Effects of improved water supply and sanitation on ascariasis, diarrhoea, dracunculiasis, hookworm infection, schistosomiasis, and trachoma.

Reviews/Analyses Effects of improved water supply and sanitation on ascariasis, diarrhoea, dracunculiasis, hookworm infection, schistosomiasis, and trachoma S.A. Esrey,' J.B. Potash,2 L. Roberts,3 & C. Shiff4 A total of 144 studies were analysed to examine the impact of improved water supply and sanitation facilities on ascariasis, diarrhoea, dracunculiasis, hookworm infection, schistosomiasis, and trachoma. These diseases were selected because they are widespread and illustrate the variety of mechanisms through which improved water and sanitation can protect people. Disease-specific median reduction levels were calculated for all studies, and separately for the more methodologically rigorous ones. For the latter studies, the median reduction in morbidity for diarrhoea, trachoma, and ascariasis induced by water supplies and/or sanitation was 26%, 27%, and 29%, respectively; the median reduction for schistoso- miasis and dracunculiasis was higher, at 77% and 78%, respectively. All studies of hookworm infection were flawed apart from one, which reported a 4% reduction in incidence. For hookworm infection, ascariasis, and schistosomiasis, the reduction in disease severity, as meas- ured in egg counts, was greater than that in incidence or prevalence. Child mortality fell by 55%, which suggests that water and sanitation have a substantial impact on child survival. Water for personal and domestic hygiene was important in reducing the rates of ascariasis, diar- rhoea, schistosomiasis, and trachoma. Sanitation facilities decreased diarrhoea morbidity and mortality and the severity of hookworm infection. Better water quality reduced the incidence of dracunculiasis, but its role in diarrhoeal disease control was less important than that of sanitation and hygiene. Introduction Water and sanitation have been the subjects of con- siderable recent attention as a result of the declara- tion by the United Nations General Assembly that the 1980s were the International Drinking-Water, Supply and Sanitation Decade (IDWSSD). A major objective of this was to improve the health of popu- lations that received the interventions. Most of the 1 Professor, Faculty of Agriculture and Environmental Sciences and Faculty of Medicine, Macdonald College of McGill University, 21111 Lakeshore Road, Ste-Anne-de-Bellevue, Canada, H9X ICO. Requests for reprints should be sent to this address. 2 Student, School of Medicine, Johns Hopkins University, Balti- more, MD, USA. 3 Student, Department of Geography and Environmental Engin- eering, Johns Hopkins University, Baltimore, MD, USA. 4 Professor, Department of Immunology and Infectious Diseases, Johns Hopkins University, School of Hygiene and Public Health, Baltimore, MD, USA. Reprint No. 5216 research on the health impacts of water and sani- tation projects has focused on the incidences of diar- rhoeal diseases, malnutrition, and mortality of young children, and evidence accumulated during the decade indicates that these rates have been reduced (22, 25). Although it is generally believed that the rates of other diseases will decrease following improvements in water and sanitation, there have been no systematic reviews of this. The present article reviews the health impact of water and sanitation interventions on the following: ascariasis (Ascaris lumbricoides), diarrhoeal diseases (including measures of nutritional status and child survival), dracunculiasis (Dracunculus medinensis), hookworm infection (Ancylostoma duodenale and Necator americanus), schistosomiasis (Schistosoma haematobium and S. mansoni), and trachoma (Chlamydia trachomatis). These diseases were chosen because they are widespread in developing countries, or because they constitute serious problems where they exist, or both (Table 1). All developing countries are affected by diarrhoeal diseases and ascariasis; the Bulletin of the World Health Organization, 69 (5): 609-621 (1991) c World Health Organization 1991 60 S.A. Esrey et al. Table 1: Incidence and effects of selected diseases In developing countries (excluding China) Estimated number of: Cases per year (x 10) Deaths per year Ascariasis' 900 20000 Diarrhoeal diseasesb 875 4.6 million Dracunculiasisc 4 - Hookworm infection 800 Schistosomiasisd 200 Trachoma" 500 -9 '-' See ref. 96, 80, 97,57, and 18, respectively. 'Usually causes debilitation rather than death. o The major disability is blindness (8 million). other four diseases are somewhat more restricted in their range, although they are all prevalent in some developing countries, except dracunculiasis, which only occurs in parts of Africa and Asia. These diseases also illustrate the variety of mechanisms through which improved water and sanitation can promote health (Table 2). Four basic aspects were considered: sanitation (i.e., human excreta disposal), water quality, personal hygiene, and domestic hygiene. Personal hygiene refers to water used for cleaning the body, including water for the face, hands, and eyes; domestic hygiene refers to water used to keep the home clean (e.g., food, utensils, and floors). Each disease is affected by one or more of these interventions. For example, provid- ing safe, potable sources of drinking-water will prob- ably prevent transmission of dracunculiasis, while Table 2: Illustration of the potential relation between water and sanitation Interventions and morbidity from selected diseases Intervention Water Water Improved for for Human drinking- domestic personal excreta water hygiene hygiene disposal Ascariasis + + + - ++ Diarrhoeal + + + + + + + diseases Dracunculiasis + + Hookworm - - - + + infection Schistosomiasis - + + + + + + Trachoma - + + + * Interventions marked with one or two pluses have an impact on a particular disease: an intervention marked "+ +" will have a stronger impact than one marked " + "; "-" = the intervention has little or no impact on reducing disease rates. For a particular disease, a package of interventions with pluses is expected to produce a larger impact than any one intervention alone. using larger quantities of water for personal hygiene will probably interrupt the spread of trachoma. The incidence, prevalence, and/or severity of all these dis- eases are reduced by water and sanitation interven- tions. Methods Computer searches of articles on each of the diseases considered were carried out using the MEDLINE and MEDLINE 500 systems, which cover the period from 1966 and 1986, respectively. For schistosomiasis, an additional computer search was performed on CAB Abstracts (period, 1984-89). For hookworm infection and ascariasis, Index Medicus was searched for 1976-86, and the Science Citation Index for 1970-84. Because a thorough review of diarrhoeal diseases was published in 1986 (22), only studies published since then were sought. For trachoma, additional references were provided by a recent review (72). References cited in the articles identified, but which were not located in the computer search, were also sought. Only studies published in English in peer-review journals were used; this methodology ex- cluded technical reports, mimeographs, and agency documents. An exception was made for studies carried out by Miller et al.,'-b because although they had been published in peer-review journals, the unpublished documents upon which they were based contained more information. Finally, the review was restricted to studies that presented data on the effect of water and sanitation conditions on one of the six diseases. Specific criteria that we have reported pre- viously were used to identify the rigorous studies from the large number reviewed (22). First, studies were divided into those that reported a statistically significant positive association between the provision of improved water or sanitation and better health and those that did not. Studies were excluded from the rigorous category if they had a single major flaw or several known or suspected minor flaws that could have biased their results. Each disease is treated in turn, with the discussion focusing on the rigorous studies. The median reductions in morbidity (i.e., inci- dence and prevalence) and a range of reductions attributable to improved water and sanitation were MMiller, F.D. *t *I. Human intestinal parasitic infections and environmental health factors in rural Egyptian communities. (EPA Research Reporting Series EPA-600/1-80-024), 1980. b Miller, F.D. et ai. Schistosomiasis in rural Egypt: a report of U.S.