Skills2Care: An innovative, cooperative learning programme for community health workers in South Africa

Background Community health workers (CHWs) hold potential to support universal health coverage and better health for vulnerable communities. They are integral to the re-engineered Primary Health Care (PHC) strategy, introduced in South Africa in 2011. This study focussed on how to train CHWs in large numbers, especially in resource-limited, rural settings. Skills2Care, a method of cooperative learning for CHWS, has been pioneered in the Eastern Cape of South Africa. Aim To determine whether Skills2Care could improve the cognitive knowledge of CHWs; to understand their response and attitude to the programme; to explore factors that enabled and inhibited learning and to consider its viability as a training method. Setting Research was conducted in 2019 in the Ngqeleni subdistrict of the O.R. Tambo district, in rural Eastern Cape. Methods A group-learning model using specifically tailored study modules in booklet format, addressing mother and baby care, was used. A facilitator promoted learning. Knowledge assessment was conducted by pre- and post-study testing using multiple choice questions. Focus group discussions and interviews explored the appropriateness and acceptability of this method, and factors enabling and inhibiting the learning. Results This method of peer group cooperative learning can significantly increase the cognitive knowledge of CHWs. Test scores indicated a significant (13%) improvement. Focus group discussions indicated that participants valued this method as it increased knowledge and boosted their confidence. Conclusion This innovative approach to district-based, continuing education suggests that CHWs could be trained in large numbers without the need for additional resources.


Introduction
Community health workers (CHWs), holding the potential to support universal health coverage, offer hope of better health for vulnerable communities. 1,2,3 They provide an opportunity for equitable access to safe, primary health services. 4 Community health workers have been described as members of the community who 'provide basic health care within their community, and are capable of providing preventive, promotional and rehabilitation care to that community'. 5 provide increases in knowledge and skills of CHWs. Traditionally, continuing education for health workers, including CHWs consists of courses run by formal trainers at large central hospitals. These courses are often expensive to run and far from health workers' families and places of work, making them inaccessible. 8 This study was conducted to determine the appropriateness of the Skills2Care learning model in improving knowledge and understanding for CHWs, so as to improve community care in rural areas. The Skills2Care content that was tested related to perinatal care of the mother and baby, and included four modules. These comprised: care of the mother, the newborn baby, exclusive breastfeeding and human immunodeficiency virus (HIV) in mother and baby. The programme is a collaboration amongst Small Projects Foundation, the Perinatal Education Programme, Bettercare and Kistefos funding agency.
Skills2Care is a learning programme modelled on the innovative Perinatal Education Programme for doctors, professional nurses and midwives, which has been shown to be highly successful. 9 Results from pre-and post-tests, with 114 midwives, demonstrated significant improvement in cognitive knowledge and understanding. In this earlier study, group-based cooperative learning without the support of a formal trainer was achieved, sparking hope that an adaptation of the programme for CHWs may be equally fruitful.
Skills2Care, being decentralised and supported through a modularised approach on-site, seemed to offer an alternate method for CHW training. Hence, this study intended to determine the effectiveness of the decentralised and facilitated learning method, to determine CHWs' response to it and to understand the factors affecting their learning. The term 'learning' was chosen as CHWs' active engagement with the learning material was required. It supported the idea of an active learning process, different from their training or teaching which may imply a more passive recipience of training.
Policy for Ward Based Primary Healthcare Outreach Teams in South Africa indicates that the country needs approximately 55 000 CHWs. 10 According to Pillay and Barron, 11 health outcomes related to CHWs in South Africa (SA) are generally accepted to be sub-optimal, especially in the areas of maternal and child health. The reasons advanced include: inadequate training, inadequate support and supervision. Posing the question of how large numbers of CHWs can be trained, especially in resource-limited settings, is imperative to ensure that they have the requisite knowledge and confidence to be effective. Given the number of CHWs that a primary health care (PHC) system ought to have, it is essential to give consideration to effective and affordable strategies for on-site training.

