Department of Tropical Medicine

Tulane School of Public Health and Tropical Medicine

1501 Canal Street Room 513 SL-29

New Orleans, Louisiana 70112

Key words: anthropology and tropical medicine, schistosomiasis control, S. haematobium, behavioral and cultural aspects of schistosomiasis control

Correspondence: B.S. Hewlett, Department of Anthropology, Washington State University, Pullman, WA 99164, e-mail hewlett@wsu.edu

Tropical disease control projects often utilize medical anthropologists to conduct KAP (knowledge, attitude and practices) or other social-behavioral studies, but these studies are usually advisory. Seldom are medical anthropologists actively involved in decision-making and implementation of tropical disease control and prevention projects. This project was different from many other tropical disease control efforts in that there was regular and extensive collaboration between tropical medicine and medical anthropology researchers. U.S. and Cameroonian medical anthropologists and tropical medicine researchers actively assisted and participated in each other's activities. For example, the Cameroonian and U.S. snail control biologists and epidemiologists participated in community meetings and helped develop community questionnaires and health education materials, while the U.S. and Cameroonian medical anthropologists assisted in the collection and diagnosis of urine and stool samples as well as in the monitoring of snail control efforts. This collaboration was also exemplified in the general administration of the project: the principal investigator was a tropical medicine researcher (the second author) while the field-director (the first author) was a medical anthropologist. Pat Rosenfield (1992) calls this regular collaboration a "transdisciplinary" approach and contrasts it with the traditional "multidisciplinary" approach where researchers work independently and seldom understand nor appreciate each other's methods and techniques.

Before describing the Northern Cameroon schistosomiasis control project it is essential to understand two complementary conceptual orientations, one primarily from tropical medicine and the other primarily from medical anthropology, that unified and motivated the design and implementation of the pilot control project. These were implicit underlying views that influenced and guided day-to-day planning and activities. The first conceptual orientation emphasized utilizing urinary schistosomiasis as an entry point to strengthening and building the primary health care system. Integrating schistosomiasis control into the existing primary health care system (horizontal rather than traditional vertical control effort) was an important explicit goal of the project, but implicitly the project was also interested in trying to utilize urinary schistosomiasis to increase utilization of and trust/faith in rural health care. Utilization rates are often low in rural clinics because health care workers often lack the training, equipment, supplies or medications to diagnose and treat endemic disease. Urinary schistosomiasis is an excellent entry point to build primary health care in this area (Northern Cameroon) because it is highly endemic, the symptoms (haematuria and painful urination) are identified by the local people as an illness that needs treatment, the diagnostic techniques are cheap and easy (reagent strips), and treatment (praziquantel) is cheap and effective (i.e., offers prompt resolution of symptoms). Reliable and efficient treatment of a common disease like schistosomiasis builds trust/faith in the local health center, which can in turn lead to an increase in utilization means the health center, which can result in the increase in money the center can use to support the diagnosis and treatment of other, less frequent, possibly more severe, diseases. This conceptual orientation was important because it reflected our genuine commitment to enhancing primary health care in general and not just promoting the control of one disease. Health care providers appreciated this orientation and it meant that schistosomiasis project resources (medical personnel, vehicles, etc.) sometimes provided training or support of other pressing community health problems (e.g., venereal disease, mengiococcal meningitis epidemic).

The second conceptual orientation originates in anthropology, but is not entirely new to tropical medicine because its basic tenets were suggested some years ago by Fred Dunn. This view emphasized building upon existing beliefs and institutions and respecting diversity between and within communities (cultural, religious, medical, individual). This orientation may seem intuitively obvious to many, but it has several implications for project design and implementation. It places individuals and communities at the center of a project. One needs to know communities very well if one is going to build upon and respect what is already there and clearly understand what types of control efforts work and which do not. It means frequent and regular visits to communities. What do elementary school students, parents, teachers and health workers already know about schistosomiasis? What local community organizations (women's groups, church groups, parent-teacher groups) might be able to incorporate schistosomiasis health education into their activities? What community institutions might incorporate or assist with a drug-delivery system? This orientation implies the project would try not introduce new institutions or positions (e.g., Community Health Workers to provide health education or community pharmacies to provide praziquantel) because it assumes communities already have informal or formal structures to build upon. This conceptual orientation also means working with and responding to diversity in individuals' abilities and personalities; some village chiefs are powerful and can mobilize communities while others are not; some health center workers respond primarily to material incentives, while others are intrinsically motivated by leaning a new technique to diagnose schistosomiasis. This does not mean the project paid some health center personnel but not others to participate in the schistosomiasis control project; it did mean that project personnel would work with, build upon and respect what existed--some health workers were motivated by having a beer or lunch with project personnel while others were motivated by learning new skills from project personnel.

