Water level regulation and control of schistosomiasis transmission: a case study in Oyan Reservoir, Ogun State, Nigeria.

Water level regulation and control of schistosomiasis transmission: a case study in Oyan Reservoir, Ogun State, Nigeria I.E. Ofoezie1 & S.O. Asaolu2 The effect of different water discharge pattems from the artificial Oyan Reservoir in Ogun State, Nigeria, on water level fluctuations and on schistosomiasis transmission was investigated between August 1990 and March 1993 to determine the impact of water level regulation on schistosomiasis transmission and control. The results show that transmission was greatly influenced by the pattern of water discharge during the hot dry season (January-April). A high discharge during this period ofno rainfall, high temperatures, and intense sunshine stimulated rapid water level fluctuations and lake draw-down, which led to significant reductions in all indices of schistosomiasis transmission, i.e. snail density, snail infection rates, human water contact patterns, and incidence of infection. Although these results support continuous water discharges from the reservoir during the hot dry season, this may run counter to current water management policies. Further investigation is therefore required to harmonize the potential benefits in this type of discharge pattem with the objectives of efficient water management in artificial reservoirs. Introduction The association between schistosomiasis transmis- sion and water resources development projects such as reservoirs, irrigation schemes, etc. is well estab- lished (1-6). Earlier attempts to develop suitable engineering methods to control transmission were hampered by many factors, including the misconcep- tion that schistosomiasis is an unavoidable accom- paniment of water development projects (7) and a shift in emphasis from control of transmission to morbidity control (8). This shift was due to im- portant advances that simplified the diagnosis of schistosomiasis (9, 10), and to availability of safe and effective oral drugs for population-based chemotherapy (11, 12). However, the inevitability of reinfection has meant that mass drug treatment can only temporarily reduce the prevalence to a level that corresponds to the site-specific environ- mental risk factors (13-15). This failure of chemo- therapy intervention to sustain the benefits of disease control has revived interest in transmission control, but within the context of an integrated con- trol approach (16). 1 Research Fellow, Institute of Ecology and Environmental Stud- ies, Obafemi Awolowo University, lie-Ife, Osun State, Nigeria. Cor- respondence and requests for reprints should be sent to this author. 2 Associate Professor and Head, Department of Zoology, Obafemi Awolowo University, Ile-Ife, Osun State, Nigeria. Currently: Direc- tor, Institute of Ecology and Environmental Studies, Obafemi Awolowo University, Ile-Ife, Osun State, Nigeria. Reprint No. 5798 Water level fluctuation has long been recog- nized as an effective control method against snails and snail eggs, and against mosquito larvae (17-19). Widescale application was, however, constrained by limited understanding of the basic relationship between such control and water release patterns, and the effect of this interplay on reservoir draw- down, human water contact patterns, and schisto- somiasis transmission during different seasons of the year. To obtain information that could enhance our understanding of the foregoing, and hence facilitate transmission control by water level fluctuation, we examined the effect of water release patterns on water level fluctuation and disease transmission in a typical man-made reservoir. The investigation was carried out in Oyan Reservoir, a medium-size artifi- cial lake in south-west Nigeria, over a period of 34 months (August 1990 to May 1993). As this reservoir is typical of many other small and medium-size reser- voirs in the schistosomiasis endemic areas of Africa, our results and follow-up studies will have a bearing on many cases concerning recommended water man- agement regimens. Materials and methods Study area and reservoir The Oyan Reservoir is located at lat. 7°14'N, long. 3°13'E near