Strengthening primary health care at district-level in Malawi – Determining the coverage, costs and benefits of community-directed interventions

Background: Community-Directed Interventions (CDI) is a participatory approach for delivery of essential healthcare services at community level. It is based on the values and principles of Primary Health Care (PHC). The CDI approach has been used to improve the delivery of services in areas that have previously applied Community-Directed Treatment with ivermectin (CDTi). Limited knowledge is available about its added value for strengthening PHC services in areas without experience in CDTi. This study aimed to assess how best to use the CDI approach to strengthen locally identified PHC services at district level. Methods: This was a comparative intervention study carried out over a period of 12 months and involving four health centres and 16 villages assigned to 1) a conventional Essential Health Package (EHP)/PHC approach at health centre level or 2) an EHP/PHC/CDI approach at community level in addition to EHP/PHC at health centre level. Communities decided which intervention components to be included in the intervention. These were home management of malaria (HMM), long lasting insecticide treated nets (LLIN), vitamin A and treatment against schistosomiasis. The outcomes of the two strategies were compared quantitatively after the intervention was completed with regard to intervention component coverage and costs. Qualitative in-depth interviews with involved health professionals, implementers and beneficiaries were carried out to determine the benefits and challenges of applied intervention components. Results: Implementation of the EHP/PHC/CDI approach at community level as an add-on to EHP/PHC services is feasible and acceptable to health professionals, implementers and beneficiaries. Statistically significant increases were observed in intervention components coverage for LLIN among children under 5 years of age and pregnant women. Increases were also observed for HMM, vitamin A among children under 5 years of age and treatment against schistosomiasis but these increases were not statistically significant. Implementation was more costly in EHP/PHC/CDI areas than in EHP/PHC areas. Highest costs were accrued at health centre level while transport was the most expensive cost driver. The study identified certain critical factors that need to be considered and adapted to local contexts for successful implementation. Conclusion: The CDI approach is an effective means to increase accessibility of certain vital services at community level thereby strengthening delivery of EHP/PHC services. The approach can therefore complement regular EHP/PHC efforts. Trial registration: The study was retrospectively registered with the Pan African Clinical Trial Registry TRN: PACTR201903883154921. Malawi is a country in Sub-Saharan Africa with a 2018 population size of 17,563,749 people and 28 districts organized into three regions: northern, central and southern [21]. The study was carried out in Mangochi, one of the 12 districts in the southern region of Malawi. The district is situated on the southern end of Lake Malawi (Fig. 1) and has a total population of 1,148,611 [21]. The district is mainly inhabited by people of Yao and Chewa ethnicities, while Islam (72%) and Christianity (28%) are the most practiced religions. Agriculture, fishing and microbusiness enterprising are the main economic activities of people in the district. Adult literacy is 49%, access to safe water supply is 73%, maternal mortality rate is 400 per 100,000 and the infant mortality rate is 169 per 1,000 [22]. Map of Malawi showing locations of Mangochi District (in red), Lake Malawi (in blue), major cities of Mzuzu, Lilongwe and Blantyre and the location of Malawi in Africa (red in the inset) (Source: Authors’ own map [20]) The study was designed as a controlled implementation study to run for 1 year. The assumption was that the district received what they needed in terms of information, materials, drugs and other supplies to allow them to satisfactorily implement EHP/PHC services in rural settings and that the study supported the district processes through a CDI based approach. In total, four health centres and 16 villages were involved in the study. Equal numbers of health centres along with their corresponding villages were randomly assigned to either the intervention or control arm of the study (Table 1). List of the involved health centres and villages in Mangochi District according to their assigned study arms 1. Saiti Tiputipu 2. Kamangazula 3. Kansiya 4. Binali 5. Kwitunji 6. Mponda 7. Sokole 8. Kasanga 1. Makunula 2. Nankamwa 3. Chimwaza 4. Mtendere 5. Itimu 6. Matenganya 7. Mbalula 8. Meso The support rendered included EHP/PHC/CDI training and supervision of district-based health professionals, health centre-based professionals attached to two health centres and community based volunteers/implementers in eight villages in the intervention arm of the study; no such training and supervision support was provided in the two health centres and eight villages involved in the control arm. For the intervention arm, the EHP/PHC/CDI approach was implemented mainly at community level on top of ongoing regular EHP/PHC activities at health centre levels in the same arm thereby complementing rather than replacing the regular health service delivery. In the control arm, the regular EHP/PHC system continued to operate at the health centre level without any EHP/PHC/CDI approach being implemented at community level. By applying these approaches, we expected to assess if CDI could strengthen ongoing EHP/PHC efforts in Mangochi District. Both the EHP/PHC and EHP/PHC/CDI strategies were implemented and evaluated over a period of 12 months. The research team identified an intervention package that was agreed with the communities and health care providers during the formative phase based on their priority health issues in the setting [20]. To be properly guided on existing policies regarding the implementation