Opening up the ‘black-box’: what strategies do community mental health workers use to address the social dimensions of mental health?

This study was set in bustling peri-urban settings of Dehradun, the capital city of the north Indian state of Uttarakhand. This sprawling periurban district has 1.96 million residents. The dominant patriarchy is evidenced in the sex ratio of 902 females born to every 1000 males while the national sex ratio is 940. Disadvantage for women is further evident in the disparate literacy rates which are 78% for women and 89% for men [18].

It was hosted by Burans, a non-profit community mental health partnership project that started in 2014 (www.​burans.​org). The three geographical communities engaged in this study were: (1) Sahaspur, a town of 100,000 people living in closely co-located and informal brick housing; (2) Kanwali road, in the centre of Dehradun city, where 80,000 people lived in informal housing, lining a large open sewer; and (3) Mussoorie, a tourist town one hour’s drive from Dehradun, where a large service-industry supports hotels, with 35,000 residents (insert map indicating these areas).

In each community, five employed team members worked with community members to promote mental health through increased knowledge, safe social spaces and partnerships for action [7, 29] and by strengthening the public mental health system. Burans works with community members who self-identify as experiencing mental distress and who are referred to as people with psychosocial disability (PPSD), which is the term preferred by people with lived experience of mental health problems in India [30]. People describing anxiety, unexplained somatic symptoms and/or depression who were able to fulfil most daily responsibilities were categorised as having a common mental disorder (CMD). People (or their caregivers) describing the loss of social networks, lack of self-care and/or auditory hallucinations or delusions and unable to fulfil their daily responsibilities were assessed as having a severe mental disorder (SMD). The majority of PPSD experience common mental health problems such as anxiety and depression and do not have access to biomedical care and do not have a formal diagnosis. Further detail of the types of problems experienced and demographics of PPSD within the Burans programme are described elsewhere [30, 32]. The CHWs working with Burans are women from the local informal urban community, who have completed high schooling and have shown evidence of engaging in their neighbourhood as change agents. Each CHW delivers a mix of individual psychosocial support to PPSD following the structured Burans ‘care plan’ which uses a coproduced recovery tool, Swasthya Labh Sadhan to support PPSD to take active steps to engage in eight domains of recovery [31]. CHWs also use validated psychometric scales to measure social participation and mental health outcomes, which have been documented in a prospective cohort study [32], CHWs also implement a range of community organisation activities as well as group facilitation for young people, caregivers using structured interventions. CHWs also support PPSD to access biomedical care by accompanying them to the Government psychiatry clinic for the first time, CHWs are trained and coached twice weekly by a social worker-trained team leader, who supports four to five CHWs. Inception training is 15 days spread over four months and includes modules in community organisation, group facilitation as well as technical topics related to mental health, active listening and problem solving and other psychosocial skills as well as advocacy and access to health care. They work eight hours a day and are paid by Burans. A published case study provides further details of the implementation team and their approach [30].

Research participants included both people with psychosocial disability (PPSD) who were registered with the Burans non-profit programme as well as CHWs working with Burans. At the time of this study, there were 1000 PPSD registered with the Burans programme. To ensure we selected a non-biased sample, we used a random number generator to identify 46 PPSD from the electronic register to invite participants in this study. Inclusion criteria were that PPSD were over the age of 17 years and had participated in meetings with the Burans team at least six times. People who were no longer resident of the Dehradun district were excluded. All selected participants consented to participate in this study.

In the second half of 2017, PP and KM interviewed seven of a total of nine CHWs about their PPSD clients using a semi-structured interview format that probed their understanding of the client’s problems, their engagement and responses to PPSD and the barriers they faced. Interviews were conducted in Hindi, and audio-recorded and detailed notes were made from recordings and translated into English. KM and PP as employees in the same organisation as the CHWs also spent several weeks shadowing the CHWs while they went about their work in the community. They sat in on conversations, watched how CHWs interacted and built connections and took written notes describing their observations of CHWs working with PPSD. They also read and reviewed the hand-written care plan records that were kept by CHWs after each home-based visit. The information collected in these care plans is described in greater detail in two published studies [30, 32].

We used a case study analysis approach, considering each CHW as a case study with the three data sources of interview notes, their written care plan records and observations of each CHW in action. Interview data were coded emergently by KM and SJ, paying attention in particular to the strategies used by CHWs and the ways that they identified and engaged with social and structural determinants of mental health.

