Study design and population

A cross-sectional descriptive study was conducted to collect qualitative information regarding educational background, knowledge, and practices of CHWs, as well as the perception of CHWs from the beneficiaries, including women of reproductive age. Focus group discussions (FGDs) for both CHWs and their beneficiaries were conducted, which included the collection of demographic data. These data were collected as part of a larger assessment of CHW capacity relating to community-based nutrition (CBN) programmes, as commissioned by the Rwanda Ministry of Health (MoH) with the support from World Bank and UNICEF Rwanda.

The study was held from 10 to 20 May 2011, in 3 districts with differing levels of food security, which was selected in accordance with the overall goal of the CHW CBN capacity study. District selection was based on assessment of food consumption levels as reported in the 2009 Rwandan Comprehensive Food Security and Vulnerability Analysis and Nutrition Survey (CFSVA). The CFSVA report divided all districts into three categories of food consumption (high, moderate, and low). One district from each of these categories was purposively selected: Gicumbi with a food consumption score above the national average; Kamonyi with a moderate food consumption score; and Ngororero with a poor food consumption score.

In each of the districts, two administrative sectors were selected via simple random sampling (using the EPIINFO randomization programme). In one sector per district, one focus group composed of twelve binomials, and one focus group of woman beneficiaries of CBN services (twelve women of reproductive age, including at least four pregnant women and four lactating women) was organized. Women are the primary beneficiaries of the CBN services offered by the CHWs, and because vaccination uptake is 98% and antenatal care is sought by over 90% of women [17], this approach minimized potential selection bias.

In the other sector selected in the district, a focus group was organized for 12 ASM, and 12 CHSA CHWs. The individual CHWs and beneficiaries were selected randomly, again with simple random sampling, from lists furnished by the local health centres. Thus, in each of the 3 districts, 108 CHWs (12 CHWs per category - ASM, binomes, and CHSA) and 36 beneficiaries (12 per district) were included in FGDs. No CHWs or beneficiaries declined to participate in the FGDs.

Data collection

Focus group guides for beneficiaries and CHWs were developed following meetings with Government of Rwanda health officials, pretested and translated into the local language of Kinyarwanda. The FGD guides included demographic data. Data were collected over a 10-day period in May 2011. The FGDs were conducted by experienced and trained enumerators, and were audio-recorded, transcribed in the local language of Kinyarwanda, and then translated into French by a professional translator.

Data processing and analysis

The translated transcripts collected from the field were assessed by two coder clerks who identified emergent codes from the FGDs. Once the final set of codes was agreed upon, they were analyzed using qualitative analysis software, Atlas.ti. Beneficiaries’ demographic characteristics, CHWs’ demographic characteristics and educational background were also recorded.

Ethical considerations

Authorization from the MoH was secured before data collection. A written request from the Maternal and Child Health Department under the MoH was submitted to district hospitals and administrative managers at the district level. Informed consent forms were read aloud to the group members of the FGD and were obtained before starting the interview. Participants were assured that all responses were anonymous, and not identifiable to any individual participant.

Participants’ profile

A total of 108 CHWs (CHSA, ASM and binome), and 36 beneficiaries (lactating women, pregnant women and women of reproductive age) were interviewed for this study.

The median age (interquartile range (IQR) of CHWs was 38 (33 to 45) with 91% being married, 4% single and the remainder either widowed or divorced. Twenty-six percent of the CHWs were male. The median years of experience were 3, although this ranged from 1 to 16. The majority of the CHWs (76%) had only a primary level of education; the average number (±standard deviation (SD)) of children per CHW was 4 (±2.2) children.

Targeted beneficiaries of community services were generally younger than CHWs, with a median age of 24 (range, 19 to 45 years). Thirty-five (96%) of the beneficiaries were farmers, and 80% were married with an average (±SD) of 2 (±1.6) children per woman.

CHW focus group discussions

CHWs identified five key areas that significantly influenced their effectiveness, including training, the role of community performance-based financing, supervision (cPBF), workload as perceived by CHWs, and motivations. In particular, findings relating to irregular trainings and confusion of the cPBF system were widespread among the FGDs, and constituted key findings of this study.

Training

CHWs were generally selected by their communities based on criteria including literacy, past experience, and social acceptance. However, most of the CHWs did not have an education in health, and were not trained prior to beginning their roles as CHWs. Newly-recruited CHWs did not undergo an initial training programme, and only received on-the-job trainings. Although the importance of trainings was well known and widely recognized as key to improving knowledge, training sessions were found to be infrequent and not equitably delivered at the community level for all three types of CHWs.

For those who received trainings, the effectiveness of the programmes was highlighted, and CHWs claimed that the trainings helped to improve their knowledge and practices. Most CHWs appreciated the trainings, and felt that the community recognized and appreciated their advanced knowledge. However, trainings were described as inconsistent and insufficient, and some groups of CHWs did not attend any training programmes beyond the basic orientation programme.

The CHWs explained that binomes, as opposed to CHSAs and ASMs, received most of the trainings for nutrition, basic health, preventive, and curative services. One CHW said: 'We were not trained, but we can see what the binomes are doing in terms of benefits of training. We can see how they are taking care of malnourished children, but this does not have to do with our work'. Throughout the FDGs, CHSA and ASM CHWs highlighted this disparity in training, and noted their own lack of access to comprehensive training programmes. In regards to their role in promoting nutrition, many CHWs reported having had some training on nutrition, but the majority expressed the need to understand different components of a balanced diet.