-Egyptian River Nile and Lake Nasser research project. (EPA Research Reporting Series EPA-600/1-78-070), 1978. 610 WHO Bulletin OMS. Vol 69 1991. Effects of Improved water supply and sanitation on six endemic diseases calculated for each disease, based on all the studies reviewed. The same calculations were also made for only the rigorous studies, the results from which more accurately represent the reductions that may be achieved by water and sanitation interventions. The results reported in some studies did not permit determination of the median reduction in morbidity, and such studies were not included in the calcu- lations. The average reduction for each disease cate- gory was expressed as the median value for all the studies considered, and not as the mean, because medians are not influenced by extreme values, while means are. The definitions of morbidity reported in the articles reviewed included incidence, point and period prevalence, infection, and indicators of sever- ity (e.g., duration). As far as possible, changes in inci- dence and prevalence have been used in the present review, but the discussion of specific studies also includes changes in infection and severity of disease. Results Ascarlasis For ascariasis, nine studies that reported a positive effect for water supply or sanitation were reviewed (3, 15, 31, 37, 63, 66, 76, 77) C together with five negative studies (28, 44, 78, 85, 89). Three studies (63, 66)c examined differences in excreta disposal facili- ties, and five (3, 15, 37, 76, 77) investigated various combinations of water supply and sanitation condi- tions. All five negative studies reported on excreta disposal facilities, while only one study (37) investi- gated the influence of water supplies alone. One study (31) reported both positive and negative find- ings; for ascariasis, significant differences were found among users of different types of excreta disposal facilities. The median reduction in morbidity esti- mated from all studies was 28%, while that for only the rigorous studies (3, 37, 76, 77) was 29% (Table 3). Water supplies with sanitation. The four rigorous studies (3, 37, 76, 77), all of which reported positive effects, investigated the combination of water sup- plies and excreta disposal facilities. The rate of morbi- dity reduction was dependent on the level of service provided, and indoor facilities were associated with larger reductions than public facilities. In the USA the prevalence of Ascaris spp. among all age groups was reduced by 71% for people with flush toilets and indoor plumbing compared with a group that had lavatories but no well-water (77). In Saint Lucia the prevalence of Ascaris spp. was reduced by 31% for a cohort of children with household water and latrines compared with a control group without such facili- ties (37). In Iran the provision of a courtyard latrine c See footnote a, p. 610. Table 3: Expected reduction In morbidity and mortality from Improved water and sanitation for selected dlseases All studiesb Rigorous studiesb nc Median reduction (%) nc Median reduction (%) Ascariasisd 11 28 (0-83)' 4 29 (15-83) Diarrhoeal diseases Morbidity 49 22 (0-100) 19 26 (0-68) Mortality 3 65 (43-79) - Dracunculiasis 7 76 (37-98) 2 78 (75-81) Hookworm infection" 9 4 (0-100) 1 4 - Schistosomiasisd 4 73 (5947) 3 77 (59-87) Trachoma 13 50 (0-91) 7 27 (0-79) Child mortality' 9 60 (0-82) 6 55 (20-82) * The reduction for each individual study was calculated, if possible, directly from the data reported, usually prevalences. However, this was not possible for studies that reported results as odds ratios or graphically. To convert the odds radios, we used prevalences, if they were known; for the data in graphical form, estimates of the reductions were made. If a study reported the results from more than one comparison, the largest reduction in disease was used to highlight the benefits of water and sanitation under optimal conditions. The results from studies that reported reductions that were not statistically significant were used. If studies showed an increase in disease, this was taken to be a reduction of zero; because medians were used, this did not affect the results. b For every disease category, all available studies that met the criteria described in "Methods" are summarized. c The number of studies may not equal the number of studies reviewed. For some studies, reductions could not be calculated because data were not available or prevalences could not be determined from the odds ratios. d Although the presence of eggs in stools was used as an indicator of the prevalence of ascariasis, hookworm infection and schistosomi- asis, data for the reduction in egg count (an indicator of the severity of these diseases) were not used in the calculations. ' Figures in parentheses are the range. 'Results from several studies were excluded from the child mortality calculations. One study (13) used the same data as two others (23, 33), and so these three were grouped as one study. Another study (4) did not isolate the effect of water and sanitation on mortality; and reduction levels could not be estimated from another study (98). WHO Bulletin OMS. Vol 69 1991. 611 S.A. Esrey et al. and public standpipes was associated with a reduction of 16% in the prevalence of ascariasis compared with a control group with no such facili- ties (76). In two studies in Iran the reduction in the sever- ity of ascariasis was greater than that of its preva- lence (3, 76). In one of these studies the baseline infection rates and egg counts in four villages that had a household latrine and a community water supply were compared with the infection rates and egg counts in three control villages (3). After 3-4 years, the prevalence of infection with Ascaris spp. in the group with latrines and water supply had decreased by 28% and the egg counts by 60%. For the control group, the prevalence of infection and the egg counts decreased by 19% and 29%, respectively. In the second study, courtyard latrines and public standpipes reduced egg counts by 62% over a 1-2- year period, but the prevalence of ascariasis dropped by only 16% (76). Water supplies. The effect of water supplies on infec- tion with Ascaris spp. was compared in two studies (37, 77). In the USA the reduction in the prevalence of infection was 37% for a group that had lavatories and indoor plumbing, and 12% for those with lava- tories and a yard well, in both cases relative to a control group that had only lavatories, but no well- water (77). In Saint Lucia the provision of household piped water supplies produced a 30% reduction in ascariasis over a 2-year period among children under 3 years of age (37). Water and sanitation plus chemotherapy. The reduction in ascariasis was greater if chemotherapy was used in conjunction with water and sanitation than if only water and sanitation were provided. In the study in Iran described above (3), four villages were supplied with a household latrine and a community water supply plus chemotherapy, four others with only chemotherapy, and three villages remained as a control group. For the group with latrines, water, and chemotherapy, the prevalence of Ascaris spp. among all age groups decreased by 79% and egg counts by 88%, while the corresponding reductions for the chemotherapy-only group were 84% and 90%. These were considerably greater than the reductions for the group with only water and sani- tation (28% and 60%, respectively). The reductions in the prevalence of ascariasis and in egg counts for the control group over the 3-4-year study period were 19% and 29%, respectively. Diarrhoeal diseases We have updated a previous review of 67 studies on diarrhoeal