Limitations were that while interviews were conducted in Hindi, the interviewers (PP and KM) did not speak the local dialect of the communities where PPSD and community workers were based, which may have limited understanding of the context. Additionally, some of the interviews risked social desirability bias as both interviewers were also involved in project operations and they may have been perceived as providing access to resources. The positive aspect of project implementation team members participating in data collection is that they had a strong understanding of the structural factors that contribute to mental health problems and had many interactions and observations of PPSD which could contribute to and enrich their analysis of data.

Priya was a 24-year-old community worker who had completed 12 years of schooling and lived with her parents in a low-income community two kilometres from the community she worked in as an employed team member. She identified PPSD through door-to-door knocking and meetings in the low-income community adjacent to her home.

Rita was an 18-year-old young Muslim woman who lived with her widowed mother and two brothers in a tiny two-room house in an informal urban community. After her fathers’ death, Rita had been required to leave school after completing 10 years of schooling to tailor clothes in order to generate supplemental income for the family. Although Rita enjoyed tailoring at home, watching her peers go to school meant Rita felt socially isolated, disadvantaged and experienced frequent concerns about the future, and stayed at home without meeting friends and felt lonely. Rita’s mother being the key decision maker after her father’s death, also faced mental health problems, which further exacerbated Rita’s condition. Rita’s status as a young, unmarried Muslim woman restricted her freedom of movement, a common experience for people like her in north India [16].

Community workers in our sample displayed a high degree of knowledge of the context of their client’s lives and were able to apply this in developing their responses. This included knowledge across a number of domains including both of more intimate spaces, like family dynamics, as well as wider forces such as environmental and social determinants. The knowledge of the context and structural health determinant was described as implicit (‘I just know’) and linked to their own experiences of living in communities that were socioeconomically similar.

The knowledge and analysis of context took place at different levels. For example, on a steamy afternoon in July, KM and Lalita, the CHW went to visit Deepak, a 25-year-old, unmarried man from the Sikh community. Lalita, the CHW squatted in a brick courtyard and gave KM a summary of Deepak’s story and family situation without any reference to notes. Her description engaged an understanding of the family history (two of Deepak’s uncle’s had gone to jail for an alleged murder), educational background (“He re-did his first grade several times but eventually he lost interest and when he was in his early teens, he started doing labouring work. Now he does any kind of mazdhuri (daily labouring) work.”).

Lalita also described the story of Deepak’s issues with substance misuse, a problem that affects his father and his two brothers. Her detailed analysis of the family dynamics included descriptions of the impact of the father’s substance misuse on the poor relationships with the sons; and the mother’s relationship with Deepak.

Lalita also reviewed and discussed the outcomes of the psychosocial interventions she had employed with Deepak (psychosocial counselling and support). She described that they did not result in significant improvement in his situation, either in terms of the standardized scores of mental well-being used, or in convincing him to seek rehabilitation but she could reflect critically on possible reasons for this.

After spending time talking with Deepak who roused himself from a dark room off the courtyard, Deepak stated he needed to head out. Back at the office, Lalita continued telling of her analysis of the situation for him. She described that one benefit for Deepak after he joined a four-month psychosocial group intervention was that he engaged with others who had higher status in the community, which she hypothesized, improved Deepak’s social capital and connections. Lalita also suggested that this outcome was tempered by factors that limited Deepak’s ability to address substance misuse which included a circle of friends who sometimes financially supported his substance misuse and the disregard and negative judgement he experienced from his parents.

In this example, the CHW, Lalita, demonstrated how she had applied a nuanced understanding of the links between family history, education, social status and economic disadvantage to engage with the interplay of structural and environmental health determinants.

Pushpa is another CHW who described her work with Sunita and her family in a town on a steep hill. She lived in a house with tarpaulin for roofing, and a dirt floor and in monsoon water ran through her house. Pushpa described how this made it very difficult for Sunita’s children to play and that the wet and dirty house led to worsening mental distress. Pushpa had visited Sunita most weeks at different times of day over a period of four or five months and so she felt she had had a clear idea of the social and economic factors which led to Sunita’s mental distress. Pragmatically, she then supported Sunita and her neighbours to file a Right to Information¹ request about why there was no water connection in their informal urban community, to prompt action by the local government.

CHWs described multiple ways that they engaged with social health determinants. For example, another CHW described negotiating with the parents of a 17-year-old PPSD client, to encourage them to allow her to attend school alongside her brothers, addressing gender relations. Another CHW described supporting a family with very low income to get access to epilepsy medicines by identifying local Government pharmacies who would provide them with free medication. Multiple CHWs described supporting families to complete the forms and doctors' assessments required in order to receive a disability pension, addressing the right to health.