Study design
The study included both qualitative and quantitative components.
The quantitative component comprised a pre-and post-study design to assess knowledge. For each of the four modules, a multiple-choice pre-test was conducted before the learning process started, and a post-test, with the same set of multiplechoice questions, was conducted at the end of the learning process. The results of the pre-test documented CHW knowledge related to the topic prior to learning, whilst the post-test assessed their knowledge after learning. The difference between the two results provided an indication of knowledge gained.
The qualitative research component used a multi-method approach, consisting of focus group discussions and interviews with key stakeholders. Where required, focus group discussions were translated into CHW home language, isiXhosa, with the support of the facilitator.
Semi-structured interviews were held with four CHWs, the director of Small Projects Foundation (SPF), the programme manager who manages and oversees the programme and the facilitator who implemented the programme. These included face-to-face and telephonic interviews. Questions for the semi-structured interviews and focus group discussions probed attitudes to the learning programme, the perceived value of the learning process and the factors which enabled and inhibited the acquisition of knowledge for the participants.

Study population
The study population comprised 63 CHWs: 49 from SPF, 4 from Siyakhanyisa (an NGO) and 10 from the Department of Health (DoH). Participants were required to complete a registration form with demographic details. Table 1 provides the demographic characteristics of the participants: depicting CHW ages, gender, length of service, educational levels and some household information.

Intervention
The study manual, Mother and Baby Care for Community Health Workers written by Prof David Woods from Bettercare, provided the content for the learning. The manual is one of a series of learning programmes for healthcare workers supporting distance-learning (https://bettercare.co.za/learn). It is intended to provide 'appropriate, affordable and up-to-date learning material for health care workers so that they can continue to learn, practice and deliver excellent patient care'. 8 Four modules included in the manual are used in this study: Care of the Mother, Care of the Newborn baby, Exclusive Breastfeeding and HIV in Mother and Baby.
Each module follows a standard format with learning objectives clearly stated at the outset, followed by its content using a question-and-answer, problem-solving style of learning, and thereafter a set of case studies, providing practical examples and applying module knowledge to real-life situations. Each module has a self-assessment test comprising 10 multiple-choice questions to determine knowledge after the module is studied.
All participants were provided with a copy of the study manual. The details of the study, its purpose and the standardised process for learning were explained to the participants by the facilitator. The facilitator was an adult educator, employed by SPF who has extensive experience working with CHWs. From previous experience, it had been learnt that the deployment of a facilitator to support CHW learning produced better results than a purely selfstudy method. The main role of the facilitator was to encourage and support group learning rather than to teach. The CHWs shared and discussed newly learnt knowledge and understanding in a spirit of cooperative learning and peer tuition.
Learning groups were established by CHWs at the clinics where they were based. Active participation was encouraged. Ten learning groups were established, one at each facility. The facilitator travelled to each group promoting learning.
Each week presented a different component of the learning process: • In week one, the pre-test was written and the material from the first module was read together by the group and discussions of the content were advanced by the facilitator. The CHEWs were encouraged to read and learn the material at home. • In week two, CHWs returned to the group, where any challenges with the learning material were discussed and clarified. Case studies from the manual, in question-and-answer style format, were presented. Community health workers were encouraged to share their own relevant case experiences. They returned home to complete their self-study of the material. • In week three, the post-test (the same as the pre-test) was conducted. The next module started with the learning cycle repeating itself for each module.
Where possible, CHWs with smart phones established WhatsApp groups through which they communicated with one another, asking questions about the study material and supporting one another.
The completed pre-test and post-tests were sent to the researcher to be marked and recorded. Results for the preand post-tests were recorded on Excel spreadsheets and analysed. In order to pass, a minimum result of 80% was required for the post-test.

Quantitative data
Quantitative data were collected for the pre-and post-tests.

Qualitative component
The researcher spent a week in the Ngqeleni sub-district, prior to the study, where introductions were made to the CHWs and the study. She also spent a week after the study and held interviews with four CHWs, the facilitator, programme manager as well as two professional nurses http://www.phcfm.org Open Access from participating clinics working for the DoH. Five focus group discussions were held with 48 CHWs: two were held two weeks after the start of the process and three were held on its completion.
The interviews and focus group discussions were held at the clinics. They were recorded, transcribed and analysed.

