998; Bird and Rieker, 1999; Courtenay, 2000). Gender bias also occurs when researchers exclusively study women in certain specific areas, such as reproductive health and family planning. Total fertility rates (TFR), for example, are automatically interpreted as women 's fertility rates; very few research studies have attempted to study men's fertility. Bird and Rieker (1999) emphasize the usefulness of research that comparatively studies both men's and women's health and urge investigating reasons for gender-related similarities and differences. Likewise, it has urged that disease control programmes "rethink the rather vague concept of 'community participation' and recognize that 7 communities are composed of women and men who relate to each other in ways that impact on their understanding of health, their health-related behavior, and ultimately their health". Most of the studies on onchocerciasis control in Africa have used research designs and data analysis that focus on information provided by male respondents. Findings from such gender- biased research lead to conclusions that are gender-blind because they hide women's perspectives and roles related to control activities under information and opinions provided by men. RELEVANCE TO THE CDTI APPROACH Community-based ivermectin treatment (CBIT) is a process wherein health providers determine the steps and the schedule to be followed. Activities are based in the community but not owned by the community. CDTI, on the other hand, is a process built on the experience of community members and thus enhances decision-making and problem solving capacity. In the CBTI approach, the community does not exercise authority over decisions on project design and implementation. Project activities, such as treatment dates and procedures, are designed outside the community without any, or with limited, contribution by the community. In the CDTI approach, the community exercises authority over decisions. Therefore, the community plans the distribution, decides on the method of distribution acceptable to them (e.g. central place, house- to-house) and who and when to distribute ivermectin. (WHO, 1996). Over the years, a number of research studies have helped APOC partners to become increasingly aware of the need for a gender-sensitive approach in the program, particularly in terms of encouraging active participation and decision-making in CDTI (Amazigo and Obikeze, 1991; Amazigo, 1994; Chippaux et al., 1995; Brieger et al., 1997; 1998; Vlassoff and Manderson, 1998; Katabarwa et al., 2001; Vlassoff et al., 2000). This "gender awareness" is gradually beginning to translate into action at the national level in a few APOC -funded projects. This is most notable in Uganda, where an operational research study on women's involvement in 8 CDTI yielded some important and intriguing findings. (Katabarwa et al., 2001). Within the context of APOC, a gender-sensitive approach implies taking the socially and culturally prescribed roles and relationships of both men and women into account when designing research studies, developing programme objectives and educational materials, and above all, when implementing the CDTI strategy. CDT['s underlying philosophy of community capacity-building and empowerment derives from the three classic models of participation: locality development, social planning, and social action (Rothman, 1979). The essential components of the three combined models are: information-sharing, consensus-building, problem-solving, equity, and occasional confrontation. There is more therefore to the polemic over "community-based" versus "community-directed" and the issues of "compliance" versus "adherence", as these concepts have direct bearing on programme impact and sustainability. Although the study of gender roles and relationships inevitably touches upon issues of power, anthropologists have argued that to frame the concept of gender solely in terms of male domination and female subordination is inappropriate, ethnocentric and even naive since status, authority and power are experienced differently across cultures (Gerson and Peiss, 1985; Muckhopadhyay and Higgins, 1988; Potash, 1989, Foster, 1999;Kar et al., 1999; Lennie, 1999). Three recent studies on the onchocerciasis control program in Uganda (Mutabazi and Duke, 1998; Katabarwa and Mutabazi, 1998; Katabarwa et al., 2000a) measured "community participation" by exclusively asking the heads of household whether they had been involved in specific activities. While the studies validate the importance of involving heads of households (mostly men) in decision making, attending health education sessions, and supporting ivermectin distributors, they yield absolutely no information about the level of women's involvement in the programme nor do they provide any insight into how women's participation in these activities could influence treatment coverage rates and sustainability. 9 In another example, a study in Nigeria assessed the impact of schoolchildren's health club on adults' knowledge about onchocerciasis and their compliance to ivermectin treatment (Shu el al,2000). The research population was, once again, almost exclusively restricted to men, and no distinction was made between the sexes in the data analysis and discussion of the findings. In all four studies, the use of the words "respondents", "community", "village meetings" and so on masks possible differences in men's and women's access to and level of information, their compliance to treatment, and their participation in decision-making about onchocerciasis control activities in their communities. In reality, the data, discussion and conclusions drawn in these studies are almost exclusively limited to men 's knowledge, compliance and participation. This gender-biased approach represents missed opportunities to obtain useful gender-specific information about the impact, outcomes, and sustainability of onchocerciasis control activities. METHODOLOGY The present study is based on an analysis of data generated from ten of the 39 projects monitored by APOC between 1998 and 2000. The ten projects, located in Cameroon, Nigeria and Tanzania, had completed their first year of implementation of the CDTI strategy. The seven CDTI projects in Nigeria are located in the states of Abia, Adamawa, Delta, Enugu, Imo, Osun and Oyo. In Tanzania, the study included one CDTI project in Ruvuma province, which covers three administrative districts: Songea, Mbina and Ludewa. The two CDTI projects in Cameroon are located in Littoral 2 and Adamaoua provinces. A total of 300 villages (thirty villages in each of the ten projects) were randomly selected during project monitoring to ensure a representative sample of communities, based upon social structure and geographical positions as identified during earlier rapid epidemiological mapping of onchocerciasis (REMO). All villages are in areas of hyper- or meso-endemic onchocerciasis. A mix of quantitative and qualitative research methods used standardized data collection instruments. In-depth interviews were held with 84 men and women CDDs (Table 1), while 109 l0 focus group discussions (FGDs) were conducted with men and women living in the communities covered by the projects (Table 2). There were 6069 respondents in the household survey in the three countries. Of these 3148 (51.9%) were femaleand292l (48.1%) were male (Table 3). The age distribution of the household survey participants is shown in Table 4. Quantitative data entry was performed using EPI-NFO software, while data analysis was performed using SPSS (version 8.0). Qualitative data was manually coded and analyzed (without computer software). FINDINGS AND DISCUSSION Gender issues on access to and levels of information Both men and women understood that height is used to determine the appropriate dosage of ivermectin, and both genders recognized the criteria for excluding people who are not eligible for treatment (e.g. under five years of age, pregnant, extremely ill). They had similar levels of information about the benefits of ivermectin treatment. Men, however, had more detailed information than women about the disease, communify roles and responsibilities in the CDTI approach, the decision-making process and the duration of ivermectin treatment. They apparently obtained this information through meetings variously described, by APOC partners, researchers, and others as "community meetings" or "village meetings". However FGDs with women revealed that community members often perceive these meetings as being exclusively for men. This perception is cultivated and underscored by the fact that most of these meetings are initiated and organized by health personnel, CDTI project staff and community leaders, who are almost always men. "The chief and his men held the meeting but we know nothing up till now." [Woman FGD participant, Osun State, Nigeria] 6(Women are not invited to meetings." [Woman FGD participant, Delta State, Nigeria] 1l Most women learned about onchocerciasis control activities and CDTI indirectly, often just before or during the distribution of ivermectin, and mostly by word-of-mouth. In Nigeria, women frequently identified the town crier as their first and major source of information about the programme. In Imo State, Nigeria, where women were exceptionally well informed about onchocerciasis and CDTI, women mentioned that a variety of public forums- including churches, schools, and village meetings attended by both sexes- had been used to disseminate information. Gender issues in decision-making This is closely linked to access to information. A Chi squared test of quantitative findings from in-depth interviews with CDDs in the three countries indicate that decision-making at village meetings has a positive effect on treatment coverage rates (12 : 16.848, p<0.001). The Chi squared test also shows that treatment rates are better when CDDs are selected at village health committee meetings (Ftest: 12.098, p<0.05). These findings are corroborated by a study in Uganda that shows that community participation in decision-making has positive influences on treatment coverage rates (Katabarwa et al., 2001). Qualitative analysis of the data however uncovers important gender issues that are masked by the quantitative findings. FGDs revealed that the real decision-makers are men, especially village chiefs and elders. Women and young people are routinely excluded from the decision-making process, either because they are not informed about the meetings or because they are not invited to attend. As noted earlier in this paper, "village assemblies" or "community meetings" are acfually misnomers for what are essentially men's meetings. A few quotes from women's FGDs in Delta State, Nigeria illustrate the general finding that women seldom participate in meetings at which decisions about onchocerciasis control activities in the community are taken: "We are women. We do not go to men's meetings. Men do not attend women's. But in the women's group there was no such thing (decision-making meeting)." t2 t'Women are not invited to meetings. I don't know whether decisions were takcn on when the drug will be distributed." "We were not at the meeting where CDDs were chosen. Women do not attend." Interestingly, in all 75 FGDs with women, there was only one village where women expressed dissatisfaction with the decisions made by men. In all other villages, women either approved of the men's decisions, or had no opinion. The one striking exception was the village of Ga'anda, located in Adamawa State, Nigeria. Women in this village complained bitterly about the choice and performance of distributors as well as the mode and period of distribution. 6'He (the CDD) does not go to all the houses. I did not get the drug because he did not come to our house and I did not even know the drug was available. We are not happy with his work.' '6We do not like the time they bring the drug. We will prefer taking the drug in the dry season. After taking the drug it is usually too uncomfortable for farming because the drug is hot and causes itching. It is better in the dry season when we just take it and rest.' In stark contrast to women's reactions, men in Ga'anda reported being satisfied with the distributors and the distribution process. In spite of the marked differences in men's and women's perceptions of CDTI process, there were no statistically significant differences in men's and women's treatment rates, nor in their rates of refusals or absenteeism. Women in Ga'anda village expressed attitudes that were far removed from the passive resignation and acquiescence normally associated with culturally prescribed gender roles. Their frustration fueled concrete suggestions and very clear ideas about how women could and should be involved in the decision-making process. In addition to replacing some of the CDDs and supervising their work, women suggested switching from the house-to-house mode of distribution to distribution at a centralized location. They also recommended increasing drug supplies to avoid shortages, and proposed a specific period for the distribution. We present here two more quotations from the FGDs to illustrate these points: "There are a lot of problems. The CDDs setect the people they treat. We only hear that treatment was done, we never get the tablets. We should be the ones to tell them when to come so that they can treat us, but they just sent a message that we should wait for them." "The CDDs should be probed. Let us know what they do with the drugs they collect, how many households treated. Let us see the names ofthose treated." 13 Gender issues in treatment-related attitudes and behaviors: women comply, men adhere Quantitative analysis of the data from household surveys conducted in Cameroon, Nigeria and Tanzania revealed no statistically significant gender differences in ivermectin treatment rates among the total eligible population in the three countries. (Figure l). Although their ivermectin treatment rates are similar, men's and women's underlying attitudes and behaviors related to treatment are nevertheless quite different. Women FGD participants frequently described their heatment behavior in terms of passive acquiescence or as simply obeying instructions (e.g."I was given the drug", "We were invited to take the drug", or "V[e were told to take the drug"). Men, on the other hand, rarely used these terms; instead they tended to express their treatment behavior as an active act of autonomous volition (e.g., "I took the drug"). Our findings show how women's and men's underlying treatment-related attitudes and behaviors demonstrate underlying gender roles and relationships prescribed by culture: women comply, men adhere- or, women do as they are told, men do as they wish. Although these gender-specific differences have no overt impact on treatment patterns, given that, in the absence of a suitable macrofilaricide, endemic communities need to take ivermectin for at least 15 years to eliminate onchocerciasis in their villages, there could be many important implications over the long-term. Within the context of CDTI, men are usually the ones who give instructions concerning treatment at the community and family levels. Women's FGD revealed the potential health hazards inherent in women's unquestioning obedience to men's instructions. Most women reported not knowing how the CDDs in their communities had been selected; some even reported not knowing the man who showed up at their home to distribute tablets (which the women assumed were ivermectin tablets). Yet in spite of lacking such important, fundamental information about the distributors and about the drug itself, women obligingly took the tablets l4 they were "invited" or "told" to take. Such gender-specific attitudes and behaviors in relation to treatment and compliance are illustrated by the following quotations taken from women's FGDs in Adamawa State, Nigeria: "They just bring it and we swallow it." "We do not know how it (the drug) was brought. Ours (responsibilify) is to swallow." "\ile were not the persons that selected the person that is distributing the drug." "A man was sent to us to give us the drug in our houses. He was the one that told us that the drug is for preventing blindness." Such unquestioning compliance, which is predominant among women is of major concern considering that many unlicensed drug vendors roam through many villages in Africa selling medicines that are often expired, improperly stored, and inappropriately administered. Although ivermectin is donated free of charge, a cost-recovery system to cover expenses related to the storage, transportation and distribution of the drug exists in Cameroon and Chad and the Adamawa CDTI Project in Nigeria. Women who do not have adequate information may not be able to distinguish between authorized community distributors and fake ones. D is s eminat io n of Info rmation Because women are less informed than men about the nature and distribution of the drug, they are more susceptible to the dangers of unintended misuse of ivermectin, such as accepting medicines that are not ivermectin to treat onchocerciasis, taking incorrect dosages of ivermectin, or taking ivermectin when there are contraindications. One disturbing example of this phenomenon was that of a young woman, quoted below, who reported taking yellow tablets whereas ivermectin tablets are white thus demonstrating a socially sanctioned "feminine" attitude and behavior through her unquestioning acquiescence: "We don't know anything about ivermectin treatment. There was no meeting held in our village discussing the treatment. We were only given some