morbidity, nutritional status, and mortal- ity (22) by including 17 more recent studies (1, 6, 7, 23, 24, 26, 27, 33, 39, 51, 52, 65, 81-83, 95, 103). The median reduction in diarrhoeal morbidity calculated from all the studies was 22%, and from the rigorous studies only, 26% (Table 3). Water and sanitation. Eleven studies examined the combined effect of water and sanitation without con- sidering the effect of each separately; the median reduction determined from the seven that provided appropriate data was 20% (Table 4). Two studies reported on specific pathogens, one on nutrition, and one on mortality. Of the 11 studies, seven were flawed (three of which reported positive impacts). In two of the rigorous studies, an average of 30% reduction in diarrhoea was associated with improved water and sanitation conditions (47, 73). A third study reported fewer malnourished children from families with a sewage system and a household bath than from families with latrines and no bath (16). In Malawi, the combination of water and sanitation was associated with a lower prevalence of diarrhoea, but the results were not statistically significant because of small sample sizes (103). Only one of the 11 studies examined mortality: among infants in Malaysia, the addition of toilets and water versus no facility was associated with an 82% reduction in infant mortality, particularly if the child was not breast-fed (33). Sanitation. The impact of sanitation was examined in 30 studies (12 of which were flawed), and 21 reported health improvements. Overall, a 22% reduction in morbidity was calculated for 11 of the 30 studies, whereas the reduction determined using data from Table 4: Expected reduction In dlarrhoeal dleae mor- bidity from Improvements In one or more components of water and sanItatlon All studies Rigorous studies Reduction Reduction n (%) n (%) Water and 7/1l b 20 2e/3b 30 sanitation Sanitation 11/30 22 5/18 36 Water quality 22/43 16 2/22 17 and quantity Water quality 7/16 17 4/7 15 Water quantity 7/15 27 5/10 20 Hygiene 6/6 33 6/6 33 ' The number of studies for which morbidity reduction calcu- lations could be made. b The total number of studies that related the type of facility to diarrhoeal morbidity, nutrition, and mortality studies. WHO Bulletin OMS. Vol 69 1991.612 Effects of Improved water supply and sanitation on six endemic diseases five of 18 rigorous studies was 36% (Table 4). Of the remaining rigorous studies, two reported on nutri- tional status and 11 on mortality. Of the studies that compared the relative importance of water and sani- tation, most reported that sanitation had a greater impact on child health, based on mortality, growth, and morbidity indicators. Some mortality studies reported that the method of disposing of excreta determined the magnitude of the health impact. Mortality was reduced to a greater extent by flush toilets than by pit latrines (2, 34, 98), which neverthe- less were associated with mortality reductions com- pared with no sanitation facilities at all. Sanitation has also been reported to produce a differential health impact depending on the presence or absence of other risk factors. For example, sanitation was most effective in reducing mortality among nonbreast-fed infants (13) and infants of illiterate mothers (23) than among breast-fed infants or liter- ate mothers. Water quality and quantity. In most of the studies reviewed, it was difficult to determine whether the differences in health conditions were due to increased amounts of water, improvements in its quality, or both; these studies have therefore been grouped together. Of the 43 studies that compared groups with different types of water supplies, 24 reported a positive impact, and for the 22 studies for which morbidity calculations could be made, a median reduction of 16% was calculated (Table 4). Of the remaining studies, four reported on patho- gens, four on nutritional status, 11 on mortality, while two had insufficient information. Fourteen of the studies were rigorous. Only two of the 22 mor- bidity studies were rigorous (median reduction, 17%). Nine of the rigorous studies reported on mor- tality and three on nutritional status. Of these, a modest benefit for particular, but not all, age groups was found. In the studies that reported a health benefit, the water supply was piped into or near the home, whereas in those that reported no benefit, the improved water supplies were protected wells (2), tubewells (74), and standpipes (27, 51, 52, 69, 98). Water quaity. Sixteen studies (10 of which reported positive effects) examined the health impacts of pure versus contaminated water supplies. Of the seven studies for which calculations could be made, a median reduction in the prevalence of diarrhoea of 17% was found (Table 4). Four of the other nine studies reported on pathogens, two on nutritional status, two on diarrhoeal mortality, while one had insufficient information. In several of the studies, impacts were found only for certain age groups. Among the seven rigorous studies, a median reduction of 15% in diarrhoeal disease morbidity was found for the four studies for which this could be calculated. Of these, one reported little or no association between the quality of drinking-water and the occurrence of diarrhoea in children (39), another reported an 8% reduction in the prevalence of Shigella spp. (87), while two found some associ- ation with child nutritional status, but not with diar- rhoea (24, 55). Because diarrhoea has many causes, drinking-water constitutes only one of many sources of infection. In areas where environmental faecal contamination is high, little or no health impact from water improvement can be expected. For example, studies in Lesotho (30) and Guatemala (79) failed to detect reductions in the prevalence of diar- rhoea following improvements in water quality. Also, a recent study in Brazil (95) found no association between water quality and diarrhoea mortality, as did an earlier study in India (104). Water quantty. Of the 15 studies that examined the effect of increased amounts of water specifically and independently of water quality, all but one reported a positive impact. The median reduction for seven studies for which this could be calculated was 27% (Table 4). Of the other eight studies, five reported on pathogens and three on nutritional status. Of the 10 rigorous studies, a median reduction of 20% was found for the five for which this could be calculated. Health benefits were greater for children whose fam- ilies used more water than for those whose families used less, but, in some instances, the differ-nc( were small or significant only for selected age groups. In Ethiopia, the prevalence of diarrhoea among under- 2-year-olds from families with higher water usage rates per person was less than that among compa- rable children from families with lower rates (32). In Lesotho, use of smaller amounts of water was asso- ciated with higher rates of infection with Giardia lamblia (26). In both of these studies, the amount of water used was more important than its source. No studies relating water quantity and mortality were identified. Hygiene. Only six studies were found that reported on the impact of hygiene interventions on diarrhoeal morbidity (1, 6, 12, 45, 82, 93). All were rigorous, and the median reduction was 33% (Table 4). Several studies focused specifically on hand- washing. In Burma, a 30% reduction in diarrhoea was reported when mothers and children were pro- vided with soap and encouraged to wash their hands after defecation and before preparing meals (6). In day-care centres in the USA (12), a handwashing regimen reduced the incidence of diarrhoea by 48% compared with a control group; and in Bangladesh WHO Bulletin OMS. Vol 69 1991. 