Analysis Quantitative
Scores for the pre-and post-tests for each participant in each module were captured on an Excel spreadsheet. The average scores for each module were calculated along with the total average for all the modules. Participants' scores of pre-and post-tests were compared.

Qualitative
An analysis of the data was collated from the interviews and focus group discussion. Key recurring themes were elicited by looking for patterns of shared meaning across the data and coding these accordingly.

Ethical considerations
Ethics approval was provided by Human Sciences Research Council: Protocol number REC7/22/08/18 on 28 September 2018. Small Projects Foundation signed a memorandum of understanding with the Provincial Department of Health, which provides for implementation and study of the Skills2Care programme.
All participants were provided with an information sheet, which described the study and gave contact details of the researcher and SPF managers. The study process, its ethical requirements as well as their voluntary participation and confidentiality were explained. Any clarifications or questions were answered. The CHWs then completed the consent form, providing their agreement to voluntarily participate in the study.
In an effort to maintain anonymity, all participants were provided a unique number that they used throughout the process for both pre-and post-tests, thus no names were utilised. The researcher used the unique numbers to match the results to the participants.

Results quantitative
The results of pre-and post-tests for each module are presented in Table 2, with the descriptive statistics displayed. Figure 1 shows the score distribution with the total average pre-and post-test scores, where 40 is the maximum score possible. Figure 2 shows the average score for each of the four modules, pre-and post-tests as well as their difference.
Inferential statistics included the test statistics (either t-test or Chi² test value), the degrees of freedom (df), the p-value and

Count
Total score the appropriate practical significance (effect size) statistics (Cohen's d) for t-tests and Cramér's V for Chi² tests.
A p-value of less than 0.05 indicates significance. Cohen's d is the statistics used to determine the practical significance, if the p-value is less than 0.05 for an inferential statistics test based on sample mean values, for example, t-test and Scheffé test. 12 Interpretation intervals are contained in Table 3. Cramér's V, as shown in Table 5, is the statistics used to determine the practical significance, if the p-value is less than 0.05 for an inferential statistics test based on sample frequencies, for example Chi² test. 12 The null hypothesis for each of the Chi² tests is that the percentage of participants with a score ≥ 80% for the post-test group is equal to the percentage for the pre-test group. Chi² tests have been used to determine whether the pass rate (pass if score ≥ 80%) improved from pre-to post-test. See Table 6 for the frequency distribution and Chi 2 results. It is interesting to note that the t-test result for Care for the Mother was not significant (p = 0.180), whilst the Chi² result for this variable was significant (p = 0.012, V = 0.32).
The results indicate that most CHWs had good prior knowledge of maternal care, information related to HIV and exclusive breastfeeding, with pre-test scores of 80% or more. The programme manager indicated that training in Growth Monitoring and Nutrition had been convened for CHWs previously by a dietician who had focussed on exclusive breastfeeding. Knowledge of these topics was confirmed in discussion with CHWs.

Qualitative results
Focus group discussions and interviews with CHWs and other key stakeholders provided the data to understand CHW response and attitude to the learning programme, and the factors promoting and inhibiting their learning. The following themes and sub-themes were elicited ( Table 7):

The value of learning
Community health workers valued the learning process for their improved knowledge, which in turn gave them more confidence and competence. They responded well to the participatory approach to learning.

Gaining knowledge improved levels of confidence
Community health workers considered that they were gaining valuable knowledge which made them feel more confident in their roles. They considered it useful and empowering: 'It was good to study so that you can know a lot. Then when I see my clients, I feel that I can tell them information that I have learned'. (CHW, female, 26 years old) 'Sometimes community members will come to your house. They want you to help them. So, these modules give you more confidence to know the right answers for them. It improves your attitude too because you know you are right'. (CHW, female, 42 years old) Community health workers repeatedly acknowledged that community members did not differentiate between themselves and the professional nurses. They explained that whilst their specific role had been explained to communities, their uniform and association with the clinic meant that their distinction as being CHWs was lost. Being equipped with improved knowledge made them more comfortable to better support communities whose expectations exceeded their competencies and qualifications: 'They don't see us as community care workers. They see us as nurses, even if we explain to them. So, they are always asking questions, like "Why are my feet swollen when I am pregnant?" Now we know and can assess and tell them that they must go to the clinic. So, it is important for us to study'. (CHW, female, 34 years old)

