Perspectives of Health Service Providers on Barriers to Accessing Perinatal Mental Health Services in Karnataka, India: a Qualitative Study

16 Background: Poor maternal mental health is a major public health concern since it adversely 17 impacts both maternal and child health outcomes. This study aimed to document the barriers to 18 utilizing perinatal mental health services in Karnataka, India, and to determine its relationship 19 with risk factors of poor maternal mental health in this context. 20 Methods: Qualitative research methods using in-depth interviews were conducted on twenty-one 21 local stakeholders who represented health service providers in various capacities: mental health 22 specialists (n = 4), gynaecologists (n = 2), government officials from the Department of Health 23 and Family Welfare (n = 2), and Department of Women and Child Development (n = 2), senior 24 state consultant to United Nations Children’s Fund (n = 1), and frontline workers (n = 10). Data 25 were analysed using a thematic framework analysis approach. 26 Results: We identified multiple barriers to service utilization operating at the levels of the health 27 system, community, family, and the individual. Health-system level barriers included lack of a 28 universal screening mechanism, poor infrastructure, poor training of frontline workers on mental 29 health issues, and inadequate time for counselling and treatment. Community-level barriers 30 included stigma and misconceptions, leading to a lack of social support. Family and individual 31 level barriers included the financial burden of availing services, lack of family and partner 32 support, and lack of empowerment and motivation in the woman to seek services. Family and 33 individual level barriers interacted with risk factors of poor maternal mental health. Based on this 34 evidence and drawing from the literature, we propose a contextualised, stepped-care model for 35 universal screening, detection, referral, and treatment of women with perinatal mental health 36 conditions for Karnataka that is integrated with the reproductive, maternal, and child health 37 (RMNCH) program in primary care settings.


Background
The health and development of a child are intrinsically linked to their mother's physical and mental well-being (1,2).Poor maternal mental health can adversely impact a range of pregnancy, birth, and child outcomes (3)(4)(5) such as higher risks of obstetric complications, preterm labour, fetal growth restrictions, low birth weight of the baby, delayed initiation of breastfeeding, and low quality of responsive caregiving (5).These, in turn, can lead to heightened irritability in the child and reduced ability for emotion regulation (6), as well as impaired cognitive, motor, and social development in the long term (7)(8)(9).12% of women globally, and around 19% in developing countries, experience depression, and anxiety in the perinatal period (10).According to recent studies, the prevalence of prenatal depressive symptoms in Karnataka, the region of focus in this study, is an alarming 36.8% (11), with similar trends being reported for the capital city of Bengaluru (12).Given the current burden and its adverse consequences on both mother and child, prevention, early detection, and treatment are essential first steps towards achieving the Sustainable Development Goals (SDGs) related to improving maternal and child health outcomes (13).
Although the need for maternal mental health services is high, the availability and accessibility of these services severely fall short of the need, especially in low-resource settings (14).Multiple factors are responsible for these lacunae.At the health-system level, important barriers include the lack of trained manpower, poor infrastructure towards providing mental healthcare in primary care settings, unlinked services leading to poor continuity of care, poorly defined referral pathways, and resistance to decentralize services (15,16).Similarly, provider-level barriers such as the lack of training and supervision of medical staff on mental healthcare, a baby-centered approach in perinatal care, and time constraints in providing quality service have been reported (15)(16)(17).Together, these factors have led to a failure to diagnose and treat upwards of 90% of those who need mental healthcare in low-and middle-income countries (LMICS) (18)(19)(20).
On the other hand, demand-side barriers including the lack of awareness about mental illnesses and their potential for treatment, lack of family or partner support to seek care, the complicated logistics of getting and attending an appointment, the financial burden of availing treatment, fear and stigma associated with a diagnosis of a mental health condition, and the disadvantaged position of women in communities, prevent women from seeking services even when they are available (16,21).Low demand for services reduces the prioritization of service provision in primary care settings (22,23), leading to poor resource allocation towards mental healthcare across decades (24,25).
To address these multilevel barriers, a holistic approach is required to meet the end-goal of providing universal, effective, and high-quality mental health services (26).Some countries have taken a first step towards integrating perinatal mental health in primary care settings, providing templates for the operationalization of this goal worldwide (27).To address system and providerlevel barriers, most of these templates follow a stepped-care approach, with mild to moderate cases being counseled or treated by minimally trained non-specialists and only severe cases being referred to specialists (28).
In India, Kerala was among the first states to launch a stepped-care approach to maternal mental health through a program called "Amma Manasu" (mother's mind), which integrated mental health screening and interventions within routine perinatal health programs (29).Similarly, Karnataka's ambitious Manochaithanya ("empowering the mind") program, although not exclusive to perinatal mental health, aims to provide mental healthcare services in primary care settings (30).With 100% coverage in all districts, this "near to home service" has largely bridged the availability and accessibility issues (31).A few additional initiatives specifically designed to promote and protect perinatal mental health is in the pipeline (32,33) making Karnataka one of the frontrunners in designing universal programs for comprehensive care of perinatal mental health across all districts in the state.
While these novel initiatives are being rolled out, it is imperative to take stock of potential barriers that may be specific to this region for the uptake of these services.This forward-looking approach will help ensure that these factors are addressed from the outset and proposed solutions to be integrated within the program's design.To this end, this study aimed to identify potential barriers to the effective utilization of perinatal mental health services in the state of Karnataka, India, from the perspective of a wide range of healthcare providers (mental health professionals, government officials leading maternal and child health programs, gynecologists, and frontline workers such as the Accredited Social Health Activists (ASHA), Anganwadi workers (AWW), and Auxiliary Nurse & Midwives (ANM).Qualitative research methods with a framework analysis approach were used to identify barriers at the health system, community, family, and individual level.Additionally, since family-level factors such as domestic violence, lack of partner support, and financial insecurity are known risk factors for poor maternal mental health (34), the second aim was to document the common causes of poor perinatal mental health to uncover any potential interactions between the identified barriers to service utilization and the causes of poor mental health.Based on this evidence, the third aim was to develop and propose an improved maternal mental health service delivery model in Karnataka's primary care settings.
The two important outputs of this manuscript would be an evidence-based framework that describes the barriers to maternal mental health service utilization in Karnataka, India and its potential interactions with the common causes of poor perinatal mental health; and the development of an integrated model for universal screening, detection, diagnosis and treatment of mental health conditions in perinatal women that is incorporated within antenatal and postnatal care in primary settings.

