Study setting and collaboration

This study took place in Mysuru (Mysore), India. Mysuru is a mid-sized city of about 1 million residents in the southwestern state of Karnataka and is home to the locally run partner organization for this project, the Public Health Research Institute of India (PHRII), which over the last 17 years has developed a strong reputation for delivering evidence-based interventions to reduce health disparities in the area. The first author co-conceptualized, designed, and oversaw field data collection in-person and remotely, while PHRII authors co-conceptualized, pilot tested, and performed data collection in collaboration with the first author.

Study population and sampling

Women were eligible for participation if they were 18 years of age and older, Hindu, married or previously married, and spoke the regional language, Kannada. Women were purposively recruited through 17 of PHRII’s existing contacts in socioeconomically diverse neighborhoods, after which we employed snowball sampling to recruit additional participants one or two degrees of separation from the original contacts. This sampling strategy aimed to recruit a sample with caste and socioeconomic variation, but with relatively uniform life stage, religion, and language characteristics. Religious identity was confined to Hinduism because religion and language spoken at home are closely entwined in this context, and ensuring that study participants spoke a common language was important for other portions of the study. Because this study was designed to assess general community attitudes and perceptions around mental health, participants were not selected based on mental health status or prior experience with mental healthcare.

Methods

From May 25, 2018 to June 21, 2018 and from August 25, 2022 to September 23, 2022, we recruited and conducted interviews in Kannada with a community-based sample totaling 54 women in Mysuru city. The initial series of interviews in 2018 were exploratory, asking women about the common causes of distress in their lives, eliciting Kannada terms and expressions for distress, and assessing where they would go for help if their distress felt unmanageable. The second phase in 2022, which was delayed because of COVID-19 disruptions, involved another round of individual interviews with a new sample of women. The research team repeated some questions from the 2018 work and added new ones exploring women’s coping, resilience, and knowledge and attitudes toward clinical mental healthcare. Knowledge, experience, and attitudes were assessed by a set of questions about how women would interpret symptoms of depression and anxiety presented in fictional vignettes, whether they would ever seek the help of a clinical practitioner if they had similar symptoms, whether they had ever sought such help, and if they knew of specific practitioners in the community. In both phases, interview questions were developed with the informal input of other researchers working in global mental health and in consultation with PHRII staff. The specific interview questions are published elsewhere [45].

Interviews occurred in women’s homes during the day when other family members were out, enabling women to speak with interviewers in private. Individual interviewing was chosen as the primary mechanism of data collection because in other study phases, women in group settings evinced some response bias and discomfort around disclosing personal experiences of distress in front of others. Interviews were conducted by a native-Kannada-speaking female member of the research team and typically lasted about one hour. Each participant was interviewed only one time.

At the conclusion of each interview, researchers administered a Kannada-translated and adapted version of the Patient Health Questionnaire-9 (PHQ-9) [46] and collected demographic data from each participant. The PHQ-9 asks participants to rate their frequency of nine depression symptoms in the previous two weeks using a Likert-type scale. A summary score was then calculated for each participant ranging from 0 to 27, indicating risk of depression. Clinical guidelines state that a score under 5 indicates no depression risk, 5–9 indicates mild risk, 10–14 moderate risk, 15–19 moderately severe risk, and 20–27 severe depression risk [47].

Ethics statement

All study procedures were pre-approved by the first author’s US-based institutional review boards (University of Alabama #18-OR-177 and University of Oregon #00000606) as well as PHRII’s India-based institutional ethical review board (#2018-04-28-43 and #2022-08-13-75). Participants provided oral informed consent after the interviewer described in Kannada the purpose, procedures, risks, benefits, and confidentiality measures associated with the research. Participants received INR 300 to compensate them for their participation in interviews, plus coverage for travel costs in case they elected to travel to PHRII offices for the meeting. Those who screened positive for any depression risk were offered free and confidential mental health consultations with a psychiatrist or social worker, a gender-matched escort to appointments, free transport, and coverage of fees.

Data analysis

Interviews were audio recorded and translated and transcribed from Kannada to English by a professional bilingual Kannada-English translator with a round of review and clarifications from the first author. Transcripts were then imported into Dedoose mixed-method analysis software [48]. Using an open coding approach with the constant comparative method, the first and second authors co-developed codebooks in an iterative fashion to reflect key themes in interview transcripts that addressed perceptions of clinical mental healthcare, distress experiences, and coping strategies [49]. These included core categories (e.g., "belief systems") as well as axial codes, which included sub-categories (e.g., "fate," “character grit”) and in vivo codes (e.g., "tension," an English term used to refer to generalized distress; [50]. The first and second authors cross-checked the other’s coding procedure for consistency.