Improved knowledge meant improved feelings of competence
Community health workers described feeling empowered by the learning: more competent and better equipped to help their clients and support their communities with their improved knowledge: 'The book has the information. So, we learn about theory. We are very happy to have that book and to read and get knowledge. Although we are practical at the clinic, but we have to have theory … The book really helped.' (CHW, male, 22 years old) 'We are comfortable now to go to visit at the homes, because we are more confident that we know our work.' (CHW, female, 24 years old) 'Their learning makes them competent. They know they can give the right advice, help their clients and refer back to the clinic or hospital when it is necessary.' (Manager, male, 40 years old) 'Building on their [CHW] knowledge is empowering. It means they never feel undermined. Building their confidence is also important to the learning. It is always exciting to watch people grow in their knowledge and confidence'. (Facilitator, female, 44 years old)

Learning improves motivation and inspires to achieve more
Community health workers described feeling more motivated going to households armed with knowledge: 'We are now excited to go to households. I wake up and feel motivated. I don't feel anxious like I would sometimes feel. Especially with the mommies and their babies'. (CHW, female, 33 years old) Some CHWs indicated that they would like to continue to learn, and Skills2Care provided the motivation and inspiration to do so:  Following the pre-determined learning process was effective. The facilitator indicated that it was important to follow the process as determined by the Skills2Care manual: 'The manual has a prescribed process. It is systematic. It has been tested with other health professionals. It worked well with the CHWs too'. (Facilitator, female, 44 years old)

Learning material was compiled in an accessible and simple way
The accessible language and format of the manual lent itself to learning. All CHWs were isiXhosa first-language speakers. Not all were fluent in English, yet repeatedly they stated that the Mother and Baby Care manual was accessible, with many indicating that they preferred to learn in English. They considered that it was written in a style that did not compromise on necessary and essential information:

Organising in groups facilitated learning
Most CHWs reported enjoying working together with their colleagues in groups as it motivated learning.. They were able to consider ideas and raise challenges amongst peers in a non-threatening environment: 'Actually, we came to enjoy the groups. We liked to get together. We shared our stories, successes and our failures, so it became a group of friends that we started to trust. So, learning became something to be enjoyed'. (CHW, female, 22 years old) Beyond the sessions with the facilitator, CHWs often studied together in groups. Groups were held during lunch hours and sometimes when the clinic was deemed to be 'quiet'. Mostly, group sessions lasted about an hour.
Collectively, CHWs were able to work through and learn the material, and considered this time beneficial: 'We were four in our clinic. We would come together on Fridays and learn together. If there was spare time at the clinic, we would go to the sister and ask for time to learn. It made us to know the manual well. We would talk to each other and encourage each other to learn'. (CHW, female, 27 years old) 'Sometimes you are shy about a term in the book, you are not sure. But in the group, you know your care workers, you feel free. You can ask your questions'. (CHW, female, 23 years old) 'I liked to learn with my colleagues. Then we develop a way of being together, so we didn't feel embarrassed to ask questions. We knew that our questions were valuable'.

WhatsApp groups not necessary for learning
Skills2Care suggested the formation of WhatsApp groups to support learning. Many CHWs did not own smartphones or they could not afford to buy data and were therefore unable to access WhatsApp. Its absence seems not to have been a deterrent to learning.

Factors that enabled Community health worker learning
It was the intangible and unmeasurable factors related to the approach to facilitation, facilitator encouragement and support that appeared to promote and motivate the learning of the CHWs most. Some of these have been described above. The more tangible factors such as a dedicated place to study, a quiet space and the availability of a smartphone to connect to a study group were seemingly less important.