yellow tablets last year to take." Gender related perceptions t5 In accordance with the Health Belief Model (Becker et al., 1977; Abraham et al., 1999) women's compliance with and men's adherence to taking ivermectin is closely linked to their perceived susceptibility to onchocerciasis and its symptoms as well as to their perceived severity of the disease and the anticipated benefits of the drug. Although overall, both men and women in all three countries reported similar benefits of treatment, particularly prevention of blindness, improvement in vision, and the expulsion of intestinal worms there were gender differences in the degrees of perceived susceptibility, severity and benefits. Women perceived themselves to be very susceptible to the symptoms of onchocerciasis- particularly blindness and other visual problems, and to skin lesions, and occasionally fertility. "There are many (side effects). But for me, it (ivermectin) cleared my eyes." [Woman FGD participant, Delta State Nigeria] "When the drug is used, it works. Some people took it and excreted worms. Others felt healthy. I haven't used it because I'm pregnant but others used and claimed it is good. I am looking forward to using my own in the next round." [Woman FGD participant Osun State Nigeria] "'We were told that the skin would be fresh if we use the drug. That the drug takes care of skin infections." [Woman FGD participant, Oyo State Nigeria] Men appeared less preoccupied by these symptoms, tending to refer more to the disease by its name (e.g. "river blindness" or "onchocerciasis"), rather than to the symptoms. They also rarely or never mentioned benefits related to the appearance of the skin. "The final goal is to avoid being blind." [Man FGD participant, Littoral, Cameroon] "We take Mectizan every year to kill the parasites." [Man FGD participant, Littoral Cameroon] These gender differences in the degree of perceived susceptibility, severity and benefits have been recorded in other studies. Women were found to be more concerned than men about the deleterious consequences of onchocerciasis such as on social acceptance, skin appearance, marriage prospects and breastfeeding (Amazigo and Obikeze, 1991; Amazigo, 1994;Brieger et al., 1997 1998; Akogln et a\.,2000; Vlassoff el a1.,2000). These more strongly felt perceptions exert a strong influence on women's compliance to ivermectin treatment. Consequently, women's agency in taking ivermectin to obtain the benefits of the drug works in combination with their t6 culturally prescribed behaviors of acquiescence and obedience. Further research in this area would help to provide more insight into the degrees to which agency and passive obedience influence women's treatment behaviors. Men are also subject to culturally prescribed attitudes and behaviors that can effect their compliance to treatment and their health. The best example of this is found in Cameroon, where serious adverse reactions (SARs), including a few resulting in coma and death following treatment with ivermectin, have been reported in areas where onchocerciasis and loiasis are co- endemic. Although SARs are rare, fear and rumors combined with insufficient public information have had a negative effect on treatment rates in Cameroon which has the lowest treatment rates in all 14 countries currently participating in APOC. (Amazigo et a1.,2001). Notably Cameroon also has the highest rates of refusals and absenteeism. Figure 2 shows that among the eligibles there is a substantial proportion of refusals in Cameroon (48.5%) compared to Nigeria (18.5%) and Tanzania(32.9%). It also shows that only in Cameroon was the refusal rate higher in men than women. Poor communication for both sexes featured prominently in NigeriaQa.9%) than in the other two countries (0.6% in Cameroon, and 5.5Yo inTanzania). Almost four out of every ten men (38.4%) eligible for treatment in Tanzania were absent. Our findings indicate that most women receiving ivermectin treatment stand on the lowest rung (Compliance), of the adapted Arnstein "ladder of participation" (Figure 3), while most men stand on the next higher rung (Adherence). Senior men in the community (particularly village chiefs and elders) are on the middle-rung (Decision-making). The number of ivermectin distributors is obviously limited, and men greatly outnumber women on this the fourth rung (Ivermectin Distribution). The study found indications of great potential for ownership in women's eagerness for additional information, in the concrete suggestions they offered when asked for their opinions, and in both men's and women's overall t7 positive perceptions of the benefits of ivermectin. Nevertheless, very few community members indeed are found on the highest rung of participation (Ownership). Most of these are senior men. During FGDs in the Littoral 2 province of Cameroon, men freely voiced their skepticism and suspicions regarding the drug. Men were particularly suspicious about the change in the size of the tablet from 6 mg to 3 -9, the fact that a medical examination was not required prior to taking the drug, and the fact that onchocerciasis control was being attempted through oral treatment rather than through vector elimination. "I was reluctant to take the product because I did not have enough information". [Man FGD participant, Littoral province Cameroon] Men's perceived treatment risks in Littoral province were thus greater than women's. The Health Belief Model would lead us to speculate that men would therefore have lower rates of compliance to ivermectin treatment. Yet surprisingly, the quantitative data from the household survey show no significant gender differences in rates of refusal and absenteeism during the distribution of ivermectin in these communities nor indeed in Nigeria orTanzania either. (Figure 2) (Amazigo et al., 2001). Further research on reasons for similarities and differences in men's and women's behaviors and attitudes to ivermectin treatment, especially in Cameroon where there are special problems in achieving adequate levels of adherence, would be extremely helpful in identifying strategies to increase and sustain treatment coverage rates in the onchocerciasis control program. Gender r'sszes in willingness to payfor treatment In Cameroon and Chad and in Adamawa state in Nigeria, the CDTI projects have instituted a cost-recovery system in order to assure the sustainability of ivermectin distribution and delivery after financial support from the APOC Trust Fund ceases. The project monitoring data from Adamawa State, Nigeria reveals that men and women have very different perceptions about the amount of the financial contribution required in this state which is 20 naira (: US 15 cents). Men bitterly complained about the 20 naira contribution, and specifically critrcized (i) the amount, (ii) 18 the fact that this amount had been imposed upon them from authorities outside of the community, and (iii) the fact that failure to pay meant that treatment would be withheld. All of these criticisms can be interpreted as perceived barriers associated with treatment. "The only problem is that community members are taxed 20 naira per household and people do object to it and as such there is low coverage." [Man FGD participant, Adamawa State, Nigeria] Women, on the other hand, were resigned to paying the imposed fee, and rarely complained about or criticized the amount. Although the same issues related to women's culturally prescribed behaviors of obedience could resurface here, the qualitative findings indicate that the reasons for women accepting to pay the fee without complaint are more closely linked to the Health Belief Model (Becker et al., 1977). FGDs show that women in Adamawa State perceive the benefits of treatment to greatly outweigh the cost: "The 20 naira we give does not matter so long as they just give us the drug before we go blind." "Each household pays 20 naira for the drug. Since it has to do with eyes, we always have 20 naira". By contrast, notwithstanding the perceived benefits of treatment, the degree of menrs perceived susceptibility to the disease seem not to be strong enough to justii/ payment of the prescribed fee hence their more strident complaints on the matter. There are numerous variations in gender roles and responsibilities concerning financial management depending on the cost, the context, and the purpose of a purchase, as well as the household variations in hierarchy, status, and influence (Potash, 1989; Rashed et al., 1999). In a research study on determinants of the use of permethrin impregnated bed nets (PIBs) for malaria control in Benin, Rashed et al., (1999) demonstrated how heads of household do not always determine how funds available within the family will be used. Men, as the usual household heads, set priorities for the use of family income, yet because bed nets did not typically fall within men's priorities, women were the ones who eventually resorted to using their personal funds to purchase the PIBs. Similar research within onchocerciasis control programs should explore gender- l9 specific issues related to the provision of both financial and in-kind incentives and other forms of compensation or support to CDDs at the level of the community. The example of gender differences in Adamawa State demonstrates how willingness to pay may vary by men's and women's perceived susceptibility to the disease, and also by gender roles and responsibilities in financial management in households. The absence of gender differences as an important issue in some onchocerciasis studies on willingness to pay (Onwujekwe et a1.,1998), is probably due to the study design, which used heads of households, mostly males, as respondents, and which utilized quantitative data collection methods to the exclusion of qualitative ones. Future studies on willingness to pay in CDTI projects should therefore incorporate gender-specifi city in their design. Gender issues in ivermectin distributors' performance Project monitoring teams interviewed a total of 84 CDDs in the three countries. Of these, only 25%o were women. Although the small sample size prevented conducting tests for statistical significance, we present here some of the more interesting quantitative findings that identiff areas that merit further research. Most of the CDDs reported having been selected during village meetings. Although data presented earlier in this paper reveals that the participants of these "village meetings" are mainly men, this type of decision-making process tends to be more favorable for the selection of women distributors than any of the other possible processes listed in the interview questionnaires (village elders' meeting, village chief, health worker, village health committee, village committee meeting, and "other"). About 620/o of women CDDs compared to 59oh of men CDDs were selected during village meetings. Women CDDs were much more likely than men to distribute ivermectin at a central place (42.9% compared to 24.60/o), while men CDDs were much more likely than women (24.6% 20 compared to 4.8oh) to use a mixed mode of distributing ivermectin (both house-to-house and centralized distribution). Out of the 84 CDDs interviewed 39 (46.4%) remembered all the topics covered during the training, with proportionally more women(52.4%) remembering the topics than men (44.4%). Type of support provided by community to CDD. Proportionally fewer women dishibutors than men distributors received community support in the form of financial or in-kind incentives or compensation for their work. More than half of women CDDs (55.6%) reported receiving no support from their communities, while less than one third of the CDD men (30.8%) said they received no support. Transportation support to collect and distribute ivermectin, was provided to 38.5% of men and 27.8%o of women. A higher proportion (38.5%) of men than women (16.7%) also received incentives in cash or kind. lYillingness to continue as a CDD Even though women CDDs received less support than their male counterparts the women (95.2%) were just as willing as the men (98.4oh) to continue their work as CDDs. Data from another study on the CDTI prograrnme in Uganda (Katabarwa et a1.,200I) indicates that in spite of receiving less support from their communities, women CDDs actually tend to perform better than male CDDs in terms of ivermectin treatment coverage rates. Gender, age and the power of ownership Qualitative data analysis indicates that most community members, male and female, do not feel in charge of the CDTI process. Responsibility for and decision-making about the selection of distributors, the timing and the mode of distribution tend to be controlled by male village elders (chiefs and their advisors). Young people, particularly young women, expressed the least degree of awareness and involvement in the CDTI prograrnme. These findings conform with cultural systems of patriarchy and gerontocracy found in many rural African societies. 2t Nevertheless, there were some exceptions to this rule that came to light during FGDs with women participating in the study. In Tanzania, a women's association had taken the initiative to collect money and purchase bicycles for their community ivermectin distributors, explaining that this had been their practical solution to assuring their access to the drug. An initiative such as this is a striking example of ownership (i.e. feeling responsible forthe success of the ivermectin distribution and treatment process, identifying problems and seeking solutions). The strongly expressed views and practical suggestions for improving ivermectin distribution and treatment by the women of Ga'anda village in Nigeria serves as another example of women's underlying agency and potential for ownership. CONCLUSIONS The main findings of this study underscore the value of a multidisciplinary research approach to providing insight into oft neglected issues like gender, which could have signif,rcant effect on the achievements of the objectives of APOC. Most of the quantitative data in this study failed to reveal important gender differences that became apparent only after qualitative data analysis. Although men's and women's treatment rates were quite similar, there were important gender-related differences in their reasons for taking ivermectin. These differences were closely associated with culturally constructed gender roles and relations. Women are generally excluded from the decision-making process in the CDTI projects. Although this exclusion has not been shown in this study to have negative effects on overall treatment coverage rates, the long-term sustainability of CDTI would clearly benefit from increasing the access of women to relevant information on CDTI processes. Our study also reveals the importance of developing effective strategies sensitive to men's concerns. For example there is a need to overcome the skepticism and suspicions regarding ivermectin treatment in areas co-endemic with loiasis in the Cameroon. Beyond communication strategies, special efforts are needed to prevent and cope with serious adverse reactions to 22 ivermectin. In this regard it is most reassuring to note that APOC's Technical Consultative Committee (TCC) has already developed a set of operational guidelines on the matter which National Onchocerciasis Control Programmes (NOCP), including and in particular that of the Cameroon, are already using, and that the TDR Task Force on Filariasis Intervention Research is actively pursuing relevant operational research on the subject. The quantitative data generated from interviews with CDDs reveal important gender differences related to CDDs' performance, including the support distributors are likely to receive from their communities. CDDs are the backbone of the CDTI strategy, therefore this important gender issue merits further investigation on a much wider scale. ACKNOWLEDGEMENTS We are indebted to the communities and their leaders for their cooperation and valuable responses, which form the basis of this study. The contributions of the National Onchocerciasis Task Forces (NOTFs) of Nigeria, Cameroon and Tanzania are gratefully acknowledged. We also thank Dr Elizabeth Elhassan and Professor Oladele Kale for their comments and suggestions on the manuscript. REFERENCES ABRAHAM, C., CLIFT, S. & cRABowsKI, r. (1999). Cognitive predictors of adherence to malaria prophylaxis regimens on return from a malarious region: a prospective study. Social Sctence and Medicine, 48, 164l -1654. AKocUN, o. B., AKOGUN, M. K. & AUDU,Z. (2000). Community-perceived benefits of ivermectin treatment in northeastern Nigeria. Social Science and Medicine, 50, l45l-1456. AK?ALA, C. oKONKwo, P. o., NWAGBO, D. & NwAKoBY, B. (1993). Comparison of three strategies for mass distribution of ivermectin in Achi, Nigeria. Annals of Tropical Medicine and Parasitologt, 87, 399 -402. 23 AMAZIGO, u. o. & OBIKEZE, D. s. (1991). Socio-cultural factors associated with prevalence and intensity of onchocerciasis and onchodermatitis among adolescent girls in rural Nigeria. (Unpublished WHO/TDR final proj ect report). AMAZIGO, u. o. (1994). Detrimental effects of onchocerciasis on marriage age and breast-feeding Tropical and T rop ical and G eo grap hical Medic ine, 46, 322-325 . AMAZIGO, U. V., OBONO, O. M., DADZIE, K. y., REMME, J., JIYA, J., NDYOMUGYENYI, R., ROUNGOU, J-B., NoMA, M. & SEKETELI, A. (2001). Monitoring community-directed programmes for sustainability: Lessons learned. Annals of Tropical Medicine and Parasitologt, ?95, (Suppl. ),???