Using Dedoose’s “descriptor x code” charts function, we analyzed relationships between stigma code frequency, PHQ score, and experience/willingness to engage with psychiatry. Dedoose’s normalize function allowed direct comparison between subgroups, which sometimes differed significantly in size (e.g., only 5 women reported having engaged with clinical care in the past). These mixed-method analyses allowed us to detect potentially meaningful differences in how frequently women discussed salient themes based on their mental healthcare usage and depression symptom severity.

Finally, we conducted basic frequency counts of how many women mentioned each of the key coping strategies we identified during the coding process. For some coping strategies, we were able to use a custom Python script to extract frequency counts. The script utilized the pandas library for data manipulation and analysis. A predefined list of search words or patterns, referred to as 'search_words,' was established. These search terms encompassed a range of coping strategies relevant to the study's objectives. For instance, the terms 'temple,' 'prayer,' 'pray,' 'pooja,' 'mantra,' 'recite,' 'mantras,' and 'God,' were used to identify religious coping strategies. The script then systematically searched for occurrences of the specified search words within the transcript and tallied the number of unique transcripts that mentioned the terms. For other coping strategies (e.g., cognitive reframing), it was not possible to identify keywords that would allow an exhaustive search using Python, so we performed hand-counts. These counts provided insights into the prevalence and distribution of coping strategies within the dataset.

Stigma

Across the sample, 50 percent of women stated that mental healthcare seeking could be stigmatized; this included two women who had sought care themselves. When asked why it would be shameful to seek mental healthcare, women stated that there is a public perception that such care is only for people with severe mental illness, which they described as involving dissociative or psychotic symptoms and referred to using the term hucca, or “madness.” That perception spanned caste, income, depression risk, and employment categories across the study sample. As one lower-caste woman who screened negative for depression risk explained,

“People will think that they are mad [if they go to a mental healthcare provider]. That might be a reason [they would not go]. They might be worried that there might be problems in their life, that people will look at them as if they are mad, or mental. They might not go because of this.”

Similarly, a general caste woman who screened positive for moderate levels of depression risk said, “If I got to [mental healthcare] doctor, I feel like they may think that I am a madwoman. That is why they are not needed for me.”

As these quotes suggest, women’s concerns about stigma mainly centered on potential discrimination from others that could occur if people saw them seeking care, as opposed to self-stigma or stigma attached to the disease state [52]. A local pediatric psychologist confirmed this, saying, “In the small city of Mysore, the biggest challenge people report experiencing before seeking professional help is the taboo or stigma associated with it. This is called huccha (madness). They think seeking a psychologist will make people label them as huccha, and it negatively reflects on them in the community.” This provider’s comment also hints at the problem of associative stigma, or the stigma that attaches to people related socially or genetically to a person labeled huccha, which in this cultural context, can damage employment and marriage prospects [53, 54].

Mixed-method analysis suggested a pattern of more frequent discussion of stigma among the small number of women who reported having engaged with clinical mental healthcare (n = 5). A normalized code x descriptor analysis, displayed in Table 2, demonstrates that those who had previously engaged with mental healthcare had higher relative frequencies on the stigma code than either those who would consider engaging (but had never done so), or those who would not consider engaging. There were 60 mentions of stigma across the 5 interviews with people who had engaged mental healthcare, in contrast to only 49 mentions of stigma across the 33 interviews with people who had not engaged mental healthcare. Keeping in mind that these results are limited by small sample sizes, it could be the case that women with direct experience of mental healthcare also have more direct experience of the stigma associated with that healthcare, or that women who had sought mental healthcare had greater stigma experience because their symptoms were more severe.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Mixed-method analysis comparing frequency of stigma mentions with participant’s experience with mental healthcare.