While the national CHW policy does not intend for ASM CHWs to deliver infants, these CHWs said that they were often confronted with women delivering before arriving at the health centre. The lack of any training on delivery was considerably problematic and viewed as a gap in their knowledge ' … because we all know that the trainings will reinforce our capacity in reducing infant and maternal mortality', said one CHW.

The CHWs also explained that they lacked training on communication skills necessary to promote behaviour change in the community. The CHWs noted that their lack of confidence contributed to difficulties in convincing beneficiaries to adopt preferred behaviour such as preparing a balanced diet or practising basic hygiene. The lack of training in communication skills was seen as a complicating factor in facilitating promotion of behaviour change in the community.

Performance-based financing

The Rwanda MoH uses community performance-based financing (cPBF), a remuneration mechanism based on outputs, to create incentives for CHWs. The CHWs are not provided with a monthly salary; however, CHWs are evaluated based on performance indicators and the resulting financial incentive is given to a cooperative of CHWs. Based on the FGD accounts, some CHWs did not clearly understand the concept of cooperatives or cPBF, and they had not received proper training on how to set up the financial mechanisms or manage the cooperatives.

According to the CHW respondents in this study, the attempts at including the individual CHWs in the management of cPBF have not been successful. Only one CHW spoke of the benefits of the CHW cooperative system, while some CHWs raised the point that transparency and accountability of cooperative managers contributed to hinder the financial success of their cooperatives. According to the respondents in the study, the workload required of CHWs was high, and for most CHWs, this engagement provided little or no financial gain. Some CHWs were unsure if the cPBF programmes were still ongoing, '… for the last 2 years, I do not know if the cPBF exists or if it was stopped', explained one CHW, and others misunderstood the system, assuming that every CHW would receive a dividend at the end of the year.

Supervision systems

While national guidelines on CHW supervision existed at the time of this study, there were discrepancies in the frequency and content of supervision. Most CHWs met with their supervisors once a month at the health centre to deliver reports, although there were no standard procedures used for field supervision. Some supervisors accompanied CHWs on home visits occasionally, although this was not consistently a policy. The majority of CHWs recognized the importance of the supervision and noted this in the FGDs, explaining that, on occasion, they learned valuable information from supervision visits. However, CHWs often primarily associated supervision to the handover of the monthly report and not always to obtaining guidance and assistance.

Workload

In terms of their workload, the CHWs considered their general range of duties to be overwhelming, and subsequently a barrier to the sustainability of the CHW system. The variable hours necessary for their work, and the unexpected crises that arose (such as epidemics), conflicted with CHW family life and their other jobs. The intensity of the work was explained as varied and unpredictable, and detracting from time necessary for families.

Many binomes felt that their range of work was too broad to allow a sufficient and quality provision of care. As a binome CHW explained, 'we have 255 households to follow up each month and each of us takes a quarter of the households to accomplish our work, to help each other'. It was found that as a result of the high workload, the three types of CHWs divided up the houses in the village, travelling only to the houses in their vicinity. This resulted in some households being visited by CHWs who did not have any training in specific areas, such as nutrition management or maternal health.

Motivation

Despite an overwhelming workload, the CHWs experienced a sense of pride in their work. Many stated that they felt they were an important part of the whole health system improvement that aimed to reduce the burden of disease in the population. They recognized that under a decentralized system, their roles were increasingly becoming critical in reducing key health indicators, such as the infant and maternal mortality rates. By playing a direct role in improving indicators, CHWs felt valued and respected in their communities.

With regard to community recognition, almost all CHWs expressed their feelings about the improving health of the population: '… I feel happy when I see someone who was sick becoming healthy'. Relationships between themselves and their community fostered support and respect, as community members routinely approached the CHWs for advice. The ability to work for oneself, and manage one’s own hours, was also a clear incentive. Beneficiaries’ relatives also recognized the importance of CHWs in the community, as one CHW relayed what a man had said to her: 'without this programme, my wife would have died at home'.

A few CHWs also shared that the programme gave them a higher status despite their gender. Some claimed that having this position had given them more authority in the household and more autonomy over household purchases. Many of the ASMs who were trained on family planning had started using family planning methods themselves, and served as role models for the women in the community.

Beneficiaries

For the women who participated in the community beneficiary FGDs, the majority made statements recognizing the benefits of the CHW system and a positive attitude towards the CHWs was generally noted. Multiple women expressed their appreciation for the CHW efforts to care for them and their families. One pregnant woman explained: 'I can see that all CHWs do the best they can … I can see that they are needed in our society', and another woman appreciated how 'they follow and support us on a daily basis'.

The beneficiaries described CHWs as their main source of educational messages about topics such as nutrition, malaria, kitchen gardens, family planning, and hygiene. Almost all beneficiaries in the focus groups seemed to know that when a child is malnourished, they should go to the CHW, and if the problem was serious, the CHW would accompany the child to the health centre and follow up on progress afterwards. They added that the CHWs explained the basics of nutrition to the families and made sure that the mothers understood. They further explained that they received frequent visits from CHWs at their houses (sometimes daily, weekly or twice a month) to check up on the health of the children and to teach them about healthy meal preparation. Most of them said that they would call the CHW or go to his/her house if they had a problem. Overall, the CHWs were perceived to be very influential in the lives of the women, with the beneficiaries only seeking out further care at the health centre for serious illnesses.