-??? ARNSTEIN, s. n. (1969). A ladder of citizen participation. Journal of the American Institute of Planners, 35,216-217 . BECKER, M.H., HAEFNER, D. P. & MAIMAN, r. e,. (1977). The health belief model of prediction of dietary compliance: a field experiment. Journal of Health and Social Behavior,l8, 348-366 BIRD, c. & zuEKER, r. (1999). Gender matters: an integrated model for understanding men's and women's health. Social Science and Medicine, 48,745-755. BRIEGER, w., OSOSANYA, o., KALE, o. OSHINAME, F. & oKE, G. (1997). Gender and ethnic differences in onchocercal skin disease in Oyo State, Nigeria. Tropical Medicine and International Health, 2, 529-543. BzuEGER, w.R., oSHINAME, F. o. & ososANYA, o. o. (1998). Stigma associated with onchocercal skin disease among those affected near the Ofiki and Oyan Rivers in Western Nigeria. Social Science and Medicine, 47,841-842. cAMERoN, C. (1996). Patient compliance: recognition of factors involved and suggestions for promoting compliance with therapeutic regimens. Journal of Advanced Nursing, 24,244- 250. 24 IcHrppAUX, J. p., GARDoN-wENDEL, N., ERNoULD, J. c. & cenooN, l. (1995). Comparison entre differentes methods de d6pistage des grossesses au cours de traitements par ivermectine d large 6chelle au Cameroun. Bulletin de la Societe Pathologie Exotique,88, 129-133. COURTENAY, w. (2000). Constructions of masculinity and their influence on men's well being: a theory of gender and health. Social Science and Medicine,50,1385-1401. DoNovAN, J. & BLAKE, D. R. (1992). Patient non-compliance: deviance or reasoned decision- making? Social Science and Medicine, 34,507-13. FosrER,l. (1999). An invitation to dialogue: Clariffing the position of feminist gender theory in relation to sexual difference theory. Gender and Society,13,43l-456. GERsoN, J. & PEISS, r. (1985). Boundaries, Negotiation, Consciousness: Reconceptualizing Gender Relations. Social Problems, 32, 317 -331. HAYNES, B. R, TAYLOR w. R. & sncxrrr, D. L. (1979). Compliance in Health Care. Baltimore, MD: Johns Hopkins University Press. KATABARwA, M. N. & rr4urasAzl, D. (1998). The selection and validation of indicators for monitoring progress towards self-sustainment in community-directed, ivermectin-treatrnent progranrmes for onchocerciasis control in Uganda. Annals of Tropical Medicine and Parasitology, 92, 859-868. KATABARWA, N. M., MUTABAZI, D. & zuCHARDS, F. o. (2000a). Controlling onchocerciasis by community-directed, ivermectin-treatment programmes in Uganda: why do some communities succeed and others fall? Annals of Tropical Medicine & Parasitologt, 94, 343-352. KATABARWA, N. M., RTCHARDS, F. O. & NoyorvrucyENyt, R. (2000b). In rural Ugandan communities the traditional kinship/clan system is vital to the success and sustainment of the African Programme for Onchocerciasis Control. Annals of Tropical Medicine and Paras itology, 94, 485 -49 5. 25 KATABARWA, M., HABoMUGISHA, P., NDYOMUGYENYI, R. & ncfrNvo, S. (2001). Involvement of women in community-directed treatment with ivermectin for the control of onchocerciasis in Runkungiri District, Uganda: A knowledge, attitude and practice study. Annals of Tr op ic al M edicine and P aras it o lo gy, 9 5, 485 -49 4 . KYNGAS, H., DUFFY, M. & KRoLL, T. (2000). Conceptual analysis of compliance. Journal of Clinical Nursing, 9, 5-12. KAR, s., PASCUAL, c. & cHICKERING, K. (1999). Empowerment of women for health promotion: a meta-analysis. Social Science and Medicine, 49, I43 I -t460. LENNIE, J. (1999). Deconstructing gendered power relations in participatory planning: towards an empowering feminist framework of participation and action. Women's Studies International Forum,22,97-I12. MUKHoPADHYAY, c. & HIGGINS, r. (1988). Anthropological studies of women's status revisited: 1977 -1987 . Annual Review of Anthropologt, 17 , 461-95. MUTABAZI, D. & DUKE, B. o. L. (1998). Onchocerciasis control in Uganda: how can self- sustaining, community-based treatment with ivermectin be achieved? Annals of Tropical Medicine and Parasttology, 92, 195-203. oNWUJEKWE, O. 8., SHU, E. N., NWAGBO, D., AKPALA, C. O. & OKONKWO, r. O. (1998). Willingness to pay for community-based ivermectin distribution: A study of three onchocerciasis-endemic communities in Nigeria. Tropical Medicine and International Health,3, 802-808. pLAyLE, J. & KEELEv, p. (1998). Non-compliance and professional power. Journal of Advanced Nursing, 27,304-311. porASH, n. (1989). ln: Gender and Anthropology: Critical Reviews for Research and Teaching. ed. Morgen, S. pp.l89-227.Washington, DC USA: American Anthropological Association. I 26 PUENTES-MARKIDES, c. (1996). Women and Health Policy in Developing Countries. Social Science and Medicine, 42, 1345-1349. RASHED, S., JOHNSON, H., DONGIER, P., MOREAU, R., LEE, C., CREPEAU, R., LAMBERT, J., JEFREMOVAS, v. & SCHAFFER, c. (1999). Determinants of the Permethrin Impregnated Bednets (PIB) in the Republic of Benin: the role of women in the acquisition and utilization of PIBs. Social Science and Medicine,49,993-1005. ROTHMAN, l. (1979). In: Strategies of Community Organization. eds. Cox, F.M., Erlich, J.L., Rothman, J., Tropman, J.E. Itasca, Illinois: F.E. peacock. sHU, E. N., ONWUJEKWE, E. O., LOKrLr, p. & OKONKWO, p. O. (2000). A health club for a community school in South-eastern Nigeria: influence on adult perception of onchocerciasis and compliance with community-based ivermectin therapy. Tropical Medtcine and Int er natio na I H e al th, 5, 222-226. VLASSoFF, C., WEISS, M., OVUGA, E.B., ENEANYA, C., NWEL, P. T., BABALOLA, S. S., AWEDOBA, A. K., THEOPHILUS, 8., coFFIE, P. & SHETABI, P. (2000). Gender and the stigma of onchocercal skin disease in Africa. social science and Medicine,50,l353-136g. VLASSOFF, c. & nNonRSoN, L. (1998). Incorporating gender in the anthropology of infectious diseases. Tropical Medicine and International Health,3, r011-1019. woRLD HEALTH ORGANIZATIoN (1996). African Programme for Onchocerciasis Control. Programme Documentfor Phase l. (Unpublished document). woRLD HEALTH ORGANIZATIoN. (1999). Community involvement in health development: a review of the concept and practice. eds. Kahssay, H.M. and oakely, p. Geneva: wHo. woRLD HEALTH oRGANIZATIoN (20004). WHO Progress Report. (Unpublished document). woRLD HEALTH oRGANISATION (2000b). Implementation and Sustainability of Community- Directed Treatment of Onchocerciasis with lvermectin: a Multicountry Study. TDR/rDE/Rp /cott/00.1 Geneva: WHO. 27 Taer,B I Number of in-depth interviews with CDDs by country and gender of distributor tors Men Women Countrv No. (%) No. (%) Total Cameroon Nigeria Tanzania 1 1 (100) 46 (77) 6 (s0) 0 (0) ts (23) 6 rso) 11 6l t2 Total 63 (7s) 2t(2s) 84 (100) Tasr,o 2 Description of 109 focus group discussions (FGDs), by participant profile and country Focus Group friscussion Number of FGDs held in each country Particioant Profile Cameroon Tanzania Nigeria Total Adult women Adolescent women Adult men Adolescent men 6 3 6 J 9 5 4 2 34 18 13 6 49 26 23 11 Total 18 Adult = l8years * Adolescent = 10-17 years TasI,o 3 Number of household survey respondents, by country and gender of respondent Country Male (%) Female (%o) Total 20 7l 109 Cameroon Nigeria Total 674 (47.2) 2002 (48.2) 7s3 (s2.8) 2149 (sr.8) t427 4t5l 246 50 2921(48.1) 3148 (s1.9) 6069 28 Tasr,B 4 Age group of household survey respondents, by country Country Age group in years 0-4 5-17 18+ No. (o/ol No. (%) No. (7o) Total Cameroon Nigeria Tanzania 1s7 (l1.0) 3ls ( 7.6) 70 fi4.3\ sr4 (36.0) 1431 (3s.s) 7s6 (s3.0) 240s (s7.8) 236 (48.t\8s 07.7\ 1427 415r 491 I Total s24 ( 8.e) 2130 (3s.1) 3397 (s6.e) 6069 Tau,B 5 Respondents who reported taking ivermectin in five villages of Adamawa State, Nigeria, by village and by sex. Village Male 7o Female (7o) Total Futu Ga'anda Kiri Kwalamba Sabon Lavi 67 (s0.4) 26 (s4.2) 4e (s3.3) 43 (s2.4) 37 (42.s\ 66 (4e.6) 22 (4s.8) 43 (46.7) 3e (47.6) 50 (57.s) 133 I 48 92 82 87 Total 222 (s0.2)' 220 (4e.8) 442 Tasr.B 6 Preferred method of ivermectin distribution by sex of CDD in Cameroon, Nigeria and Tanzania Mode of distribution Male CDD Female CDD Both sexes House to house (H-H) Central location Both H-H & 33 (54.1) ts (24.6) r3 (21.3) 1t (s2.4) e (42.9) 44 (s3.7) 24 (2e.3) t4 07.1)t( Total 61 (100) 2t (100) 82 (100) 29 I FIGURE 1 Number and proportion of males and females treated with ivermectin in Cameroon, Nigeria, and Tanzania !,o G oL L o .ct E z 1600 1400 1200 1000 800 600 400 200 0 Emale Ifemale Cameroon Nigeria Country Tanzania T T 30 FIGURE 2 Reasons for non-treatment in Cameroon, Nigeria and Tanzania, by sex .2 0.6 .45.0 40.0 35.0 30.0 25.0 20.0 15.0 10.0 5.0 0.0 3E.4 l.b..nt lraf us.l Enot lnlo.m.d 2 3.3 I 7.! 1.1 3i4.3 32.0 21.1 29.E 25.9 ll.l. - F.m.l.c a meroon 15.1 7.Et.1 1.,4 0.0 tt'1.1. -.. F.mal.Nrgefla Mala - - FamelaI anzanta 31 FIGURE 3 Gender-specific schema of ladder of participation in CDTI, based on Arnstein's (1969) concept. OWNERSHIPIFItrtr IVERMECTIN DISTRIBUTIONII I * DECISION-MAKINGI tr tr tr tr tr ADHERENCEIIIII* COMPLIANCE*****I Legend: *: women; A: men; D: senior men. 32 MONITORING CO MMUNITY-DIRECTED TREATMENT PROGRAMMES FOR SUSTAINABILITY: TESSONS FROM THE AFRICAN PROGRAMME FOR ONCHOCERCIASIS CONTROT (APOC) MONITORING COMMUNITY-DIRECTED TREATMENT PROGRAMMES FOR SUSTAINABILITY: LESSONS FROM THE AFRICAN PROGRAMME FOR ONCHOCERCTASTS CONTROL (APOC). AUTHORS: U. v. Avezrco,t o. M. onoNo,2 K. Y. DeozIE,3 J. REMME,4 J. JryA,5 R. NovorraucyENyl,6 J-B. Rormcou,T M. NoMA,l aNo A. SeretBLtl ADDRESSES OF AUTHORS: 1 African Programme for Onchocerciasis Control (APOC), Ouagadougou, Burkina Faso 2 Brown University, Providence, RI, USA 3 01905 Osu - Accra, Ghana 4 TDR, World Health Organization, Geneva, Switzerland 5. National Onchocerciasis Task Force (NOTF) Secretariat, Lagos, Nigeria 6. National Onchocerciasis Task Force (NOTF) Secretariat, Kampala, Uganda 7. World Health Organization, AFRO, Harare,Zimbabwe Addres s for Coruespondence Dr. Uche V. Amazigo African Programme for Onchocerciasis Control (APOC),01 B. P. 549 Ouagadougou 01, Burkina Faso Tel: (226) 34.29.59; (226) 34.29.s3 Fax: (226) 34.28.75 E-mail: amazigouv@oncho.oms.bf Short running title: Monitoring CDTI for sustainability L ABSTRACT Community-directed treatment (ComDT) strategy is a new approach, adopted in 1997 by the African Programme for Onchocerciasis Control (APOC) for large-scale distribution of ivermectin (Mectizan@). Participatory monitoring of 39 community-directed ivermectin treatment (CDTD projects was undertaken from 1998-2000 with a focus on process implementation of the strategy and predictors of sustainability. Data from 14,925 household interviews in 2314 villages, 183 complete treatment records, 382 focus group discussions (FGDs), interviews with 669 community leaders, 757 trained community-directed dishibutors (CDDs) of ivermectin and 146 health personnel in 26 projects in four countries were analyzed. The data show that CDDs dispensed ivermectin to 65.4Yo of the total population, (71.2% of eligible population) with no significant gender differences (p>0.05) in coverage. Treatment coverage ranged from 60.20/o in Cameroon to 76.90/o in Uganda. There is no significant relationship between the provision of incentives to CDDs and treatment coverage (p>0.05). Refusal rate was highest in Cameroon (29.2%). Overall, community decision-making at village meetings was highest in selection of CDDs (72.1%) and lowest for distribution period (37.9%). Lessons learned include the need to improve communication strategies, address issues on absentees and refusals, community ownership and de-emphasising incentive. Investigating predictor indicators of sustainability, enables APOC to understand the proximate determinants of project performance, and how to initiate appropriate changes. 2 INTRODUCTION The development of Community-Directed Treatment with Ivermectin (CDTD was pioneered by the UNDPAVoTId Bank/WHO Special Programme for Research and Training in Tropical Diseases (TDR) in 1996 (WHO, 1996a). It was in response to the need for an ideal cost-effective and affordable method for large-scale ivermectin treatment, of populations in onchocerciasis endemic countries in Africa. It is of paramount importance that the drug delivery method should be sustainable given that health services are not optimally developed in terms of human and material resources in target endemic countries. It had become obvious that previous mobile and programme-directed community-based drug delivery systems would not sustain the long-term ivermectin distribution needed to reduce the ocular morbidity of onchocerciasis and its skin manifestations (Boatin et al., 1998). Among the various programme options for disease control, CDTI is widely regarded as a "ttmely and innovative strateglt ... which could be used as a model in developing other community based programmes, and is also a potential entry point in thefight against other diseases of public health tmportance" (WHO, 2000a). The African Programme for Onchocerciasis Control (APOC) established as a health development partnership between affected communities, the private sector (especially Merck & Co., Inc.), donors, non-governmental development organizations (NGDOs), and l9 participating governments, through the agency of their Ministries of Health (MoH), adopted the strategy of CDTI to control onchocerciasis in 1997 . By the end of 2000, APOC has awarded 63 grants , 84o of which are specifically devoted to field operations to establish sustainable ivermectin treatments using the CDTI approach. Given the multi-sectoral nature of its operations APOC instituted a programme implementation monitoring processes which has made it possible to assess the performance, and by inference, the commitment of partners with respect to expectations and goals of the Programme. Absence of such commitment reduces the prospects of 5 sustaining CDTI programmes. Monitoring CDTI Projects (f 998-2000) The CDTI approach and partnership model of APOC requires that communities assume full ownership and responsibilities for planning, implementing and overseeing a variety of measures for the control of onchocerciasis through the distribution of ivermectin in the communities. These include selecting and supporting ivermectin dishibutors, conducting household censuses, collecting the drug, determining the period and method of drug distribution, record-keeping and self-monitoring as well as undertaking an evaluation of the distribution processes at the community level. The community is undoubtedly the lead-stakeholder in the APOC partnership although the direct financial costs of its contributions are kept minimal and affordable. APOC funds each project for five years, and decreases its funding within the period in inverse proportion to the support provided by the other partners, (NGDO and MoH). In the first two years all the partners, (MoH, NGDOs and APOC) assist the community in terms of responsibilities clearly defined in the CDTI strategy. From the third to fifth years, the community is expected to have intemalized the CDTI philosophy of programme ownership having taken responsibility for drug distribution from inception to evaluation and should be able to treat at least 650/o of its total population annually. The ultimate goal is that, by the sixth year, the MoH, NGDOs and the affected community would have developed structures and procedures for establishing a sustainable drug delivery system that would be completely independent of external support for upwards of 15 years. Monitoring CDTI consists of periodic collection of information on selected indicators of programme objectives in order to assess the level at which key activities are being carried out and to determine whether programme expectations are being met with respect to the establishment of CDTI projects in the target population. The implementation process is assessed from the perspective of predictors of sustainability, usually 12 months after the establishment of projects, 4 and in particular the reorientation of extant mobile and community-based treatment programmes to the CDTI approach. As defined by APOC, sustainability indicators of CDTI programmes include community involvement, partnership commitment, treatment coverage, integration into the health systems and cost per person treated (WHO,1996b). Given that sustainability can only be truly determined after several years of implementation, periodic assessment for predictors of sustainability is important. Therefore, the main focus of CDTI project monitoring is to assess treatment coverage and the implementation process for the presence of predictors of three sustainability indicators namely, community involvement, partnership commitment and integration into the health systems. APOC introduced independent monitoring of CDTI implementation in 11 projects in Malawi, Nigeria, Sudan and Uganda, in 1998. The number increased in 1999 to 16 in Cameroon, Chad, Nigeria and Tanzania. [n 2000, 12 projects were monitored in Central African Republic (CAR), Nigeria and Uganda. A total of 39 (73%) of 53 CDTI projects in existence by December 2000 have been monitored. External independent scientists carried out the first set of monitoring exercises. From 1999, APOC transformed external independent monitoring to participatory partners' monitoring. The activity became more participatory and inclusive, involving communities, district and front-line health personnel and representatives of NGDOs. This transformation promotes the related processes of devolution, skills transfer and information sharing. The tools were designed to assess if participation was active or merely passive and to investigate participation as a means of empowering the community, identify strengths and weakness in projects and areas on which attention and resources should be focused. This paper describes the experience of APOC in the large-scale use of the community- directed approach and underlines the significance of first year monitoring of programme success and sustainability. 