Setting and Participants
In-depth interviews (IDIs) were conducted from August 2019 to April 2020 in Bangalore, Karnataka, to understand the perspectives of a range of local stakeholders relevant to maternal mental health (Table 1).Participants were recruited through a purposive sampling technique to capture the maximum diversity in views.They were four mental health experts (two practicing psychiatrists, one clinical psychologist involved in diagnosing and treating mental health disorders in women during the perinatal period, and one expert in mental health epidemiology).
We also interviewed two gynecologists involved in antenatal care in public hospitals of Bengaluru, two officers each from the Directorate of Health and Family Welfare Services, Government of Karnataka (GoK) and Department of Women and Child Development, GoK, one senior state consultant to United Nations Children's Fund (UNICEF), and ten frontline workers (four AWWs, four ASHAs, and two ANMs) from Bengaluru, Karnataka.All frontline workers were females with an average age of 40.7 years (range: 24-57 years) and at least secondary education levels.Among the eleven specialists, five were females.The minimum qualification of the specialist group was a master's degree.

Ethical Considerations
Ethical approval for this study was granted by the institutional ethics committee of the Indian Institute of Public Health-Hyderabad, Bengaluru campus (IIPH-B).Participants were approached directly or through email with a request for an in-person or telephonic (out-of-station participants) interview, which was expected to be completed within one hour.The objectives of the study were described to the participants at this stage, and voluntary informed consent, which specifically requested an audio recording of the interview for later transcription, translation, and analysis, was obtained.All interviews were conducted at a time and language convenient to the participant and in a location that maintained the privacy of the conversation.Data from these interviews were strictly used for research purposes only.Confidentiality was ensured by handling audio recordings by key research personnel only, storage in password-protected files, and removing personal identifying information from interview transcripts before sharing data with the larger analysis team.

Data Collection
An initial draft of the semi-structured interview guide was prepared by the research team.Prespecified probes were included to encourage discussion of relevant ideas that may be missed in the initial response.Based on the participant's preferences, interviews were conducted either in English or Kannada by a multilingual, trained research staff.The interview began by collecting basic demographic details, followed by open-ended semi-structured questions to elicit elaborated responses.The topic guide included questions related to stakeholders' professional experiences and contact with women during the perinatal period, causes and impact of perinatal depression on mother, child, and the family, and the availability, access, and utilization of mental health services in public health facilities in Karnataka.The bulk of the interview was geared towards discussing stakeholder perceptions on the barriers to the utilization of available and upcoming services related to maternal mental health in the state of Karnataka.Data was collected and analyzed iteratively until saturation was reached.During this process, the interview guide was modified as necessary for subsequent interviews to clarify questions and add probes based on feedback from previous interviews.