While these conceptual orientations emphasize decisions from "below" rather than from "above", it was clear that political, social and economic decisions in the U.S. and in the Cameroonian capital (both very distant from Cameroon communities) dramatically influenced what the communities and project personnel could or could not do.

One cannot understand the schistosomiasis control project described below without understanding these conceptual orientations. They provided the motivation and guiding forces in organizing and implementing the day-to-day activities of the schistosomiasis control project. It is our impression that both of the above described orientations contributed substanially to the success of the project and increased the community members' confidence and trust in local health care and the diagnosis and treatment of schistosomiasis.

The Cameroon Pilot Control Project

The Cameroonian Ministry of Health (MOH) selected Kaelé subdivision (population of about 100,000) in the Far North Province for a "pilot" schistosomiasis control to help it formulate a national plan for the control of schistosomiasis. Kaelé is the largest town in the subdivision (population 30,000) and is about 750 km north of Yaoundé, the capital of Cameroon. The MOH requested and received technical assistance from USAID to conduct the pilot control and establish a national control program. Tulane University's School of Public Health and Tropical Medicine (TU) was already involved with the MOH and USAID in developing the MOH's capacity to conduct research on schistosomiasis and other tropical diseases so USAID requested TU provide technical assistance for the Kaelé pilot control.

Kaelé subdivision was selected because a national survey of 5th grade students indicated that 85% of all schistosomiasis cases were in the North and Far North Provinces (Ratard et al. 1990). Subdistrict prevalences for urinary schistosomiasis in school-age children averaged 35%, but Kaelé subdistrict was one of the highest at 55%. The project focused on 14 villages in the subdivision with a health center or a "health house."

The control project consisted of 4 components: health education, snail control, diagnosis and treatment, and cost-recovery. The first three components were developed to lower morbidity by trying to interrupt the schistosome life cycle at four vantage points---diminishing human contact with infected water sources, reducing human urine and fecal contamination of water sources, killing the parasite in the human host through chemotherapy, and removing or reducing the number of intermediate hosts (snails). Elementary school children were targeted for health education as this groups had the highest prevalence rate. The two overriding goals of the project were: 1) heavy urinary schistosomiasis infection (>50 eggs/10ml) should be reduced by 75%, and 2) 70% of the elementary schoolchildren should have an understanding of schistosomiasis transmission. Detailed descriptions and specific objectives of each component are described elsewhere (Cline and Hewlett 1996).

The explicit project design was similar to many other integrative (i.e., horizontal) and comprehensive schistosomiasis control projects, but was relatively unique in that 1) diverse approaches to health education and community involvement were central and perceived as the driving forces for all other components (diagnosis and treatment, snail control, cost recovery), and 2) elementary schoolchildren and other patients had to pay for diagnosis (screening at schools or at clinic) and possible treatment. Other schistosomiasis control projects have targeted schoolchildren, but we are unaware of other control efforts in which students or patients had to pay for diagnosis and possible treatment.

Health education was the central component of the project. Simple, but culturally sensitive and appropriate health education methods were needed to build upon local peoples understanding about the transmission of the disease and the availability of new, cheap and easy diagnostic methods and treatments. Local people would not seek treatment or support a snail control program if they did not understand the consequences of not being treated or perceive the social or economic costs to be greater than the benefits.

The construction of pit latrines and water pumps are other methods to reduce schistosomiasis transmission and were encouraged locally and nationally, but the project was not sufficiently funded to subsidize the construction of these important items. The project worked with UNICEF and Save the Children, both of whom were involved with establishing community co-financed wells in the region. The project was also involved at "structural" or outside the community (called extracommunity by Dunn) levels as it aimed at establishing a sustainable regional and national drug-delivery (in particular, praziquantel) and reagent-strip delivery systems. With the assistance of Mr. Kondji Kondji, we also advocated and supported national efforts to increase the training, emphasis and value placed on health education of nurses and physicians. Nurses that ran the primary health care centers had little background in methods, materials or philosophies of health education and often did not feel that health education was part of their job, in large part, because it was not an integral or valued part of their training as a nurse.

Ecological and Cultural Overview of Kaelé Subdivision

Kaelé subdivision is located at about 12oN latitude and lies at the very northern limits of the Sudanic climatic zone. The area receives between 900-1500 mm of rain annually and is 200-500 m in elevation. The landscape is relatively flat with a few basalt outcrops and the vegetation is dominated by thornbrush, semidesert scrub and a few trees (baobob and acacia). The rainy season starts in June and usually ends by late-August. Critical and intensive agricultural activities take place during this period. There are few permanent water bodies in the region, but temporary ponds and streams (maayos) are numerous during, and for some months after, the rainy season (Greer 1993). Daytime temperatures are rather mild during the rainy season (25-30oC), but are relatively warm the remainder of the year. It can be cool (20oC) in the months of December and January and this time of year is often referred to as the dry-cold season. This season contrasts with the hot-dry season that follows in February-March when daytime temperatures are often above 40oC.