Abeokuta, the capital city of Ogun State, Nigeria (see 20 for a detailed description of Bulletin of the World Health Organization, 1997, 75 (5): 435-441 © World Health Organization 1997 435 I.E. Ofoezie & S.O. Asaolu the area). The reservoir was established in 1984 with, lying adjacent, two resettlement communities named Abule-titun and Ibaro. An outbreak of urinary schistosomiasis occurred in the two communities within 4 years of the reservoir being established, probably as a result of the impoundment (20). The dam has an embankment crest length of 1044m, a height of 30.4m, four spillway gates (each 15m wide and 7m high), and three outlet valves (each 1.8m in diameter). The reservoir has a surface area of 40 kM2, a gross storage capacity of 270 million m3, and a dead storage capacity of 16 million m3. The reservoir is used for breeding fish, sprinkler irrigation, and water supply to downstream end-user Water Boards in Abeokuta and Lagos metropolitan cities. Water discharge is by a regulated opening of the gates and valves. The gates and valves are usually operated at 8.00 a.m. when the percentage opening for each day is set and the water level is read off an automated gauge. Water discharge during each opening period was read off an operating chart provided by the construction engineers. Transmission studies The transmission of urinary schistosomiasis was assessed by the presence, number, and the rate of Schistosoma infection of Bulinus globosus, the local snail intermediate host; by the human water contact pattern; and by the incidence and intensity of infec- tion among schoolchildren in the two communities. The detailed methods used to carry out these inves- tigations are described below. Snail density. The relative density of B. globosus was determined every month in each of six human water contact sites using a standard technique involving 30 passes of Hairston drag scoop, supple- mented by a manual search for 30 person-minutes. The number of snails per month was the total col- lected in all sites during each month's survey. The snails were examined for patent Schistosoma infec- tion using the shedding method (21). Human water contact pattern. People's activities in the six human water contact sites were studied once every 3 months between 7 a.m. and 7 p.m. from February 1991 to November 1992. Information was collected on all persons entering and/or leaving the water (sex, age, activity, time of entering and leaving the water, and the parts of the body exposed). Each exposure type was assigned an activity coefficient for estimating the relative index of exposure (RIE), i.e. the product of activity coefficient and duration of contact (22). Human infection pattern. Urine samples were col- lected from 118 schoolchildren (5-14-year-olds) in Abule-titun and Ibaro primary schools in April/ May 1991. The samples were examined for S. haematobium eggs using the sedimentation-by- gravity concentration method (23, 24). A total of 77 children who were found to be uninfected by this method were re-examined in April/May 1992 to determine the incidencea and intensity (eggs/10 ml urine) of any new infection during the 1991-92 transmission season. All the children found infected in the first and second examinations were treated immediately after the second examination and re- examined 4 weeks later in order to identify those children who had stopped excreting eggs. Those children who became negative by treatment were followed as cohort A, while those who had remained uninfected since 1991 were followed separately as cohort B. Members of the two groups were re- examined in April/May 1993 to assess the incidence and intensity of new infection during the 1992-93 transmission season. The geometric mean intensity of infection was calculated after transforming egg counts to log(x + 1) to include zero counts and to normalize the variance. Differences in incidence were examined using x2 tests, while Student's t-test was used to com- pare intensities of infection during the two transmis- sion seasons. Results Water discharge and their ecological consequences The pattern of water discharge in Oyan Reservoir since impoundment in 1984 was highly seasonal and regular every