613 S.A. Esrey et al. (45), a 35% reduction in the incidence of diarrhoea caused by Shigella spp. occurred following an inter- vention to promote handwashing. Other studies have examined not only hand- washing, but combinations of handwashing and other hygienic behaviours. For example, an educa- tional intervention in Bangladesh (82) emphasized proper handwashing before preparing food, de- fecating away from the house at a proper site, and suitable disposal of waste and faeces; the investiga- tors reported a 26% reduction in the incidence of diarrhoea. Another study in Bangladesh provided hand pumps and health education to promote per- sonal and domestic hygiene (1); a greater than 40% reduction in the incidence of diarrhoea was found among groups with good hygiene practices (irrespec- tive of whether or not they received the interven- tion), compared with individuals with poor practices. The reduction attributable to the intervention itself was 17%. In Guatemala the incidence of diarrhoea was reduced by 14% following a programme to promote health awareness and hygienic behaviour (93). Drwauncullasls Seven studies (11, 21, 36, 41, 54, 75, 94), two of which were rigorous (21, 36), examined the impact of improved water supplies on the prevalence of dra- cunculiasis. All reported a positive impact, with similar median reductions being reported for all studies combined and for the rigorous studies only (76% and 78%, respectively). water source. The two rigorous studies (21, 36), both of which were from Africa, compared the use of borehole water and of water from unimproved sources. In Nigeria (21), the impact of a UNICEF- assisted rural water project that provided boreholes and hand pumps, along with health education, was evaluated. Prior to the intervention 8600 subjects, and 3 years after its installation, over 10000, were examined. The study comprised 20 serviced and five unserviced communities. Use of borehole water reduced the incidence of dracunculiasis by 81%. The greatest reduction in the prevalence occurred in those villages where the boreholes were conveniently sited and close by, but the effect was less if the wells were distant. The impact was also less dramatic in villages where the water was unpalatable or the supply erratic. A study in Uganda reported data from a pre- intervention survey (36). Over 2000 people were interviewed in the north-west of the country to determine the relationship between water use and disease. The results were analysed by season because the attack rate of dracunculiasis peaked twice during the year. During the rainy season the attack rate among borehole users was one-tenth of that among non-users, while during the dry season it was two- fifths that of non-users. Seasonal transmission. A number of studies on dra- cunculiasis have investigated the seasonal nature of its transmission. In West Africa, attack rates corre- lated positively with periods of below-average rain- fall or the dry season, when the copepod vectors reached their highest density in ponds that were drying up (9, 20). In Nigeria, the prevalence of the disease remained high if the piped-water supply was intermittent or unreliable during the dry season (40). The problem of unreliable tube-wells or boreholes which fail, particularly during the dry season, is fre- quently cited as an important factor that contributes to transmission of dracunculiasis (11, 84). Hookworm Infection Eleven studies (3, 15, 17, 31, 44, 63, 77, 78, 86, 89)d were reviewed that related hookworm infection to excreta disposal facilities, with or without water supplies. Five studies reported positive findings, two involving water supply and sanitation (15, 77) and three sanitation only (17, 44, 89). Only one study, which examined the influence of sanitation facilities in conjunction with chemotherapy, was considered to be rigorous (3). The median reduction in the prevalence of the disease calculated from all studies was 4%, which was the same as that calculated from the rigorous study alone (3). Excreta disposal. In the study in Iran referred to under ascariasis (3), villages where sanitation improvements were carried out had a 4% decrease in the prevalence of hookworm infection, while there was a 26% reduction in egg counts among those infected. For the group that received the sanitation and chemotherapy intervention, the reduction in prevalence of infection was 69%, while the egg count was reduced by 88%; for the chemotherapy-only group, the reductions were 73% and 87%, respec- tively; and for the control group, 11% and 12%. However, each cohort began with a different preva- lence of hookworm infection, e.g., at the outset 77% of the sanitation cohort and 44% of the control group were infected. Schlstosomlasls Twelve studies were found that examined water and sanitation facilities and rates of schistosomiasis (8, d See footnote a, p. 610. 614 WHO Bulletin OMS. Vol 69 1991. Effects of Improved water supply and sanitation on six endemic diseases 29, 38, 43, 46, 49, 50, 53, 59, 64, 86)&. All four of the rigorous studies examined the degree of protective efficacy of water supplies (8, 43, 50, 59). Extensive water supply and washing facilities were provided to the study populations. One study also examined the effect of water supply and the provision of latrines (8). For all studies combined, the median reduction in morbidity was 69%, and for the rigorous studies only, 77% (Table 3). Water supplies. The presence of piped-water in the home (43, 50) was associated with larger reductions in the prevalence of schistosomiasis than that pro- duced by community water supplies (59). In Saint Lucia over a 5-year period following the provision of household piped-water and community washing and showering facilities, the overall prevalence of schisto- somiasis was reduced by 27% and that among child- ren aged under 10 years of age, by 59% (43). In south-east Brazil, children aged 5-14 years were 2.3 times more likely to be infected if they had no piped- water in their home (50). In Zimbabwe the prevalence of S. mansoni among schoolchildren who lived on communal lands without a piped-water supply was 4.8% and of S. haematobium, 4.4% (59). Among children who lived on the same lands, but with piped water, the preva- lence of S. mansoni and S. haematobium was 0.8% and 0.4%, respectively. In Brazil the severity of schistosomiasis was reduced more than was its prevalence (50). Children aged 5-14 years were 7.3 times more likely to have splenomegaly (an indication of severe schistoso- miasis) if they had no piped water in their home. Malaria, which can also cause splenomegaly, was not a confounder, because its prevalence in the study area was low. The risk of severe infection (7.3) was much greater than the risk of the prevalence of schis- tosomiasis (2.3). Sanitation and water supplies. One study in north-east Brazil reported the prevalence of S. mansoni before and after a water, sanitation, and health education campaign (8). In the treatment villages, latrines were built for each house, and communal taps, laundry facilities, showers, latrines, and hand pumps were installed. Over a 7-year period, children under 14 years of age exhibited a net drop in the prevalence of schistosomiasis, and this was 77% greater in the treatment than in the control villages. Water contact. One of the primary mechanisms through which improved water supplies have an @ See footnote b, p. 610. impact on schistosomiasis is by reducing human contact with infected water. A distinction should therefore be made between possession of an improved water supply and having contact with pathogen-laden water. In Saint Lucia (42, 43) and south-east Brazil (50), access to improved supplies that included laundry and shower facilities was associated with reduced contact with infected waters and thus reduced infection. In Saint Lucia, this led to an 82% reduction in the frequency that people had contact with infected water and a 96% reduction in water-contact time. In Kenya, the installation of boreholes, without laundry or shower facilities, failed to reduce the reliance on high-risk marshes and ponds for bathing (46). Water suppiles and chemotherapy. Reinfection can rapidly diminish or negate health improvements brought about by community-wide schistosomiasis programmes devoted exclusively to drug therapy (10, 101). Drug treatment combined with improved sources of water produces a greater reduction in the prevalence of schistosomiasis than that resulting from provision of