Another example of a CHW addressing social determinants included engaging in dialogue with people in the neighbourhood to increase awareness about the medical condition of their young neighbour with epilepsy to reduce stigma and name calling.

In-depth knowledge of the context in some instances may have been negative for PPSD. For example, a community worker who knew a client and his family well judged that they were not likely to benefit from psychosocial support based on their family history, describing the criminal activities of his uncle and father. In this instance, family knowledge potentially led to prejudice and reduced psychosocial support.

CHWs described using a variety of practical and relational tools to build trust with a PPSD, which required a nuanced understanding of the local context and power relations. Relationships had typically been developed through multiple visits over many months and the ‘thickness’ of relationships was evident for example, Lalita’s description of Deepak’s family outlined above and her description of the range and quality of different relationships of Deepak without reference to any notes:

“Deepak travels a lot for work, he has been working to sell semi-precious stones and astrology for the last few years. This is his main income ( ) he’s married. They are expecting their first child. ( ) But because of travelling a lot he sometimes feels lonely and then he befriends others who are also travelling for work. ( ) He has described how he feels peer pressure from them to use mind altering substances. ( ) and then he also has unstable relationships with his family members and household.”

CHWs described that trust building happens over a long time period through multiple communications which facilitates a tailored response. In one situation, the CHW connected with a village head (known as a pradhan) who had an alcohol overuse problem. The process of building this relationship illustrated the practical and relational tools used, E.g., sharing a brochure, discussing strategies to reduce use, active listening and problem solving to build a relationship of trust as described below:

“Through our regular meetings I listened well and held dignified and respectful encounters. Then we developed action plans which Sanjay could do in the next week. So (using these) he could chart his progress and see that, and at same time he felt trustworthy and could complete his plans. A key part of this was our relationship in his house (and where we) showed told him how well he was doing.”

A further approach that improved outcomes, included supporting PPSD to build relationships with peers as a CHW working with a young man described in the quote below:

“Through an increased number of peer friendships and being a group member, I think Sunil’s self-image improved. He told me “Mujhe bahut izzit milte hai jab session chalet hai.” (I feel respected when I attend the sessions).

The strong relationships between CHWs and PPSD emerged from this study as a key mechanism supporting outcomes. Positive relationships have been identified as core to mental health recovery. Recently theorised approaches to relational care in health services suggest that relationships can have causal effects when enabled by the following conditions: firstly, each person has a personal and social identity known to the other. The fact that CHWs are well known and live in or near the communities they work means that they are known and can use their personal experiences to facilitate the relationship [26]. This close relationality brings important challenges. Chase [11] drawing on ethnographic research in Nepal explores what happens in the context of task-shifting when the boundaries between CHWs and community members are fluid. This fluidity was seen in our study of the bespoke and responsive care provided by CHWs, where they described unique approaches with each client that engaged with their circumstances and were less systematised. For example, focussing on self-image with Ravi, and supporting Deepak in strategies to say ‘no’ to substances with peers. Secondly, relational care is primarily characterised by caring rather than a focus on activity or performance (outputs and outcomes). This was evident in this study, for example, when the CHW describes Sanjay feeling affirmed due to the relationship rather than perceiving the relationship as transactional or conditional on performance. Thirdly, CHWs described how trust grew with a growing duration of the relationship, as described in the example with Sanjay, the village leader with alcohol use problems.

At the same time, there were problems in the investment and proximity of CHWs in relationships. CHWs were far from impartial in their relationships, and in multiple instances they showed negative judgement of men drinking alcohol, talked of persuading men to reduce their alcohol use and were frequently observed to be ‘siding’ with the women in the households of substance abusing men. This concurs with Chase’s observations that the practice of CHW can be “judgement laden and directive” and that CHWs can often use personalised over professional approaches [11]: pp. 8).

This study illustrates examples of ways that CHWs offer care explicitly tailored to local contexts. Evidence-based practices developed in one place are often not appropriate, feasible or effective in a different context [21]. A large number of recent publications have identified the importance of a socially and culturally informed approach to community mental health that engages with the local context to develop endogenous solutions [4, 21]. Our dataset had various illustrations of how CHWs creatively adapted their approaches to account for local contextual and cultural aspects. For example, CHWs tailored their interventions and relationship building to account for local context and power relations within communities. Chase [9] discusses this in relation to lay psychosocial counsellors in Nepal. She illustrates how these workers flexibly use the concept of ‘psychosocial’ in their practice. For example, she discusses how CHWs developed symptom lists and descriptions that were flexibly tailored to social contexts and individual histories. This was in contrast to publicly available materials describing complaints that psychosocial counsellors could address.