The ecological landscape is relatively homogenous by comparison to the heterogeneous nature of the cultural and linguistic landscape. Three very distinct ethnolinguistic ethnic groups--the Guiziga who speak a Chadic language, Fulbe who speak a West Atlantic language and Moundang who speak an Adamawan language--are the predominant ethnic groups in the relatively small (70 km by 100 km--the size of many counties in the Western U.S.) Kaelé subdivision. The Moundang are the most numerous and the Fulbe a minority (about 25%) among the 100,000 inhabitants of the subdivision. Each village is represented predominately by one ethnolinguistic group. Fulbe is the lingua franca (i.e., trade language), French is taught in public and private elementary schools and Arabic is taught in the Koranic schools. Most of the Fulbe are Islamic while the majority of the Moundang and Guiziga identify themselves as Christian (Protestant or Catholic). Most Guiziga and Moundang attend private (Catholic or Protestant) or public schools while many Fulbe attend village Koranic schools. Several of the Moundang and Guiziga villages have Catholic or Protestant supported primary health care clinics. The Fulbe moved into the area in the 1800's as part of a Holy War and are the politically dominant group in Northern Cameroon, but the Moundang are the politically dominant in the Kaelé subdivision. Intermarriage between the different ethnolinguistic groups is relatively common so there is often an awareness of cultural differences (e.g., different views towards schistosomiasis).

Fulbe houses are usually rectangular and are surrounded by a large wall, while Moundang and Guiziga homes are round and usually do not have walls surrounding them. The lives of men and women are more separate among the Fulbe than among the Moundang and Guiziga. Fulbe women usually do not participate in agricultural tasks because of Islamic restrictions on women's public movements; men and women have separate latrines in the residential compound and men and women wash their own clothes. Women collect the water and firewood in all groups. All groups practice patrilineal descent and patrilocal post marital residence.

Millet is the principle food crop, supplemented by corn, peanuts and cow peas. Onions are gown in irrigated gardens and, together with cotton, constitute the main cash crops. Goats, sheep and cattle are also kept by many villagers. The Fulbe are the prime cattle producers and onion cultivators, while the Guiziga and Moundang are better known for their cotton cultivation.

Cultural Contexts of Urinary Schistosomiasis

Before control project began, Helen Regis, a TU graduate student in anthropology, conducted a one-year ethnographic study of schistosomiasis in one Fulbe village (Regis 1997). This was part of the first, primarily research, phase of the TU schistosomiasis project . Her detailed ethnographic data provided the basis for more systematic ethnographic studies in randomly selected villages that represented all primary ethnic groups in the subdivision. This paper will not describe the details of the ethnographic and KAP methods that followed the Regis study, but will summarize some of the local beliefs and practices that were eventually incorporated into the schistosomiasis control effort.

Urinary schistosomiasis is known throughout the area by its most marked symptom, haematuria, or blood in the urine, and is viewed by all ethnic groups as an illness that needs treatment. The Fulbe call urinary schistosomiasis cille naange or sun urine. They explain that when they spend long hours walking or working in the sun, their urine becomes red like the sun. While anyone can get the illness, children are especially susceptible because they spend so much time out in the sun, especially during the hot-dry season, when cille naange is said to be most common. Consequently, parents often tell their children to get out of the sun. Some Guiziga and Moundang indicated that staying out in the sun too long could cause red urine (i.e., urine with blood), but they were more likely to suggest that drinking dirty water (i.e., especially, when wells get low during the hot-dry season) or drinking water from two very different sources (e.g., well and temporary pond) as causes of red urine (tetchoume sjmi in Moundang kwanay babaran in Guiziga). Guiziga and Moundang also frequently distinguished "red urine" from "white urine". White urine was more painful, more dangerous, sexually transmitted and not as common as red urine. Red urine was much more common, not as dangerous and less likely to be sexually transmitted; it acted slowly as one often continues to be sick even after the hot-dry season when the rains start and there may not be more blood in the urine. Back and kidney pain were also associated with red urine. Red urine is urinary schistosomiasis and white urine is usually gonorrhea. The Fulbe viewed the disappearance of blood after the hot-dry season as evidence of healing.

All groups felt that children were at a greater risk because they either stayed out in the sun too long or were more likely to drink dirty water or mix water from different sources. Young children (boys and girls ages 4-15) are responsible for grazing goat and sheep herds, usually at temporary bodies of water some distance from the village. Children are likely to be out in the sun long periods of time, swim in the temporary bodies of water because it is so hot and drink from these bodies of water since no other water sources are available.