water facilities alone (43, 60, 64, 68). Programmes that adopt a multifaceted approach have repeatedly been successful (64, 67, 90), and the provision of adequate water supplies to those who have been treated with drugs can prevent reinfection (43). For example, in Saint Lucia oxamniquine caused a precipitous drop in the rates for infection with schistosomiasis and a 4-year follow-up pro- vided no evidence for reinfection (43). Trachoma Sixteen studies on the role of water, sanitation, and/or hygiene in the reduction of trachoma were reviewed. Thirteen reported positive effects (5, 14, 19, 35, 56, 58, 61, 62, 71, 88, 91, 92, 99) and three, nega- tive (48, 70, 102). Five of the rigorous studies were positive (5, 61, 91, 92, 99) and two, negative (48, 102). The median reduction in trachoma calculated for all studies was 50% and for the rigorous studies only, 27%. Distance and time to water source. Four studies (5, 61, 92, 99) reported a 30% median reduction in tra- choma that was associated with shorter distances or time to water sources. In China (Province of Taiwan) the prevalence of trachoma among people with household water connections was 45% less than among those whose water source was over 500 m away (5); and in India those who had a water supply within 200 m exhibited 30% less trachoma than those who obtained water from a more distant source (61). In Malawi there was 26% less trachoma among WHO Bulletin OMS. Vol 69 1991. 615 S.A. Esrey et al. children whose water was less than 5 minutes away than among those who required a trip of more than 1 hour to obtain water (92). In the United Republic of Tanzania, 26% fewer households had trachoma that affected all the children if the source of water was less than 30 minutes away, compared to a water source that was more than 2 hours distant (99). Two other studies (one in Morocco (48) and one in Mexico (91)) failed to find such an association between the prevalence of trachoma and distance to water supply. Distance to water has been taken to be a proxy for the amount of water used, but the above- mentioned Tanzanian study did not find a direct association between the distance to water and the amount of water brought into the household (99). Also in the study in Morocco there was no corre- lation between distance to the source and the per capita use of water (48). Similarly, several studies of diarrhoeal diseases have reported no significant association between improved water or distance to water and the amount of water used (27,51, 100). Personal hygiene. In India people with good hygiene practices had 79% less trachoma than those whose practices were poor (61). In Mexico a study of the relationship between the prevalence of trachoma and a variety of possible risk factors reported a signifi- cant association only for the frequency of face- washing (91). Children who washed their faces seven or more times a week had 69% less trachoma than those who washed their faces less frequently; however, the importance of this aspect of personal hygiene was not confirmed in a follow-up investiga- tion in Mexico near the original study site (102). The different findings in these two studies might be accounted for by the different conditions: in villages included in the follow-up study, almost twice as many children washed their faces frequently as did those in the original villages. In the United Republic of Tanzania observa- tions of hygiene practices indicated that a child was 1.7 times more likely to have trachoma if all the children in the family had unclean faces (99). Also, distance to water was related to the proportion of children with unclean faces. There were 14% fewer households where all the children had unclean faces when the source of water was less than 30 minutes away, compared with more than 2 hours away. Conclusions The results of this review indicate that improvements in one or more components of water supply and sanitation can substantially reduce the rates of mor- bidity and severity of ascariasis, diarrhoeal diseases, dracunculiasis, hookworm infection, schistosomiasis, and trachoma. Despite the mix of both positive and negative studies, the overwhelming evidence is in favour of positive impacts, with the exception of hookworm infection, for which the impact was negli- gible. The reduction in morbidity from the rigorous studies ranged from 26% for diarrhoea, 27% for tra- choma, and 29% for ascariasis to 77% for schisto- somiasis and 78% for dracunculiasis. For the rigorous studies the median reduction in diarrhoea- specific mortality was 65% and in overall child mor- tality, 55%, which suggests the important role that water and sanitation play in enhancing child sur- vival. Studies that reported reductions for one disease (or even more than one) most probably underesti- mated the total effect of water and sanitation in improving health. This was particularly true if several diseases that are affected by water and sani- tation were prevalent simultaneously in the interven- tion area. Even though studies often report on only one indicator of health for a selected age group, other health indicators may also be changing for other age groups. In addition to reducing the incidence or preva- lence of disease, improvements in water and sani- tation can be expected to affect other aspects of health. When infection rates are reduced by chemo- therapy, as was the case for some parasitic diseases, water and sanitation facilities prevented infection rates from increasing again to pretreatment levels. Furthermore, the severity of infection was often reduced more than that of the incidence or preva- lence. For example, the reductions in egg counts for ascariasis, schistosomiasis, and hookworm infection were greater than those in the incidence or preva- lence of these diseases. In addition, reductions in childhood diarrhoea and overall mortality rates were greater than for diarrhoea incidence or prevalence, which suggests that the severity of diarrhoea was reduced more than its incidence. The expected reductions in disease rates provide a guide. The studies reviewed were of variable quality, and therefore reductions may be smaller or larger depending on several factors; for example, the success with which an intervention was installed, the overall health status of the recipients, and the pres- ence or absence of environmental factors that can also reduce transmission of disease agents. If the interventions do not work, break down, remain underused, or do not change behaviour sufficiently, they will probably result in small impacts at best. No study, irrespective of how well it is conducted, can produce substantial impacts under such conditions. On the other hand, considerable impacts will prob- ably result if the intervention was successfully imple- WHO Bulletin OMS. Vol 69 1991.616 Effects of Improved water supply and sanitation on six endemic diseases mented, properly utilized, and the population disease rates were high enough for benefits to be measured. Results published since 1986 on diarrhoeal dis- eases concur with previous findings (25), and the rig- orous studies conducted in the last few years show a greater reduction in disease: while this may reflect better studies, it may also indicate better-conceived interventions. Interventions to improve excreta dis- posal and water quantity, which are associated with better hygiene practices, produce greater impacts than improvements in water quality. This is particu- larly so in highly contaminated environments where diarrhoea rates are high. Because the use of more water is not automatic following the installation of water supplies, hygiene education is a necessary part of the intervention. The following recommendations can therefore be made: - to achieve broad health impact, greater attention should be given to safe excreta disposal and proper use of water for personal and domestic hygiene rather than to drinking-water quality; - sanitation facilities should be installed at the same time as water facilities when faecal-related diseases are prevalent; - access to the water supply should be as close to the home as possible, in order