of the EHP/PHC/CDI approach at community level the final choice of intervention components also depended on consultations with key health personnel and other professional stakeholders. The following intervention components were finally included in the study: (i) home management of malaria (HMM) and fever to children under 5 years of age, (ii) distribution of long lasting insecticide treated nets (LLIN) to children under 5 years of age and pregnant women, (iii) vitamin A distribution to children under 5 years of age and (iv) treatment of urinary schistosomiasis using Praziquantel (PZQ) for those above 5 years of age. The study involved a total of five district-based officers consisting of the District Environmental Health Officer, District Community Health Nurse, and three Coordinators for malaria, Neglected Tropical Diseases (NTD) and diarrhoea. These were incorporated into the study as trainers and supervisors after being briefed and trained by the research team on the overall aims of the study, principles and processes of the CDI approach, and on available intervention components of the study. In turn the five district-based officers identified and provided training for two health centre based health workers, who were either a Medical Assistant or a Nurse in charge, and two Senior Health Surveillance Assistants from the two participating health centres under the intervention arm of the study to serve as trainers and later as supervisors for community based CDI implementers. The two health centre based staff in turn engaged their respective villages through community meetings from where 32 community based volunteers (one volunteer per each of the four intervention components in each of the eight intervention villages) were identified, trained and assigned roles as CDI implementers of the selected interventions. At every stage at health centre and community levels, both the research and district teams participated in the trainings as observers to ensure adherence to the study protocol. These health centre based staff later continued to supervise and support CDI implementers in their respective villages throughout the implementation period. No briefing and training were offered out to the staff from the corresponding two health centres and eight villages participating in the control arm of the study. Two implementation partner organizations involved in delivery of health services at district level, Icelandic International Aid Agency (ICEIDA) and Amref Health Africa (AHA) were identified and consulted at the beginning of the study. During the implementation of the CDI process at community level, the health services, implementation partners and the community played the following roles: The study employed a mixed-method approach to data collection focusing on quantitative data for coverage and cost estimates during baseline and follow-up, and qualitative data for assessing intervention benefits and evaluating processes. Data were collected at district, health centre and village levels in the intervention and control arms of the study. Research assistants collected the necessary data for the study from the involved health professionals, implementation partners, CDI implementers and beneficiaries using 19 data collection instruments previously used by the research group in a 2008 multi-country study [13]. An additional file shows the instruments that were used during data collection [see Additional file 1]. The instruments consisted of survey questionnaires administered to household heads or representatives at community level for determining intervention coverage, and questionnaires administered to relevant health programme coordinators at district level, in-charges at health centres, and leaders at community level, for determining intervention costs. Moreover, Health Management Information System (HMIS) records were reviewed to establish the burden of disease and coverage data at district, health centre and village levels using checklists; Interview guides were used to conduct in-depth interviews with professionals at district and health centre levels, implementation partners at district level and CDI implementers at community level about their perceptions on benefits and critical factors. Finally, focus group discussion guides were used to conduct group interviews with beneficiaries about their perceptions on using the CDI approach. Using these tools data were collected in both intervention and control areas before (baseline) and after (follow-up) introducing the intervention. All the proceedings of the key informant in-depth interviews and focus group discussions were recorded using digital audio recorders. Table 2 summarizes the methods, purposes, sources and quantities of data collected in the study. Methods, purposes, sources and amount of data collected in the study Quantitative data collected through survey questionnaires and checklists were processed and analyzed using statistical software Epi Info™ version 7.2.1. Analysis involved calculation of percentages, tabulations and frequencies to estimate coverage of individual intervention components. Furthermore, statistical significance tests using Chi Square were performed on differences in delta values (i.e. differences between baseline and follow-up) for each intervention component between intervention and control groups. The analyses of costs and benefits data were carried out using the following procedures: Qualitative data consisted of textual and audio data, including transcripts of key informant in-depth interviews, transcripts of focus group discussions, field notes on observations and other intervention-specific insights, notes and reports from meetings. Transcripts were translated into English and were entered in the computer using standard word processing software. A computer-assisted qualitative content analyses of the data using Atlas-Ti 8, a qualitative data analysis software programme (GmbH 2016) were conducted. Data were analyzed using open coding to come up with cross-classification and retrieval of categories of texts by theme.