A similar situation occurs for the whole family at the end of the rainy season. Dry millet is planted during this season in fields that are several kilometers from the house because it only grows in very specific kinds of soil. Every able bodied person in the household assists in the planting of this staple crop (children are usually kept out of school to do this). Everyone drinks and bathes in the temporary ponds near the dry millet fields because there are no other sources of water.

People generally distinguish five different types of water sources in the environment: deep wells in the village, temporary shallow wells dug in the dry sand beds of the maayo, flowing water in the maayo following a rain, low points in the maayo that collects water and temporary ponds in depressions in the outlying savanna areas. People tended to prefer to obtain their drinking water from the temporary shallow wells dug in the dry sand beds and preferred to do their wash in flowing water, if possible. Women in all groups did the strenuous task of providing water for the household; many women said this was their most demanding daily task so efficiency was very important to them--i.e., locating water sources close to home.

The Fulbe were most likely to build latrines in their compounds, while relatively few Guiziga and Moundang built latrines near their homes; people urinated and defecated in fields close to their home.

Before describing the anthropological contributions to the different components of the project it is necessary to briefly define and clarify some terms. Anthropologists often talk about doing "ethnography"--that is, the study of a culture. To conduct an ethnography an anthropologist utilizes several methods (see Koss chapter for more detail)--sometimes one utilizes qualitative participant observation methods, such as, going to the fields with villagers, attending a market-day and informally interviewing people while participating and observing these activities, and sometimes one utilizes precoded standardized questionnaires of a random sampling of villages (e.g., KAP studies). In the section that follows, the term "ethnographic studies" encompasses a range of anthropological methods.

Linking Health Education to Culture and Ecology

Local communities were dealing with urinary schistosomiasis long before the TU pilot project started. The role of the project therefore was to build upon existing knowledge and practices related to schistosomiasis rather than to try and completely replace and denigrate indigenous knowledge with a Western biomedical model. Ethnographic studies provided the basis for incorporating local community knowledge and practices into health education. The following list and descriptions summarize the cultural beliefs and practices that were built upon, left alone, or targeted for change.

--indigenous terms for the illness

--perception as an illness that needs treatment

--haematuria and painful urination are the most common symptoms of the illness

--children are at greater risk because of greater exposure to sun or greater

chance of drinking dirty water

--dirty water causes the illness

--keep children out of mid-day sun

--the illness is seasonal

--the distinction between red and white urine

--use of latrines by Fulbe

All ethnic groups had indigenous terms for the illness, knew the primary symptoms, recognized that it needed treatment and knew that children were at greater risk. This made health education substantially easier as indigenous terms for the illness could be utilized in health education discussions, people were already used to seeking traditional treatments for the illness, it was easy for people to see the value of their knowledge, and they already understood why health education targeted school-age (5-19 year-olds) children. This contrasts to other cultures where urinary schistosomiasis is perceived as a normal part of adolescent development and not an illness that needs treatment (Kloos 1995, Huang and Manderson 1992).

Local people also described a seasonality for the illness. Most people felt it was primarily an illness of the hot-dry season because the sun is especially hot during this season or there was a greater possibility of drinking dirty water because of the limited water supplies during this season. In order to build upon this the health education program and messages were divided into two seasons--"season of symptoms" and "season of transmission." The hot dry season (February-April) was characterized as the "season of symptoms" and built upon existing knowledge and perceptions that this is the time of year haematuria and painful urination were most common. This is also the time of year many people are most willing and able to purchase praziquantel for treatment because they receive a lump sum cash payment for their cotton in this season. Health education during this season focused on seeking treatment. The second season for health education was end of the rainy season (August-September) and was called the "season of transmission" because temporary streams and ponds were common and frequently utilized for planting millet, washing clothes, bathing and swimming. It also coincides with the start of the school-year. Health education messages during this time of the year focused on decreasing contamination behavior.

People knew that children were at greater risk because children spent more time in the hot sun and were more likely to drink dirty water. Consequently, parents understood why health education focused on school-age children and were usually willing to pay for their children's diagnosis and treatment. Fulbe parents also said they tried to keep children out of the mid-day sun to prevent cille naange which often resulted in keeping children away from swimming holes in the middle of the day. This was utilized in health education as cercariae concentrations are highest in the middle of the day. Moundang and Guiziga were more likely to identify dirty water as a cause of the illness. Health education emphasized their existing knowledge about the importance of dirty water in the transmission of the illness, but indicated it was from contact with dirty (contaminated by human urine or feces) water rather than drinking dirty water.

