The role of CHW provides increased agency and status in communities and can be an empowering experience [8, 10, 28, 39,40,41,42,43]. Our findings also reveal the dynamic and conditional nature of this empowered position, which can be altered by policies and shaped by multiple intersecting factors. Both districts in our study area had greater numbers of women working as APEs, which is at odds to what is seen in the country as a whole and particularly the dynamic in the north of the country. This is an obvious limitation; however, exploring their experiences helps guide suggestions for the recruitment of women and encouraging retention of men and women. It further serves to highlight the importance of context: how policies play out within countries varies, not just between countries. Our findings showed differences not just by districts, but by individuals. APEs are not a homogeneous group, and gender, age, geographical location and marital status all converge to influence both recruitment and retention of APEs. The conceptual framework (Fig. 1) used to frame our results, highlights this complexity: APEs’ experiences were influenced by other axes of inequity, which were important in shaping APEs’ outside livelihood opportunities, and national context, which included the historical impact of war and resource availability during times of famine. We present recommendations for policy to support recruitment and retention based on our findings within key themes from the conceptual framework. These are summarised below in Table 2.
Modular, flexible training
Experiences of training were heavily influenced by gender roles that dictate household chores and child-rearing be predominantly carried out by women. The four-month training programme away from home impacts some women who may find it difficult to eschew their domestic roles, which often include being the primary caregivers to children. Conversely, men’s ability to fulfil the ‘provider’ role for their family was challenged as they rely on the minimal subsidies set out by the Mozambican government and cannot take part in additional income-generating activities for these months. The expectation that the CHW role should fit in with income-generating activities not being met has also been reported in Ghana . Experiences of the training programme were also shaped by intersecting factors—such as marital status. In the case of single mothers who take on both caring and ‘breadwinning’ roles for their families, the pressure to provide caused particular stresses during training. External factors such as political context and resource availability also influenced training experiences, as demonstrated by the discussion of famine at the time of training. The financial stresses taken on by APEs during training illustrates their economic and social vulnerability—something which has also been noted in work from Nicaragua and Ethiopia [40, 43]. This vulnerability further highlights their unique embedded position in communities, occupying the same socio-economic bracket as their neighbours, but with slightly elevated status.
To support the training of this cadre and ensure that men and women are not limited by their domestic roles, we suggest that residential training be accompanied by childcare provision and a formal salary, with an option to complete the training in shorter modules to allow for breaks at home. Modular learning would also support opportunities for further income generation, which limited men in particular. Residential, modular training with on-site childcare facilities has been rolled out in India to support the female CHWs (known as ASHAs) with childcare obligations, whilst building an immersive and empowering learning experience .
Gender transformation within communities
Gendered intra-household bargaining and male decision-making power have been well documented with regard to limiting women’s autonomy over health-seeking behaviour in low- and middle-income countries, including Mozambique [26, 45,46,47]. Our research extends this literature to show how gendered intra-household bargaining also shapes APEs’ own experiences and their livelihood negotiations. In these cases, patriarchal hierarchies within the family impact both women’s access to healthcare and women’s choices in delivering healthcare. Married women, in particular, may have less power over their decision to become an APE as their often-limited autonomy led to difficulties in the recruitment due to the four-month residential training programme. Women’s ability to influence and make decisions within their households, therefore, needs to be considered from a human resource perspective.
It is vital to challenge the patriarchal notion that men should make decisions about whether their female household members can work outside the home. Challenging this notion needs to occur in communities with community leaders and APE supervisors who may inadvertently reinforce this notion as they are often called upon to liaise with husbands in instances where spousal consent to work outside the home was sought. The values and ideals of the supervisors are likely to influence this discussion—necessitating a training module for supervisors in how to approach discussions that have important implications for women’s intra-household bargaining power. Work within communities with the help of community leaders may also help to gain husbands’ acceptance, but importantly, also to transform harmful attitudes that see men threatened by women’s earning power or employment status. Challenging patriarchal norms around livelihood decision-making should occur simultaneously alongside supporting and being responsive to the strategic interests of APEs in these settings .
The success of gender transformation within the community comes from an Indian CHW programme which conceptualised CHWs as activists to empower the poor and marginalised. This vision was carried through to selection, training, ongoing support and systems of accountability and remuneration and resulted in the empowerment of women in the community . They could access information about their rights, partook in community decision-making and understood violence against women as a social rather than personal issue .
