There is a growing body of evidence that in many developing countries school-based health services such as treatment of schistosomiasis and intestinal nematode infections can be delivered at low cost (World Bank 1993; Partnership for Child Development 1999) and can contribute to improving children’s general well-being, growth (Stephenson et al. 1989; Stoltzfus et al. 1998), nutritional status (Stephenson et al. 1993; Beasley et al. 1999), cognitive ability (Drake et al. 2000) and school attendance (Simeon et al. 1995).
In recognition of the benefits to children of deworming, many countries in areas where helminth infections are prevalent are now advocating national policies for school-based helminth control. At present many school-based deworming programmes are small or at the pilot stage of implementation. The challenge is now to expand these programmes and take them to national scale. Evaluating the acceptability of school-based health interventions at the local level is essential for identifying and addressing areas of concern and for achieving long-term sustainability. Many of the organizations which now advocate school-based health programmes emphasize the importance of acceptability to, and involvement of key stakeholders such as teachers, parents and the wider community (UNESCO, UNICEF, WHO, World Bank 2001).
For school-based health programmes that deliver interventions such as regular, annual or six-monthly treatment with praziquantel for schistosomiasis and albendazole for intestinal nematodes, long-term sustainability will depend on a number of factors such as community knowledge and perception of the disease and its consequences –, i.e. perceived justification for the intervention, perceived efficacy and benefits of treatment, acceptability of the programme and of new means of treatment delivery (e.g. using teachers to administer drugs) and willingness to contribute financially to continuance of the programme. Whilst evaluation of community acceptability is widely recognized as an important component of programme evaluation, for school-based health programmes this has only been examined for small-scale and experimental projects (Nwaorgu et al. 1998; Shu et al. 2000; in Nigeria, Magnussen et al. 2001a,b in Tanzania). This paper presents the results of an evaluation of community perception of two large-scale, government-run (in collaboration with the Partnership for Child Development), school-based health programmes delivering anthelmintic drugs to primary school children, in the Volta Region of Ghana (three intervention districts and two comparison districts) and the Tanga Region of Tanzania (three intervention and three comparison districts).
Subjects and methods
The Partnership for Child Development (PCD) was established in 1992 to conduct and promote operations research on efficacy, feasibility and cost-effectiveness of school health and nutrition programmes, at an operational scale, and to conduct research on child development issues. The Partnership for Child Development was established in response to the growing number of children who were surviving to reach school age (Partnership for Child Development 1997).
In 1994, Partnership programmes were initiated in the Volta Region of Ghana (80 442 children in 577 schools in the intervention Districts of Jasikan, Kpandu and Hohoe) by the Ghana Partnership for Child Development (GPCD) and in the Tanga region of Tanzania (110 000 children in 352 schools in the intervention Districts of Tanga, Muheza and Korogwe) by the Tanzania Partnership for Child Development (Ushirikiano wa Kumwendeleza Mtoto Tanzania – UKUMTA). In both countries the basic interventions included health education and annual mass treatment for all pupils with albendazole for intestinal nematodes (Ascaris lumbricoides, Trichuris trichiura and hookworm) and school mass praziquantel treatment for pupils in those schools where 50% or more of pupils were infected with Schistosoma haematobium, as estimated by self-reported blood in urine (Ansell et al. 1997). Praziquantel was also given selectively to other pupils if they reported blood in urine. Teachers were trained in the administration of drugs and other related intervention activities, such as use of ‘tablet poles’ to determine the correct dose of praziquantel for each child. Praziquantel is given as a dose of 40 mg/kg body weight. Since weight and height are correlated, height can be used as a proxy for weight in determining the dose required (Hall et al. 1999). All treatment was provided free of charge, and the main programme implementers, the teachers, were provided with per diems for attending health and training workshops, but were not paid any extra for their subsequent work in administering drugs to the children. Government officials played an important role in programme implementation in both Ghana [School Health Education Programme (SHEP) Coordinators and Circuit Officers], and Tanzania (Ward Education Officers), by providing the direct link between schools and the programme at the district level, and were paid expenses for this work. Research carried out as part of these programmes into diagnosis and treatment of schistosomiasis and intestinal nematode infection, health and nutritional status, cost and efficacy, has been reported elsewhere (http://www.child-development.org/publications.htm lists documents and references).
Community Perceptions Surveys to investigate stakeholders’ views about the programmes were conducted in Ghana in 1997 and Tanzania in 1996, after the first round of treatments. The aims of these surveys were to investigate perceptions of stakeholders at the local community level (children, parents and teachers) of key aspects of the programme that affect acceptability and sustainability, i.e. knowledge of the programme and of the programme aims and activities, knowledge of symptoms and consequences of helminth infection, perceived benefits and improvements in children’s health, acceptability of teachers administering drugs, the perceptions of teachers of their role in the programme, suggestions for programme improvement and willingness of parents to contribute to costs of treatment. No prior information about costs of treatment was given to parents, as detailed information on areas such as cost of delivery was not available at the time of this survey.
Data collection and selection of participants
Selection of schools and participants for the survey was designed to obtain representative samples and coverage, using a combination of structured and random sampling. Knowledge, perception and attitude evaluation took place using a combination of methods including questionnaires, focus groups, interviews and meetings. In order to make methods appropriate for each country, the study designs varied for Ghana and Tanzania, and thus presentation of results differs for the two countries. Note that in both countries, for the interviews with children it was found that many were somewhat inhibited in giving their views, so data from the children presented here should be interpreted with this in mind. In Tanzania (but not in Ghana) informal discussions with children were the most satisfactory method of eliciting information. The study concentrated primarily on obtaining the views of teachers and parents.
Ten schools were selected from each district in the intervention areas (Tanga, Muheza and Korogwe Districts), the total of 30 schools including schools from three categories (rural, urban and remote) in each. The aim was to interview the head teacher and the domestic science (school health) teacher in each – the implementers of UKUMTA – as well as parents and children.
Focus group discussions, in-depth interviews and informal interviews were used to gather data. Three questionnaires were designed for use with teachers, children and parents. These were translated into Kiswahili, reviewed and revised after field testing. Focus group discussions were tested but children were found to be unwilling to volunteer information except to answer specific questions, so the main use of these was to identify patterns which were later followed-up in informal discussions.
Children were selected from within each school using random number tables: three girls and three boys were selected from Standards 1 and 2 (age 8–10 years) and from Standards 3 and 4 (age 12–14 years), for interview. We interviewed 356 children.
For interviewing parents from each village, we ensured that representation was as unbiased and varied as possible. Before interviewing parents in the community, the permission and assistance of the village chairman was sought. Parents for interview were then selected by taking the third house on the left of the house of the chairman. If there was no parent in that house, the next house would be visited until a parent was located. After the first interview, the next house visited would be at least 3 houses further on. In some villages parents had to be visited in the fields. Our aim was to interview 5 parents in each of the 30 schools, and this target was met in most cases (Table 1).
Community perception surveys in Ghana and Tanzania: samples interviewed
Focus group discussions and interviews were held with parents and teachers in the three intervention districts Jasikan, Kpandu and Hohoe. We selected 75 small, medium and large schools giving a balanced representation of rural, remote and urban schools. For each school the aim was to interview the head teacher, the school health teacher and a parent who was a member of the Parent Teachers Association (PTA) (Table 1). Separate questionnaires were designed for each category interviewed, field tested and revised accordingly. Answers from the questionnaires were used to direct and explore areas of concern and interest in follow-up interviews and discussions, conducted in the local language. In Ghana children were not included in this analysis as pilot interviews revealed that they would only answer ‘what was expected of them’, so it was difficult to obtain meaningful information.
No consistent differences were found in either Ghana or Tanzania in the views of parents and teachers, according the size of school or whether it was from an urban rural, or remote area, and thus overall results are presented.
Community perception of health problems
Baseline surveys of 41 schools in Tanzania (in 1996) and 85 schools in Ghana (in 1994) revealed prevalence rates of intestinal nematodes of 63% in Ghana and 86% in Tanzania. The prevalence of S. haematobium infection was 13% in Ghana and 58% in Tanzania (Partnership for Child Development 1998). Against this background, most parents in Ghana did not perceive worm infection as a major health problem for children. When asked what the most common health problems suffered by their children were, only 4.1% of parents in Ghana listed ‘worms’ as a health problem; headache, fever and malaria being the most commonly cited problems.
In Tanzania parents were asked to report symptoms or problems associated with worm infection and schistosomiasis in children. For worm infection the most commonly cited symptoms were loss of appetite (67%), weight loss (52%) and abdominal pain (47%). For schistosomiasis more than 90% of parents reported blood in urine as a symptom, problems urinating (22%), anaemia (18%) and fatigue (16%).
Community awareness of the school health programme
Interviews with parents in both Ghana and Tanzania revealed some problems of communication between the school and parents regarding the school health programmes in the first year of implementation. In Ghana, for example, although all parents interviewed had heard of the GPCD programme, either from their children, the PTA, the church or village gong beater (whose task it is to announce news to the village), they were much less well informed about the actual content of the programme. Although all parents knew that their children received tablets at school, only 57% knew beforehand that these tablets were specifically for treating schistosomiasis or intestinal nematodes.
Discussions with parents in Ghana also revealed that the level of communication between parents and the school varied between villages. In one village, for example, the head teacher was also the village ‘community leader’, and had called a meeting to explain to parents that their children would receive drugs at school the next day, and that the children should eat something beforehand. In another village, parents only knew that their children were to receive drugs the next day when the children informed them that they needed to take a cup to school (for water to help them swallow the pills).
In Tanzania, nearly all parents interviewed had heard of the deworming programme, although less than half knew the programme name, UKUMTA. Again, most had received the information from their children, with a smaller proportion having heard from teachers. In contrast to Ghana, nearly all parents interviewed (96%) knew that their children were being given tablets for treatment of intestinal nematodes or schistosomiasis.
Although the predominant reactions to the deworming programme in both countries were very positive, in-depth interviews in some villages revealed concern about lack of communication and representation of all parties at meetings. In Tanzania, for example, there was concern that not all schools had convened a meeting before treatment to explain the programme to parents, and that for schools that did convene meetings, women (as the primary caregivers in these communities) were under-represented. In Ghana, there was also concern about lack of coordination and collaboration between the health and education sectors in implementation of the programme. For example, in Hohoe District, it was claimed that only one Ministry of Health (MoH) staff had been invited to monitor the administration of drugs. It was suggested that Community Health Nurses could play a greater role in administration, monitoring and supervision of drug administration and could give teachers training in first aid so that they could better deal with problems such as side-effects of the drugs experienced by some children (see below).
Perception of the teachers’ role in the programme
In both Ghana and Tanzania, the vast majority of teachers (both head teachers and health teachers) indicated that they were happy with their involvement in the programme and would be pleased to continue with it (96% in Ghana and 98% in Tanzania). They perceived the benefits of the programme as upgrading teachers’ knowledge of health issues, improvements in children’s health, and improvement in children’s awareness of cleanliness. Teachers did, however, feel that involvement in the programme increased their work load. In Ghana, for example, head teachers estimated that they spent approximately 11 h during the year on programme-related activities (preparation, drug administration, monitoring, paper work). An additional 3 days were spent on attending training courses in the first year.
Teachers in Tanzania were asked to list their activities for the programme. These included giving drugs, preparing food (for children before treatment with praziquantel), health education, measuring children’s height (for determining correct doses of praziquantel) and collecting unused tablets, along with mobilizing pupils and the community.
When asked about problems encountered with the programme, teachers cited disruption of classes, increased work load, side-effects experienced by children, absenteeism on treatment day (for reasons such as sickness, working in the fields or looking after relatives at home), some parents not giving permission for treatment, and lack of incentives for their involvement with the programme. Both teachers and Circuit Officers (education sector personnel who have responsibility for teaching practices, standards and supplies in a Circuit of about 20 schools) also indicated that if treatment days took place at busy times during the school year, because of exams or teacher assessments, then this added to the burden of work.
Suggestions for improvement of the programme made by teachers included giving them non-financial incentives for their involvement, such as a certificate to acknowledge their contribution. It was also suggested that teachers and their families should be included in the drug treatment and that more courses should be run for parents and teachers to educate them about helminth infections and treatment. Treatment should be timed to avoid school exam terms.
When parents were asked about the acceptability of teachers providing treatment at school, most parents in Ghana (85%) and Tanzania (96%) were happy with this arrangement. Those who expressed discontent, would have preferred a nurse or health worker to be present to supervise the administration of drugs by teachers, or to administer the drugs.
In Tanzania most children (73%) were also happy with teachers administering drugs and preferred being treated in school, as travel to hospital was inconvenient and the hospitals were generally overcrowded. One pupil summed it up: ‘Before issuing the drugs the teacher teaches us the subject, she explains to us how to use the drug … and informs us that if you do not change your behaviour you will contract the disease again’.
Perceptions of the side-effects of treatment
Side-effects following treatment with albendazole and praziquantel, although mild and not considered serious, may nevertheless be unpleasant for children and a cause of concern for parents. Intensity of side-effects is associated with intensity of infection, with essentially no side-effects being experienced by those free from infection (Olds et al. 1999). About one-third of parents in both Ghana (28%) and Tanzania (31%) reported that their children had experienced side-effects after treatment, most commonly dizziness, vomiting, stomach-ache, fatigue and weakness. For treatment with praziquantel, it was recommended that children be fed before treatment to minimize side-effects, but no connection was found between not getting food beforehand, which was the case for a small number of children, and getting side-effects afterwards. In fact, children who felt unwell, from whatever cause, on the day of treatment, are likely to ascribe this to side-effects of the drugs.
In Ghana, where 18 children were taken to local clinics for treatment of side-effects (aspirin for headaches and topical lotion for rashes), problems had arisen about who was responsible for paying for these treatments. It had previously been decided at the regional level that the MoH would be responsible for this cost, because the side-effects were considered a consequence of a public health programme. But in some areas this decision was not well understood and was not put into practice by the health team. This resulted in strong objections from parents, although the MoH did eventually reimburse the costs of treatment.
Reactions to the benefits of the deworming programmes in both Ghana and Tanzania were positive and enthusiastic, with a general wish that the programmes should continue. In Ghana, parents – particularly mothers – reported improvements in their children’s health and reported the visible benefits as the expulsion of worms in stools. In Tanzania parents were also pleased that their children were being treated in school and emphasized that without the programme they had difficulty in obtaining quality drugs, and would incur costs because of the time lost and distances that had to be travelled to the clinics/hospitals. In Tanzania 90% of children reported that they felt healthier after treatment, with benefits including losing a swollen belly, improved ability to concentrate and improved appetite.
Willingness and ability to pay
As parasite control resources are finite and programmes typically last for years, the willingness and ability of communities to pay for continued treatment is critical to the sustainability of programmes. In both Ghana and Tanzania more than 90% of parents said that they would be willing to pay towards the cost of continuing the deworming drug intervention programme. In Tanzania, 97% of parents were willing to pay the full cost of drug treatment for their children once external funding was withdrawn (although details of costs were not specified). In Ghana 90% of parents were willing to pay between US$ 0.02 and US$ 2.27 for annual treatment of each child (mean of US$ 0.66). Those unwilling to pay usually had too many children to be able to afford it. The favoured strategy for payment was inclusion in school fees, although the level of schools fees varied considerably between individual schools.
Teachers, however, were more dubious about whether parents would pay for treatment, with only half the teachers in Tanzania believing that parents would pay. Teachers also suggested that there would be an inequality of ability to pay, with some parents being unable to afford the drugs. In Ghana it was also suggested that there would be a difference between better educated, ‘enlightened’ parents, who would be prepared to pay, and poorer, less informed communities, in which school fees may already be a burden.
Although the different survey techniques adopted in Ghana and Tanzania make direct comparison of the two countries inappropriate, these surveys nevertheless revealed common areas of perceptions of the two programmes. Most teachers were positive about their role in the programme, including administration of drugs, and parents fully accepted their role. The benefits of the programme were apparent to teachers, parents and children in terms of improved health and well-being, and most parents indicated a willingness to pay for continuation of drug treatment, although it was difficult to determine the precise level of contribution that parents were willing to pay.
The evaluation also highlighted areas that needed addressing, such as educating communities on the importance of worm infection as a chronic health problem (compared with more acute or life threatening illnesses such as malaria), issues of payment for treatment of side-effects, local collaboration between the health and education sectors, and communication between parents and schools. The importance of communication between school and community for effective coverage and compliance was also noted by Magnussen et al. (2001a), who found that in a school-based schistosomiasis treatment programme in Tanzania, the headmasters’ and teachers’ attitudes and involvement with children’s health were important factors influencing the commitment, interest and communication between the school and the community.
Knowledge of consequences of intestinal helminth infection and schistosomiasis is an important component of support for control programmes. In a study of a schistosomiasis control programme in northern Cameroon, Hewlett & Cline (1997) found that local people must perceive that the benefits of treatment outweigh the consequences of not being treated (both health and economic costs) if they are to seek treatment and support control programmes. There is now evidence of the potential consequences of chronic helminth infection on growth, nutritional status, and cognition (Stephenson 1987; Stoltzfus et al. 1997; Watkins & Pollitt 1997; Drake et al. 2000), and the benefits of treatment in alleviating these problems (Stephenson et al. 1989, 1993; Simeon et al. 1995; Stoltzfus et al. 1998; Beasley et al. 1999; Nokes et al. 1999). Targeting children for treatment is also of benefit to the local community in reducing overall prevalence of infection, as children can act as a major source of transmission (Bundy et al. 1990). Improved health education within the local community would help reinforce the messages of consequences of infection and benefits of treatment, especially as parents and children already perceived the benefits of treatment in terms of improved well-being of the children after deworming.
Anthelmintic treatment, which is cheap, and simple to deliver, can also provide a starting point for wider school health programming, so that others issues can be addressed, including malaria prevention education, HIV/AIDS prevention, improved school sanitation and hygiene (WHO 1996).
Implementation of the deworming programme through schools, with teachers treating pupils, was acceptable to the large majority of parents, teachers and pupils interviewed. In Tanzania, Magnussen et al. (2001a,b) also found that after initial training, teachers could successfully diagnose and treat schistosomiasis and uncomplicated malaria in school children, and this was acceptable to parents. The main concern among a minority of parents in this survey was that teachers did not have adequate training and that children were not examined individually. An assessment by Nwaorgu et al. (1998) of how a Nigerian community would perceive a hypothetical school-based schistosomiasis and helminthiasis control programme revealed similar tendencies: Although most respondents said they would agree to have teachers (e.g. headteachers) manage such a programme and to administer the drugs, many also wanted health workers to be involved or provide supervision.
This apprehension by some parents could be addressed through improved communication, reassuring parents that teachers had received adequate training as part of the programme. The teachers’ suggestion of receiving a certificate to acknowledge their role in the programme could also serve a dual function here: as a recognition of the teachers’ contribution and an incentive, and as a reassurance to parents, the community and health workers that teachers had been properly trained to treat their children. Closer collaboration between health and education workers at the local level would also ensure that drug treatment was properly supervised – and perceived to be so – and that the issue of side-effects could be dealt with more effectively.
Willingness (and ability) to pay for treatment is an important step for the change from externally funded programmes with treatment being given free of charge, to self supporting, sustainable programmes that are locally funded. This applies even when the cost of drugs is low or the drugs themselves are free (as in onchocerciasis control programmes) where distribution and administrative costs must still be met. In a survey of three villages in Nigeria, Onwujekwe et al. (1998) found that more than 90% of heads of households were willing to pay for ivermectin distribution costs. Factors affecting the amount that people were willing to pay included education and occupation, average monthly expenditure on health care, presence and manifestations of onchocerciasis and types of saving scheme.
In our study, more than 90% of parents indicated a willingness to pay for their children’s treatment, with the average amount indicated by parents in Ghana of 0.66 US$. Parents in Tanzania, although indicating a willingness to pay for the cost of treatment, did not give figures, so it is not possible to assess how realistic their willingness to pay was in relation to actual costs of drug treatment. Parents’ willingness to pay for school-based drug treatment of their children has also been demonstrated in a small scale school-based schistosomiasis control programme in Tanzania (Magnussen et al. 2001a). In the second year of the programme, a cost recovery system was introduced. Parents paid the full price of a single-dose treatment with praziquantel (0.22 US$ per tablet) when their children were diagnosed with schistosomiasis (through examination of a urine sample when the child complained of blood in urine).
A detailed cost analysis of delivery of albendazole and praziquantel in the school-based programmes in Ghana and Tanzania was conducted by the Partnership for Child Development (1999). The total financial costs per treatment per child (cost of drugs plus delivery) of a combination of mass and selective treatment for schistosomiasis using praziquantel, which also involved prior screening at the school level, was US$ 1.22 in Ghana and US$ 0.79 in Tanzania. The costs of treating intestinal nematodes with albendazole which was given as a fixed dose to all children was US$ 0.24 in Ghana and US$ 0.23 in Tanzania. The largest component of this cost in most cases was the drugs themselves, with delivery comprising a smaller proportion (the exception being cost of delivery of praziquantel in Ghana, which was slightly higher than the cost of the drug).
If payment for anthelmintic treatment of children was added to school fees (which was the favoured method of payment reported by parents), it is hard to assess what percentage increase this would mean for most families, as fees vary considerably between schools, and sending a child to school may include a range of additional costs other than enrolment (such as uniform, equipment, PTA fees). In Ghana parents reported paying primary school fees that ranged from US$ 0.06 to > US$ 5 per child.
Another possible scheme of contribution might be through school health insurance. In Vietnam, for example, school health insurance has been successfully introduced as a means of paying for school-based health interventions such as deworming (Carrin et al. 1999). Timing of payment may also be an important factor, as Hewlet & Cline (1997) found that people were more likely to be able to pay for treatment at certain times of year, such as when extra money was available from the harvest. Whichever payment scheme is introduced, the problem still remains of parents who are unable to pay, especially if they have a large number of children needing treatment. Onwujekwe et al. (1998) also make the point that expressing willingness to pay is not the same as actual payment. This issue could only be clarified by continued evaluation and monitoring of programmes once local payment had been introduced, and this is an area that warrants further investigation.
The evaluation of willingness to pay presented here is a preliminary survey, intended to give an indication, rather than a detailed analysis, of what level of payment for treatment parents were willing to consider. Accurate figures for family income and expenditure are notoriously hard to assess and costs of treatment per child will vary depending on a number of factors. These include number of children included in the anthelmintic treatment programme, current costs of purchasing the drugs and costs associated with delivery. Costs of school fees (and other school related expenditure), family income and expenditure also vary, at both the individual and community level. This is especially true now that HIV/AIDS is having an increasing impact on both the income and expenditure of many families. A more detailed cost analysis of all these factors would be required for a programme implemented at national scale. Subsidies through government, NGOs or community funds for low income households would almost certainly be necessary to cover the costs of treatment (see Onwujekwe et al. 1998) for discussion of contributions to community funds for treatment).
Our study reports the results of an evaluation of local community perception of large school-based programmes in Ghana and Tanzania, which indicates that stakeholders in local communities were positive about the deworming programme and potentially willing to pay for its continuation. However, it is also crucial that for programmes to be sustainable and go to scale nationally, all levels of involvement in policy making, administration and implementation at district, regional and national level support it. Although a large part of this paper is concerned with the perceptions of the local community (parents and teachers), interviews with circuit officers in Ghana, who were involved with the programme at subdistrict level, also indicated enthusiasm and support. At the national level, the programmes described here were implemented with a Memorandum of Understanding between the Ministries of Health, Education and other related organizations involved. Just as community acceptance of programmes at the local level is the bottom line for their sustainability, so collaboration and cooperation between Ministries and other key stakeholders is the first step for adoption and success of programmes for school-based helminth control at the national level.
This study also provides important lessons both for issues that need addressing in the continuation, improvement and sustainability of school-based helminth control programmes and in the need for continuing evaluation, especially of whether expressed willingness to pay for treatment of children is translated into action. The aim of our study was not to provide rigorous statistical analyses of programmes, but rather initial insights into perceptions of acceptability of unique large-scale programmes and into areas of concern in their implementation. Future work also needs to address the issue of how best to assess the perception of children (the main target and beneficiaries of the programmes) of their health and their involvement in the programme. Differences between communities in terms of communication and perceptions, indicate the need for surveys to provide sufficient coverage to encompass these differences, and the need to better understand the reasons for them.
The governments of both Ghana and Tanzania have now formulated initial plans for national scale school-based health programmes, including anthelmintic treatment, and the lessons learnt from the programmes described here have been useful in their planning, cost analysis and impact epidemiology. The present study provides complementary insights about how such programmes are perceived by parents and teachers, and gives an indication of parents’ willingness to pay for their children to be treated for intestinal nematodes and schistosomiasis. How such national programmes are taken to scale, and how they perform, will be watched with interest.