The distinction between red and white urine made the Moundang and Guiziga was also utilized in health education, primarily to emphasize the extent of their existing knowledge, in particular, their ability to recognize schistosomiasis and distinguish it from other types of urinary infections.

--children swimming in streams and temporary bodies of water

--herbal treatments

It was decided that decreasing water contact would not be emphasized in health education. It is hot year-round in the arid north and it is not reasonable or feasible to ask people, children, in particular, to stay out of the water when the temperature is often over 40 degrees C (over 100 degrees F). The project or community could not provide another alternative to cooling off, bathing and washing clothes, so the health education focused instead on contamination behavior--urinating and defecating far away from the ponds and rivers. This was especially important in Islamic communities where villagers preferred to clean themselves with water after elimination.

It was also decided not to encourage or discourage the use of traditional herbs as it was not possible to investigate their pharmacological components. Villagers noted a variety of herbal treatments for red or sun urine, but tamarind seeds (crushed and boiled) were most frequently mentioned as an herbal treatment, especially among the Guiziga and Moundang. A recent study of traditional plants utilized to treat urinary schistosomiasis by Zimbabwe healers (Ndamba et al. 1994) indicates that plant extracts from other Leguminosae, such as the tamarind tree, were lethal to adult schistosomes. Many of these plants have tannins, which produce a red color, and the authors suggest symbolic links between tannins and haematuria, the primary diagnostic feature. Many people reported that traditional herbal treatment decreased symptoms, but usually did not cure the illness.

--lack of knowledge about the role of snails as intermediate hosts

--lack of knowledge about the role of water contact

--lack of knowledge of the role of excretory behavior in contaminating water

sources with schistosomiasis

--perception of hot-dry season as primary season of transmission

--perceptions of severity

--perceptions of cost for diagnosis and treatment

--perceptions of the use of latrines

While local people had some understanding of the causes, transmission and treatment of urinary schistosomiasis, there were aspects of the disease that people had limited or no understanding. Not surprisingly, none of the groups recognized the role of snails in the transmission of the illness. It was necessary to add this knowledge if local people were to support a snail control effort. Many people knew dirty water contributed to transmission, but thought it was the drinking of the water rather than standing or swimming in the water that transmitted the illness. Many people knew that defecating or urinating in the water could transmit parasitic illnesses, but there was no link between excretory behavior in or near water sources and urinary schistosomiasis. Health education emphasized urinating and defecating far from water sources and the use of latrines if available. All groups thought the hot-dry season was the most likely season of transmission, so health education was divided into the two explicit seasons already described above.

The local perceptions of severity as they influenced treatment-seeking behavior are discussed in the next section.

Previous to the project many people did not seek health center diagnosis and treatment because the health center did not have the means to diagnose or treat the disease, or because they knew the treatment was extremely expensive. Few health clinics had microscopes and those that did often did not know how to use them. Only two of the villages had praziquantel in their village pharmacies, but it was prohibitively expensive ($12-$20 for an adult treatment) for most villagers. Consequently, health education messages emphasized that diagnosis (especially with reagent strips) and treatment (praziquantel in each village pharmacy) were cheap, easy and locally available.

Health education training was provided to all health workers (about 75) and elementary school teachers (about 250) in the subdivision. Teachers were encouraged to work with their local health center and writing and drawing competitions on schistosomiasis were held to engage students. Extensive community input was utilized to develop a schistosomiasis flipchart, brochure, poster and curriculum, but the trainings in the use of the health education materials emphasized the methods of presenting these materials rather than letting the materials speak for themselves--i.e., adapting the materials to the age, sex and abilities of the group; listening to people's explanations for llnesses, building upon local knowledge and practices and providing simple, clear and culturally sensitive health education messages.

Each health center also developed written health education action plans for the season of transmission and the season of symptoms. Health education plans helped health center personnel identify and schedule health education for a diversity of community groups--public, private and Koranic schools, churches, traditional healers, women's groups, cooperatives, traditional chiefs, political leaders, etc.

Anthropological Contributions to Other Components

Anthropological contributions to health education were of primary importance because health education was assumed to be the driving force of other componets, but anthropology also contributed to the other components of the study--diagnosis and treatment, snail control, and cost-recovery.

Treatment-seeking was an area of concern. Diagnosis was cheap (less than 20 cents) and relatively easy, especially with the reagent strips, but the praziquantel treatment for an adult was more expensive (about $1.50). The price of praziquantel was based upon what it costs to purchase the drug from a wholesaler and deliver the drug to the community. This was a substantially lower price than previously available, but ethnographic follow-up of non-treated cases indicated that perceptions of severity and cost were important considerations in treatment-seeking.