Facilitation of the process was beneficial
The learning of the CHW is successful with the support of a facilitator. The need for a facilitator, as compared to a self-study method, had previously been determined by SPF in an earlier pilot study. Key stakeholders confirmed this earlier finding:

An encouraging, participatory adult-education approach promoted learning
The facilitator was clear that a participatory, adult education approach was important. She explained that the active participation of the CHWs is a successful with adult learning: 'It's about building on the knowledge that adults have. As an adult educator you always assume that the people you are teaching have knowledge. So, you don't feed them information, it must come from them. When they are told the information, they don't value the information they have, they are afraid of the information they know'. (Facilitator, female, 44 years old) The support and encouragement from the facilitator were important: 'Mam G always believed in us. When we came to difficult topics, then we talked amongst ourselves and she guided us to the right answers. It was nice'. (CHW, female, 28 years old) The facilitator explained that she needed to be flexible in her approach: 'We know group discussion will always be effective. But then someone doesn't follow, and then you use one-on-one method. You can't be like a professional nurse, standing there and lecturing in front of people.

Support of nurses at the clinics assisted learning (and was appreciated)
At all the clinics, CHWs had encountered professional nurses who had been supportive and helpful. If concepts required clarification, CHWs would ask the nurses. They had been willing to help and provided explanations as required, which was much appreciated: Most improvement was evident in the module Care of the Newborn Baby. The CHWs had limited prior knowledge of the topic, and their group average score increased from 54% to 89%. These results demonstrate the possibility of introducing new topics for CHW learning and their potential for success. These results were similar to those documented for professional nurses. 13 Schneider et al. noted that many CHW programmes focussed on maternal and child health. 2 This programme followed that trend. They emphasised that programmes, having been donor-led, needed to be integrated into the national health programmes and local PHC systems. Whilst this programme has indeed been included in the PHC system, advocating for its implementation at provincial and national departments with emphasis on local needs is essential if it is to be implemented more widely.
Whilst prior Skills2Care experience had shown that the facilitator was integral to the process, 14 this study confirmed the nature of the facilitation being important. An understanding of the principles of adult education, adopting an encouraging approach with CHWs and nurturing participation were deemed important. The attitude of the professional nurses in this study indicated their willingness to support CHW learning. They may well provide a clue to the necessary support for CHW learning at health facilities in the future.
Gaining new knowledge provided CHWs with a sense of confidence and improved their competence. This newly gained competence is 'an important driver to empowerment'. 15 Community health workers hope and belief in themselves and their ability, triggered their thinking towards further learning and their career development.
That CHWs reported valuing the programme and enjoyed the learning was encouraging and suggests that this method of continuing education may be as successful with CHWs as it has been with professional nurses and midwives.
The recent Oxfam report on health workers 16 and the chapter by Vale and Di Maurio 17 both attest to the value that CHWs bring to the health system. Both argue that their intimate knowledge of local communities and of the households within the communities was invaluable yet undervalued by the health system. Empowering CHWs with knowledge and understanding may contribute to building their confidence and providing an avenue in which CHWs are emboldened to speak up and challenge the system so as to better understand and appreciate the value that they offer the health system.
The evidence gathered in this study points to the programme's appropriateness for improved CHW knowledge, and the results demonstrate that the model has been successful. It suggests the value of using a decentralised, group learning approach on a larger scale. The Skills2Care method can possibly be applied in all settings, not just rural.
The small study sample with only four modules for learning in this study were its main limitations. It nevertheless provides some insight into what may be possible with a larger programme. It holds the promise of expanded programmes of learning for CHWs with larger numbers and a wider range of topics. The study did not examine the costs and cost-effectiveness of the programme in a rural setting.

Conclusion
Community health workers are integral to health systems, and PHC especially in low-and middle-income countries. Their ongoing training needs to be appropriate, affordable and prioritised if they are to be properly recognised by the health system and their benefit are to be maximised. This study showed that the on-site, cooperative Skills2Care learning programme, with the support of a facilitator, was a success and well received by the CHWs. A comparison of pre-to posttest results demonstrated a significant improvement in cognitive knowledge with three of the four modules. It showed that a participatory approach to facilitation to support and encourage CHW learning is important. Health facilities in remote, rural areas in particular may benefit from the Skills2Care programme. The engagement of nursing staff at the clinics may provide further support for continuing education for CHWs. Advocacy within the Health Department for the ongoing implementation of this project will enable it to realise its potential.