Kirmayer and Pedersen [21] discuss four ways that culture contributes to social health determinants and in this study we identified two of these in CHW interactions. First, they identify the importance of “categories of identity that disadvantage specific groups”. CHWs in this study identified and addressed name-calling using their local knowledge to respond via the cultural beliefs and drivers for stigma and discrimination. Second, Kirmayer and Pedersen [21] argue that culture may provide systems of understanding that may worsen or moderate particular hardships. CHWs in this study supported the development of peer relationships and activities that shifted understanding of poverty and social exclusion for Rita (by joining the sewing group). At the same time, the solutions the CHWs proposed were conforming and non-transformational, and fully proscribed by the dominant cultural and gender relations, for example, the only suggestions CHWs provided for income generation for women with mental health problems was to work in sewing or in a beauty parlour.

CHWs in this study described that they supported mental health by engaging with families and communities to act on structural health determinants such as access to clean drinking water and sanitation or income generation which built on their in-depth knowledge of the context as well as skills in accessing government entitlements. A recent scoping review [2] described how CHWs can increase equitable health outcomes in LMIC, through advocacy for social and structural transformation. This review also found that CHW in particular gave additional time and resources to families who were most disadvantaged, recognising that they had fewer resources and supports. This approach suggested that CHWs were more successful in increasing equity in service delivery than in health outcomes. Further research is needed to examine whether or how CHWs support equity.

One of the counter-arguments to the approaches that engage with the ‘social’ outlined above is that these bespoke approaches do not entirely materialize just from the CHWs and their interactions. They are likely to be linked to the ‘altruistic personalities’ described in Allen and Rushton [3] among individual CHWs who may be more likely to be compassionate, pro-social and socially connected than other community members. The values, shaped through the training and ethos of the host organisation, are also likely to inform what CHWs do and how they do it [22]. There is additionally, something that happens in the interactions with PPSD that might be adaptive or bespoke. One potential way of looking at this is to conceptualize the approaches in terms of ‘thickness’ of interactions, and a layering process that takes place with different elements playing roles in a bricolage way. Additional methods which include ethnography are likely to be required to untangle and understand in-depth how CHWs develop contextualised micro-adaptations that support improved mental health outcomes among PPSD.

The micro-innovations shown by CHWs in this study are non-linear and thus dependent on elements we have identified above including local knowledge, time, and relationships. For example, when the CHW role played conversations to build Rita’s social confidence and social connection (the sewing group) and to support Rita to pursue further education. A recent review showed that in addition to the health system and economic factors, the key community-level factors that influenced CHW performance and operations included socio-cultural factors such as gender norms and values, perceived safety and security as well as the education and knowledge level of the target group [22]. In considering which of these elements could be scaled, and how in thinking about social interventions, further attentive qualitative research is needed to examine in greater depth the particular interventions or particular ways of being employed by community health workers.

This study sourced data from community health workers active in a regional community mental health programme set in Uttarakhand, North India. The process of using cases that were randomly selected by 46 PPSD for qualitative interviews provides a breadth of identities as well as ensures a mix of eventual outcomes. A limitation of the study is that we did not directly interview PPSD meaning our findings primarily rely on CHWs’ interpretations of what happened in a given situation. This was mitigated somewhat by both KM and PP triangulating findings from CHW interviews with participant observation over several weeks of shadowing CHWs. Both PPSD, family and community members may have interpretations that differ substantially. This points to the value of future ethnographic research examining the interrelationships between different stakeholders in these care processes.

The female identity of data collectors (KM and PP) as well as all women CHWs is likely to have influenced responses by CHWs, and may have limited representation of the approaches used by men CHWs and men PPSD.

Transferability to other contexts was given attention with the inclusion of a wide range of PPSD and was also strengthened with the use of case studies which provided thick description of CHW-client interactions. Although some of the findings may be applicable to other community-based mental health programs in South Asia, the context of Northern India is unique and therefore generalizations from this research should be examined alongside context descriptions. Lastly, dependability and confirmability were addressed by reflexive discussions between researchers, CHW participants, and notes on the research process, ensuring consistency across data collection and analysis processes.