Schistosomiasis was an illness that needed treatment: many said they could eventually die from it if it was not treated, and that they would get treatment as soon as they could afford it. But many people indicated that traditional treatments temporarily relieved symptoms (red urine and/or pain on urination) and some people said symptoms went away on their own after diagnosis. Some mothers mentioned they had five or more children and could not presently afford treatment. While cost was a factor, most said they would obtain treatment (or provide it to their children) after the cotton harvest--the prime cash income activity in the area. Villagers usually received a lump sum payment for their cotton in January or February and this is, in part, why treatment-seeking was emphasized in the season of symptoms--farmers were more likely to have the cash. As mentioned above, this is also the time of year when symptoms are most common.

Severity was also a concern as people were willing to wait months before seeking treatment. Consequently, health education training and materials were modified to discuss the consequences of not being treated. Bladder cancer and possible decreased fertility were mentioned, but greater emphasis was placed on decreased fertility or potential infertility because cancer is seldom diagnosed in the area and fertility is central to marriage stability and social status (i.e., kinship resources--children, in particular--are central to economic and social well-being) and is of concern to both men and women.

Ethnographic studies also contributed to the snail control effort. Snail control was attempted late in the project, in part, due to fact that many people (Fulbe, in particular) did not link transmission to water. Community-wide health education had to precede any attempt at mollusciciding.

Ethnographic studies indicated the fish and snails killed by the molluscicide were not of much economic or social importance. Villagers indicated they were most likely to use the shells of the snails as toys or for washing their pans; the small fish in the ponds were eaten at the end of the rainy season, but they were not regular or important elements of the diet. When the ponds were actually treated there were few comments about dead fish or snails.

The greatest concern of local people both before and after mollusciciding had to do with the quality and nature of the water. The treated ponds were not sources of drinking water, but could be utilized for swimming and bathing. Some health workers and villagers were not interested in snail control because of previous negative experiences with DDT treatments of ponds for mosquito control several years ago; it was only after the ponds were treated did the people find out that DDT was dangerous to their health and that they should not bath or swim in this water. Would the molluscicide do the same? Only two villages attempted community-based snail control and the one with some previous experience with DDT experienced less community support. The community in general wanted snail control, but not as much money was raised to treat ponds and fewer ponds were treated in this community. Ethnographic studies (participant observation) also indicated the potential transmission importance of ponds near millet fields a few kilometers from the village that were planted at the end of the rainy season. Previous to this it was assumed that most transmission was taking place in streams and ponds next to the village (water collecting and washing sites). Health education included statements about the risks during this season.

Several of the anthropological contributions to cost-recovery have already been described. Ethnographic studies evaluated how much people were willing to pay for diagnosis and treatment and when they were most likely to pay. But ethnographic studies were also important for helping to identify what structures in each community could be build upon to manage the drug delivery and cost-recovery systems. Some villages already had small village pharmacies while in other villages health personnel provided drug delivery.

Impact of community-centered health education

Four villages (of the 14 in Kaelé subdivision with health centers) were selected as "assessment" villages. A fifth village in another subdivision was selected as a "control" for health education. The "control" village was provided reagent strips and praziquantel at the same price as villages in the assessment villages, and the health center personnel received the same training as health workers in the assessment villages. The control village also received the health education materials (flipchart and poster). School-teachers in the village did not receive health education training nor did the health clinic staff develop health education plans for the community. The control village (for ethical reasons) was provided the same capacity for diagnosis and treatment as the assessment villages and therefore it could not serve as a control for the parasitologic impact of diagnosis and treatment.

Tables 1-3 summarize some of the significant differences between "assessment" and "control" village school-age children's knowledge about schistosomiasis. Table 1 indicates that children in the assessment villages were significantly more likely than control village children to recognize the role water contact (swimming or bathing in the maayo) in contracting the disease, and were less likely to associate it with walking in the hot sun than children in the control village. Drinking dirty water was similarly perceived as a source of the disease in both the control and assessment villages. Table 2 indicates assessment village children clearly understood the role of snails in the transmission of the disease while Table 3 illustrates an increased awareness of the of role of the rains rather than the sun in the transmission in of the disease.

While adults were not targeted as frequently as were children, health clinic staff did include all segments of the community in health education action plans. Children were also encouraged to bring the brochures home and share their information from school with their parents and others at home. Tables 4 and 5 indicate adults had a less marked change in schistosomiasis knowledge in comparison to children. Table 4 indicates adults in assessment villages better understood the role of water contact that adults in the control village, while Table 5 indicates an increase in knowledge about the role of the rains, end of rainy season, in particular, in the transmission of the disease while the control village adults were more likely to link the disease to the hot-dry season. Assessment village adults' knowledge of the role of snails in transmission was also double that of adults in the control village (50% to 25%).