Training on selection for community leaders
Nepotism in selection, which has also been shown in the Democratic Republic of the Congo , was linked to giving income-generating opportunities to regard to young men. This has implications on the sustainability of the programme as it was reported they often leave for better-paid opportunities. Community leaders lead the selection process. Therefore, it would be beneficial to conduct thorough training with them around the roles and responsibilities of APE, aims of the programme and the importance of clear accountability mechanisms at the community level. This will help to ensure all community members are consulted on the selection of candidates, open up more opportunities for female candidates and avoid investment in candidates who are not committed to the role.
Remuneration and social security
Given APEs’ essential role in the health care sector, there is a duty to ensure that they are appropriately and adequately supported by the health sector once through the recruitment process. Studies in Zimbabwe, Uganda and Mozambique have linked low and irregular pay and increasing workloads to poor retention and motivation amongst CHWs [15, 44]. Our results demonstrate a clear sentiment of APEs who felt conflicted—happy in their roles but also struggling to support their families within the realities of the working environment that sees increasing tasks and delayed subsidies. In particular, female APEs spoke movingly of feeling undervalued and unappreciated. Although there is no distinction in policy between male and females with regard remuneration, the experiences of women may have been influenced by their gendered positions which afford them relatively limited options for alternative sources of employment when compared with men, who were reported to travel to South Africa for mining opportunities.
Our findings corroborate literature from Mozambique and Ethiopia by Maes and Kalofonos  who used a life histories approach with CHWs to explore recruitment and retention. They report that CHWs feel undervalued and underpaid by the government—expressing frustration when organisations benefit from CHW work whilst the CHW’s aspirations for socio-economic advancement go unfulfilled. This sentiment was powerfully voiced by our female APEs describing themselves as ‘slaves of the ministry’. The study also critically notes that CHWs’ desire to earn money is beyond their own self-interest but in order to support, or in some cases, start a family . This was a notion that was continuously expressed by both male and female APEs. Yet, despite their frustrations, APEs remain committed to their communities via a sense of duty to their community. Our respondents had internalised a common discourse of their work being priceless—and therefore unpaid—only remunerated with mental satisfaction. They suggest altruistic motives for remaining in the programme and used phrases like ‘serving our country’ and ‘there is no way a life is bought’. Similar sentiments have been expressed by volunteer CHWs in Nicaragua  and Ethiopia, where volunteers experienced psychological distress at not supporting their families, yet continued to volunteer out of obligation , highlighting the moral duty of care CHWs hold.
This highlights the complex issue of voluntary work. Voluntary work has been touted as a way to ensure intrinsic motivation  and sustainability; NGOs may also see their work as ‘ticking the sustainability box’ if it relies on the work of volunteers . The reality of limited remuneration, however, is that can be a hindrance to empowerment for many CHWs across contexts  and may prevent CHWs from being taken seriously . Many CHWs also risk impoverishment as they care for their communities at their own expense [49, 51]. Unpaid work is often taken on by those who can least afford it as it is viewed as a conduit into a paid position [12, 34, 41, 43], as voiced by our respondents who saw APE work as an entry point to a career in healthcare. Volunteerism is also inherently gendered, and women are often vertically segregated to unpaid, informal positions as they have more limited economic opportunities . Recent findings from Afghanistan illustrate this, where male CHWs were praised for their volunteer work, but it was seen to be expected of women . The assumption poor women can set aside time for voluntary work is problematic . Further, labelling the work as voluntary reinforces the perception that women have domestic duties they need to work around  and bolsters the ‘double burden’ women face of juggling paid work with family responsibilities as expressed by our respondents.
In the Mozambican context, volunteerism may reinforce harmful gender norms as men quit for higher-paid positions, but women remain in the role suggesting their labour is cheaper and less valuable. However, care needs to be taken if moving to formalised employment. In India, a reported benefit of keeping the CHW role voluntary is that educational requirements could be relaxed for selection . This would need to be thought through in the Mozambican context of interrupted education as a result of war and women’s lower level of literacy. Further, alongside offering formal employment needs to be strict selection criteria to ensure it is not just men who get selected for paid work, reinforcing gender power relations.