year (Fig. 1). Usually, the reservoir gates are closed from the onset of the dry season in November to the onset of the rainy season in April. During this period, only a limited discharge is al- lowed through the outlet valves. As rainfall in- creases, the gates are gradually opened, from 2% to 30%. Correspondingly, the quantity of water dis- charged increases gradually, reaching a maximum at peak rainfall (either in July or September). With the onset of the dry season, the gates are again gradually closed, reducing the discharge to a minimum in the hot dry season (January to March). a Incidence was calculated as i = 100 (p/n), where n = number of children not infected in the first examination, and p = number infected in the second examination. 436 WHO Bulletin OMS. Vol 75 1997 Fig. 1. Monthly variations in the water level, in the number of discharge days per month, and in the quantity of water discharged each month from Oyan Reservoir, from January 1984 to March 1993. water leveml 1 l. DI i g 0aD0ni.~LIfL J FMAMJJ ASONDJ FMAMJJ ASONDJ FMAMJJ ASONDJ FMAMJJ ASONDJ FMAMJJ ASONDJ FMAMJJ ASONDJ FMAMJJ ASONDJ FMAMJJ ASONDJ FMAMJJ ASONDJ FM 1984 1985 1986 1987 1988 1989 1990 1991 1992 93 Year and month I an ...i iin [ - -70 -60 -50 -40 C CD 30 Il 20 ~o-10 0 .-0 i Table 1: Data on water level fluctuations and November 1992 meteorological factors in Oyan Reservoir from August 1990 to Rate of No. of Maximum draw-down/ Year and Rainfall Temperature days Water level draw-down water rise month (mm) (OC) gates open (m elevation) (m) (m/day) 1990: Aug. 30 36.3 0 58.4 0 0 Sept. 184.6 36.0 0 59.8 0 0 Oct. 105.8 35.6 26 62.9 0 0 Nov. 0 36.0 16 62.9 0 0 Dec. 115.5 36.3 2 62.8 3 0.10 1991: Jan. 0 36.2 0 62.8 8.6 0.18 Feb. 33.0 35.9 15 62.6 21.1 0.45 Mar. 45.7 37.4 31 62.0 37.3 0.52 Apr. 165.1 38.8 30 61.3 43.7 0.21 May 92.6 38.1 31 60.8 47.1 0.11 Jun. 118.8 37.7 30 61.8 36.6 -0.35 Jul. 189.4 38.9 31 62.8 12.9 -0.76 Aug. 68.9 38.7 27 62.3 1.0 -0.38 Sep. 189.7 38.6 25 62.8 0 -0.03 Oct. 148.4 38.2 26 62.8 0 0 Nov. 0 37.8 10 62.6 8.9 0.30 Dec. 0 37.6 29 62.2 25.9 0.55 1992: Jan. 0 39.0 31 60.9 47.0 0.68 Feb. 0 42.1 29 58.8 84.4 1.34 Mar. 12.3 41.0 31 57.8 112.4 0.90 Apr. 93.3 39.2 30 57.3 115.6 0.10 May 105.2 36.5 21 56.8 94.1 -0.69 Jun. 165.3 34.5 0 56.8 51.0 -1.44 Jul. 169.5 32.9 13 57.2 8.4 -1.37 Aug. 29.3 31.1 0 58.6 0.6 -0.25 Sep. 259.4 33.3 8 61.1 0 0 Oct. 138.5 35.8 31 62.8 0 0 Nov. 10.2 37.2 11 62.4 0 0 WHO Bulletin OMS. Vol 75 1997 800 - 0- -I 600- Co (o cn c o a ct 400- Xcnz 0. o- Z6 n i- LRLMR-a-9URAIIIsza--J..1111I.Allia - ,L_s.in l-so os-au -- -In INN,, D iina I. Nomem I 437 I.E. Ofoezie & S.O. Asaolu This pattern was altered in 1992 when, unlike the preceding years, the gates were opened every day from December 1991 to April 1992. This gave rise to a high discharge pattern throughout the hot dry season when water was usually conserved. Table 1 shows that this high discharge may have been prompted by a drought which occurred between December 1991 and April 1992, when no single epi- sode of rainfall was recorded and the temperature was exceptionally high. As a result, the mean monthly lake level in the hot dry season of 1992 was more than 3m lower than its 1991 level and the water level fluctuated at 1.03m/month (3.4cm/day), which was twice its fluctuation rate in 1991. These occurrences had important ecological consequences: the horizontal shoreline at the water contact sites receded by 134cm/day compared with 52cm/ day in 1991, snail density was 351 in the hot dry season of 1991 compared with 173 for the same pe- riod of 1992, and no infected snails were found in 1992 compared with a rate of 8% in 1991. The rapid lake draw-down and consequent fast receding shore- line also caused the water contact sites to shift regu- larly, thereby altering the depth and topography of the contact sites. Consequently, the frequency of human water contact decreased from 206 in the hot dry season of 1991 to 173 over the same period of 1992; other decreases were in the mean duration from 14.4 minutes/contact in 1991 to 9.6 minutes/contact in 1992 and in the relative index of exposure from 1463.3 in 1991 to only 723.9 in 1992 (Table 2). Incidence and intensity patterns The annual