Results produced using a normalized code x descriptor analysis in Dedoose mixed-method analysis software.

https://doi.org/10.1371/journal.pmen.0000142.t002

Instead of going to mental health clinics, women often expressed a preference for seeking help from mahila sanghas, or women’s empowerment organizations. One general caste woman with mild depression symptoms said, “These mahila sanghas should help me secretly because my respect in the society should not go, and all my family issues and problems should not be disclosed to everyone. They can help me by suggesting how others can be punished or how I should lead my life. … Mahila sanghas will give solutions by maintain the confidentiality.” Another general caste woman with mild depression risk reported that if she were experiencing depression or anxiety, “I would go to mahila sanghas. One of my in-laws works there. They will actually guide people in problems to have a way out. They will tell people individually like where to go, or what to do, or what to study.” Likewise, a third woman with no depression risk said that if someone were suffering, they could go to such an organization.

“She must tell everything without hiding anything. She must tell every scene what has happened with her. There are many mahila sanghas and NGOs to support. In olden days, if there were any problems, it was necessary to solve by ourselves. Otherwise, one must commit suicide; that was the mentality. Now it is not like that. People are educated; they can share it with family members or neighbors. Or else they can go to mahila sanghas. Through them they can approach the police and solve the problem.”

Perceptions that services are not relevant to their needs

Besides concerns about stigma related to care-seeking, women said they would decline mental healthcare because they felt they could handle their problems on their own. Women differentiated between the kind of person who might need mental healthcare and the kind who would not, emphasizing that some people are inherently “weak-minded” while others are resilient and strong. Those who were “weak-minded” were prone to descend into madness when over-stressed and would require outside help to manage their distress. “People who are mentally weak will usually go to counselor,” explained one general caste woman with no depression risk. “If they do not go [on their own], their family will take them. …Some people cannot find answers on their own, but when somebody else tells them the reason for their problem, they will realize quickly.” Women in this study almost universally characterized themselves as “strong-minded,” and they often relied on this as a reason for why they would not want to seek mental healthcare. Women positively characterized “strong-minded” people as careful and deep thinkers, acting with self-conviction or self-efficacy to “manage” things without help. This personal resilience was perceived as a valuable quality; several women articulated the idea that adversity begets strength. One woman, who had one of the highest depression risk scores in the study, used an idiom to explain that idea.

“My mother has taught me that if we suffer now, then in the future we can grow strong. The stone that suffers when it is struck by an artist will someday become a lovely sculpture. So, I think I can handle all this [depression or anxiety symptoms]. I have handled many big things in my life, and so I could handle this also.”

Even when women’s distress would be considered relatively severe by psychiatric standards, they described relying on their qualities of “strong-mindedness” to manage distress on their own. One woman who had considered suicide in the past and screened positive for moderate depression risk explained, “I felt like committing suicide; I cried and cried a lot. Then thought of my two kids. Who will look after them if I take any bad decision in a hurry? I consoled myself, took courage on my own.” Through this process of remembering her responsibilities and attachments to her children, she was able to turn her mind away from suicide without outside help. She perceived this as a positive approach for handling her symptoms.

Other women felt they did not need mental healthcare because they viewed their problems as resulting from external stressors that could not be addressed by clinical intervention. “When there are any stressors at home, or if there are any problems, it is from money,” one said. Another explained, “[A main] reason for stress is money. Not only for me; money is the thing which causes stress to every human being. It may be [a major] problem for some, and small for others. But money is the reason for stress.” Women were acutely aware that mental health clinicians could not solve their money problems.

There were no discernable patterns between women’s willingness to engage with clinical care and their depression symptom scores in mixed-method analysis. In other words, women’s general lack of interest in clinical care did not appear to be patterned by their own levels or personal experiences of distress. It might have been the case that women would seek mental healthcare under different conditions than those we assessed in the study, but they generally agreed that the kinds of distress we were talking about in this study—specifically, symptoms of clinical depression and anxiety—were not an appropriate reason to seek it.

Coping strategies

Since most women did not desire mental healthcare, we explored what strategies they used to manage their distress instead. Women’s coping strategies were diverse, and we were able to group them into 5 major categories: cognitive reframing, somatic release, self-talk, mindfulness activities, and social engagement (see Table 3, below). There was some overlap between categories; for example, visits to temples could serve both a religious coping and a social engagement function; breathwork could be considered mindfulness or somatic release. Below we describe the nature and frequency of each category with exemplar quotes.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Categories of coping strategies used by women in the study, and frequency with which they were mentioned.