Similar patterns in adult's knowledge were found in another small but independent evaluation of the project (Bausch and Cline 1995). Adults in an inadvertently omitted community were significantly less likely to identify swimming in ponds as a risk factor and to identify snails as important factors in its transmission by comparison to adults in the center of the village.

Table 6 summarizes some of the changes in the utilization of the health centers for schistosomiasis. There was a six-fold jump in the number of reported cases, but even more importantly the total number of consultations at the health centers increased.

Overall prevalence of infection declined from 21% to 7% (67% reduction) in the four assessment villages between 1991 and 1993, and the number of heavy infections (>50 eggs/10 ml of urine) in assessment villages declined 65% (23 to 8) (see Cline and Hewlett 1996 for details of parasitologic evaluations). In the fortuitous study of the omitted community, the prevalence of schistosomiasis infection in school-aged children in the omitted community was 71% while it was only 7% 2 km away in the part of the village that received health education. High intensity infection was 1% in the village center and 26% in the outlying community (Bausch and Cline 1995).

Finally, it is important to point out that local people paid for their diagnosis and treatment. The project did not subsidize diagnosis or treatment so all costs (with the exception health workers salaries) associated with control were supported by the community.

Many people from many different disciplines contributed to the success of the Kaelé schistosomiasis control project. This paper has highlighted the anthropological contributions and the extensive collaboration between anthropology and tropical medicine researchers.

Philosophical and conceptual frameworks from both disciplines were complementary and provided the motivation and organizational framework for day-to-day activities--build upon what already exists and utilize urinary schistosomiasis to build PHC. This framework defined the community-centeredness of the project. The 14 Kaelé communities held meetings at the beginning and various points during the project and a KAP was administered at the start of the project, but no new volunteer positions ("community health workers") were established to help implement the program. These meetings, the KAP and the establishment of CHWs are often the measures of "community participation." But community-centered or community-based means much more than this--in order to respond to communities and incorporate communities into the tropical disease control it means knowing the communities very well and responding to a diversity of political, social, cultural, economic and personality types. This means frequent interactions with various segments of the community. Hielscher and Sommerfeld (1985:481) have a similar view and state "There is no single, universal way to motivate community participation. Success largely depends on factors that differ from village to village, e.g., the authority of the village chief, the internal cohesion of the village community, the general economic situation and economic stratification."

Indigenous knowledge of the Fulbe, Moundang and Guiziga made this urinary schistosomiasis control project relatively easier than others in that all groups had an indigenous term for the illness and perceived it as an illness that needed treatment. Several others African cultures utilize haematuria, in part, to evaluate the coming of age in men (Akogun 1991, Bello and Idiong 1982, Nash et al. 1982). Among the Luguda, a man's family may have to pay four times the regular brideprice if he has not demonstrated "manly [red] urine (Akogun 1991)." These and several other reasons are often given for suggesting that urinary schistosomiasis is not regarded as a priority in many communities (Kloos 1995, Huang 1992). While urinary schistosomiasis is viewed as an illness needing treatment in the Kaelé area, it is also true that families have several other health needs besides urinary schistosomiasis and there were several instances where individuals did not seek treatment because of costs (of drug, getting to clinic, other competing family expenses), perceived benefits (would they get reinfected?), perceived severity (will it go away on its own) and the availability of traditional treatments. Discussing possible decreased fertility seemed to engage men and women in health education sessions and reportedly impacted perceptions of the illness.

The project had some rather obvious weaknesses. First, relatively little energy was invested in structural or "extracommunity" change. Governmental and non-governmental agencies were approached about building more wells, construction of latrines was encouraged, and an effort was made at the national level to increase and place greater value on health education in PHC training, but most of the time emphasis was placed on change at the individual and community level. Second, with all of the emphasis on community-centeredness, we were surprised to discover a community 2 km from a village with a health clinic had been omitted from project activities. The project focused on the center of the villages--this is where the health clinics and most schools are located. This points to two problems--the project probably did not serve the poorest of the poor and intracommunity political, religious, and other forms of diversity can and does lead to village inequality in health care delivery and services.

This was not the first schistosomiasis project to try and integrate schistosomiasis control into PHC (Sibiya 1986), nor was it the first to use health education as the driving force for other components (Tanner 1989, Kloos 1995), but it was is relatively distinct in a few ways.

This project appears relatively unique in that people had to pay for diagnosis and treatment. This meant parents had to pay for the diagnosis and possible treatment for four or more children. We are unaware of other African schistosomiasis control projects where this has been attempted. Most projects provide free mass screening and treatment or mass treatment with praziquantel to school-age children along with health education (Chandiwana et al. 1991).