incidence of new S. haematobium infections among schoolchildren in the two schools investigated is summarized in Table 3. Incidence among untreated children decreased from 66.2% during the 1991-92 transmission season to 10.0% in the 1992-93 season. Among 79 treated children the incidence by the end of the 1992-93 season was only 39.2%. Compared with the 1991-92 season, the level of reduction achieved in the 1992-93 season among the treated and untreated children was significant. The geometric mean intensity of infection de- creased from 25.5 eggs/lOml urine in 1991-92 to 1.3/ 10ml and 6.3 eggs/lOml urine among the untreated and treated children, respectively. These changes were highly significant (P < 0.01). Individual intensities were also heavier during the 1991-92 season, with more than 80% and 25% of the infected children, respectively, excreting over 50 eggs/10ml and 500 eggs/lOml urine compared with 66% and 3% in the 1992-93 transmission season. Table 2: Seasonal variation in the indices of water contact and in the number and infection pattern of B. globosus at the water contact sites in Oyan Reservoir in 1991 and 1992 Total Relative B. globosus No. of duration index of Season contacts (min) exposure No. collected No. infected % infection 1991: Hot dry 206 2974 1463.3 351 28 8.0 (Jan.-Mar.) Early rain 229 2412 939.3 288 10 3.5 (Apr.-Jun.) Rain 131 1036 439.8 833 17 2.0 (Jul.-Sept.) Early dry 155 1390 480.4 682 NEa NEa (Oct.-Dec.) P-value <0.001 <0.001 <0.001 <0.001 <0.001 1992: Hot dry 173 1653 723.9 124 0 0 (Jan.-Mar.) Early rain 142 1543 444.2 146 1 0.7 (Apr.-Jun.) Rain 121 913 241.6 671 21 3.1 (Jul.-Sept.) Early dry 83 817 252.6 402 15 3.7 (Oct.-Dec.) P-value <0.001 <0.001 <0.001 <0.001 <0.001 Total 1240 12738 4985.1 3497 92 2.8 a NE: not examined. WHO Bulletin OMS. Vol 75 1997438 Schistosomiasis transmission and water level regulation Table 3: Incidence and intensity of S. haematobium infection among schoolchildren examined in Abule-titun and Ibaro primary schools during the 1991-92 and 1992-93 transmission seasonsa 1991-92 1992-93 Untreated Untreated Treated Total Subjects (a) (b) (C) (d = b + c) No. negative 77 20 79 99 No. positive 51 2 31 33 Incidence (%) 66.2 10.0 39.2 33.3 Intensity (eggs/10ml urine): Geometric mean 25.5; 0-1672' 1.3; 0-64 6.6; 0-1584 4.6; 0-1584 No. infected with: <50 eggs/10ml urine 6 (19.4)c 1 (50) 10 (32.3) 11 (33) >50 eggs/10ml urine 25 (80.6) 1 (50) 21 (67.7) 22 (66) <500 eggs/10ml urine 23 (74.2) 2 (100) 30 (96.8) 32 (97) >500 eggs/10ml urine 8 (25.8) 0 (0) 1 (3.2) 1 (3) a Results of statistical comparison of 1991-92 and 1992-93 infection patterns: Incidence Geometric mean intensity a x b: X2 test = 16.15, P < 0.001 a x b: t test = 4.99, P < 0.001 ax c' X2 test = 6.99, P < 0.01 a x c: t test = 2.96, P < 0.01 a x d: X2 test = 11.96, P < 0.001 a x d: t test = 3.81, P < 0.001 b Figures in italics are the range. c Figures in parentheses are percentages. Discussion The consequences of the unusual water discharge pattern in 1992 were not at all expected. The emer- gency measure was taken by those responsible for the Oyan Dam Project to offset the effects of the drought on the downstream-receiving water boards. The dramatic health and ecological impact of this action was therefore unplanned and totally unexpected. As shown in our results, the reservoir draw- down was more highly pronounced than hitherto and produced significant declines in all indices of schistosomiasis transmission snail abundance, snail infection rates, and human water contact patterns. Subsequently, the incidence and intensity of new infections among schoolchildren declined significantly, suggesting that a fall in transmission potential may have occurred between the 1991-92 and 1992-93 seasons. The exceptional drought con- ditions, which led the reservoir management to change the conventional water release pattern, thus provided a rare opportunity to compare the health impacts of two opposing water release patterns. The usual policy of conserving water dur- ing the dry periods poses greater risks to human health, while the unconventional recommendation to allow continuous releases of water at such periods appears to protect human health. However, the