Our findings also linked limited autonomy over livelihood decisions to the inadequate financial gain from the role. Some husbands deemed the role of little value due to the minimal contribution to the household economy. Sen  describes how women’s contribution to the household may be diminished by gendered ideologies that characterise women’s income as ‘supplementary’ to that of the male ‘breadwinner’. Intra-household bargaining positions have also been shown to be influenced by an individual’s ability to pay, or perceived contribution to household livelihoods, amongst community members in Ghana . Outside earnings therefore provide women with psychological and practical leverage and increase their decision-making power by increasing their perceived contribution to households .
The call for greater subsidies and formal employment rights, including social security, from the APEs should be heeded. Working at the lower end of the health system hierarchy, APEs are vulnerable to poverty and are overcoming damaging past events involving famine and conflict . The provision of fair wages reduces food insecurity and helps progress towards Sustainable Development Goal eight for decent work and economic growth. It can provide men, but importantly women, an entry point into the formal work sector and employment rights such as insurance. Further, payment helps APEs to secure their livelihoods, contributing to empowerment - which has been argued to be an essential prelude to CHWs being committed to, and effective in, enacting their roles as health promoters and agents of change . Formally paying both male and female APEs may also pay dividends to the broader health agenda and contribute to the economic development of the country- with; maternal income has been shown to lead to improved child health, one of the aims of the APE programme [45, 55].
Further training opportunities, scholarships and career progression
Some younger female APEs spoke of their choices in choosing domestic work over APE work, but others felt limited in their choices due to their educational level. They articulated their desire to further their education in order to make them eligible for formal employment, better their futures and receive the recognition from the health system they felt they were owed. Developing further training opportunities, building a career structure and allocating sponsored places in higher education schemes for those eligible may be one way to further the transformation of this cadre and contribute to the economy and health workforce. A recent review of government reports by Percival et al. found no evidence that the Ministry of Health prioritised gender equity in its overall human resource strategy and that the promotion of women was not part of the recruitment process for the APE programme . This highlights a key gap to address within the APE policy; however, the capacity of the health sector to absorb human resources trained in health will need to be considered and alternative pathways under ‘community health’ included. Success in increasing the number of women in supervisory roles was shown via gender-transformative programming by the President’s Malaria Initiative Africa Indoor Residual Spraying Project across African settings. Giving hiring priority to qualified women applying for supervisory positions helped to increase the percentage of women in supervisory roles from 17% in 2012 to 46% in 2015 . Providing further career opportunities for eligible APEs may also encourage the aspirational younger generation to join, impacting the sustainability of the programme and eventually increasing the number of qualified health workers in the country. This may also particularly benefit women to progress their education and help to transform broader societal norms.
Provision of health posts
Our research revealed interesting findings around policies on the place of work that have gendered implications for women. Policy dictates that the APEs do not operate out of a health post, but within communities—often using their own or community members’ homes. These policies however disproportionately impact on female APEs who are often not the primary decision-makers within their homes—if they treat clients out of their homes, they may be subject to family disapproval. This disapproval can lead to attrition, but it may also place women at risk of domestic violence, as voiced by a male APE. Violence can be considered one of the ‘most graphic expressions of unequal household power relations’  and can further limit women’s autonomy. Provision of a dedicated health post to work out of is essential to improving the safety of female APEs and legitimacy and recognition of the cadre.
Our study has some limitations that need to be considered; firstly, whilst we set out to explore the reasons behind the low levels of female participation in the APE programme, in Maputo Province, the ratio of women was reported to be much higher than is seen in the North and Central zones of the country. As this is a qualitative work, our findings cannot be generalised to the rest of the country but provide important insights into some of the considerations in recruitment and retention of male and female APEs. Secondly, it would have been beneficial to hear not just from current APEs but from those who had left the programme to explore their lived experiences and the reasons behind their attrition. The sample size of the single key informant was small but adds depth of knowledge and insight into the national context. Another potential limitation was the need to combine the male and female APEs into one FGD due to the limited numbers of APEs in Moamba. We feared this may have created some gendered power dynamics. However, the APEs presented themselves as a team and spoke openly about personal and community gender issues, and we elicited some rich responses this way. Finally, it is important to consider our position as researchers and the impact that our position as outsiders of the community, living in Maputo city and beyond, may have had on respondents. Despite this, participants generally seemed open and willing to talk to us and wanted to share their stories to enact change.