Second, it is one of the few urinary schistosomiasis projects to try and evaluate the impact of health education. An assumption of the project was that the ability of the community to participate in control activities rested on the delivery of health education. One of the major criticisms of health education in schistosomiasis control projects is that seldom is the impact of health education evaluated (Kloos 1995). This project evaluated health education in two ways--one intentional one inadvertently--and both were consistent in that villages with community and school health education experienced lower prevalence and intensity of infection. Having cheap and simple diagnostic and treatment methods was not enough to motivate change. These methods of evaluating health education are very limited in that it is not clear what aspects of health education actually contributed to the lower prevalence--e.g., did it increase treatment-seeking or did it decrease contamination behavior? We suspect the later, especially given the short duration of the project, but it will be important in the future to evaluate precisely how health education influences schistosomiasis morbidity.

Third, it is one of the few urinary schistosomiasis control projects not associated with a water development project (i.e., dam construction or irrigation scheme). The temporary ponds and maayos are "natural" parts of the landscape. Water development projects have generally led to relatively recent and dramatic increases in prevalence and intensity of urinary schistosomiais. The high prevalences in the Kaelé region do not appear to be recent and may explain, in part, why local peoples have a relatively good understanding (from a Western biomedical perspective) of the illness--i.e., the high prevalences do not appear to be a recent development and consequently people have been adapting (biologically and culturally) to the disease for some time.

Finally, it is probably one of the few tropical disease control projects to utilize aspects of the "transdisciplinary" approach advocated by Rosenfeld (1992). Social scientists, epidemiologists, biologists and tropical medicine researchers with the project activevly participated in each others activities and this led to to increased cooperation, better communication and appreciation of each other’s methods and techniques. As we become more familiar with each others disciplines it may be possible to merge theories, concepts and approaches to solve a common problem.

References Cited

Akogun, O (1991) Urinary schistosomiasis and the coming of age in Nigeri. Parasitology Today 7,62.

Bausch D & Cline BL (1995) The impact of control measures on urinary schistosomiasis in primary school children in Northern Cameroon: A unique opportunity for controlled observations. American Journal of Tropical Medicine and Hygiene 53, 577-580.

Bello CSS & Indiong DU (1982) Schistosoma urethritis: Pseudo-gonorrhoeal disease in Northern Nigeria. Tropical Doctor 12, 141-142.

Chandiwana SK, Taylor P & Matanhire D (1991) Community control of schistosomiasis in Zimbabwe. The Central African Journal of Medicine 37, 69-76.

Cline BL & Hewlett BS (1996) Community-based approach to schistosomiasis control. Acta Tropica 61,107-119.

Dunn FL & Janes CR (1986) Introduction: medical anthropology and epidemiology; In Anthrhopology and Epidemiology (eds. CR Janes et al.), Reidel Publishing Co, Dordrecht, pp 3-34.

Hielscher S & Sommerfeld J (1985) Concepts of illness and the utilization of health-care services in a rural Malian village. Social Science and Medicine 21, 469-481.

Huang Y & Manderson L (1992) Schistosomiasis and the social patterning of infection. Acta Tropica 51,175-194.

Kloos H (1995) Human behavior, health education and schistosomiasis control: a review. Social Science and Medicine 40, 1497-1511.

Nash TE, Cheever AW, Ottesen EA, & JA Cook (1982) Schistosoma infection in humans: perspectives and recent finding. Annals of Internal Medicine 97, 740-754

Ndamba J, Nyazema N, Makaza N, Anderson C & KC Kaondera (1994) Traditional herbal remedies used for the treatment of urinary schistosomiasis in Zimbabwe. Journal of Ethnopharmacology 42, 125-132.

Ratard CR, Kouemeni LE, Bessala MME, Ndamkou CN, Greer GJ, Spilsbur J & Cline BC (1990) Human schistosomiasis in Cameroon: I. Distribution of schistosomiasis. American Journal of Tropical Medicine and Hygiene 42, 561-572.

Rosenfield PL (1992) Social determinants of tropical disease, In Tropical and Geographic Medicine (eds. KS Warren & AAF Mahmoud ), McGraw-Hill, NY, pp. 197-205.

Sibiya JB (1986) Community participation and health education in Botswana. Tropical Medicine and Parasitology 37, 168-169.

Table 1. Children’s explanations for the causes of urinary schistosomiasis.

this utilized explanation:

Table 2. Children’s knowledge of intermediate host for urinary schistosomiasis.

identified intermediate host:

Table 3. Children’s knowledge about season most likely to get urinary

schistosomiasis.

identified season of greatest risk:

Table 4. Adults’ explanations for the causes of urinary schistosomiasis.

gave this explanation*:

Table 5. Adults’ knowledge about season most likely to get urinary

schistosomiasis.

identified season of greatest risk: