Journal of Biosocial Science http://journals.cambridge.org/JBS Additional services for Journal
of Biosocial Science:
Email alerts: Click here Subscriptions: Click here Commercial reprints: Click here Terms of use : Click here
RE-ASSESSING COMMUNITY-DIRECTED TREATMENT: EVIDENCE FROM MAZABUKA DISTRICT, ZAMBIA H. HALWINDI, P. MAGNUSSEN, S. SIZIYA, D. W. MEYROWITSCH and A. OLSEN Journal of Biosocial Science / Volume 47 / Issue 01 / January 2015, pp 28 - 44 DOI: 10.1017/S0021932014000170, Published online: 15 May 2014
Link to this article: http://journals.cambridge.org/abstract_S0021932014000170 How to cite this article: H. HALWINDI, P. MAGNUSSEN, S. SIZIYA, D. W. MEYROWITSCH and A. OLSEN (2015). REASSESSING COMMUNITY-DIRECTED TREATMENT: EVIDENCE FROM MAZABUKA DISTRICT, ZAMBIA. Journal of Biosocial Science, 47, pp 28-44 doi:10.1017/S0021932014000170 Request Permissions : Click here
Downloaded from http://journals.cambridge.org/JBS, IP address: 203.64.11.45 on 15 Mar 2015
J. Biosoc. Sci., (2015) 47, 28–44, 6 Cambridge University Press, 2014 doi:10.1017/S0021932014000170 First published online 15 May 2014
RE-ASSESSING COMMUNITY-DIRECTED T R E A TMENT: EVI DEN C E F R O M MA Z A B U K A DISTRICT, ZAMBIA H. HALWINDI*1, P. MAGNUSSEN†, S. SIZIYA‡, D. W. MEYROWITSCH§ and A. OLSEN† *Department of Public Health, School of Medicine, University of Zambia, Lusaka, Zambia, †DBL-Centre for Health Research and Development, University of Copenhagen, Frederiksberg, Denmark, ‡Department of Clinical Sciences, School of Medicine, Copperbelt University, Ndola, Zambia, §Section of Health Services Research, Department of Public Health, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark Summary. Cross-sectional surveys with carers, health workers, community drug distributors (CDDs) and neighbourhood health committees were conducted to identify factors associated with utilization of community-directed treatment (ComDT) of soil-transmitted helminths in children aged 12–59 months in Mazabuka district, Zambia. The surveys took place in December 2006 and December 2007. In addition child treatment records were reviewed. The factors that were found to be significantly associated ( p < 0.05) with treatment of children by the CDDs were: (1) the perception of soil-transmitted helminth infections as having significant health importance, (2) the communitybased decision to launch and subsequently implement ComDT, (3) the use of the door-to-door method of drug distribution, (4) CDDs being visited by a supervisor, (5) CDDs receiving assistance in mobilizing community members for treatment, (6) CDDs having access to a bicycle and (7) CDDs having received assistance in collecting drugs from the health centre. Despite the effectiveness of ComDT in raising treatment coverage there are factors in the implementation process that will still affect whether children and their carers utilize the ComDT approach. Identification and understanding of these factors is paramount to achieving the desired levels of utilization of such interventions.
Introduction There is a major gap between the development of new health interventions and their delivery to communities in the developing world (Madon et al., 2008). Many poten1
Corresponding author. Email: [email protected]
28
Re-assessing community-directed treatment
29
tially effective disease control programmes have had only limited impact on the burden of disease because of inadequate implementation, which results in poor access, even to very simple and affordable products (TDR, 2007). Among the various programme options for disease control, community-directed treatment (ComDT) is regarded as a timely strategy that could be used as a model for public health programmes (Amazigo et al., 2002b). The ComDT approach was developed as a response to the need for a cost-effective, affordable and sustainable method for the mass treatment of onchocerciasis with ivermectin in endemic countries of Africa (World Health Organization, 1996; African Program for Onchocerciasis Control, 1998). It is based on the principle of active community participation such that communities themselves plan and carry out treatment of their members. The process empowers community members to make decisions and direct drug distribution for a sustained period of time. The term ‘community’ in this paper is defined as geographical areas of local neighbourhoods and villages consisting of people linked by social ties and who share common perspectives. The community members share common values, intent, beliefs, resources, preferences, needs and risks. However, they also show diversity in these same characteristics and a number of other conditions affecting their identity and degree of cohesiveness. These communities are nested, such that one community can contain another – for example a geographic community may contain a number of ethnic or political communities. Several factors have been identified that have led to successful ComDT. These include adoption of collective decision-making by community members, avoidance of paying monetary incentives to community drug distributors (CDDs), health education in mobilizing other community members to take part, and the use of the local system of administration, which can be strong in rural settings (Katabarwa & Mutubazi, 2000; Rowley et al., 2000; Katabarwa et al., 2000, 2002a, 2002b, 2004, 2005; Ndyomugyenyi & Kabatereine, 2003; Panter-Brick et al., 2006). Factors at community, household and individual levels have been identified that are significantly associated with treatment coverage of ComDT interventions, in addition to implementation issues. These include the method of distribution, ethnicity, sex, age, past experience with ComDT and perceptions about personal risk of the disease (Brieger et al., 2002; Nuwaha et al., 2005). The principal barriers to implementing community-directed interventions are usually social and health system constraints (TDR, 2008). These include: the acceptability of the intervention, a reluctance on the part of health workers to empower community implementers to manage drug administration, a shortage of supplies, delays in the process of transferring drugs through the various levels of the programme, transport-related problems and the issue of distance and the absence of health workers at the health centres when needed by CDDs (Amazigo et al., 2002a; Burnham & Mebrahtu, 2004; TDR, 2008). A study was implemented where the current health facility approach of delivering anthelminthic drugs to children aged 12–59 months was compared with an approach where ComDT was added to the health facility approach (Halwindi et al., 2010). As part of this study, factors that were associated with treatment of children by CDDs in the ComDT approach were investigated and the results are reported in this paper.
30
H. Halwindi et al. Methods
Study area and population The study was conducted in the catchment area of Chivuna rural health centre, in Mazabuka district, where the ComDT approach was implemented. Chivuna lies 63 km from Mazabuka town centre, and 40 km from the nearest hospital (Monze General Hospital). This area was selected mainly because of the low proportion of under-fives attending child health week programmes (25 to 40%) and the presence of soil-transmitted helminth (STH) infections in the area (Halwindi et al., 2011). Chivuna rural health centre has an estimated catchment population of 15,551 (2400 under-five children) inhabiting 58 villages. All villages are accessible by road except in the rainy season when some of them become difficult to reach because of bad roads and a lack of bridges across some of the streams. The furthest village in the catchment area is 18 km from the health centre. Like many rural areas in Zambia, the population in Chivuna is sparsely and unevenly distributed in small and scattered villages. This uneven and dispersed distribution of the population creates serious problems for the provision of social services in these areas. Levels of health and education provision are consequently adversely affected. A small number of households have access to piped water and electricity, mainly those living in institutional houses at the clinic and schools and also a few commercial farmers. Formal education levels are low and rarely go beyond 7th grade, the last year of primary school. The majority of the population are peasant farmers who depend on subsistence agriculture for food. The majority of adults own their land, which is given to them by the village headmen. A significant factor determining the level of agricultural development has been identified as closeness to the railway, which generally corresponds to closeness to markets and supplies of agricultural supplies (Atack & Margo, 2011). This consequently affects the patterns and systems of agricultural production. Hybrid maize production dominates market-related small-scale agriculture, while other crops, including local maize, ground nuts, sweet potatoes and sorghum, dominate the subsistence agriculture. The study area is very homogeneous in terms of ethnicity, with more than 90% of residents being native Tonga-speaking people. The other 10% is made up people from different ethnic groups within Zambia, notably the Luvale, Ngoni, Lozi and Bemba. There are cultural differences between the different groups, but they are not very pronounced partly because of inter-marriages between different ethnic groups. The people from different ethnic groups live peacefully together. There isn’t any reported marginalization of minority ethnic groups in terms of health and other social services. The local leadership comprises the chief, the ngambela (messenger for the chief ), senior headmen, village headmen, village messenger, village secretary and the village council. The chief and headmen play a critical role as gatekeepers to the communities. The village headman and council are responsible for day-to-day administration and decision-making in the village. Women participate in many village activities but they do not play a significant role in decision-making at village level. The village leaders are usually the important decision-makers on issues of community development. Once the leaders agree on a programme the villagers will in most cases accept it. Matters that involve the active participation of all community members are usually first considered by the leaders, and the final decision is made by the community through the village council or community meetings.
Re-assessing community-directed treatment
31
There are a number of community-based organizations in the area, but the neighbourhood health committee is the main one. The catchment area for the rural health centre is divided into eleven zones and each zone has a neighbourhood health committee made up of a representative from each village. The functions of the neighbourhood health committee include provision of basic health education to the communities, and co-ordination of all health activities at community level, acting as a link between the community and the rural health centre. The health centre supervises the neighbourhood health committees. In most cases, community members are either Christians or not religious. The majority of the Christians belong to the Catholic faith. This is because of the presence of the Catholic mission in the area. There is a parish in the area that runs the only secondary school and provides a lot of support to the local health centre. This has resulted in a strong influence on the community’s religious affiliation to Catholicism. The Catholic priests also play an important role of local leadership. However, their role is more inclined to matters that relate to religion and the welfare of community members. Other Christians mainly belong to other mainstream denominations like the Anglican Church, Seventh Day Adventist Church, United Church of Zambia, Baptist Church and new apostolic churches. Implementation of the ComDT approach Sensitization meetings for implementation of ComDT were held with health authorities at national, district and health centre levels, and with neighbourhood health committee chairpersons and village leaders. The details of the problem of STH infections and the necessity and the benefits of treatment were discussed. The concept of the ComDT approach was introduced; that is, the villagers were given full responsibility for selecting, organizing and implementing their own method of distributing drugs to children in their villages. The community members were informed that the drugs would be provided for free by the rural health centre. Following these initial meetings, eleven community meetings were held, and the issues debated with the health workers, neighbourhood health committees and village leaders were subsequently discussed with community members. At these meetings the communities selected the CDDs and the method, venue and time of drug distribution. Two methods of drug distribution to children were used by the CDDs: a door-to-door method and a ‘central place’ method. In the door-to-door method, the CDDs moved from one household to another to deliver the drugs to the children. In the ‘central place’ method treatment was done at a specific location in the community chosen by community members. A total of 29 CDDs participated in drug distributions. The CDDs were ordinary community members chosen by communities to carry out drug distribution. All the selected CDDs received training on STH infections, how to administer a single dose of 500 mg of mebendazole (Johnson & Johnson) and 200 IU of vitamin A supplementation (0.3 mg retinol), and on record keeping. During each child health week in June and December children were treated on two different occasions. First, treatment was offered at the health centre and outreach posts during child health week. Second, children who were missed on this first occasion were treated by CDDs within the villages the following week. Only children who had not received treatment from the health centres were treated by the CDDs.
32
H. Halwindi et al.
The two drug delivery approaches are described in detail elsewhere (Halwindi et al., 2010). The drugs used by the CDDs were supplied by the district health office and were collected from the district pharmacy by the rural health centre and stored there. The rural health centre was supplied with enough drugs to cater for all eligible children in the catchment area. The supply of drugs for the study was not prioritized but the drugs were brought to the rural health centre using the normal supply chain. Study design and sampling For the main study, the ComDT intervention was implemented and monitored between June 2006 and December 2007. Four treatment rounds (child health weeks 1, 2, 3 and 4) were monitored during this period. After determining the treatment coverage for June and December 2005 for all the rural health centres in Mazabuka district, Chivuna and Magoye rural health centres were selected based on comparable low treatment coverage, equal-sized population of under-five children and comparable staff situation. Chivuna rural health centre was then randomly selected as the intervention area where the ComDT approach was implemented simultaneously with the regular work of the rural health centre, thereby becoming the HF þ ComDT area. Magoye rural health centres were then the control area where only the health facility approach was implemented. Villages in each catchment area were grouped into local geographical areas based on the existing neighbourhood health committee zones. Five local geographical areas were created in each catchment area. Each local geographical area was comprised of two to three neighbourhood health committee zones. A census was conducted in each local geographical area to record all children aged 12–59 months living in the area. The stratified sampling method, using local geographical area as stratum, was used to select the sample of carers to be interviewed. In each local geographical area 100 eligible children were randomly selected by systematic sampling, where children were chosen at regular intervals from the sampling frame. The sampling frame was the list of all eligible children who were registered during census and thus a total of 500 children were sampled from each study area. The present study consisted of two cross-sectional surveys conducted in December 2006 and December 2007, during the treatment rounds for child health weeks 2 and 4. On the two occasions, 371 and 427 carers of the randomly selected children were interviewed, respectively. The results from the health facility area (Magoye rural health centre) do not contribute to the results discussed in this paper. Data collection Ten trained field assistants and the first author conducted the interviews. Data were collected using semi-structured questionnaires on factors related to treatment of children by CDDs. Data were also collected from four health workers, eleven neighbourhood health committee chairpersons and 29 CDDs. Simple random sampling was used to select the children, whose carers were interviewed. The purposive sampling method was used to select the health workers that were interviewed. The health workers
Re-assessing community-directed treatment
33
that were selected and interviewed were the clinical officer, the Maternal and Child Health (MCH) co-ordinator, and two nurses. All CDDs and neighbourhood health committee chairpersons were interviewed. The first author interviewed the health workers, neighbourhood health committee chairpersons and CDDs, while the field assistants interviewed the carers. Interviews with the health workers were conducted in English, but interviews with the CDDs, carers and neighbourhood health committee chairpersons were conducted in the local language, Tonga. All interviews were conducted at the place of choice of participants. Treatment records were also reviewed to determine the total number of eligible children for treatment and the proportion of children treated by CDDs. Quality control was assured by training the field assistants on general interview techniques and in administering the specific questionnaire. All questionnaires were pre-tested and necessary adjustments were made prior to actual data collection. The factors investigated for their effect on utilization of ComDT were perceptions of STH infections, the decision-making process, drug supply and distribution method, community awareness about ComDT, supervision of CDDs and evidence of support for CDDs. Data analysis Microsoft Office Excel (Microsoft Corp., Redmond, WA, USA) and EpiData 3.2 (EpiData Association, Odense M, Denmark) software was used for data entry. Data were analysed using STATA 9.2 (StataCorp, College Station, TX, USA). The main outcome variable was treatment of children by CDDs. The differences in proportions were tested for significance using the chi-squared test. Analysis of the qualitative data was done manually using a systematic text analysis procedure (Ulin et al., 2005). The text from open-ended questions was reduced to specific thematic categories. Descriptive codes were assigned to the thematic categories and sub-categories were generated that were used to describe and summarize the findings. For data generated from semistructured interviews, each group of questions was first analysed by the ‘Question analysis’ method (Morse & Field, 1995), whereby the interviews were first sorted by question number and then ‘content analysis’ (Morse & Field, 1995) was applied on the data from each of the questions. Ethical considerations The research proposal was reviewed and approved by the University of Zambia Ethics Committee (Ref.: 003-01-06) and the Danish National Committee on Biomedical Research Ethics (Ref.: 2006-7041-83) before the research was carried out. In addition, the Ministry of Health (MoH) in Zambia gave permission for the study to be conducted. At implementation, an explanation of the project was given to the communities during community meetings, and to the carers, health workers and neighbourhood health committee chairpersons during the interviews. Informed consent was also obtained from the carers for their participation in interviews. Participants were told that they were free to withdraw from the study at any time. They were informed that the information they provided was going to be treated as confidential and no participant’s identity would be associated with the published data.
34
H. Halwindi et al. Results
Proportions of participants treated during child health week The proportions of children treated by CDDs during the treatment rounds for child health weeks 1, 2, 3 and 4 were 29.0% (306/1053), 37.3% (442/1184), 58.1% (289/497) and 53.0% (406/766), respectively. There was a significant increase over time in the proportion of children treated by CDDs from child health week 1 to 4 ( p < 0.005). In this paper the results of child health weeks 2 and 4 are compared in order to assess the change over time of different parameters (Table 1). Child health week 2 was used instead of child health week 1 to allow for the community members and CDDs to appreciate the ComDT approach before evaluation. Factors associated with treatment of children by CDDs Table 2 summarizes the quantitative results for factors associated with treatment of children by CDDs. Community awareness of the ComDT programme. The levels of awareness in the community about the ComDT programme, that is the proportion of carers interviewed who knew about the ComDT programme, were moderate. Fifty-three per cent (195/ 371) and 60.8% (260/427) were aware of the ComDT programme during the treatment round for child health weeks 2 and 4, respectively. However, the level of awareness Table 1. The number and proportion of children treated by CDDs during treatment rounds for child health weeks 2 and 4, Mazabuka district, Zambia Child health week 2
Zone
Total number of children eligible for treatmenta
1 2 3 4 5 6 7 8 9 10 11 Total
142 112 98 121 97 99 78 117 107 101 112 1184
a
Child health week 4
Number (%) of children treated by CDDs 71 21 74 62 25 5 40 40 34 70 0 442
(50.0) (18.8) (75.5) (51.2) (25.8) (5.1) (51.3) (34.8) (31.8) (69.3) (0.0) (37.3)
Total number of children eligible for treatmenta 35 66 19 65 86 55 98 90 81 96 75 766
Number (%) of children treated by CDDs 0 27 0 48 20 28 90 32 20 84 57 406
(0.0) (40.9) (0.0) (73.8) (23.3) (50.9) (91.8) (35.6) (24.7) (87.2) (76.0) (53.0)
The total number of eligible children was the number of children that remained untreated after conducting the health facility treatment component during child health week. These were the children who were targeted for treatment by the CDDs. The proportion of children treated was therefore calculated as ‘the number treated divided by total number of eligible children’.
Table 2. Factors associated with the treatment of under-five children by community drug distributors, Mazabuka district, Zambia Child health week 2 Variable
Number (%) treated
270 156
152 (56.3) 114 (73.1)
459 594
p-value
Total number of children
Number (%) treated
p-value
of Biosocial Science:
Email alerts: Click here Subscriptions: Click here Commercial reprints: Click here Terms of use : Click here
RE-ASSESSING COMMUNITY-DIRECTED TREATMENT: EVIDENCE FROM MAZABUKA DISTRICT, ZAMBIA H. HALWINDI, P. MAGNUSSEN, S. SIZIYA, D. W. MEYROWITSCH and A. OLSEN Journal of Biosocial Science / Volume 47 / Issue 01 / January 2015, pp 28 - 44 DOI: 10.1017/S0021932014000170, Published online: 15 May 2014
Link to this article: http://journals.cambridge.org/abstract_S0021932014000170 How to cite this article: H. HALWINDI, P. MAGNUSSEN, S. SIZIYA, D. W. MEYROWITSCH and A. OLSEN (2015). REASSESSING COMMUNITY-DIRECTED TREATMENT: EVIDENCE FROM MAZABUKA DISTRICT, ZAMBIA. Journal of Biosocial Science, 47, pp 28-44 doi:10.1017/S0021932014000170 Request Permissions : Click here
Downloaded from http://journals.cambridge.org/JBS, IP address: 203.64.11.45 on 15 Mar 2015
J. Biosoc. Sci., (2015) 47, 28–44, 6 Cambridge University Press, 2014 doi:10.1017/S0021932014000170 First published online 15 May 2014
RE-ASSESSING COMMUNITY-DIRECTED T R E A TMENT: EVI DEN C E F R O M MA Z A B U K A DISTRICT, ZAMBIA H. HALWINDI*1, P. MAGNUSSEN†, S. SIZIYA‡, D. W. MEYROWITSCH§ and A. OLSEN† *Department of Public Health, School of Medicine, University of Zambia, Lusaka, Zambia, †DBL-Centre for Health Research and Development, University of Copenhagen, Frederiksberg, Denmark, ‡Department of Clinical Sciences, School of Medicine, Copperbelt University, Ndola, Zambia, §Section of Health Services Research, Department of Public Health, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark Summary. Cross-sectional surveys with carers, health workers, community drug distributors (CDDs) and neighbourhood health committees were conducted to identify factors associated with utilization of community-directed treatment (ComDT) of soil-transmitted helminths in children aged 12–59 months in Mazabuka district, Zambia. The surveys took place in December 2006 and December 2007. In addition child treatment records were reviewed. The factors that were found to be significantly associated ( p < 0.05) with treatment of children by the CDDs were: (1) the perception of soil-transmitted helminth infections as having significant health importance, (2) the communitybased decision to launch and subsequently implement ComDT, (3) the use of the door-to-door method of drug distribution, (4) CDDs being visited by a supervisor, (5) CDDs receiving assistance in mobilizing community members for treatment, (6) CDDs having access to a bicycle and (7) CDDs having received assistance in collecting drugs from the health centre. Despite the effectiveness of ComDT in raising treatment coverage there are factors in the implementation process that will still affect whether children and their carers utilize the ComDT approach. Identification and understanding of these factors is paramount to achieving the desired levels of utilization of such interventions.
Introduction There is a major gap between the development of new health interventions and their delivery to communities in the developing world (Madon et al., 2008). Many poten1
Corresponding author. Email: [email protected]
28
Re-assessing community-directed treatment
29
tially effective disease control programmes have had only limited impact on the burden of disease because of inadequate implementation, which results in poor access, even to very simple and affordable products (TDR, 2007). Among the various programme options for disease control, community-directed treatment (ComDT) is regarded as a timely strategy that could be used as a model for public health programmes (Amazigo et al., 2002b). The ComDT approach was developed as a response to the need for a cost-effective, affordable and sustainable method for the mass treatment of onchocerciasis with ivermectin in endemic countries of Africa (World Health Organization, 1996; African Program for Onchocerciasis Control, 1998). It is based on the principle of active community participation such that communities themselves plan and carry out treatment of their members. The process empowers community members to make decisions and direct drug distribution for a sustained period of time. The term ‘community’ in this paper is defined as geographical areas of local neighbourhoods and villages consisting of people linked by social ties and who share common perspectives. The community members share common values, intent, beliefs, resources, preferences, needs and risks. However, they also show diversity in these same characteristics and a number of other conditions affecting their identity and degree of cohesiveness. These communities are nested, such that one community can contain another – for example a geographic community may contain a number of ethnic or political communities. Several factors have been identified that have led to successful ComDT. These include adoption of collective decision-making by community members, avoidance of paying monetary incentives to community drug distributors (CDDs), health education in mobilizing other community members to take part, and the use of the local system of administration, which can be strong in rural settings (Katabarwa & Mutubazi, 2000; Rowley et al., 2000; Katabarwa et al., 2000, 2002a, 2002b, 2004, 2005; Ndyomugyenyi & Kabatereine, 2003; Panter-Brick et al., 2006). Factors at community, household and individual levels have been identified that are significantly associated with treatment coverage of ComDT interventions, in addition to implementation issues. These include the method of distribution, ethnicity, sex, age, past experience with ComDT and perceptions about personal risk of the disease (Brieger et al., 2002; Nuwaha et al., 2005). The principal barriers to implementing community-directed interventions are usually social and health system constraints (TDR, 2008). These include: the acceptability of the intervention, a reluctance on the part of health workers to empower community implementers to manage drug administration, a shortage of supplies, delays in the process of transferring drugs through the various levels of the programme, transport-related problems and the issue of distance and the absence of health workers at the health centres when needed by CDDs (Amazigo et al., 2002a; Burnham & Mebrahtu, 2004; TDR, 2008). A study was implemented where the current health facility approach of delivering anthelminthic drugs to children aged 12–59 months was compared with an approach where ComDT was added to the health facility approach (Halwindi et al., 2010). As part of this study, factors that were associated with treatment of children by CDDs in the ComDT approach were investigated and the results are reported in this paper.
30
H. Halwindi et al. Methods
Study area and population The study was conducted in the catchment area of Chivuna rural health centre, in Mazabuka district, where the ComDT approach was implemented. Chivuna lies 63 km from Mazabuka town centre, and 40 km from the nearest hospital (Monze General Hospital). This area was selected mainly because of the low proportion of under-fives attending child health week programmes (25 to 40%) and the presence of soil-transmitted helminth (STH) infections in the area (Halwindi et al., 2011). Chivuna rural health centre has an estimated catchment population of 15,551 (2400 under-five children) inhabiting 58 villages. All villages are accessible by road except in the rainy season when some of them become difficult to reach because of bad roads and a lack of bridges across some of the streams. The furthest village in the catchment area is 18 km from the health centre. Like many rural areas in Zambia, the population in Chivuna is sparsely and unevenly distributed in small and scattered villages. This uneven and dispersed distribution of the population creates serious problems for the provision of social services in these areas. Levels of health and education provision are consequently adversely affected. A small number of households have access to piped water and electricity, mainly those living in institutional houses at the clinic and schools and also a few commercial farmers. Formal education levels are low and rarely go beyond 7th grade, the last year of primary school. The majority of the population are peasant farmers who depend on subsistence agriculture for food. The majority of adults own their land, which is given to them by the village headmen. A significant factor determining the level of agricultural development has been identified as closeness to the railway, which generally corresponds to closeness to markets and supplies of agricultural supplies (Atack & Margo, 2011). This consequently affects the patterns and systems of agricultural production. Hybrid maize production dominates market-related small-scale agriculture, while other crops, including local maize, ground nuts, sweet potatoes and sorghum, dominate the subsistence agriculture. The study area is very homogeneous in terms of ethnicity, with more than 90% of residents being native Tonga-speaking people. The other 10% is made up people from different ethnic groups within Zambia, notably the Luvale, Ngoni, Lozi and Bemba. There are cultural differences between the different groups, but they are not very pronounced partly because of inter-marriages between different ethnic groups. The people from different ethnic groups live peacefully together. There isn’t any reported marginalization of minority ethnic groups in terms of health and other social services. The local leadership comprises the chief, the ngambela (messenger for the chief ), senior headmen, village headmen, village messenger, village secretary and the village council. The chief and headmen play a critical role as gatekeepers to the communities. The village headman and council are responsible for day-to-day administration and decision-making in the village. Women participate in many village activities but they do not play a significant role in decision-making at village level. The village leaders are usually the important decision-makers on issues of community development. Once the leaders agree on a programme the villagers will in most cases accept it. Matters that involve the active participation of all community members are usually first considered by the leaders, and the final decision is made by the community through the village council or community meetings.
Re-assessing community-directed treatment
31
There are a number of community-based organizations in the area, but the neighbourhood health committee is the main one. The catchment area for the rural health centre is divided into eleven zones and each zone has a neighbourhood health committee made up of a representative from each village. The functions of the neighbourhood health committee include provision of basic health education to the communities, and co-ordination of all health activities at community level, acting as a link between the community and the rural health centre. The health centre supervises the neighbourhood health committees. In most cases, community members are either Christians or not religious. The majority of the Christians belong to the Catholic faith. This is because of the presence of the Catholic mission in the area. There is a parish in the area that runs the only secondary school and provides a lot of support to the local health centre. This has resulted in a strong influence on the community’s religious affiliation to Catholicism. The Catholic priests also play an important role of local leadership. However, their role is more inclined to matters that relate to religion and the welfare of community members. Other Christians mainly belong to other mainstream denominations like the Anglican Church, Seventh Day Adventist Church, United Church of Zambia, Baptist Church and new apostolic churches. Implementation of the ComDT approach Sensitization meetings for implementation of ComDT were held with health authorities at national, district and health centre levels, and with neighbourhood health committee chairpersons and village leaders. The details of the problem of STH infections and the necessity and the benefits of treatment were discussed. The concept of the ComDT approach was introduced; that is, the villagers were given full responsibility for selecting, organizing and implementing their own method of distributing drugs to children in their villages. The community members were informed that the drugs would be provided for free by the rural health centre. Following these initial meetings, eleven community meetings were held, and the issues debated with the health workers, neighbourhood health committees and village leaders were subsequently discussed with community members. At these meetings the communities selected the CDDs and the method, venue and time of drug distribution. Two methods of drug distribution to children were used by the CDDs: a door-to-door method and a ‘central place’ method. In the door-to-door method, the CDDs moved from one household to another to deliver the drugs to the children. In the ‘central place’ method treatment was done at a specific location in the community chosen by community members. A total of 29 CDDs participated in drug distributions. The CDDs were ordinary community members chosen by communities to carry out drug distribution. All the selected CDDs received training on STH infections, how to administer a single dose of 500 mg of mebendazole (Johnson & Johnson) and 200 IU of vitamin A supplementation (0.3 mg retinol), and on record keeping. During each child health week in June and December children were treated on two different occasions. First, treatment was offered at the health centre and outreach posts during child health week. Second, children who were missed on this first occasion were treated by CDDs within the villages the following week. Only children who had not received treatment from the health centres were treated by the CDDs.
32
H. Halwindi et al.
The two drug delivery approaches are described in detail elsewhere (Halwindi et al., 2010). The drugs used by the CDDs were supplied by the district health office and were collected from the district pharmacy by the rural health centre and stored there. The rural health centre was supplied with enough drugs to cater for all eligible children in the catchment area. The supply of drugs for the study was not prioritized but the drugs were brought to the rural health centre using the normal supply chain. Study design and sampling For the main study, the ComDT intervention was implemented and monitored between June 2006 and December 2007. Four treatment rounds (child health weeks 1, 2, 3 and 4) were monitored during this period. After determining the treatment coverage for June and December 2005 for all the rural health centres in Mazabuka district, Chivuna and Magoye rural health centres were selected based on comparable low treatment coverage, equal-sized population of under-five children and comparable staff situation. Chivuna rural health centre was then randomly selected as the intervention area where the ComDT approach was implemented simultaneously with the regular work of the rural health centre, thereby becoming the HF þ ComDT area. Magoye rural health centres were then the control area where only the health facility approach was implemented. Villages in each catchment area were grouped into local geographical areas based on the existing neighbourhood health committee zones. Five local geographical areas were created in each catchment area. Each local geographical area was comprised of two to three neighbourhood health committee zones. A census was conducted in each local geographical area to record all children aged 12–59 months living in the area. The stratified sampling method, using local geographical area as stratum, was used to select the sample of carers to be interviewed. In each local geographical area 100 eligible children were randomly selected by systematic sampling, where children were chosen at regular intervals from the sampling frame. The sampling frame was the list of all eligible children who were registered during census and thus a total of 500 children were sampled from each study area. The present study consisted of two cross-sectional surveys conducted in December 2006 and December 2007, during the treatment rounds for child health weeks 2 and 4. On the two occasions, 371 and 427 carers of the randomly selected children were interviewed, respectively. The results from the health facility area (Magoye rural health centre) do not contribute to the results discussed in this paper. Data collection Ten trained field assistants and the first author conducted the interviews. Data were collected using semi-structured questionnaires on factors related to treatment of children by CDDs. Data were also collected from four health workers, eleven neighbourhood health committee chairpersons and 29 CDDs. Simple random sampling was used to select the children, whose carers were interviewed. The purposive sampling method was used to select the health workers that were interviewed. The health workers
Re-assessing community-directed treatment
33
that were selected and interviewed were the clinical officer, the Maternal and Child Health (MCH) co-ordinator, and two nurses. All CDDs and neighbourhood health committee chairpersons were interviewed. The first author interviewed the health workers, neighbourhood health committee chairpersons and CDDs, while the field assistants interviewed the carers. Interviews with the health workers were conducted in English, but interviews with the CDDs, carers and neighbourhood health committee chairpersons were conducted in the local language, Tonga. All interviews were conducted at the place of choice of participants. Treatment records were also reviewed to determine the total number of eligible children for treatment and the proportion of children treated by CDDs. Quality control was assured by training the field assistants on general interview techniques and in administering the specific questionnaire. All questionnaires were pre-tested and necessary adjustments were made prior to actual data collection. The factors investigated for their effect on utilization of ComDT were perceptions of STH infections, the decision-making process, drug supply and distribution method, community awareness about ComDT, supervision of CDDs and evidence of support for CDDs. Data analysis Microsoft Office Excel (Microsoft Corp., Redmond, WA, USA) and EpiData 3.2 (EpiData Association, Odense M, Denmark) software was used for data entry. Data were analysed using STATA 9.2 (StataCorp, College Station, TX, USA). The main outcome variable was treatment of children by CDDs. The differences in proportions were tested for significance using the chi-squared test. Analysis of the qualitative data was done manually using a systematic text analysis procedure (Ulin et al., 2005). The text from open-ended questions was reduced to specific thematic categories. Descriptive codes were assigned to the thematic categories and sub-categories were generated that were used to describe and summarize the findings. For data generated from semistructured interviews, each group of questions was first analysed by the ‘Question analysis’ method (Morse & Field, 1995), whereby the interviews were first sorted by question number and then ‘content analysis’ (Morse & Field, 1995) was applied on the data from each of the questions. Ethical considerations The research proposal was reviewed and approved by the University of Zambia Ethics Committee (Ref.: 003-01-06) and the Danish National Committee on Biomedical Research Ethics (Ref.: 2006-7041-83) before the research was carried out. In addition, the Ministry of Health (MoH) in Zambia gave permission for the study to be conducted. At implementation, an explanation of the project was given to the communities during community meetings, and to the carers, health workers and neighbourhood health committee chairpersons during the interviews. Informed consent was also obtained from the carers for their participation in interviews. Participants were told that they were free to withdraw from the study at any time. They were informed that the information they provided was going to be treated as confidential and no participant’s identity would be associated with the published data.
34
H. Halwindi et al. Results
Proportions of participants treated during child health week The proportions of children treated by CDDs during the treatment rounds for child health weeks 1, 2, 3 and 4 were 29.0% (306/1053), 37.3% (442/1184), 58.1% (289/497) and 53.0% (406/766), respectively. There was a significant increase over time in the proportion of children treated by CDDs from child health week 1 to 4 ( p < 0.005). In this paper the results of child health weeks 2 and 4 are compared in order to assess the change over time of different parameters (Table 1). Child health week 2 was used instead of child health week 1 to allow for the community members and CDDs to appreciate the ComDT approach before evaluation. Factors associated with treatment of children by CDDs Table 2 summarizes the quantitative results for factors associated with treatment of children by CDDs. Community awareness of the ComDT programme. The levels of awareness in the community about the ComDT programme, that is the proportion of carers interviewed who knew about the ComDT programme, were moderate. Fifty-three per cent (195/ 371) and 60.8% (260/427) were aware of the ComDT programme during the treatment round for child health weeks 2 and 4, respectively. However, the level of awareness Table 1. The number and proportion of children treated by CDDs during treatment rounds for child health weeks 2 and 4, Mazabuka district, Zambia Child health week 2
Zone
Total number of children eligible for treatmenta
1 2 3 4 5 6 7 8 9 10 11 Total
142 112 98 121 97 99 78 117 107 101 112 1184
a
Child health week 4
Number (%) of children treated by CDDs 71 21 74 62 25 5 40 40 34 70 0 442
(50.0) (18.8) (75.5) (51.2) (25.8) (5.1) (51.3) (34.8) (31.8) (69.3) (0.0) (37.3)
Total number of children eligible for treatmenta 35 66 19 65 86 55 98 90 81 96 75 766
Number (%) of children treated by CDDs 0 27 0 48 20 28 90 32 20 84 57 406
(0.0) (40.9) (0.0) (73.8) (23.3) (50.9) (91.8) (35.6) (24.7) (87.2) (76.0) (53.0)
The total number of eligible children was the number of children that remained untreated after conducting the health facility treatment component during child health week. These were the children who were targeted for treatment by the CDDs. The proportion of children treated was therefore calculated as ‘the number treated divided by total number of eligible children’.
Table 2. Factors associated with the treatment of under-five children by community drug distributors, Mazabuka district, Zambia Child health week 2 Variable
Number (%) treated
270 156
152 (56.3) 114 (73.1)
459 594
p-value
Total number of children
Number (%) treated
p-value