latter policy is associated with a high rate of water loss, which goes against efficient water management in reservoirs. These findings point to a conflict between the need for water conservation and health protection in reservoir management. Resolution of this conflict requires further studies to shift the balance towards minimal water loss and maximum health benefits. These studies demand both engineering and biologi- cal inputs and, as emphasized by Ackers & Smith (25), call for multidisciplinary collaboration in disease research. One option for collaborative research is the de- velopment of decision-making criteria for water- release patterns. In this regard, the levels of trans- mission associated with a spectrum of releases will be determined and used to formulate health mitigat- ing and operationally sound water release guide- lines. The other option is to reduce the marsh potential, i.e. the ability of the draw-down area to retain water, at identified transmission sites. This will involve application of various environmental man- agement measures such as modifying the shore- line to enhance the effects of reservoir draw-down, subdividing the reservoirs in compartments so that water can be pumped from one part to the other, causing water-level fluctuation with minimal overall water loss, and/or constructing multiple dams downstream to conserve water releases upstream (see 7,19 and 26 for further information on environ- mental management techniques). Although the mul- tiple dam option has been used to great benefit in the Nile delta region (27, 28), detailed studies are still required to understand the cost-effectiveness of the various possibilities. WHO Bulletin OMS. Vol 75 1997 439 I.E. Ofoezie & S.O. Asaolu Acknowledgements We are grateful to the following: Ogun Osun River Basin Development Authority, Abeokuta, Ogun State, Nigeria for permission to undertake this study in Oyan Reservoir; Dr N.0. Christensen and Dr H. Madsen, Danish Bilharziasis Laboratory, Charlotteniund, Denmark, for their useful comments; Dr E.l. Obiajunwa and Dr O.B. Obioh, Center for Energy Research and Development (CERD), Ile-Ife, for assistance in word processing; and the World Health Organization for financial assistance. Resume Regularisation du niveau de l'eau et lutte contre la transmission de la schistosomiase: le cas du reservoir d'Oyan, Etat d'Ogun, Nigeria L'effet des diff6rences de d6bit du r6servoir artificiel d'Oyan, Etat d'Ogun, Nig6ria, sur les fluctuations du niveau de l'eau et la transmission de la schisto- somiase a 6te 6tudie entre ao0t 1990 et mars 1993. L'etude avait pour objectif de determiner l'impact de la regularisation du niveau de l'eau sur la transmis- sion de la schistosomiase et la lutte contre cette maladie. Les r6sultats montrent que la transmission est fortement influencee par le d6bit du reservoir pendant la saison chaude et seche (janvier a avril). Pendant cette p6riode ou les precipitations sont absentes, la temp6rature 6lev6e et l'ensoleillement intense, les d6versements d'eau a debit 6lev6 pro- voquent une fluctuation rapide du niveau de 1'eau et une vidange du lac, avec comme consequence une diminution sensible de tous les indices de transmis- sion de la schistosomiase, a savoir la densite de gasteropodes, le taux d'infection des gast6ropodes, les modalites des contacts homme-eau et l'inci- dence de l'infection humaine. Bien que ces r6sultats plaident pour un d6versement continu pendant la saison chaude, cette pratique irait a l'encontre des politiques actuelles en matiere de gestion des resources en eau. 11 est donc n6cessaire de poursuivre ces travaux afin de parvenir a un 6quilibre entre les benefices potentiels de ce mode d'exploitation des r6servoirs et les objectifs d'une gestion efficace des ressources en eau. References 1. Jenkins DW. Bilharzia defeats development. Lakartidningen, 1972, 69: 2822-2827. 2. Waddy BB. Research into the health problems of manmade lakes, with special reference to Africa. Transactions of the Royal Society of Tropical Medi- cine and Hygiene, 1975, 69: 39-50. 3. Gilles HM. Health implications of man-made water resources. African health, 1980, 3: 19-21. 