to foster the use of larger amounts of water for hygiene practices; -water supply and health programmes should emphasize hygiene education to encourage the use of more water for personal and domestic hygiene; sanitation facilities should be culturally appropri- ate to ensure their use; use of facilities is essential during critical seasonal transmission periods for diseases such as dracun- culiasis; and water and sanitation programmes should com- plement those in other sectors (e.g., chemo- therapy) to reduce disease rates. Isues for future research Despite the accumulation of knowledge during the IDWSSD, questions remain about the health bene- fits associated with water and sanitation. Many of these issues could be examined using well-established epidemiological methods (22) in areas where water and sanitation systems are working and used by the recipients. Rigorous studies should focus on the fol- lowing issues: -the maximum travel distance/time that will result in appropriate use of an improved water supply under a variety of socioeconomic and environ- mental conditions; - the minimum quantities of water that are neces- sary to produce positive health impacts under a variety of environmental conditions (e.g. peri- urban, rural, etc.); - the hygienic behaviours most conducive to better health; - appropriate methods for introducing and reinfor- cing behavioural change; - the conditions under which water and sanitation facilities are likely to be sustained; - the possibility of increasing the amount of water used and of changing behaviours in the absence of interventions to install piped supplies; - the conditions (including environmental, cultural, and level of development) under which installa- tion of water and sanitation facilities are likely to produce the greatest health benefits; and - the health indicators that are most susceptible to change and most related to disease. Acknowledgements This study was supported by the Water and Sanitation for Health Project (WASH). A more detailed report, (Health benefits from improvements in water supply and sani- tation: survey and analysis of the literature on selected diseases), including an annotated summary of each study, can be obtained from WASH, 1611 North Kent Street, Room 1001, Arlington, VA 22209-2111, USA. We thank S. Blumenfeld, J. Burns, S. Cairncross, J. Kalber- matten, L. Laugeri, A.D. Long, R. Middleton, J. Newman, D. Okun, F. Reiff, P. Roark, M. Scott, H. Taylor, J. Tomaro, J.E. Turner, and J. Walker for their comments. D. Campbell is thanked for his relentless efforts to iden- tify and obtain literature for the present article and K. Leccisi and B. Reddaway are thanked for their editing skills. Resume Amelloration de l'approvisionnement en eau et de l'assalnissement: consequences sur l'ascarldlase, la dlarrhee, la dracunculose, I'ankylostomiase, la schistosomiase et le trachome On analyse ici les resultats de 144 etudes portant sur l'impact de l'amelioration de l'approvi- sionnement en eau et des installations d'assai- nissement sur l'ascaridiase, les maladies diar- rheiques, la dracunculose, I'ankylostomiase, la schistosomiase et le trachome. On a choisi ces affections parce qu'elles sont tres repandues, constituent des risques graves pour la sante et illustrent bien la diversite des mecanismes par lesquels I'am6lioration de l'approvisionnement en WHO Bulletin OMS. Vol 69 1991. 617 S.A. Esrey et al. eau et de l'assainissement peut proteger les popu- lations. On a calcule pour 1'ensemble des etudes la diminution mediane du nombre de cas de chaque maladie qu'elle entraine et l'on a tire des etudes les plus rigoureuses sur le plan methodolo- gique une deuxieme valeur de la diminution me- diane. Les resultats montrent que l'approvision- nement en eau et l'assainissement entrainent une amelioration de la sante des populations concer- nees. En ce qui concerne la diarrhee, le trachome et l'ascaridiase, la diminution mediane de la mor- bidite (c'est-a-dire de l'incidence et de la prevalence) a ete respectivement de 26%, 27% et 29% dans les etudes rigoureuses. La diminution correspondante pour la schistosomiase et la dra- cunculose a ete plus importante, a savoir 77% et 78%, respectivement. Toutes les etudes sur l'ankylostomiase ont ete defectueuses, a 1'exception d'une seule dans laquelle on a trouve une diminution de 4% de la prevalence. Pour l'ankylostomiase, I'ascaridiase et la schistosomiase, on a observe une diminution plus importante du nombre d'oeufs que de la preva- lence. Les interventions pratiquees au niveau de l'approvisionnement en eau et de l'assainissement on 6galement eu un impact important sur la survie des enfants. Neuf etudes ont indique une diminu- tion mediane de 60% de la mortalite infantile ge- nerale, et on a calcule qu'elle etait de 55% dans les six etudes rigoureuses; malheureusement, aucune de ces dernieres etudes ne s'est precise- ment interessee a la diminution de la mortalite due aux maladies diarrheiques. Dans la plupart des pays en developpement un meilleur approvisionnement en eau et des efforts d'assainissement peuvent donc avoir des consequences importantes et demontrables sur la sante dans tous les groupes d'age. Les diminu- tions observees dans la gravite de la maladie sont parfois plus importantes que celles relevees dans son incidence ou sa prevalence; cependant, on neglige trop souvent l'importance de cet impact. References 1. Alam, N. et al. Mothers' personal and domestic hygiene and diarrhoea incidence in young children in rural Bangladesh. International journal of epi- demiology, 18: 242-247 (1989). 2. Anker, R. & Knowles, J.C. An empirical analysis of morbidity differentials in Kenya at the macro- and micro-levels. Economics, development, and cultural change, 29: 165-185 (1980). 3. Arfaa, F. et al. Evaluation of the effect of different methods of control of soil-transmitted helminths in Khuzestan, south-west Iran. American journal of tropical medicine and hygiene, 26: 230-233 (1977). 4. Ascoll, W. et al. Nutrition and infection field study in Guatemalan villages, 1959-1964. IV. Death in pre- school children. Archives of environmental health, 15: 439-449 (1967). 5. Aaaad, F.A. et al. Use of local variations in tra- choma endemicity in depicting interplay between socio-economic conditions and disease. Bulletin of the World Health Organization, 41: 181-194 (1969). 6. Aung Myo Han & TheIn Hlalng. Prevention of diar- rhoea and dysentery by handwashing. Transactions of the Royal Society of Tropical Medicine and Hygiene, 83: 128-131 (1989). 7. Baltazar, J. et al. Can the case-control method be used to assess the impact of water supply and sani- tation on diarrhoea? A study in the Philippines. Bul- letin of the World Health Organization, 66: 627-635 (1988). 8. Barbosa, F.S. et al. Control of schistosomiasis mansoni in a small north east Brazilian community. Transactions of the Royal Society of Tropical Medi- cine and Hygiene, 65: 206-213 (1971). 9. Belcher, D.W. et al. Guinea worm in southern Ghana: its epidemiology and impact on agricultural productivity. American journal of tropical medicine and hygiene, 24: 243-249 (1975). 10. Benstad-Smith, R. et al. Evidence for the predis- position of individual patients to reinfection with Schistosoma mansoni after treatment. Transactions of the Royal Society of Tropical Medicine and Hygiene, 81: 651-654 (1987). 11. Bhatt, A.N. & Palan, K.H. Guinea-worm infection in Banaskantha District of Gujarat-some important epidemiological aspects. Indian journal of medical sciences, 32: 1-4 (1978). 12. Black, R.E. et al. Handwashing to prevent diarrhoea in day-care centers. American journal of epidemi- ology, 113: 445-451 (1981). 13. Butz, W.P. et al. Environmental factors in the relationship between breast feeding and infant mor- tality: the role of sanitation and water in Malaysia. Americanjournalofepidemiology,119:516-525(1984). 14. Calmcross, S. & Cliff, J.L. Water use and health in Mueda, Mozambique. Transactions of the Royal Society of Tropical Medicine and Hygiene, 81: 51-54 (1987). 