5 METHODS In each site monitors first held briefing sessions with policy-makers and programme partners before commencement of data collection. Thereafter, at debriefing sessions, members of the National Onchocerciasis Task Force (NOTFs), monitors and policy-makers reviewed the findings of the monitors, and discussed and developed strategies for improving subsequent distributions. The Sample and Sampling Procedure All the sites used a standard protocol and set of guidelines, developed in consultation with social and bio-medical scientists, for the selection of sample villages for the study. At each site a total of 30 villages were to be drawn from among communities eligible, (as determined by REMO) for CDTI, with proportionate representation to reflect disease prevalence and varying distances from health facilities Data collection procedure The technique used for monitoring the CDTI projects consisted of a set of eight qualitative and quantitative data collection instruments. This included focus group discussions (FGDs), a series of in-depth and semi-structured interviews, and a standardized household survey form that had been developed and validated in earlier studies sponsored by the TDR Task Force on Onchocerciasis Operational Research (OOR). The registers of events kept by CDDs in relation to distribution of ivermectin were examined. From these records data on treatment coverage, eligibility, refusals, absenteeism and serious side effects (SSE) were extracted by the monitors for comparison with records kept at other levels of programme implementation. Monitors interviewed or held discussions with the following five broad groups of people: leaders and members of the community, (adult men and women and the youth), health personnel, policy-makers, project managers and the NGDOs assisting with the projects. A fundamental consideration in the design of the instruments was to assess whether the activities and commitment of partners promoted progressive community involvement in all 6 aspects of CDTI including ivermectin procurement, distribution, record keeping, reporting and the selection and support of CDDs. The instruments also helped in the assessment of the role and contributions of the health care system and the support from NGDOs. This report is based on a survey of 26 of the 39 projects monitored in four APOC countries (Cameroon, Nigeria, Tanzania and Uganda) from 1998-2000. (Table 1) Each monitoring team randomly selected five households in each of six randomly selected villages. In each selected household they measured the height of all eligible members of the household and checked from the register how many tablets every one of such persons received from the CDD. This enabled the monitors to verify the accuracy of the dose of ivermectin administered by the respective CDDs. The study protocol also enabled the monitors to determine the proportion of those not eligible for treatment (principally children under-five years of age, pregnant women and very sick persons) who might have been heated in error by CDDs. To promote community participation, monitoring teams engaged the services of community members like community health personnel, agricultural workers, teachers and the religious leaders as resource persons and facilitators of the monitoring exercise. In a very few instances, because of unforeseen local logistical problems, monitoring teams were unable to adhere strictly to the standard protocol. Such departures were however very few and far between and are not deemed to detract from the main findings and conclusions of this study. The data collected were entered using EPI INFO software version 6.02 and were converted to SPSS-PC files. Data analysis was performed with SPSS version 8.0 for windows. RESULTS As shown in Table 1, household surveys were conducted in 2314 villages. Household members numbering 14925 were interviewed and treatment records of the households were examined on 7 the spot. A total of 382 FGDs were held. One-on-one interviews and discussions were held with 669 community leaders, 757 CDDs, and 146 health personnel. Treatment Coverage Treatment coverage is defined by APOC as the proportion of people treated with ivermectin, in a given CDTl-eligible community. Distribution of Treated population by project and country. The age and sex distribution of the 14925 household members interviewed showed a normal distribution. The mean age of the study population (all sites) was 24, ranging from 0 to 101 years. Table 2 shows the number of persons treated and the treatment coverage by country, based on a total population of 14,714 households respondents. The overall treatment coverage rate for all26 projects was 65.40/o. In 14 projects (54%) a treatment coverage rate of more than 650/owas achieved. The breakdown shows that in the first year of operations of the 26 monitored CDTI projects, this (65%) minimum desirable level of coverage was reached in one out of the three projects (33%) in Cameroon; nine out of the 18 projects (50%) in Nigeria; one out of the two projects (50%) in Tanzania and all the three projects (100%) in Uganda. The highest coverage (85%) was recorded in Kabarole district, Uganda, while the lowest (41%) was in Plateau State, Nigeria. Uganda had a mean treatment coverage rate of 76.9%o (range 71.7-85.5); Nigeria, 64.9% (range 41.9-82.0); Tanzania, 62.9% (range 58.0-67.0) and Cameroon, 60.2%o (range 52.5-67.2). Data from both the household surveys and the CDD registers indicate that Uganda with the highest coverage rate (76.9Yo) which also had the lowest absenteeism, refusal and "not informed" rates. Figures 1 and 2 illustrate the treatment coverage rates for projects monitored, using household surveys, in 1999 and 2000, respectively. Five out of the 14 projects (35.7%) monitored in 1999 and eight out of the 11 (72.7%) projects monitored in 2000 respectively achieved 8 coverage rates of over 650 during their first year of CDTI implementation. The substantial increase in coverage rates between projects monitored in 1999 and 2000 reflects a satisfactory trend and improvement in the CDTI implementation process. The Chi squared test of household surveys revealed no statistically significant gender differences (P (X'<13.885) :0.95) in ivermectin treatment rates overall in the five countries, with 49%o males and 5lo/o females. Figure 3 provides a comparison of treatment coverage, absenteeism, and refusal rates from two sets of data - the ivermectin treatment registers kept by CDDs and the household survey carried out by the project monitors. The difference in rates between the two sets of data in each country was: 0.3% in Nigeria, 5.lo/o in Uganda, 8.3 % in Tanzania and 16.3 oZ in Cameroon. Contrary to what was obtained elsewhere, the treatment coverage recorded by CDDs in Cameroon (43.9%), was very much lower than that obtained by the monitors (60.2oh) from household survey. Household survey data gave higher coverage rates than CDD register data in Uganda and Cameroon. The reverse was the case in Tanzania. The difference of 0.3% between the two estimates in Nigeria is negligible. It is noteworthy that Cameroon had the lowest coverage rate and the highest absentee and refusal rates. The significance of these findings is addressed in the Discussion section of this paper. Geographical (treatment) coverage. Geographical (treatment) coverage is the proportion of communities in a (REMO-defined) CDTI area where ivermectin distribution is undertaken in a given year. Out of a total of 669 hyper- and meso-endemic (CDTl-eligible) villages surveyed, ivermectin distribution took place in 630 (94.2%). The geographical coverage rates recorded, in descending order, were: Uganda,l00Yo,Tanzania,98.3o , Cameroon 95.6% and Nigeria 89.6%. Reasons for not taking the drug 9 The reasons for not receiving or taking the drug included ineligibility, absence at time of distribution and refusal. In the household population of 14,714 surveyed, 5,087 (34.6%) were not treated. Of thesel, 455 persons (10.0%) were absent during the distribution period, while 526 (3.7%) persons refused treatment. Among those who were not treated, 4910 persons had complete data available (Table 2). The result by country showed the proportions of absentees are fairly consistent for all countries (average of 29.1%). A breakdown by project showed consistent rates for projects in Uganda andTanzania and intra-project variations of absentee rates in Nigeria and Cameroon (Table 2). A rather unexpected feature of the report from both Uganda and Taruania, particularly Uganda, is the unusually high proportion of ineligibles recorded, 62.8% and 42.9o/o respectively. A breakdown of the basis of ineligibility showed consistent rates for projects in both countries indicating an unusually high proportion of under-fives. This may be due to a change in population structure due to either migration or the considerable impact of HIV/AIDS in the last decade. The highest rate of refusal (29.2%) was in Cameroon (Table 2) where the qualitative data revealed that the fear of serious side effects based on past experience following the use ivermectin, in some communities where onchocerciasis and loiasis co-exit, contributed to high refusal rates. In Littoral II Province, Cameroon where 52%o of those not treated had refused ivermectin (Table 2),men in particular recalled their experience with diethylcarbamazine (DEC) in the treatment of filariasis. Noting that the side effects of this drug were severe, the men were "hesitant" to take Mectizan. The comment of a male FGD discussant is worthy of note. Said he "Ese ce qu'on ne peut pas avoir le produit pour tuer les moutmouts? ("[s it not possible to have a product that will kill blackflies?"). In Nigeria 12.4oA of the people, respectively, apparently did not take the drug because they were not informed in advance about treatments. l0 Effect of drug shortage and untimely supply and collection on treatment coverage Among the communities surveyed, an average of 29.4oh experienced shortages in the supply of ivermectin ranging from ll.lYo in Uganda to 33.9Yo in Nigeria (Table 3). The degrees of shortages were not determined in every instance although some CDDs were reported to have devised coping strategies to deal with drug shortage. "This year (1999), half the people in the village did not receive Mectizaz@." (Village Head). The situation posed problems for CDDs who tried to cope in several ways as exemplified by the following revealing observation by an independent monitor in Nigeria. "They (CDDs) were only treating the "indigenous" and even then, only male adult heads and not all members of the household' [Independent Monitor, Nigerial In 175 communities with complete data for cross-tabulation analysis showed a significant relationship between shortage and coverage, p:0.005. Participants in the FGDs frequently mentioned instances of late delivery of ivermectin by health services and NGDOs to agreed collection points. Late ivermectin supply was experienced by an average of 21.7 Yo of the communities surveyed. This ranged from 33.3o/o in Cameroon, to 16.1%o in Nigeria. Tanzania and Uganda recorded 30.4 % and 27.8%o respectively (Table 3). However, quantitative data showed no statistically significant association between late supply of ivermectin to communities and treatment coverage. Out of the 669 communities surveyed, 6lYo collected ivermectin at their own cost from the health facilities, some located as far as 30 kilometers from the communities. The proportion of communities which collected ivermectin from health facilities was highest in Tanzania (78.3%), followed by Cameroon (65.4%), Nigeria (59.1%) and Uganda@a.Vo). " When we know thot the drug has come, we will make arrangemerrt to go and collect the medicine." [FGD, Tanzania] il In some communities focus group discussants maintained that it is the responsibility of the government to ensure that the drugs are delivered to the communities or that the community is too poor to take on the responsibility of collecting the drug. Incentives to CDDs The study examined whether a correlation exists between the provision of incentives (cash or in- kind) by communities to CDDs, their performance and treatment coverage. Out of the 757 CDDs surveyed, 52.7% did not receive any kind of incentive, monetary or in-kind; 26.6% received some form of incentives and 20.7% received transport support. The result show no significant association between giving incentives to CDDs and treatment covera1e (X2:1.089; p>0.05). Providing incentives to CDDs did not improve treatment coverage. It is remarkable that in Nigeria, 47 out of ll2 (42%) communities sampled, did not provide any community support, or monetary or in-kind incentives to their CDDs, yet in these communities treatment coverage rates were generally higher (70%o and above) than in communities where CDDs received incentives (Figure 4). "CDD support is from the community purse. They used to collect moiley but we discouraged it because it is not a good thing to give them money" (Village Head, Nigeria). "The only support given to CDDI by the community is exemption from communifit work However, this exemption is not restricted to those engaged in Oncho control services but it is extended to all those involved in other Primary Health Care aclivities" (FGD, Tanzania). "Government should give allowances to CDD*. l/'e are too poor to assist them" (FGD, Uganda). Some communities exempt CDDs from communal work as a form of incentive, while in others the thinking is that "He is our sonr "llhy should we pay him." Also, some CDDs reasoned that their own family members benefit from their services and that makes expectation and demand for incentives difficult. Willingness to continue as ivermectin distributors (CDD) l2 Information on willingness to continue as a CDD was available for 164 CDDs. Willingness to continue to serve as CDDs had no statistically significant association with provision of incentives. Both quantitative and qualitative data indicated a high proportion of CDDs who did not receive financial rewards or in-kind assistance from community and other partners are willing to continue as ivermectin distributors. Community-directed distributors: To contain the workload of CDDs, communities are advised to select 2-3 CDDs to serve about 250 persons, but a community can of course appoint any number of CDDs it chooses. The monitoring results show there were 1584 CDDs in the 603 villages surveyed for which complete data was available. There was an average of 2.6 CDDs per village while the average village population was 500-600. Eight out of every ten CDDs were males. The proportion of communities who changed CDDs after the last distribution was highest in Nigeria (20.7%), followed by Uganda (15.4%), T anzania (I0%), and Camer oon (3 .4%). Attributes of CDDs Occupation' The occupations of the CDDs in the study group were as follows: farmers 59o/o, civil servants 9.1%o, teachers 7.6oh, health workers, 2.loh, other occupations, 22.2%. Training: There are over 67,000 trained CDDs in ten countries. Monitoring data indicate that each community in Cameroon has approximately 1.8 CDDs, 2.2inNigeria,3.1 in Uganda and 7.0 in Tarrzania. CDDs were trained by district and front-line health personnel and NGDO partners on how to administer the correct dosage of ivermectin and to recognize and apply exclusion criteria. The study showed that they not only had good knowledge of and adhered very well to exclusion criteria but also administered appropriate doses of ivermectin. Out of 9,627 persons treated, only I .7%owas under five years of age. Community Participation and Programme Ownership 13 CDTI promotes active community participation and decision-making in ivermectin distribution as a means of improving access to drug and promoting a sense of responsibility and ownership. The study showed that communities participated in decision-making in ivermectin distribution although the extent and form of participation varied. The data allowed for examination of some predictor indicators of sustainability- community decision-making process on selection of CDDs and choice of period and method of drug distribution. Selection of CDDs. The level of community involvement in the selection and support of CDDs was high. .. When the message came to us through the health center, a village meeting was held...after which we selected CDDs. It is our sons who we know and trust that are handling it... The CDDs are doing fine." [Mele FGD, Nigerial The proportion of communities that selected CDDs at a meeting of the entire village or only at community elders meeting or village committee meeting was lowest in Cameroon (66.7%) and highest in Uganda (89%) (Table 4). Method of distributton. The order was reversed in respect of community decision making on the mode of distribution (house-to-house, central or both) which ranged from 33.3 %o inCameroon to 82.3% in Uganda. (Table 4). Decision-making on method of distribution by community showed no association with treatment coverage. Period of distribution. The results indicate that of the three sustainability indicators on community participation, communities are least likely to exert influence over the period of distribution. Only 379% of the communities surveyed made input into decision-making on period of treatment (Table D; 26% of the communities took the decision at village meetings. In 16 projects monitored in 1999, only 28.60/o of communities decided the period of distribution. FGDs with men, women and youths revealed that health personnel and NGDO partners decided on the period of drug distribution and delivered the drugs to CDDs during training or t4 later in their homes, "in order not to delay distribution, or if they had been going to the communities anyway. " (FGD, Oyo State, Nigeria). FGDs showed that in some communities, even though only village heads or village leaders made decision on behalf of the communities 65%o coverage rate and above was achieved. The results, summarised in Tables 2 and 4, show that in Uganda where 82.3% of communities decided on the period of drug distribution at a village meeting, treatment coverage was much higher than in Nigeria, Tanzania and Cameroon where in more than 50% of the communities health workers decided the period of treatment. PATNERSHIP COMMITMENT Health Education of communities The partners on the NOTF (MoH and NGDO project personnel) have the responsibility to educate and sensitize communities on the importance of every eligible member of the community swallowing the correct dose of ivermectin every year for several years for effective control of the disease. Analysis of the interviews with 699 village leaders showed that 86.3Yo of the villages surveyed received health education, ll.5% did not and2.2o/o of the leaders did not know whether health education was given or not. The rates of the health education provided were, in ascending order: CAR (66.70/o), Cameroon (83.3%), Nigeria (84.4%) and 100% in both Tanzania and Uganda. Project Funding A vital indicator of sustainability is the political will on the part of Ministries of Health to fund CDTI. This study showed that all the endemic local governments in Zamfara State project, Nigeria, which recorded 78.4% treatment coverage, released counterpart funding to support health education and mobilization of communities and supervision of CDDs. By contrast under- funding of some projects in Cameroon,Tanzania and Nigeria was associated with low treatment coverage. l5 INTEGRATION INTO THE HEALTH SYSTEMS Supervision of CDD by Health Personnel With the exception of the Cross River State, the CDTI project in Nigeria and the Kabale, Kabarole and Kisoro projects in Uganda, the overall quality of supervision by health staff of the CDTI projects were found to be generally inadequate. This inadequacy was reflected in the relatively poor perfornance of the CDDs concerned. Some of the deficiencies include, incomplete treatment records, poor records of minor and severe side effects and not following up to treat those absent during distribution of ivermectin. Although the quality of supervision was generally poor, the data showed that overall, supervision by health personnel, is significantly associated with coverage (X2:4.375; p<0.05), especially in Nigeria (X2 = 9.624, df : 2 P<0.005). This level of significance was not obtained in other countries. The data shows, that more supervisory visits during distribution improved achieved treatment coverage. The study found that some degree of integration of onchocerciasis activities have been achieved in some study sites; also that coordinators at district/LGA levels are involved in other health care activities such as immunization, malaria, leprosy and tuberculosis control. "We combine oncho control activities with supervision of other heahh activities such as collection of health records and supervision ofclinics." (Oncho. Coordinator). In Ruvuma district in Tanzania, monitors observed "districts are being empowered to continue the programme by integrating onchocerciasis into PHC activities. In some sites however, the level of integration was unsatisfactory. DISCUSSION The ONCHOSIM simulation model for onchocerciasis estimates that. depending on the prevailing level of endemicity, an annual ivermectin treatment coverage rate of at least 65%o of the total population of the community, for a period ranging from 15 to 25 years, is required to l6 eliminate the disease as a public health problem (Plaisier et al., 1990). It also envisages that 100% geographic coverage rates would be attained and maintained. Coverage The results of this study show that more than 50o/o of the projects monitored achieved 65%o or more treatment coverage and the trend over time is upwards. The highest coverage rates were achieved in the countries with the lowest absenteeism and refusal rates. The study also confirms the findings of a multi-country study (WHO, 2000c) that the main reason for not taking ivermectin tablets was ineligibility which in this study was 30.7yo, followed by absenteeism (29.1%) and outright refusals I0.7%. These proportions are fairly consistent for all countries, except in Cameroon where refusals top the list with 29.2%. The Cameroon situation has been attributed to failure to allay fears of side effects and the discontent, especially among men, with having to pay under the cost-recovery system that is in operation in the country. (Clemmons el. aI.,2001). FGDs with men and women in Littoral II Province, Cameroon, strongly indicate that a fear of side effects coupled with a negative response to payment for treatment and reported mismanagement of funds under the cost-recovery scheme are major factors militating against higher treatment coverage in Cameroon. (Ngoumou et al., 1996, Amazigo et a/., 1998). One assumption is that the cost-recovery policy or the process of its implementation in some districts may influence the accuracy of the number of treatments reported by CDDs in Cameroon. As one independent monitor observed, the fact that the number of persons reported by a CDD as having received treatment determines the amount (in cash) he has to account for and return is a temptation for the CDD to under-report treatment coverage so as to keep back some of the monies collected. A study to examine whether cost-recovery policy has an impact on the number of treated persons reported by a CDD is clearly desirable. It is significant that overall, the cumulative proportion of absenteeism and refusal 39.8% is more than that of the ineligibles (30.7%). It is recognized that a proportion of absentees are 17 persons whose absence is a reflection of their unwillingness to take the drug. In order words refusals. This implies that to maintain a satisfactory level of treatment coverage effort must be directed at reducing both the refusal and absentee rates. Studies have shown that the imposition of payment mechanisms results in an unwillingness to pay by recipient communities (Onwujekwe et al., 2000). This would also account for the high rate of absenteeism (30.2Yo) among those not treated in Cameroon. (Table 2). It is tantamount to 'refusal by absenteeism' and calls for improved and more effective community mobilisation and health education. (Clemmons et a1.,200I). Lack of information was another main reason for not receiving treatment. There is therefore a need to remobilize communities adequately and to intensiff the use of IEC materials especially in Nigeria, with the highest recorded proportion (12.4%) in this category. This study shows that another determinant of low treatment coverage is drug shortages in the communities during the period of treatment. The shortage of drug means that a number of eligibles who turn up for treatment would be denied it. It also leads CDDs to compromise the accuracy of dosages in order to ensure that as many people as possible are treated by cutting back on the amount of the drug that individuals are entitled to. Akogun et al., (2000) in a study in Takum, Nigeria observed that drug shortage was the commonest reason for non-treatment of eligible individuals, and that it encourages alteration of approved dosages by CDDs who wish to cope with the shortage problem. Shortages also tend to contribute to absenteeism as communities are unable to predict when enough ivermectin would be available. Community participation There are many dimensions to community involvement in CDTI. Most of these centre around community participation in decision-making process and perception of ownership of the programme. In the CDTI strategy, CDDs are selected by community members. The community determines the criteria for selection and exercises authority over the selection of CDDs. It also t8 provides support to CDDs and changes those CDDs whose performance is deemed unsatisfactory. In a study of CDTI implementation in Uganda, Katabarwa et al., 2000a) found programme success to be significantly associated (P < 0.05) with community involvement in: (1) decisions about prograrnme execution; (2) CDD selection; (3) attendance at health education sessions; and (4) rewarding CDDs in kind. Braide et al., (1990) in a study in Cross River State (CRS), Nigeria, also made a case for the role of community participation in the sustainable control of onchocerciasis. The strength of CDTI lies in the communities' appreciation of the benefits of the programme, the drug ivermectin and their continued willingness to participate in its implementation. (Clemmons e, al., 2000). In most of the projects in the four countries visited, communities appreciated the benefits of ivermectin in improving health. "The drug distribution programme is a success because we are healthter... A measure of success is to check our eyes. Those who had eye ailments are now feeling better. " However, they also perceived the prograrnme as being owned by government or another external-donor programme. In Abia State, Nigeria (monitored in 1999), a focus group discussant observed that "We are not aware of ownership, but we understand the importance of the drug and are ready to cooperate" (Okeibunor et al.,1999). This statement which is common to several projects monitored, reflects the undeveloped perception of ownership in many projects. This is not unexpected in the first year of transition from a programme-driven to a community-directed approach. Innovative information, education and communication (IEC) strategies and country specific plans need to be developed. Incentives A special feature of the CDT strategy of APOC partnership is the decision that incentives and support to CDDs are the responsibilities of the community, but the issue of providing CDDs with specific incentives remains a thorny one. National partners in the implementation of CDTI, (MoH l9 and NGDOs) are encouraged to discontinue provision of cash or in-kind incentives and during health education to sensitize communities to assume this responsibility. Evidence from Uganda traces the root cause of the "the incentive problem" to the former practice of NGDOs in giving erstwhile Community-Based Distributors (CBDs) cash as incentive for drug distribution under the (pre-APOC) Ivermectin Distribution Programme (IDP). The discontinuation of paid incentive under the CDTI has clearly alienated some of the distributors, especially those who had been CBDs. In the pre-APOC era, communities perceived the support of CBDs as the financial responsibility of external prograrnme sponsors. Besides there are other contemporary control prograrnmes, like the poliomyelitis eradication prograrnme where substantial sums of more are paid out to community-based operatives. The CDTI distributors feel cheated for being left out of the control programmes bonanza. Currently, about one third (34 %) of the communities have made the transition to the APOC CDTI strategy with regards to CDD support, and thus provide support to CDDs (WHO, 2000b). However, data from this study do not show a significant relationship between the provision of incentives to CDDs and heatment coverage rates. Indeed in some project sites the coverage rates attained by CDDs who did not receive direct material incentive or reward in cash or in kind were as good as, and sometimes better, than for CDDs who received incentives or rewards. (Amazigo et al., 2001, Clemmons et al., 2001). Kajubi (2000) also confirmed this pattern in a Uganda study. This, it seems, is particularly so, if CDDs are selected along lines of kinship and clan affiliations (Katarbawa et a\,2000b). Traditional communities live their lives in a homogeneous moral and spiritually charged environment, where filial piety is associated with non-quantifiable blessings from ancestors and other spiritual patrons of the community. The prayers of elders are regarded very highly and are part of a broad mix of moral rewards. The very performance of their duties gains respect for the CDDs. A quasi-experimental study in Ibadan, Nigeria, found that 32 out of 33 CDDs from both study and control groups indicated a 20 willingness to continue as CDDs. The conclusion of the study team was that "an enhanced sense of respect is an important non-monetary incentive for CDDs to continue their work" (WHO, 2000b). While qualitative findings in this study show this is true for most communities, evidence monitored in Ido Local Government Area (LGA) of Southwest Nigeria pointed in another direction. In that LGA, with a reported coverage rate of llo/o,the low level of CDD morale may be part of a wider complex of administrative inefficiency and frictional politics at the LGA level. ln the words of a CDD at Omi Adio (Ido LGA), "I cannot work withoutfinancial reward'. This attitude may be fuelled by widespread suspicion that prograrnme managers in the LGA had been paid huge sums of money that they then decided to keep for themselves (Obono et al., 2000). This is a recurring theme whenever different programmes in the same or adjacent localities adopt different means of incentives or compensation for local 'volunteers'. This suspicion is heightened where there are other contemporary control programmes, like the poliomyelitis eradication programme in which substantial sums of more are paid out to community-based operatives. The CDTI distributors in some instances have felt cheated. Such CDDs and their communities require more sensitization. Communities ought to be left to handle the issue of incentives in a way they ordinarily would in comparable situations. Another determinant of coverage is decision-making on the period of distribution. The timing of drug distribution is still subject to influences beyond the community's control. Since the 27 projects were monitored only one year after the take-off of CDTI, it may be premature to conclude that this is a strong sustainability factor negatively affecting treatment coverage. Nevertheless, conununity participation in decision-making was lowest for deciding on period of distribution (37.9%). Further, focus group discussants linked non- participation of communities in deciding the month and time of ivermectin treatments to the high rate of absenteeism (10.0%). We therefore suggest an in-depth structured study of the impact of this variable which is necessary for decision on policy. 2r \ This study points also to the role of traditional institutions in community decision-making process as well as a relationship of trust between the community and the health system, where single individuals or a trusted group may be empowered by communities to take decisions on their behalf. The lessons here are firstly, that CDTI strategies ought to respect variability among different communities, in as much as they represent a broad cultural spectrum. An allowance should be made for community ingenuity, with respect to how certain aspects of programme implementation are carried out and what aspects of its commitment may be delegated through strategic negotiation with individuals, haditional leaders, groups or the health system. Just as it is misleading in some traditions, to equate the views of community leaders as identical with the views of the entire community (Onwujekwe et al., 1999), it is also incorrect to ignore situations where the voice of the community leaders are accepted as the voice of the people. Training The training received by CDDs is of potential wide ranging benefit as the mass of CDTl-trained local persons at the community level constitute an invaluable pool of human resource that could be mobilised and recruited for other health and development prograrnmes. Training of health personnel and CDDs was found to be generally effective, and to follow CDTI guidelines, though inadequate in a few cases. This study however indicates that the quality of training with respect to record keeping is still generally poor. This is mainly as a result of literacy and numeracy problems at the community level. Poor govemment funding was partly responsible for inadequate training of CDDs. At times, because of inadequate funding more than 30 CDDs were trained at one centre in a single day. There is also a perceived need to further increase the number of CDDs per community so as to reduce their workload (Amazigo, 1999). CONCLUSION Results of the monitoring exercises show clearly that APOC is well on the way to achieving it primary objective of establishing an ivermectin distribution system at a treatment coverage level I 22 of 650 or more. The study has provided useful insight into critical sustainability indicators such as community participation in decision-making processes, CDD motivation, access to the drug and the relative effect of these on treatment coverage. It is note worthy that contrary to programme managers' long standing perception about incentives, the study results indicate that incentives to CDDs may not (at least in the short run) be a major programmatic issue if health education of communities on the immediate and long-term benefits of ivermectin are clearly communicated and understood. This study has also shown clearly that, independent and partners' participatory monitoring is vital for sustaining community-directed treatment programmes if adopted as an integral aspect of project implementation. The transition from independent to participatory partners' monitoring, in tandem with APOC's devolution of responsibilities to NOTFs, signifies strongly that APOC's ultimate withdrawal from the partnership will not signal the collapse of the control programme. Monitoring should continue to serve as a means of assessing the quality of Partnership commitment, including the crucial processes of drug distribution at the community level. Participatory monitoring of CDTI exercises can be used to determine whether programme interventions are improving health outcomes in the target communities, and to modify these interventions over time. AcxNowT,EDGEMENTS We are grateful to the communities and their leaders who participated in this study. We would like to thank Professor Oladele Kale for his valuable comments and suggestions on the manuscript. We thank also the members of the 26 monitoring teams for their assistance in the collection of the data for this study. 23 REFERENCES AKOCUN, O. 8., AUDU,2., WEISS, M.G., ADELAKUN, A. O., AKOH, J. I., AKOGUN, M. K., REMME, H. & KALE, o. o. (2001). Community-directed treatment of onchocerciasis with ivermectin in Takum, Nigeria. Tropical Medicine and International Health, 6, 232-243. AMAZIGo, u., NoMA, M., BoATIN, B. A., ETYA'ALE, D. 8., SEKETELI, A. & DADZIE, v. r. (1998). Delivery Systems and Cost Recovery in Mectizan Treatment for Onchocerciasis. Annals of Tropical Medicine and Parasitologl, 92, (Suppl. l), S23-S31. AMAzIGo, U. (1999). Community selection of ivermectin distributors. Community Eye Health,12,39-40. AMAZIGo, U.V., BRTEGER, W.R., KatnBARwA, M., AKocuN, O., NrEp, M., BonrrN, B, N'Doyo, J., NoMa, M. & SEKETELI, A. (2001). The challenges of community-directed treatment with ivermectin of the African Programme for Onchocerciasis Control. Annals of Tropical Medicine and Parasitology, ?95,, (Suppl. ) ???-??? BOATIN, B. A, HOUGARD, J. M., ALLEY, E. S., AKPOBOUA, L. K., YAMEOGO, L., DEMBELE, N., sEKETELI, A. & DADZIE, K.y. (1998). The impact of Mectizan on the transmission of onchocerciasis. Annals of Tropical Medicine and Parasttologt,92, (Suppl. 1), 546-60. BRATDE, E. r., oBoNo, o. M., & nnssry, s. e. (1990). Community Participation in the control of onchocerciasis in Cross River State, Nigeria. Acta Leidensia,59r 427432. CLEMMONS, L., AMAZIGO, U. V., BISSECK, A-C., MSLTYA-MPANJU, J., NOMA, M., OYENE, U., EKPO, U. & snxerelr, A. (2001). Gender issues in community-directed treatment with ivermectin (CDTI) of the African Programe for Onchocerciasis Control (APOC). Annals of Tropical Medicine and Parasitologt. ?95, (Suppl . ), ??? -???. KAJUBI, J. B. (2000). Mission Report to Kabale and Kisoro Districts, 20-26 November 2000. AFRO Memorandum: Communication to Oladapo Walker. Ref. PiKAKUBI (Unpublished document). KATABARWA, N. M., MUTABAZT, D, & zuCHARDS, F. o. ln. (2000a). Controlling onchocerciasis by community-directed ivermectin treatment programmes in Uganda: why do some 24 communities succeed and others fail? Annals of Tropical Medicine and Parasitology, 94, 343-352 KATABARWA, N. M., RICHARDS, F. o., AND NDYoMUGYENYI, R. (2000b). In the rural Ugadan communities the traditional kinship/clan system is vital to the success and sustainability of the African Programme for Onchocerciasis Control. Annals of Tropical Medicine and Parasitology, 94, 485-495. NGouMoU, P., ESSoMBA, R. o. & GoorN, C. (1996). Ivermectin-based Onchocerciasis Control in Cameroon. World Health Forum, 17,25-28. OBONO, o. M., KoRVE, K. N., oLAMIru, F. & ErcRAru, n. (2000). Independent Monitoring of CDTI Projects in Oyo State, Southwest Nigeria, 0l September 22 September 2000. wHo/APoc/cs/00.2. oKEIBUNOR, J., OKOROAFOR, C. U., OYENE, U., NNORUKA, 8., TAMBALA, p., OGBU-PEARSE, p. E. & OLAMIJU, n. (1999). Report of Independent Monitoring of Community-Directed Treatment with lvermectin in Imo/Abia States, Nigeria 20 September-2} October, 1999. wHo/APOClCSl99.t oNwUJEKwE, o., SHU, E. & oKoNKwo, P. (1999). Can community leaders' preferences be used to proxy those of the community as a whole? Journal of Health Service Research and Policy,4, 133- 138. oNwUJEKwE, o., SHU, E. & oroNrwo, n. (2000). Community financing of local ivermectin distribution in Nigeria: potential payment and cost-recovery outlook. Tropical Doctor,30, 9t-94. PLAISIER, A. P., OORTMARSSEN, G. J. van., HABBEMA, J. D. F., REMME, J. & ALLEY, E. S. (1990). ONCHoSIM: a model and computer simulation program for the transmission and control of onchocerci asis. Computer Methods and Programs in Biomedicine, 31, 43-56 25 woRLD HEALTH oRGANISATIoN ( 1996a). Community-Directed Treatment with lvermeclin. Report of a multi-country study TDR/AFR/RP/96.1. Geneva: WHO wORLD HEALTH ORGANISATION (1996b) African Programme for Onchocerciasis Control. Programme Documentfor Phase I. (Unpublished Document). woRLD HEALTH oRGANISATIott (2000a). African Programme for Onchocerciasis Control Report of the mid-term (Phase l) External Evaluation of APOC. wHo/APoc/csi00.3. Ouagadougou: APOC woRLD HEALTH oRGANISATIoN (2000b) Implementation and Sustainability of Community- Directed Treatment of Onchocerciasis with lvermectin: A Multi-country Study (Final Report). rDR/IDE/RP/cor/00. 1. Geneva: WHO. 26 TaeI-B I Distribution of 26 CDTI projects monitored between 1998 & 2000 in four countries SAIo. Project Name Country Year Moni- tored No. HH surv- eyed No. No. HH CDDs memb- interv- interv- iewed iewed No. No. No. Village Health FGDs leaders person- condu- interv- interv- cted iewed iewed l.) 3. 4. 5. 6. 7. 8. 9. 10. I l. 12. 13. 14. 15. 16. t7. r8. 19. 20. 21. 22. 23. 24. 25. )6 90 98 105 90 81 90 78 t20 t21 96 103 73 90 71 Kaduna Abia Adamouwa Anambra Ebonyi Enugu FCT Abuja Imo Littoral 1l Mahenge Nasarawa Osun Plateau Ruvuma SW Province Adamawa Bomo Delta Edo Kabale Kabarole Kano Nebbi oyo Yobe 7 emfqrq Nigeria Nigeria Cameroon Nigeria Nigeria Nigeria Nigeria Nigeria Cameroon Tanzania Nigeria Nigeria Nigeria Tarzania Cameroon Nigeria Nigeria Nigeria Nigeria Uganda Uganda Nigeria Uganda Nigeria Nigeria Ni oeri a I 998 t999 t999 r999 1999 1999 1999 1999 1999 t999 t999 1999 t999 1999 t999 2000 2000 2000 2000 2000 2000 2000 2000 2000 2000 92 92 89 90 106 90 79 400 s82 8ll 369 777 562 474 661 620 586 752 551 778 491 532 750 564 607 373 579 482 708 548 491 521 192 NA 32 28 30 43 3l 29 35 43 59 30 29 27 65 36 30 16 30 28 8 6 49 5 24 30 22 NA 30 30 27 30 29 32 30 31 30 23 30 28 30 3l 30 33 30 29 4 5 30 4 3l 30 22 NA t4 6 5 6 l3 5 t2 7 3 J J 2 7 t3 J 5 5 lt 3 2 6 4 5 7 4 NA 18 l8 l8 18 l8 l8 18 18 18 t2 t2 t2 t8 l8 l8 20 l8 l8 NA NA l8 NA 18 I8 20 45 81 90 63 83 ) TOTAL (for26 proiects in 4 countries. 1998-2000) 2314 14925 757 669 146 382 27 TASI-B 2 Reasons for not receiving ivermectin, treatment by country TREATMENT Reasons for not being treated COVERAGE Number lneligible Absent Refusal Not Others COUNTRY Not treated lnformed % Pop N otto olto otlo otto olto CAMEROON ADAMAOUA LITTORAL II SW PROVINCE TOTAL 67.2 52.5 58.6 810 619 532 266 288 220 42.9 16.7 21.4 33.8 28.1 28.6 5.6 52.1 27.7 0.4 0.3 12.7 17.3 2.8 9.5 60.2 1961 774 27.0 30.2 29.2 3.9 9.7 NIGERIA ABIA ADAMAWA ANAMBRA BORNU DELTA EBONYI EDO ENUGU FCT-ABUJA lMo KADUNA KANO NASARAWA OSUN oYo PLATEAU YOBE ZAMFARA TOTAL 73.2 75.7 61.3 70.2 58.5 56.7 64.6 52.4 62.6 61.1 72.0 66.6 71.7 66.9 64.2 41.9 82.0 78.4 64.9 s3B 750 367 564 607 736 373 511 460 660 397 707 749 525 491 775 521 384 10115 146 182 135 138 251 319 120 229 162 250 103 240 214 174 163 445 94 75 3440 23.3 24.2 21.5 50.0 21.9 30.7 16.7 22.7 37.0 20.8 48.5 28.8 29.9 11.5 17.8 18.9 77.7 46.7 27.2 61.0 49.5 25.2 39.1 45.4 37.6 25.0 31.4 29.0 32.4 31.'l 15.4 15.9 46.6 36.2 6.1 5.3 24.0 29.8 10.3 1.6 9.6 0.0 15.5 12.9 0.8 3.9 6.8 33.2 2.9 0.0 0.9 3.4 6.7 1.8 0.0 8.0 7.3 0.0 13.2 28.1 0.0 4.4 0.0 56.7 14.8 5.6 0.0 3.9 37.1 0.0 9.8 31.9 17.8 0.0 2.7 12.4 5.5 11.5 15.6 10.9 12.7 1B.B 0.8 27.1 21.6 13.6 13.6 18.8 53.3 28.7 7.4 55.5 17.0 18.7 23.3 TANZANIA MAHENGE RUVUMA TOTAL 67.0 58.0 62.9 581 488 1069 192 200 392 47.4 38.5 42.9 24.5 22.5 23.5 8.9 12.0 10.5 2.6 2.0 2.3 16.7 25.0 20.9 UGANDA KABALE KABAROLE NEBBI TOTAL 75.1 85.5 71.7 76.9 575 447 547 1 569 140 23 141 304 65.0 65.2 60.3 62.8 25.0 26.1 26.2 25.7 4.3 4.3 0.7 2.6 0.0 0.0 1.4 0.7 5.7 4.3 11.3 TOTAL FOR 4 couNTRlES 65.4 14,714 4,910 30.7 29.1 10.7 9.5 20.0 TABLE 3 Percentage of communities that experienced shortage or late supply of ivermectin Country %o with shortage 7o Late Supply Yes Yes Cameroon 29.6 33.3 Nigeria 33.9 16.l Tanzania 21.7 30.4 Usanda ll.l 27.8 Total 29.4 21.7 Tnsr,B 4 Community participation in decision making on sustainability indicators Predictors of Sustainability Country 7o of communities where decision was made Community Village Health Other meeting leader worker Distributionperiod Cameroon Nigeria Tanzania Uganda o/o Total 24.0 34.9 34.7 82.3 37.9 20.0 6.4 0 0 6.9 s6.0 54.1 56.6 r 1.8 50.6 0 4.6 8.7 5.9 4.6 Distribution method Cameroon Nigeria Tanzania Uganda o/^ Totzl 33.3 45.4 60.9 82.3 49) 7.4 t2.7 8.7 5.9 l0 59.3 35,5 13.0 11.8 33.9 0 6.4 11.4 0 6.2 Selection of CDDs Cameroon Nigeria Tanzania Uganda o/o Total 66.7 70.4 74.0 88.8 72.1 11.1 17.4 4.3 5.6 t3.7 11.1 t2.2 2t.7 5.6 t2.6 11.1 0 0 0 1.6 29 71 7 732 608 586 61 1 61.3 62 6 52.4 525 41.9 80 70 60 .50 zU =k40 ultF 20 10 0 z U6 lolzlu f UF 5 o- =fi o z f o o uJ oz u,I = zlo o l lo For t6 z lt(J z ot o- =o =l lt tz o6 tU o = Fig. I Treatment Coverage by Project monitored in 1999 =) fr o E J I I I i I I 1 I Fig. 2 Percentage Treatment coverage by project monitored in 2000 00 t 5,5 a2 t0 70 60 50 /40 30 20 t0 72 C) (dq) H o s oitt o*o .+.o ood J *J'* ,;.t' """ \o +.' 6oI {' f.* PROJECT 751 7l-7to.2 aa Z aa.a 31 0 Fig. 3 Comparison of mean treatment coverage, absentee and refusal rates by household survey and CDD treatment registers in four countries H/H survey E =* 100 80 60 40 20 0 CDD register H/H suwey CDD register H/H survey cDo register H/H survey CDD register UGANDA NIGERIA TANZANIA CAMEROON E Mean coverage !% Absentees tr% Refusal71.8 71.2 64.7 t.t r.9 11.5 1.1 76.9 65 62.9 60.2 43.9 t.6 t.6 2.70.5 32 { Ftc. 4 Incentives to CDDs and treatment coverage Treatment coverage and incentives Sample of '1 12 communities in Nigeria 100 0 N= 31 3,1 Transport Financial incentive No incentive Support that CDD received from community G80o\ ; o) 860 oo E40 EG o)F20 o 33 CO M MUNITY-DIRECTED TREATMENT WITH IVERMECTTN (cort) sTRATEGy oF Apoc AND rrs POTENTIAL FOR PROVIDING ADDITIONAT HEATTH SERVICES OF THE POOREST POPULATIONS COMMUNTTY-DTRECTED TREATMENT WITH IVERMECTIN (CDTI) STRATEGY OF APOC AND ITS POTENTIAL FOR PROVIDING ADDITIONAL HEALTH SERVICES TO THE POOREST POPULATIONS AUTHORS: M. HouenA,l E. BRAIDE,2 E. EluasseN,'U. v. AMAZrco,a B. Ltese,s B. BeNroN,s M. Notrre,a D. Eryn'aLB,6 K. Y. Daozrc,i o. o. KarB8 eNo A. SBrrtelra ADDRESSES OF AUTHORS: 1. Academy of Medical Sciences and Technology, Khartoum, Sudan 2. Department of Biological Sciences, University of Calabar, Nigeria 3. Sight Savers International (SSI), Kaduna, Nigeria 4. African Programme for Onchocerciasis Control (APOC), Ouagadougou, Burkina Faso 5. The World Bank, Washington D. C.20433, USA 6. World Health Organization, Geneva, Switzerland 7. 01905 Osu - Accra, Ghana 8. Department of Preventive and Social Medicine, University College Hospital, Ibadan, Nigeria Addres s for Corresp ondence Dr. Azodoga S6k6t6li Director, African Program for Onchocerciasis Control (APOC),01 B. P. 549, Ouagadougou, Burkina Faso Tel: (226) 34.22.77 Fax: 0.56\ 34.48.00 E - * oi l?ts6 k a e t i a@p n c t t o. Short running title: Health care services potential of CDTI 1 ABSTRACT Since its inauguration in 1995 APOC has made significant progress towards achieving its objective of establishing sustainable community-directed ivermectin (Mectizan@) delivery system (CDTI) in onchocerciasis endemic areas outside OCP countries. In the year 2000, the Programme in partnership with Governments, Non-Governmental Organizations and Endemic Communities, has succeeded in treating 20,298,138 persons in49,654 communities in 63 projects in [4 countries. Besides the distribution of ivermectin, the prograrnme has strengthened Primary Health Care (PHC) through capacity building, mobilization of resources and empowerment of communities. The community-directed treatment approach is a model that can be adopted in developing other community-based health programmes. The approach has also made it possible to bring some measure of interventions to the poor in some other health care prograrnmes like in malaria, eye care, maternal and child health, nutrition and immunizationprogrammes to the poor. Thus CDTI presents, at all stages of its implementation, a unique window of opportunity for promoting the functional integration of health care activities. For this to be done successfully and in a coordinated manner, adequate funding of CDTI within PHC is as important as an effective sensitization of policy makers, health workers and communities on the value of integration. These must be accompanied by appropriate training at all levels. Evaluation of experiences in integration of health services particularly at community level is crucial to the success of the integration. 2 INTRODUCTION The search for simple effective methods of ivermectin delivery in onchocerciasis endemic countries became necessary after the donation of Mectizan@ (ivermectin, MSD) by Merck & Co. Inc., in 1987 "to all that need it for as long as needed". In this regard, a meeting was organized by the WHO in Geneva in 1991 to develop strategies for ivermectin distribution through the Primary Health Care (PHC) System. This meeting was attended by representatives of six major onchocerciasis endemic countries, eight non-governmental organizations (NGOs), and Merck & Co., Inc. Participants noted that though there was no barrier to quickly getting ivermectin from the donor to governments yet there were difficulties in getting the drug to the vast majority (80%) of the target populations because of lack of political will and of awareness at the national level and limited resources and infrastructure particularly at the periphery. It was therefore decided that the PHC infrastructure be strengthened for large-scale distribution of ivermectin. This was to be achieved by including the drug in national lists of essential drugs and making it available at all health care levels for distribution by community workers to the most badly affected communities. The Community-Based Ivermectin Treatment (CBIT) approach evolved from these recommendations and was adopted by NGDOs supporting ivermectin distribution in endemic countries. However, the CBIT approach was soon found to be costly, not participatory and relatively ineffective in terms of level of treatment coverage (WHO, 1996a). The African Programme for Onchocerciasis Control (APOC) was set up in 1995 with the objective of establishing sustainable community-directed systems for ivermectin delivery covering 50 million people in 19 countries which fall outside the scope of the ongoing OCP. The systems are expected to become self-sustaining by 2007. In 1997, the Programme adopted the Community-Directed Treatment with Ivermectin (CDTI) approach after a multi-centre study showed that endemic communities are capable of handling ivermectin distribution by themselves J and record a consistently higher treatment coverage rate than when the distribution was organized and directed from outside by Programme Managers. (WHO, 1996b). Using this approach, an impressive totalof 20,298,138 persons in49,654 communities in 63 APOC-funded projects in l4 countries received ivermectin treatment in the year 2000 (WHO, 2000a). The essence of CDTI is that endemic communities targeted for mass treatment with ivermectin, are encouraged to take responsibility for organizing the distribution of the drug themselves. This includes mobilizing those eligible for treatment, taking collective decisions on who is to distribute (Community-directed distributor (CDD) selection), when to distribute (taking cognizance of local seasonal variation and cultural practices, festivals and events and how to distribute (whether house to house or from a central location). Under CDTI, collection of the drug from the nearest health facility or a collection point that has been previously agreed with the supervisory health care personnel as well as providing the means of doing so is the responsibility of the community. So also is the provision of incentives or compensation to CDDs, monitoring their performance and ensuring proper keeping of records and follow-up. In implementing the CDTI strategy, APOC has mobilized resources to provide training in a variety of disciplines, thereby building the capacity of the PHC system. Participating communities are empowered to take on responsibilities for decision-making and support of the prograrnme. A unique opportunity exists for utilizing CDTI as a model for planning various developmental programmes and as an entry point for implementing other health interventions. This paper discusses the CDTI strategy of APOC and its potential for facilitating the provision of additional health services to the poorest populations. CDTI and Health Care Delivery Systems in APOC countries Health care delivery in APOC countries, as in other developing countries, is mainly through primary, secondary and tertiary structures and facilities at community, regional and central levels respectively. Tertiary health facilities provide mainly curative health care. At the regional level, 4 secondary facilities act mainly as referral centers for health posts, which provide simple curative and preventive health services at community level. In Cameroon, Ethiopia, Liberia, Malawi, Nigeria, Sudan, Tanzania and Uganda, Government provides about 40%-60% of health services while private practitioners, missions, non-governmental organizations and traditional healers supplement health care service. Health care delivery systems in APOC countries are based on the PHC philosophy "to reduce dtsease morbidity and mortality through methods and technologlt that are scientifically sound, socially accepted, universally acceptable and participatory" (WHO, 1978). Most of these countries have had to introduce health reforms realizing that nearly a quarter of a century after the Alma Ata declaration, the objectives of PHC have not been achieved. Healthcare delivery systems in developing countries have tended to deteriorate progressively and are grossly ill equipped to cope with the present demands. Some of the comparative inequities between developed and developing worlds are summarised in Table 1. It is evident that in spite of the PHC philosophy, health care in most developing countries does not favour the poor (WHO, 2000b). Health care personnel are concentrated at the tertiary and secondary facilities in urban areas. Resource allocation is more in favour of curative than promotive and preventive health care. Emphasis is on personal medical care rather than on broad public health interventions. The poor cannot meet high cost of treatment and have to travel far distances to health facilities. Beyond these, the risk factors that contribute to the ill health of the poor have not been adequately addressed. There is a wide gap between policy and practice that is aggravated by inadequacies in human, physical and financial resources. Most of the reforms being introduced in some countries are designed to address these inadequacies and also to promote community participation and focus on major health problems of the poor. Unfortunately what PHC set out to achieve is what the poor still lack in many developing countries - accessible and affordable basic health care. This is the situation that obtains in most CDTI project areas. 5 The introduction of CDTI has provided an opportunity for strengthening PHC. Five years after the commencement of CDTI, significant progress has been made in structural and functional integration of CDTI into the PHC systems in onchocerciasis endemic countries (Table 2). Multi- centre studies have shown that practical integration of CDTI into PHC has begun. CDTI coordinators and CDDs are involved in activities other than CDTI. (Table 3) (WHO, 1999a, 2000a). CDTI officials deliberately handle other health problems during ivermectin distribution. "We combine onchocerciasis control activities with supervision of other health acttvities such as collection of health records, and supervision of health clinics" CDTI Coordinator, Plateau, Nigeria. An APOC-funded multi-centre study in four CDTI project site in four countries - Kisoro in Uganda, SWl Province in Cameroon, Raja in Sudan andZamfara in Nigeria - investigated the extent to which CDDs are involved in other health and developmental activities, the similarities between CDTI and these activities, and the effect of additional responsibilities on their performance in CDTI. The results of the study show that many CDDs are involved in other health and developmental activities without any apparent adverse effect on their performance in CDTI. Many were in fact selected, as CDDs because of their past satisfactory performance in community service (Tables, 4 & 5). The absence of conflict in CDD responsibilities was shown (particularly in Uganda) to be attributable to the following similarities between CDTI and the other activities: o All are community-based activities aimed at serving the people. o All are voluntary community services. o It is feasible to integrate the activities at community level. Besides, communities have traditional arrangements for avoiding such conflicts which, they indicated did not occur often. "CDTI work comes up once a year and hardly coincides with other activities ". (WHO, 2000d) 6 (CDD Bingi, Maru LGA,Zamfara State, Nigeria). "The programme belongs to the community and the community can exclude CDDs from other activities in case both periods clash". (Village Head, Tudun Masu, Bukkuyum LGA, Zamfara State, Nigeria). (WHO,2000c ) The CDDs are even willing to take on other tasks. One CDD in Sudan comments "...and when they call me 'doctor, doctor', I really feel happy. Can I be trained so that I should not only treat onchocerciasis but other health problems in the village". (CDD, Sudan) (WHO, 2000 e) It is evident that in spite of added responsibilities, CDDs effectively perform their CDTI task. "I go for training, collect drugs, call a meeting with the community to decide on time and mode of distribution, distribute the drugs based on height, return the balance ofdrugs to the supervisor, take the census of the people in the community. Shaibu, the second CDD records the distribution" CDD Anka Zamfara ,o 2000c It is interesting to note that treatment coverage of target population in the study areas are quite high; 60-84% in Kisoro, Tl-76%inZamfara, and in Sudan, coverage is reported to have increased from 40o/o to 58Yo (WHO, 2000c; d; e; f,).What comes out clearly from these observations is that performance of CDDs in CDTI may not be adversely affected if the tasks are complimentary and if communities are allowed to decide on the nature and scope of integration. Therefore, the fear of overloading CDDs may be unfounded but could become a reality if implementers exclude communities in the plan for integration. Reports from other projects confirm the findings of the multi-centre study. In PlateauArtrassarawa, Nigeria, CDDs are being used in immunization. In Jigawa and Cross River States, Nigeria, and South West Province in Cameroon they act as case finders for cataract, childhood blindness and refractive errors. In Sudan, CDTI officials have arranged for and implemented numerous "extra-CDTI" health and 7 developmental activities. In Raja, Sudan,292l operations have been conducted in two surgical camps in Raja while in Wau with three surgical camps, 299 operations have been conducted. CDDs in Tanzania, are involved in Vitamin A distribution, maternal and child health, immunization, home visits, health education and simple curative services. In Borno and Adamawa States of Nigeria, plans are on to use CDDs to distribute Vitamin A supplements along with ivermectin. There is a plan in Ethiopia to integrate CDTI into EPI, malaria control, family planning and other PHC programmes. It is clear that the "passive" practical integration observed in the study and evolving in most CDTI projects has arisen not from deliberate programme design but from the decision-making role assigned to the communities. It is predictable that CDDs will continue to be called on to perform other tasks in their communities. Thus introduction of new health programmes are easier to contemplate where there are trained CDDs, and existing programmes can and are being reinforced with the use of CDDs. The collaborative and consolidated use of CDTI resources has begun in many participating countries. For example, motorcycles and bicycles secured through CDTI projects are made use of for immunization during National Immunization Days (NIDs). Coordinators and CDDs involved in other health activities use CDTI motorcycles and bicycles for these activities in most countries. In Sudan, it was possible to quickly immunize 300 children in Diam Zubir using CDTI bicycles. The most noticeable impact that CDTI has made on peripheral health workers is in the "movement" of health workers from the health centers to the communities during ivermectin distribution and routine CDTI supervision. Other health programmes manned by these staff benefit immensely from such closer interaction with the communities. Alleviation of Diseases of the Poor For CDTI to be of maximum benefit to the poor, it must contribute to the solution of the major health problems of the poor. Onchocerciasis by itself is a very debilitating disease, which causes 8 blindness and severe itching and exerts serious social and economic impact on endemic communities. It is estimated that about 50 million people are at risk of infection and 15.5 million are infected in 19 countries. Onchocerciasis endemic areas are characteristically communities "at the end of the road" in very remote inaccessible areas with inadequate health facilities and services. Health problems common to all CDTI areas are indeed health problems of the poorest. WHO (2000b) identifies HIV/AIDS, malaria, tuberculosis, maternal mortality, water borne diseases, respiratory infections and vaccine preventable diseases as the "top killers" of the poor. About 95o/o of all HIV infected persons live in developing countries where 95Yo of all death from the disease also occurs. More than 40o/o of the world's population live in malaria-endemic areas which are usually areas of social and environmental crisis and weak health systems. About 90% of one million persons, who die yearly, live in sub-Saharan Africa where one in five childhood deaths is caused directly by malaria. Indirectly, malaria contributes to illness and deaths from Acute Respiratory Infections (ARI), diarrhoea, malnutrition etc. Tuberculosis and poverty are closely linked. A close association exists between probability of becoming infected and developing clinical disease with malnutrition, overcrowding, poor circulation, inadequate sanitation. There has been an increase of l3oh in TB cases between 1993 and 1996 (WHO, 2000b). It is significant that 30% of AIDS death result from TB. Malnutrition accounts for l5Yo of the global burden of disease and is associated with 49Yo death of children, mainly in developing countries. Most common nutritional disorders are related to def,rciencies of iron, Vitamin A, iodine and protein energy. Maternal mortality remains a major problem in developing countries with an estimated 585,000 women dying yearly from complications of pregnancy and childbirth. In Africa, the risk of such deaths is one in 16. About 2 million lives are lost yearly from water and sanitation related diseases that are also responsible for 1,5 million bouts of illness in children under 5 years of age, and for 75 million Disability Adjusted life Years (DALYS) lost in 1998. One billion 9 persons have no access to safe water and 3 billion with no access to improved sanitation are vulnerable to diarrheal diseases. In 1998, Respiratory infections accounted for 3.5 million deaths among people of all ages worldwide. Two million of these deaths were in children under 5 years and 99%o of the deaths were in developing countries. Factors responsible are mainly environmental (poor housing etc). Vaccine preventable diseases still abound among the poor. In spite of the elaborate immunization programmes one in five children is not fully immunized against diphtheria, whooping cough, tetanus, polio, measles and TB (UNICEF, 2001). WHO recorded a decline in mortality in the 20th century as a result of income growth, improved education, improved food intake, better access to sanitation, increased access to new knowledge, access to new drugs and improved immunization coverage. However, this decline is not significant in developing countries because of death, suffering and general poverty resulting from the major diseases earlier mentioned. This decline may become significant in CDTI areas if prevention and control of these major diseases are integrated into CDTI in a coordinated manner. Opportunities for Integration The potentials of CDTI as an entry point for other health services to the poor derive from the fact that it is based on and commensurate with the philosophy of PHC. CDTI is targeted at one of the main health problems of the poorest communities. The strength of CDTI lies in the fact that it is evidence-based, participatory and uses already existing health structures and personnel. It is for example a vivid demonstration of the empowerment innate in CDTI that apart from taking on the basic (CDTI) responsibilities, listed in the Introduction to this article, the communities have on their own volition exercised the right to make changes to the distribution process to suit themselves, for example to change any CDD whose performance is deemed unsatisfactory, or to modify and redirect the mode or period of distribution. This is the first time a health intervention involving mass treatment is being successfully directed by communities. Indeed, CDTI is one of the few health interventions implemented in 10 real partnership with communities. Other health programmes can take advantage of this and confidently involve communities in programme activities to the extent to which CDTI has done. The nature of onchocerciasis requires that treatment with ivermectin should continue for at least 10-15 years from first treatment. This means that, provided they are sustained and properly maintained, the structures and facilities built around CDTI will be available for use by other health programmes for a long time. Health problems requiring long-term interventions can preferentially benefit from CDTI. APOC through CDTI has fostered a solid inter-sectoral, multi-disciplinary partnership with obvious benefits. Duplication of activities is avoided thereby allowing for optimal use of resources Responsibilities are clearly assigned at all stages of implementation. For example, in drug procurement, storage, and distribution, responsibilities are as shown in the box. Drug-related responsibilities of partners in CDTI: . Drug Requisition D CommunitiesAtOTFs. o Production and shipping D + + D D D Merck Sharp & Dohme (MSD) o Clearing (storage) NOTFsA{GDOs o Transportation to Health Facility S tatelD istri ctlGovernment o o o Collection from Health Facility Communities Cost of distribution in community Communities Cost to individual patient NIL 11 Beyond these major partnerships, a few projects have begun to develop partnerships at the local level. The Cross River State CDTI project in Nigeria is successfully collaborating with the National Park, Forestry Department, Oil Palm and Rubber estates and Community Based Organization during ivermectin distribution in the State. The partnership has resulted in the National Park having a budget line for support of some CDTI activities and other PHC progralnmes in remote communities (WHO, 1999b). The Government of Sudan (GOS) project has collaborated with the Red Crescent, the German Red Cross, the Academy of Medical Sciences and Technology as well as local schools in implementing CDTI in Raja (WHO,20009) Drug management The procurement, delivery and storage of drugs are generally efficient in all CDTI projects. Apart from a few instances, stock taking and requisition is fairly accurate with minimal pilferage, shortage and delayed supply. About 61% of communities collect ivermectin from collection points (Amazigo et al., 2001). At the district level, ivermectin is stored and collected from health facilities manned by PHC staff who are in charge of other health care programmes. This has led to local capacity building on stock management, supervision of ivermectin distribution and handling of adverse effects. Willingness of communities to pick up the drug from collection points is a unique opportunity for health system to involve communities in handling drugs for other health interventions. Capacity building CDTI has resulted in considerable capacity building in participating countries. A total of 67,188 CDDs and 13,795 health workers have been trained on community mobilization, conduct of census, drug dosage determination, management of side effects, exclusion criteria, record keeping and stock management in CDTI (Table 6). In addition to these, a total of 60 persons have been trained in monitoring, 53 in financial management, and 27 in t2 vector elimination (WHO, 2000a). These large numbers of trained personnel are available to other health programmes within an integrated health care delivery system. CDTI has succeeded in mobilizing 49,654 communities towards optimum participation in a health intervention. If pursued in the right way through community empoweffnent and all it entails, it would be relatively easy to mobilise the same communities for other health interventions. An important lesson learnt from CDTI, which is useful for other health interventions, is that provision of relevant and comprehensive information to communities enhances community participation. It has been observed that communities that score high on community perception, expectation and satisfaction are supervised by health officers who provide adequate information on CDTI. (WHO, 2000h). CDTI Data Base Proper implementation of CDTI automatically provides accurate census data useful for planning, implementation and monitoring of other health programmes. Most health prograrnmes require such data that are most often unavailable. In the process of updating registers, CDDs interact with community members and are in a position to identify and offer help in solving other health problems among community members. The CDTI mode of drug distribution has been shown to be effective and efficient. It can serve as a model for use by other health programmes. Similarly, the referral component of CDTI has the potential of strengthening referral facilities that can be used for other health interventions. Health Surveillance Potential Record keeping and reporttng skills acquired by the CDDs are also useful for other health intervention programmes. Informations generated during CDTI are equally useful for planning other interventions. Ability of community members