4. Hunter JM, Rey L, Scott D. Man-made lakes and man-made diseases. Social science and medicine, 1982, 16: 1127-1145. 5. Jewsbury JM, Imevbore AMA. Small dam health studies. Parasitology today, 1988, 4: 57-59. 6. Hunter JM et al. Parasitic diseases in water resources development. The need for intersectoral negotiation. Geneva, World Health Organization, 1993. 7. Muller M. The engineer in the control of schisto- somiasis. Progress in water technology, 1978, 11: 167-172. 8. The control of schistosomiasis: report of a WHO Expert Committee. World Health Organization, 1985 (WHO Technical Report Series, No. 728). 9. Peters PAS, Kazura JW. Update on diagnostic methods for schistosomiasis. In: Mahmoud AAF, ed. Bailliere's clinical tropical medicine and communi- cable diseases. Volume 2: Schistosomiasis. London, Bailliere Tindall, 1987: 419-434. 10. Feldmeier H. Diagnosis. In: Jordan P, Webbe G, Sturrock R, eds. Human schistosomiasis. Wallingford, CAB International, 1993: 271-303. 11. Wegner DHG. The profile of the trematodicidal com- pound, praziquantel. Arzneimittel Forschung, 1984, 34: 1132-1136. 12. The use of essential drugs: with report of the WHO Expert Committee. Geneva, World Health Organiza- tion, 1995 (WHO Technical Report Series, No. 850). 13. Ouma JH, Wijers JB, Arap-Siongok TK. The effect of repeated targeted mass treatment on the preva- lence of schistosomiasis mansoni and the intensity of infection in Machakos, Kenya. Annals of tropical medi- cine and parasitology, 1985, 79: 431-438. 14. Wilkins HA. Reinfection after treatment of schis- tosome infections. Parasitology today, 1989, 5: 83-88. 15. Gryseels B et al. Epidemiology, immunology and chemotherapy of Schistosoma mansoni infections in a recently exposed community in Senegal. Tropical and geographical medicine, 1994, 46: 209-219. 16. The control of schistosomiasis: second report of the WHO Expert Committee. Geneva, World Health Or- ganization, 1993 (WHO Technical Report Series, No. 830). 17. Jobin WR. Control of Biomphalaria glabrata in a small reservoir by fluctuation of the water level. American joumal of tropical medicine and hygiene, 1970, 19: 1049-1054. 18. Klumpp RK, Chu KY. Ecological studies of Bulinus rohlfsi, the intermediate host of Schistosoma hae- matobium, in the Lake Volta, Ghana. Bulletin of the World Health Organization, 1977, 55: 715-730. 19. Pike EG. Engineering against schistosomiasisl bilharzia: guidelines towards control of the disease. London, Macmillian, 1987. 20. Ofoezie IE et al. A study of an outbreak of schis- tosomiasis in two resettlement villages near Abeokuta, Ogun State, Nigeria. Joumal of helminthology, 1991, 65: 95-102. 440 WHO Bulletin OMS. Vol 75 1997 Schistosomiasis transmission and water level regulation 21. Frandsen F, Christensen NO. An introductory guide to the identification of cercariae from freshwater snails with special reference to cercariae of trematode spe- cies of medical and veterinary importance. Acta tropica, 1984, 41: 181-202. 22. Chandiwana SK. Community water contact patterns and the transmission of Schistosoma haematobium in the highveld region of Zimbabwe. Social science and medicine, 1987, 25: 495-505. 23. Snail control in the prevention of bilharziasis. Geneva, World Health Organization, 1965 (WHO Monograph Series, No. 50). 24. Richards FO Jr et al. An evaluation of quantitative techniques for Schistosoma haematobium eggs in the urine preserved with carbolfuchsin. American journal of tropical medicine and hygiene, 1984, 33: 857-861. 25. Ackers GL, Smith DH. Design and management of development projects to avoid health hazards. Journal of tropical medicine and hygiene, 1988, 91: 115- 120. 26. Mock JF. Engineering measures for reservoirs and irrigation to reduce waterborne diseases. Proceedings of the Symposium on Scientific Procedures applied to the Planning, Design and Management of Water Resources, Systems. Hamburg, IAHS Publica- tions, 1983: 503-520. 27. Abdel-Wahab MF et al. Changing patterns of schistosomiasis in Egypt 1935-1979. Lancet, 1979, 2: 242-244. 28. Michelson MK et al. Recent trends in the prevalence and distribution of schistosomiasis in the Nile Delta region. American journal of tropical medicine and hygiene, 1993, 49: 76-87. WHO Bulletin OMS. Vol 75 1997 441