15. Chandler, A.C. A comparison of helminthic and protozoan infections in two Egyptian villages two years after the installation of sanitary improvements in one of them. American journal of tropical medi- cine and hygiene, 3: 59-73 (1954). 16. Chrlstlanson, N. et al. Family social characteristics related to physical growth of young children. British journal of preventive and social medicine, 29: 121-130 (1975). 17. Cort, W.W. et al. A study of reinfection after treat- ment with hookworm and Ascaris in two villages in Panama. American journal of hygiene, 10: 614-625 (1929). 18. Dawson, C.R. et al. Guide to trachoma control. Geneva, World Health Organization, 1981. 618 WHO Bulletin OMS. Vol 69 1991. Effets of Improved water supply and sanitation on six endemic diseases 19. Dawson, C.R. et al. Severe endemic trachoma in Tunisia. British journal of ophthalmology, 60: 245- 252 (1976). 20. Edungbola, L.D. & Wafts, S.J. Epidemiological assessment of the distribution and endemicity of guinea worm infection Asa, Kwara State, Nigeria. Tropical and geographical medicine, 37: 22-28 (1985). 21. Edungbola, L.D. at al. The impact of a UNICEF- assisted rural water project on the prevalence of guinea worm disease in Asa, Kwara State, Nigeria. American journal of tropical medicine and hygiene, 39: 79-85 (1988). 22. Esrey, S.A. & Habicht, J.-P. Epidemiologic evidence for health benefits from improved water and sani- tation in developing countries. Epidemiologic reviews, 8: 117-128 (1986). 23. Esrey, S.A. & Habicht, J.-P. Maternal literacy modi- fies the effect of toilets and piped water on infant survival in Malaysia. American journal of epidemi- ology, 127: 1079-1087 (1988). 24. Esrey, S.A. et al. Drinking-water source, diarrheal morbidity, and child growth in villages with both tra- ditional and improved water supplies in rural Lesotho, southern Africa. American journal of public health, 78: 1451-1455 (1988). 25. Esrey, S.A. et al. Interventions for the control of diarrhoeal diseases among young children: improv- ing water supplies and excreta disposal facilities. Bulletin of the World Health Organization, 63: 757-772 (1985). 26. Esrey, S.A. et al. The risk of infection from Giardia lamblia due to drinking water supply, use of water, and latrines among preschool children in rural Lesotho. International journal of epidemiology, 18: 248-253 (1989). 27. Esrey, S.A. et al. Infection, diarrhea, and growth rates of young children following the installation of village water supplies in Lesotho. In: Tate, C.L., ed. Proceedings of the International Symposium of Water-Related Health Issues, Atlanta, GA, Novem- ber 1986. Bethesda, MD, American Water Resour- ces Association, 1987, pp. 11-16. 28. Eyles, D.E. et al. A study of Endamoeba histolytica and other intestinal parasites in a rural west Ten- nessee community. American journal of tropical medicine and hygiene, 2: 173-190 (1953). 29. Farooq, M. et al. The epidemiology of Schistosoma haematobium and S. mansoni infections in the Egypt-49 project area. Bulletin of the World Health Organization, 35: 319-330 (1966). 30. Feachem, R.G. et al. Water, health and develop- ment. London, Tri-Med Books, 1978. 31. Feachem, R.G. et al. Excreta disposal facilities and intestinal parasitism in urban Africa: preliminary studies in Botswana, Ghana and Zambia. Trans- actions of the Royal Society of Tropical Medicine and Hygiene, 77: 515-521 (1983). 32. Frell, L. & Wall, S. Exploring child health and its ecology. Acta paediatrica Scandinavica (suppl), 267: 1-180 (1977). 33. Habicht, J.-P. et al. Mother's milk and sewage: their interactive effects on infant mortality. Pediatrics, 81: 456-461 (1988). 34. Halnes, M.R. & Avery, R.C. Differential infant and child mortality in Costa Rica: 1968-1973. Population studies, 36: 31-43 (1982). 35. Hardy, D. et al. The cytology of conjunctival smears from Aboriginal schoolchildren at Yalata, South Australia, after improved hygienic conditions and treatment with oxytetracycline and systematic sul- phormetoxine. American journal of ophthalmology, 63: 1538-1540 (1967). 36. Henderson, P.L. et al. Guinea-worm disease in northern Uganda: a major public health problem controllable through an effective water programme. International journal of epidemiology, 17: 434-440 (1988). 37. Henry, F.J. Environmental sanitation infection and nutritional status of infants in rural St. Lucia, West Indies. Transactions of the Royal Society of Tropical Medicine and Hygiene, 75: 507-513 (1981). 38. Howarth, S.E. et al. Worms, wells and water in western Madagascar. Journal of tropical medicine and hygiene, 91: 255-264 (1988). 39. Huttly, S.R.A. et al. The epidemiology of acute diar- rhoea in a rural community in Imo State, Nigeria. Transactions of the Royal Society of Tropical Medi- cine and Hygiene, 81: 865-870 (1987). 40. llegbodu, V.A. et al. Source of drinking water supply and transmission of guinea worm disease in Nigeria. Annals of tropical medicine and parasitol- ogy, 81: 713-718 (1987). 41. Johnson, S. & Joshl, V. Drancontiasis in Rajasthan. VI. Epidemiology of dracontiasis in Barmer District, western Rajasthan, India. International journal of epidemiology, 11: 26-30 (1982). 42. Jordan, P. Epidemiology and control of schistosomi- asis. British medical bulletin, 28: 55-59 (1972). 43. Jordan, P. et al. Value of individual household water supplies in the maintenance phase of a schisto- somiasis control programme in Saint Lucia, after chemotherapy. Bulletin of the World Health Organi- zation, 60: 583-588 (1982). 44. Khalil, M. The pail closet as an efficient means of controlling human helminth infection as observed in Tura Prison, Egypt, with a discussion on the source of Ascaris infection. Annals of tropical medicine and parasitology, 25: 35-54 (1931). 45. Khan, M.U. Interruption of shigellosis by hand- washing. Transactions of the Royal Society of Tropi- cal Medicine and Hygiene, 76: 164-168 (1982). 46. Kholy, H.E. et al. Effects of boreholes on water uti- lization in Schistosoma haematobium endemic com- munities in coast province, Kenya. American journal of tropical medicine and hygiene, 41: 212-219 (1989). 47. Koopman, J.S. Diarrhea and school toilet hygiene in Cali, Colombia. American journal of epidemiology, 107: 412-420 (1978). 48. Kupka, K. et al. Sampling studies on the epidemi- ology and control of trachoma in southern Morocco. Bulletin of the World Health Organization, 39: 547-566 (1968). WHO Bulletin OMS. Vol 69 1991. 619 S.A. Esrey et al. 49. Lima e Costa, M.F.F. et al. A clinico-epidemiological survey of schistosomiasis mansoni in a hyper- endemic area in Minas Gerais State (Comercinho, Brazil). 1. Differences in the manifestations of schis- tosomiasis in the town centre and in the environs. Transactions of the Royal Society of Tropical Medi- cine and Hygiene, 79: 539-45 (1985). 50. Lima a Costa, M.F.F. et al. Water-contact patterns and socioeconomic variables in the epidemiology of schistosomiasis mansoni in an endemic area in Brazil. Bulletin of the World Health Organization, 65: 57-66 (1987). 51. Llndskog, U. et al. Childhood mortality in relation to nutritional status and water supply-a prospective study from rural Malawi. Acta paediatrica Scandina- vica, 77: 260-268 (1988). 52. Llndskog, U. et al. Child health and household water supply: a longitudinal study of growth and its environmental determinants in rural Malawi. Human nutrition: clinical nutrition, 41C: 409-423 (1987). 53. Logan, J.W.M. Schistosomiasis in Swaziland-a comparative study of three irrigated estates. Journal of helminthology, 57: 247-253 (1983). 54. Lyons, G.R.L. Guineaworm infection in the Wa dis- trict of north-western Ghana. Bulletin of the World Health Organization, 47: 601-610 (1972). 55. Magnanl, R. et al. Evaluation of the provincial water project in the Philippines. Washington, DC, Interna- tional Statistical Program Center, Bureau of the Census, U.S. Department of Commerce, 1984. 56. Malcuk, J.F. A study of trachoma and associated infections in the Sudan. Bulletin of the World Health Organization, 35: 262-272 (1966). 57. Markell, E.K. et al. Medical parasitology. Philadel- phia, W.B. Saunders, 1986. 58. Marshall, C.L. The relationship between trachoma and piped water in a developing area. Archives of environmental health, 17: 215-220 (1968). 