https://doi.org/10.1371/journal.pmen.0000142.t003

Stigma toward clinical encounters, not toward distress

Social stigma toward mental disorders is often cited as a key reason for lack of mental healthcare seeking in research conducted in LMIC settings [4–14, 16]. Stigma was indeed an important driver of women’s rejection of mental healthcare in this study. But unlike most of the extant literature, which describes stigma that attaches to individuals for being mentally ill [15, 55], women in this study associated stigma with the act of seeking clinical care, not with states of distress. This is similar to the findings of Koschorke and colleagues, who report that service users across seven LMIC settings experienced discrimination from family and community directly related to their care-seeking behaviors (for instance, one woman in Tunisia reported, “Even my best friend, said ‘you’re crazy, you go to [name of psychiatric hospital], your siblings are crazy, your whole family is crazy. Who would marry into your family? Who would marry you? If you’re a crazy person who seeks treatment at [name if psychiatric hospital], who would marry you?” [6]. Our mixed-method analyses revealed a similar pattern, where women in our study who had attended clinical mental healthcare in the past mentioned stigma far more frequently than those who had not sought care; the association between clinical care and stigma was also a common theme in the qualitative analyses.

Women did report that stigma attaches to individuals with severe symptoms associated with psychosis (e.g., hallucinations, delusions), but they did not associate this stigma with individuals experiencing symptoms of common mental disorders (CMDs) like depression and anxiety. Stigma arose primarily because of the perception that clinical mental healthcare is for those who are “mad,” which in this context usually referred to people with dissociative and/or psychotic symptoms. While other scholars have noted a similar pattern of stigma around care-seeking [6], to our knowledge, this is one of the first studies to find that the stigma of care-seeking eclipses the stigma of CMD-type distress.

Interventions to increase access to mental healthcare often involve a psychoeducation component designed to convey the idea that common mental disorders are disease states for which effective clinical treatments exist [56, 57]. The intent of these programs is usually to destigmatize mental illness and increase mental health literacy, but by connecting people with distress to care systems that are stigmatized, psychoeducation could paradoxically increase stigma in this cultural context. That phenomenon has been observed in some Western settings where disease-model psychoeducation has reproduced stigma by reifying mental illness as a static state [18, 58].

These findings demonstrate that simplistic assumptions about the way stigma operates should not overshadow careful study of where stigma comes from and what it attaches to. For our study participants, distress itself was not the problem; the stigmatized clinical setting was the problem.

Given the stigma that attaches to clinical mental healthcare in this setting, study participants suggested alternatives where women could seek support without stigma. They strongly endorsed women’s empowerment organizations, or mahila sanghas, as a positive alternative to clinical settings. They described these organizations as welcoming, confidential, and trustworthy, and often recommended that if a woman was experiencing more distress than she could handle, she should make a visit to a mahila sangha as a first or second step for seeking help.

Perceptions that services are not relevant to their needs

In biomedical psychiatry, symptoms are generally considered severe if they interfere with everyday functioning (e.g., a woman is so depressed that she can’t get out of bed in the morning). But women in this study conceptualized “severe” not in terms of disability, but in terms of “madness.” They distinguished between people who were “weak-minded” and “strong-minded”; weak-minded people were prone to mental illness and even madness, while strong-minded people were self-reliant and generally viewed positively. Women normalized their distress through belief systems that included viewing suffering as an inherent part of life and valuing suffering as a source of personal growth and strength. Given that women almost always described themselves as “strong-minded,” they generally did not consider themselves candidates for mental health treatment, even when their distress was severe. Our finding echoes the results of previous studies addressing attitudes toward clinical mental healthcare in South Asian populations, which report lack of perceived need as a key driver of mental healthcare non-engagement [31, 59]. Moreover, several previous studies report lack of perceived need as a key deterrent to mental healthcare seeking in other LMIC settings, suggesting that this is not an India-specific phenomenon [4–7, 9, 10].

In some cases, women did not see themselves as candidates for mental health treatment because their distress arose primarily from social conditions such as poverty, and they knew a clinician could not amend these causes. This is similar to the findings of some previous work in India, which argues that mental healthcare may lack relevance in contexts where people’s distress arises from basic needs insecurity [31].

A second guiding question of this research was, What do women do to manage distress instead of engaging clinical mental health services?

We found that women described a diverse array of dispositions and techniques to manage distress. These consisted exclusively of self-directed activities they engaged at home or in their communities, not in clinical spaces, likely a result of the stigma associated with care-seeking that we described above. Since this study was cross-sectional, we could not determine whether women’s coping activities actually resulted in better mental health outcomes, but they reported that they found these strategies effective for managing distress.