to keep records is an incalculable asset and all community-based public health activities. l3 Monitoring By the end of December 2000, 39 CDTI projects had been monitored. Recently a community self-monitoring procedure was introduced and is being tried out in a number of projects. It entails empowering the community, through the provision of skills and knowledge, for monitoring and reporting events and activities in the community. It has been found to be useful in promoting community ownership and capacity building. Lessons learned during monitoring of CDTI can be used in other programmes. Resources. There has been a gradual build up of resources provided by the partners in CDTI project sites. Notable are transport facilities including vehicles, motorcycles and bicycles, as well as office equipment like computers, photocopiers, fax machines and audiovisuals. Since CDTI is operating within national health systems, these resources have further strengthened these systems. An opportunity exists therefore for other health progralnmes to share these facilities. However, as expressed during the Eleventh Session of the Technical Consultative Committee (TCC) of APOC, and reiterated in the Report of the External Mid-term Review (WHO, 2000, "While APOC Programme document (Phase I of 1996) calls for integration generally, it does not give APOC governing bodies, including TCC, a mandate to allow use of APOC resources to support activities other an onchocerciasis". APOC would need to address this need in Phase 2. Conlidence buildtng. CDTI has succeeded in building up confidence among community members in the health system, by bringing a health intervention close to them - a privilege most onchocerciasis endemic communities have never enjoyed in the past. In Raja, Sudan, ivermectin is the only drug in the health kit unlike in other places where ivermectin is merely one additional drug in the health kit. This can inspire confidence in other health programmes and can contribute to the enhancement of the relationship between communities and health workers. Thus communities have been re-orientated to accept the responsibility for their well being and provided with an opportunity for becoming more actively involved in solving their many other health problems. t4 Advocacy. For any CDTI project to successfully act as entry point for other health programmes, it must operate with all elements of CDTI in place. Therefore, recommendations made after monitoring and evaluation exercises should be implemented. These include sustaining drug delivery, ensuring adequate funding, intensifying community mobilization, improving coverage as well as improving quality of training, record keeping and supervision. Prerequisite for this is the need to sensitize policy makers, health workers and communities on the value of integration and the potential of CDTI as entry point for other health programmes. Stakeholders meetings will be useful channels for sensitization. Funding. Most CDTI projects at present depend on APOC funding and are not adequately budgeted for within PHC. Support for CDTI by Governments with few exceptions is limited to provision of physical facilities and salaries for personnel. Funds are usually not provided for activities such as training, mobilization, monitoring and supervision. In some projects, CDTI is reflected as a one line item in the budget but no funds are thereafter released. In others, funds are released outside PHC. If projects are adequately budgeted for and funded within PHC then funding related weaknesses in CDTI will be removed and the judicious use of PHC funds realizable. Training. Joint training of all cadres of health workers on CDTI and the control of other major health problems of the poor, will promote and enhance integration. The more participatory the trainings are, the more opportunities the health workers will have to work out best ways of collaborating during mobilization, health education, drug procurement, reporting, referral, monitoring and supervision in implementing health programmes. Evaluations of experiences in integration Evaluation of health services at all levels is crucial to planning. It is particularly important to carefully study the ways by which communities, without external influence, have promoted integration of health services. Experiences in Cross River State, Nigeria, (WHO, 20007) in 15 integration are of particular interest. This project actively promoted integration and inter-sectoral collaboration of PHC programmes, through meetings and workshops for state programme managers of NPI, nutrition, the Bamako Initiative, HIV/AIDs, malaria, Local Government Area (LGA), PHC coordinators and LGA CDTI team leaders. The aim was to update their skills in communication; share experiences on opportunities provided by CDTI to other health prograrnmes and adopt a standard protocol for community entry, mobilization and education. The outcome is better planning in channelling of resource and better coordination of PHC progralnmes. An attestation to that fact is the report by the PHC coordinator in Obudu LGA, who affributed the success to the National Immunization Day (NID) in the area to the adoption of the CDTI strategy. Stakeholders meetings already conducted at the state capital and communities in health districts have been found to be useful in improving collaboration among partners, and providing a forum for feedback on CDTI activities. As suggested in a report by Tarimo (WHO, 2000k) the potential for the use of schools in promoting health has not been fully exploited in developing communities. The CDTI programme is yet to take full advantage of this potential. Distribution of ivermectin is predominantly household-based. The option of effecting distribution through schools is hardly ever considered. However, health education and community mobilization messages can be effectively transmitted through schools. School age children constitute a significant proportion of the population in developing countries and the proportion is increasing with the success of the child survival prograrnmes of UNICEF and other Agencies. This means that the largest gathering of children will continue to be in schools. Though the highest rates of overall morbidity and mortality in children, from all causes, are recorded in age group five and below, the school age group also bears a heavy burden of disease, which hinders their growth and development. School-based health interventions have the potential of being cost effective. There are more schools than 16 health clinics and more teachers than health workers. Teachers and pupils can be used as channels for passing health messages to parents, peers, siblings and community members. CONCLUSIONS The activities of APOC in the first phase of its operations have had a profound effect on the health services of participating countries. Attitudes at all level have become more favourably disposed towards extending health care to the poorest and most neglected communities where characteristically onchocerciasis is highly prevalent and whose populations have to cope with myriad health problems hitherto ignored by the authorities. As a result of the empowernent that has come through the operations of CDTI the communities are beginning to realise and develop their latent potential for taking effective charge of their health and development problems. To consolidate this trend it is necessary that the partnership that is the driving force behind APOC continues to invest in the venture, and that the integration of CDTI into existing health care programmes be pursued with vigour by the Ministries of Health (MoH) and their partners. Lapses in budgetary provisions and inadequacy of support that have compromised the effective execution of CDTI and consequently treatment coverage in some projects need to be addressed during APOC Phase 2. When these and other issues identified in this paper have been successfully addressed then the establishment of a self-sustainable programme would be a reality. ACKNOWLEDGEMENTS The authors are grateful to Dr. B. Boatin, Director, Onchocerciasis Control Programme (OCP) in West Africa for his valuable comments and suggestions on the manuscript. We appreciate the co-operation and valuable information received from National Onchocerciasis Task Forces (NOTFs) of Sudan, Nigeria, Cameroon and Uganda. We are also grateful to Mrs. Patricia Mensah of APOC for her secretarial assistance. T7 REFERENCES AMAZIGO, U.V., OBONO, O.N., DADZIE. K.Y., REMME, J., JIYA, J., NDYOMUGYENYI, R., ROUNGU, J-B., NoMA, M. & sEKETELT, A. (2001). Monitoring community-directed treatment programmes for sustainability: lessons from African Programme for Onchocerciasis Control (APOC). Annals of Tropical Medicine and Parasitologt, ?95, (Suppl.),???-??? uNrrED NArroNS cHTLDRENS FUND (2001). The State of the World's Early Childhood, Children. New York: UNICEF. ??? woRLD HEALTH oRGANISATIoN ( 1978) Report of the International Conference on Primary Health Care, Alma Ata. 1978. Geneva: WHO woRLD HEALTH oRGANISATToN (1996o). African Programme for Onchocerciasis Control. Programme Documentfor Phase 1. (1996-200I), (Unpublished Document) woRLD HEALTH oRGANISATIott (1996b). Report of a Multi-Country Study on Community- Directed Treatment with lvermectin. toUtrn/np/96.1. Geneva: WHO. woRLD HEALTH oRGANISATIoN (1999c). Progress Report, Joint Action Forum, Fifth session, The Hague, The Netherlands,1999. wuo/nPoc/Mc/99. 4. Ouagadougou: APOC. woRLD HEALTH oRGANrsATroN (1999b). Cross River State, Ntgeria, Project Technical Report. Apoc/NIG/NorF/cRo/99. 1. Ouagadougou: APOC. woRLD HEALTH oRGANISATToN (2000a). Joint Actton Forum Report sixth session, Yaoundd, 2000. wHo/Apoc/MG/ 00. 1 6. Ouagadougou: APOC. woRLD HEALTH ORGANISATTOT.T (2000b), Health - A precious Asset. Geneva: WHO. woRLD HEALTH oRGANIZATIoN (2000c). Report on Community-Directed Distributors Involvement in other Health and Development Projects in Zamfara State, Nigeria 2000. wHo/Apoc/cs/00.6. Ouagadougou: APOC. 18 woRLD HEALTH ORGANISATION (2000d). Report on Assessment of Community-Directed Distributors' (CDDs) Involvement in other Health or development activities in Kisoro District, South Western Uganda, 2000. wttoiapoc/cs/00.7. Ouagadougou: APOC. WORLD HEALTH ORGANIZATIoN (2000e). Report on Assessment of Community-Directed Distributors (CDDs) Involvement in other Health or development activities in Sudan. wHo/Apoc/cs/00.8. Ouagadougou: APOC. wORLD HEALTH ORGANISATIoN (2000r. Report on CDDs Involvement in other Health Care and Development Acttvtties in Cameroon, 2000. wHo/APoc/cs/00.9. Ouagadougou: APOC. woRLD HEALTH ORGANISATIoN (20009). Government of Sudan Project Technical Report. wHo/Apoc/suD/ NT/00. l. Ouagadougou: APOC. woRLD HEALTH oRGANISATIoN (2000ft). Reports of Independent Monttortng of CDTI Projects in U ganda. wHo/Apoc/cs/00.4.Ouagadougou: APOC. WoRLD HEALTH oRGANISATIoN (2000r). Report of Mid-term (Phase I) External Evaluation of APOC. wHo/Apoc/cs/00.3. Ouagadougou: APOC. woRLD HEALTH oRGANISATIoN (2000j). Cross River State Project Technical Report. Apoc/NIG/NorF/cRo/00. I . Ouagadougou: APOC. woRLD HEALTH oRGANISATIoN (2000k). Final Report on the Sustainability of Community- Directed Treatment of Onchocerciasis with lvermectin. wno/epoc/cs/00.1. Ouagadougou: APOC. t9 Tau,B I in Health Care Adapted from WHO, 2000b, Health: a precious assel Tasr,B 2 n of CDTI into the Health Source: APOC 2000 Partner's Meeting (Unpublished document). Tlsr,r 3 Proportion of onchocerciasis coordinators at local level involved in control activities for other diseases in four APOC countries No. in sample oZ of Oncho. Coordinators involved in control of other diseases Country CNo. projects) Malaria Epilepsy TB Leprosy Aids of Nigeria Malawi Usanda le0 (16) 400 (2) 9/2). 96 99 67 46 100 100 43 98 11 33 0 0 26 0 0 Developed Countries Developing Countries o I nurse per 130 persons 1 nurse per 5000 personso . I pharmacist for 2000 - 3000 persons o 1 pharmacist per I million persons o I course of antibiotics - 2-3 hrs wages 1 course of antibiotics - 1 month wagesa o I year treatment for HIV - 4-6 months wages o 1 year treatment for HIV - 30 years wages o MaioriW drug cost - reimbursed o Drug cost not reimbursed o Government expenditure on health - high o Government expenditure on health -low Actions Countries Onchocerciasis control activity as a line item in PHC budget Democratic Republic of the Congo (DRC), Nigeria, T anzania, Uganda. Oncho. control activities/unit integrated in Disease surveillance/PHC Programme CAR, Cameroon, DRC, Ethiopia, Liberia Malawi, Nigeria, T anzania, Uganda. Ivermectin included in Essential Drug List (EDL) CAR, Cameroon, Chad, Nigeria, Tanzania, Uganda. Ivermectin integrated in National Drug Delivery System Cameroon, Liberia, T anzania, Uganda. Total s99 (20) 262 (43.7) 246 (41) Lsz (2s) NB. Some Coordinators are involved in more than one activity - (Source WHO, 1999a) 26 (4.3) 33 (5.5) 20 Tasln 4 Other Health Assignment of CDDs Some CDDs are simultaneously involved in more than one activity Source - WHO (2000), APOC Report Multi-centre Study. Activities Number (%) of CDDs involved Kisoro, Uganda N:47 swl, Cameroon N:50 Raja, Sudan N:60 Zamfara, Nigeria N:284 Traditional Medicine | (2) Community Health Work 34 (72\ STDS Control | (2\ 4 (8) Family Planning 6 (13) Malaria Control 4 (e\ 1s (3) Polio Control s (11) Village Health Committee | (2) Water & Sanitation r (2) Traditional Birth Attendant 4 (e\ rB (DOrS) 2 (4) Vitamin A Distribution 6 (10) Immunization le (38) l8 (30) e0 (2) Guinea Worm Eradication 4 (7\ s8 (20) HIV AIDS Control e (18) 2 (3) First Aid 2 (4\ 2s (42) Environmental Sanitation 4 (7\ Leprosy Control 6 (r2\ | (2\ Nutrition L (2) r (2) Epanutin 29 (48) PHC 8 (3) 2l Tanle 5 Develo ment Activities of CDDs Source - WHO/APOC (2000), Report on Multi-centre Study. Tesr-B 6 CDD and Health personnel trained under CDTI in 8 countries in 2000 Countrv CD Health worker Cameroon CAR Liberia Malawi Nigeria Sudan Tanzarria Ueanda 3,724 4,290 1,,525 1,7l0 44,723 1,232 748 7.236 1,218 534 tt2 420 9,305 t26 1,452 628 Total 67.188 13.795 Number (%) CDDs involvedActivities Kisoro, Uganda N=47 sw1, Cameroon N:50 Raja, Sudan N:60 Zamfara, Nigeria N=284 Community Agricultural Schemes 17 (33) 16 (3 l)Community Leadership Credit & Saving Schemes 11 (21) 9 (r1\Health Care Schemes (Engozi) Labour It (23',) t0 Q2\Church Leadership/Ivlembership Water Supply 5 (10) 8 (13) 3 (6) 11 (18)Education (PTA) Villaee Development Committee 6 (12\| (2\Maintenance of Health Center rt (23) 14 (23)NGO/CBO Work e (3)Vigilante Islamic First AID r21 (43) 2s2 (88) Construction (Road, School, Mosque, market, Drainage) 17 6\Community Secretary 22 ACRONYMS USED IN THE ARTICTES OF THE SUPPTEMENT t

Acronyms used in the articles of the Supplement AFRO APOC ATP CAR CBD CBHP CBM CBIT CDD CDTI CHAL CHW CIH ComDT CSA DEC DRC EAC EPI ERR FCT FGD FLHF GIS GOS HNI HOD IEC IDP ITN African Regional Office of WHO African Programme for Onchocerciasis Control Annual Transmission Potential Central African Republic Community-Based Distribution Community-Based Health Pro gramme Christoffel Blinden Mission Community-Based Ivermectin Treatment Community-D irected D istributor Community-Directed Treatment with Ivermectin Christian Health Association of Liberia Community Health Worker Community Involvement in Health Community-Directed Treatment Committee of Sponsoring Agencies Diethylcarbamazine Democratic Republic of the Congo Expert Advisory Committee (of OCP) Expanded Programme of Immunisation Economic Rate of Return Federal Capital Territory (of Nigeria) Focus Group Discussion Font Line Health Facility Geographical Information System Government of Sudan HealthNet International Head of Department Information, Education and Communication Ivermectin Distribution Programme lnsecticide Treated Nets TRD JAF JPC KAP LGA LF LOCT MDP MITOSATH MoH MOU MSST MTP NGDO NID NOCP NOTF NPI OCCGE OCP OLS ONCHOSIM ooR OR ORSTOM Institut de Recherche pour le Developpement Joint Action Forum (of APOC) Joint Programme Committee (of OCP) Knowledge, Attitude, Practice Local Government Area Lymphatic Filariasis Local Onchocerciasis Control Team Mectizan@ Donation Programme Mission to Save the Helpless Ministry of Health Memorandum of Understanding Motion Sensitivity Screening test Monthly Transmission Potential Non-Governmental Development Organisation National Immunisation Days National Onchocerciasis Control Programme National Onchocerciasis Task Force National Programme on Immuni zation Organisation de Coordination et de Coop6ration pour la Lutte contre les Grandes End6mies Onchocerciasis Control Programme in West Africa Operation Lifeline of Sudan Onchocerciasis (computer) Simulation Model Onchocerciasis Operational Research (Task Force of TDR) Operational Research Institut Frangais de Recherche Scientifique pour le D6veloppement en Coop6ration (now IRD) Preparatory Assistance to Governments (Mission for) Prevention of Blindness Programme (of the WHO) Primary Health Care Rapid Epidemiological Assessment Rapid Epidemiological Mapping of Onchocerciasis Serious Adverse Effects Serious Adverse Reactions PAG PBL PHC REA REMO SAE SARs SHM SOCT SSE SSI STP TCC TDR TFR UNDP UNICEF wHo WR wzu Stakeholders Meeting State Onchocerciasis Control Team Serious Side Effects Sight Savers International Short-Term Professional Technical Consultative Committee (of APOC) UNDPAVoTId BankAVHO Special Programme for Research and Training in Tropical Diseases Total Fertility Rate United Nations Development Programme United Nations Children's Fund World Health Organisation World Health Organisation Country Representative World Resources Institute