59. Mason, P.R. et al. Piped water supply and intestinal parasitism in Zimbabwean schoolchildren. Trans- actions of the Royal Society of Tropical Medicine and Hygiene, 80: 88-93 (1986). 60. Mason, P.R. & Tswana, S.A. Single-dose metrifon- ate for the treatment of Schistosoma haematobium infection in an endemic area of Zimbabwe. Amer- ican journal of tropical medicine and hygiene, 33: 599-601 (1984). 61. Mathur, G.M. & Sharma, R. Influence of some socio- economic factors on the prevalence of trachoma. Indian journal of medical sciences, 24: 325-333 (1970). 62. Mlsra, K.K. Safe water in rural areas. International journal of health education, 18: 53-59 (1975). 63. Moore, H.A. et al. Diarrheal disease in Costa Rica. American journal of epidemiology, 82: 162-184 (1965). 64. Negron-Aponte, H. & Jobin, W.R. Schistosomiasis control in Puerto Rico. American journal of tropical medicine and hygiene, 28: 515-525 (1979). 65. Pickering, H. et al. Social and environmental factors associated with the risk of child mortality in a peri- urban community in the Gambia. Transactions of the Royal Society of Tropical Medicine and Hygiene, 80: 311-316 (1986). 66. Plmentel, D. et al. Aspects of schistosomal ende- micity in three Puerto Rican watersheds. American journal of tropical medicine and hygiene, 10: 523-529 (1961). 67. Pltchford, R.J. Findings in relation to schistosome transmission in the field following the introduction of various control measures. South African medical journal (suppl), October 8: 3-16 (1966). 68. Pltchford, R.J. Further observations on bilharzia control in the eastern Transvaal. South African medical journal, April 18: 475-477 (1970). 69. Popkin, B.M. Time allocation of the mother and child nutrition. Ecology of food and nutrition, 9: 1-14 (1980). 70. Portney, G.L. & Hoshlwara, I. Prevalence of tra- choma among southwestern American Indian tribes. American journal of ophthalmology, 70: 843-848 (1970). 71. Pratt-Johnson, J.A. & Wesels, J.H.W. Investigation into the control of trachoma in Sekhukuniland. South African medical journal, 32: 212-215 (1958). 72. Prost, A. & Negrel, A.D. Water, trachoma, and con- junctivitis. Bulletin of the World Health Organization, 67: 9-18 (1989). 73. Rahaman, M.M. et al. The Teknaf health impact study: methods and results. Paper presented at the International Workshop on Measuring Health Impacts of Water and Sanitation Programmes, Cox's Bazaar, Bangladesh, 21-25 November 1983. 74. Rahman, M. et al. Impact of environmental sanita- tion and crowding on infant mortality in rural Bang- ladesh. Lancet, 2: 28-31 (1985). 75. Reddy, C.R.R.M. et al. Epidemiology studies on guinea-worm infection. Bulletin of the World Health Organization, 40: 521-529 (1969). 76. Sahba, G.H. & Arfaa, F. The effect of sanitation on ascariasis in an Iranian village. American journal of tropical medicine and hygiene, 70: 37-39 (1967). 77. Schlilessmann, D.J. et al. Relation of environmental factors to the occurrence of enteric diseases in areas of eastern Kentucky. Washington, DC, U.S. Public Health Service, 1958 (Public Health Mono- graph No. 54). 78. Scoft, J.A. & Barlow, C.H. Limitations to the control of helminth parasites in Egypt by means of treat- ment and sanitation. American journal of hygiene, 27: 619-648 (1938). 79. Shiffman, M.A. et al. Field studies on water, sani- tation and health education in relation to health status in Central America. Progress in water tech- nology, 11: 143-150 (1978). 80. Snyder, J.D. & Merson, M.H. The magnitude of the global problem of acute diarrhoeal disease: a review of active surveillance data. Bulletin of the World Health Organization, 60: 605-613 (1982). 81. Stanton, B.F. & Clemens, J.D. Socioeconomic vari- ables and rates of diarrhoeal disease in urban Bangladesh. Transactions of the Royal Society o1 Tropical Medicine and Hygiene, 81: 278-282 (1987). 620 WHO Bulletin OMS. Vol 69 1991 Effects of Improved water supply and sanitation on six endemic diseases 82. Stanton, B.F. & Clemens, J.D. An educational inter- vention for altering water-sanitation behaviors to reduce childhood diarrhea in urban Bangladesh: II. A randomized trial to assess the impact of the inter- vention on hygienic behaviors and rates of diar- rhea. American journal of epidemiology, 125: 292-301 (1987). 83. Stanton, B.F. et al. Educational intervention for altering water-sanitation behavior to reduce child- hood diarrhea in urban Bangladesh: impact on nutritional status. American journal of clinical nutri- tion, 48: 1166-1172 (1988). 84. Stelb, K. & Mayer, P. Epidemiology and vectors of Dracunculus medinensis in northwest Burkina Faso, West Africa. Annals of tropical medicine and parasi- tology, 82: 189-199 (1988). 85. Stephenson, L.S. et al. Evaluation of a four year project to control Ascaris infection in children in two Kenyan villages. Journal of tropical pediatrics, 29: 175-184 (1983). 86. Stephenson, L.S. et al. Water, sanitation, and knowl- edge about urinary schistosomiasis in a Kenyan coastal community: a study combining ethnographic and survey techniques. Cornell international nutri- tion monograph series, 16: 70-191 (1986). 87. Stewart, W.H. et al. The relationship of certain environmental factors to prevalence of Shigella infection. IV. Diarrheal disease control studies. American journal of tropical medicine and hygiene, 4: 718-724 (1955). 88. Sutter, E.E. & Ballard, R.C. Community participation in the control of trachoma in Gazankulu. Social science and medicine, 17: 1813-1817 (1983). 89. Sweet, W.C. et al. A study of the effect of treatment and sanitation on the level of hookworm infestation in certain areas in Panama. American journal of hygiene monograph series, 9: 98-138 (1929). 90. Tamelm, 0. et al. Control of schistosomiasis in the new Rahad irrigation scheme of central Sudan. Journal of tropical medicine and hygiene, 88: 115-124 (1985). 91. Taylor, H.R. et al. The ecology of trachoma: an epi- demiologic study in southern Mexico. Bulletin of the World Health Organization, 63: 559-567 (1985). 92. Tlelsch, J.M. et al. The epidemiology of trachoma in southern Malawi. American journal of tropical medi- cine and hygiene, 38: 393-399 (1988). 93. Torun, B. Environmental and educational interven- tions against diarrhea in Guatemala. In: Chen, L.C. & Scrimshaw, N.S., ed. Diarrhea and malnutrition: interactions, mechanisms, and interventions. New York, Plenum Press, 1982, pp. 235-266. 94. Udonsl, J.K. Control of endemic dracontiasis by pro- vision of water supply in rural communities of Imo State, Nigeria. Public health, 101: 63-70 (1987). 95. VIctora, C.G. et al. Water supply, sanitation and housing in relation to the risk of infant mortality from diarrhoea. International journal of epidemi- ology, 17: 651-654 (1988). 96. Walsh, J. Estimating the burden of illness in the tropics. In: Warren, K. & Mahmoud, A., ed. Tropical and geographical medicine. New York, McGraw- Hill, 1984. 97. Watts, S. Dracunculiasis in Africa in 1986: its geo- graphic extent, incidence, and at-risk population. American journal of tropical medicine and hygiene, 37: 119-125 (1987). 98. WaxIer, N.E. et al. Infant mortality in Sri Lankan households: a causal model. Social science and medicine, 4: 381-392 (1985). 99. West, S. et al. Water availability and trachoma. Bul- letin of the World Health Organization, 67: 71-75 (1989). 100. White, G.F. et al. Drawers of water: domestic water use in East Africa. Chicago, University of Chicago Press, 1972. 101. Wilkins, H.A. et al. Resistance to reinfection after treatment of urinary schistosomiasis. Transactions of the Royal Society of Tropical Medicine and Hygiene, 81: 29-35 (1987). 102. Wilson, M. et al. The epidemiology of trachoma in Chiapas (Mexico). Revue internationale du tra- chome, 64: 133-142 (1987). 103. Young, B. & Briscoe, J. A case-control study of the effect of environmental sanitation on diarrhea mor- bidity in Malawi. Journal of epidemiology and com- munity health, 42: 83-88 (1987). 104. Zaheer, M. et al. A note on urban water supply in Uttar Pradesh. Journal of the Indian Medical Associ- ation, 38: 177-182 (1962). WHO Bulletin OMS. Vol 69 1991. 621