The two most common coping strategies in this study were social engagement and mindfulness/religious activity. These results align with a large body of work in the medical social sciences demonstrating the importance of social integration for mental health [60] and the value of religious and spiritual resources for coping in resource-constrained settings [61]. These findings also align with the long history of research documenting the role of faith healers who may function in positive ways as front-line healthcare workers for both physical and mental health needs across India, and as a bridge between patients and medical providers when needed [62–64].

During data analysis, the research team noticed that many of the coping strategies women described resemble current evidence-based movements in psychology and talk therapy; this was a post-hoc observation rather than a guiding principle for the identification of coping themes. For instance, women’s discussion of somatic release mirrors recent trends in psychotherapy such as Somatic Experiencing [65] and Tension and Trauma Releasing Exercises [66], trauma therapies that differ from cognitive approaches by encouraging clients to focus on bodily sensations and physical release of tension. These relatively new, “bottom-up” or “body-first” approaches to emotional processing like Somatic Experiencing have positive impacts on symptoms and perceived wellbeing among traumatized and non-traumatized individuals [67].

Likewise, women’s strategy of self-talk and solution-focused thinking mirrors acceptance and commitment therapy (ACT), a form of pragmatic therapy that emphasizes changing one’s self-talk to support positive actions rather than the more traditional cognitive approach that involves changing thought patterns [68]. ACT appears to be as effective for treating depression and anxiety as cognitive therapy, but it functions through different mechanisms [69].

Women’s discussions of changing their mindsets are reminiscent of cognitive reframing, a key skill taught in cognitive-behavioral therapy which encourages people to shift their mindsets to see a situation in a different perspective [70].

Finally, women’s endorsement of social interaction as an effective coping strategy echoes the decades of research demonstrating that social connectedness buffers mental health while social isolation damages it [60]. An approach to stress management focused on social engagement also mirrors the research on interpersonal emotional regulation, which is a key mechanism of social support for people with depression and could be harnessed therapeutically in group sessions [71–73].

We point out these similarities between women’s coping strategies and existing therapies not to imply that women have no need for mental health support, but rather to challenge the common assumption that people’s autochthonous coping strategies are distant from (and therefore perhaps less effective than) clinical approaches. The similarity between women’s copings strategies and therapy best practices implies that women in this cultural context might be receptive to teaching and learning these kinds of therapies in more structured modalities, as long as they did not rely on clinical settings. These results could provide a road map for developing culturally acceptable interventions that incorporate preexisting ways that people manage their distress outside of clinical encounters.

Reimagining interventions that build on women’s existing strategies

The diverse coping strategies women identified in this study support a conclusion that cultural psychologists have long made: that people’s coping strategies should be taken seriously as sources of accessible, low-stigma interventions [43]. Not everyone would necessarily have mastery of all the coping strategies identified in this study, but these could be taught and learned in culturally-acceptable settings by leaders such as community health workers or nurses, who could be physically located outside clinical settings and might work with groups of individuals in a manner similar to task-sharing interventions in India [30]. Given that many of women’s existing strategies resemble existing therapy practices, interventions could use culturally modified versions of existing training manuals or could develop culturally derived materials for use in specific cultural contexts. An approach like this could stand alone, or it could be paired with additional innovative treatment modalities to comprise a wellbeing-focused psychosocial intervention [56]. Given women’s apparent disinterest in interventions focused narrowly on clinical symptoms in this study, more holistic interventions focused on wellbeing and social integration might meet greater success.

Based on our results, any intervention would need to seriously consider working outside clinical settings, possibly in a women’s empowerment organization or mahila sangha, which women were familiar with and endorsed as trustworthy and comfortable settings. Clinical settings provide obvious logistical advantages for researchers, and current best practices even suggest that interventions in LMIC settings should build on existing health infrastructure [74]. Women’s discomfort with clinics in this context, however, suggested that the benefits of convenience and cost-savings from working in clinics might involve serious tradeoffs in terms of women’s willingness to engage. Our results suggest that research and interventions that take place outside clinic settings would have significantly higher uptake than those confined to clinical settings.

Limitations

Our results should be interpreted in the context of the fact that no study participants were suffering from acute mental health crises, even though many did report symptoms of depression. Because of the sampling strategy, Christian, Muslim, and other religious minority women were not represented, nor were youth. Therefore, the results are not generalizable to Indian women in general or to people under the age of 18, who in this context may be especially vulnerable to common mental disorders [75].