Data Analysis
Data analysis followed a thematic approach using the framework analysis method.This involved the following steps: Transcription: Audio-recorded interviews were transcribed in the same language as the interview.Kannada transcripts were further translated into English and referred to for clarifications as needed.Interviews conducted in English were directly transcribed in English.
Independent quality checks were done by the research team members to ensure error-free transcription and translation by listening back to random segments of the audio recordings and reading the transcripts simultaneously.
Familiarization with the data: Members of the research team thoroughly read and re-read each transcript to familiarize themselves with the whole data set and independently develop an understanding of the initial themes that emerged from participant responses.Developing a working analytical framework: As a first step, the interview guide and familiarity with the data were used to generate an initial list of codes that were mutually agreed upon by the research team.Interviews were coded to extract data on two main themeswhat are the barriers to accessing mental health services in Karnataka, and what are the most common causes of poor maternal mental health in the state.Additional notes were taken of evidence demonstrating an interaction between these two broad themes.This initial list was used to code a small set of transcripts and generate new codes where necessary.This exercise was done independently by each member of the analysis team and then discussed in the presence of the whole team to update the initial code list and finalize the definitions for each included code.This formed the initial working analytical framework.NS, MK, and DJ then collectively coded all the remaining transcripts using the initial framework.New emerging themes and codes were noted and discussed during regular team meetings before including them within the framework.Codes were grouped into themes and sub-themes relevant to the research questions as analysis progressed.
Creation of the final framework matrix: NVIVO 12 software (QSR International Pty Ltd) was used to digitally code, compile and chart data within the final framework developed in the previous step.The outcome was a matrix that cross-tabulated the raw data (as quotes) for each code and the participants.This matrix was used to summarize and interpret the data for each theme.Relationships between themes were identified, which informed the interactions between the barriers to accessing perinatal mental health services in Karnataka, the causes of poor maternal mental health, and its impact on the mother, child, and family.This evidence-based framework, along with evidence from the literature on successful maternal mental health service delivery platforms in other countries, was then used to develop and propose an improved model for maternal mental health service delivery for Karnataka.This improved model aimed to address the common perceived barriers summarized in the framework and integrate universal screening of poor maternal mental health in public health facilities by non-specialists to identify women who could benefit from the above services.

Results
A synthesis of the study's findings is presented as a framework in Fig. 1 and described below.
Please refer to Table 2 for illustrative examples of quotes for each category, theme, and subthemes listed in the framework.
As demonstrated in Fig. 1, common barriers to perinatal mental health service utilization could be categorized into the following three levels: Health System, Community, and the Family and Individual.

Lack of screening services:
There is no mechanism to systematically screen every woman for mental health conditions during the perinatal period, even though this phase marks a period when women proactively make multiple contacts with the health system.Antenatal and postnatal visits are child-centric and largely overlook mental health issues.An unfortunate consequence of the absence of a systematic screening facility is the lack of understanding of the true burden of perinatal mental health problems in the state, as highlighted in Fig. 1.Without an accurate estimate of the burden, the design and implementation of effective perinatal mental health service delivery platforms would be hard to achieve.
In terms of the format for screening, some stakeholders were of the opinion that mental health screening should involve open-ended conversations by trained providers that indirectly probe the mental health status and associated risk factors during the perinatal period, rather than standardized questionnaires that solely require "ticking boxes." Lack of adequate infrastructure: The quality of the health system is not uniform across the state.
Women in rural areas are particularly vulnerable since the availability of even the most basic maternal and child health services are far from optimal.Mental health services take a backseat when more elementary services are hard to access, although they are likely to add to the stresses associated with pregnancy and childbirth.

Lack of sensitization and training of frontline workers:
Although frontline workers are poised to identify signs of poor mental health and their associated risk factors since they make multiple contacts with perinatal women over extended periods of time, they typically overlook or avoid conversations around mental health since they are not sensitized or trained to screen, monitor and refer women with mental health conditions to appropriate facilities.This poses a major barrier towards identifying mothers with mental health issues and placing them on a referral pathway when needed.However, counselling support for mental health is starting to be included as a topic during ASHA trainings (see quote 1.1.4in Table 2).However, it is yet to receive the same level of emphasis as common physical disorders during pregnancy.This lack of training was highlighted by several misconceptions among frontline workers about mental health conditions during pregnancy.For example, many considered anxiety and depression as mental states that could be voluntarily controlled and no need for any counselling or treatment support.
Time constraints due to heavy caseloads: Heavy caseloads of health professionals, especially in remote areas, present a major barrier to dedicating adequate time to assess mental health or to counsel women on common causes and risk factors of depression, stress, or anxiety.Frontline workers also expressed their concerns about managing multiple programs that often went beyond women and child health, which left them with limited time to attend to and solve for every woman who came to seek their help.This is particularly important for mental health issues such as depression.The frontline worker may have to proactively reach out to mothers who refuse to engage and spend adequate time building a rapport and trust before treatment and counselling facilities can be discussed.

Community-level barriers
Stigma is associated with a mental health disorder: Stigma is one of the most significant and complex barriers that prevent women and their families from seeking support when the mother is mentally unwell, to avoid discussions and gossip amongst community members about the family's mental health issues.

Lack of awareness about the need for and potential of treatment:
The level of awareness about mental health issues was perceived to be very poor in the general population, which likely contributed to the stigma surrounding the issue.Mental health experts opined that the general public often wrongly assume that mood swings or feeling sad is part of life, even if experienced for extended periods or those which impact daily functioning.Concerns were raised about family perceptions of perinatal depression and anxiety being considered a consequence of physiological changes brought about by pregnancy and, therefore, not needing any follow-up diagnosis or treatment.This poses a major barrier to seeking support, even if services are available.
Lack of awareness about service availability and access: Even if mental health issues are identified, and the family is willing to access services, there is a general lack of awareness about local availability of mental health services and how to access them.

Lack of social support:
As a result of the stigma, the lack of acknowledgment that mental health problems require treatment, and the lack of awareness about available services, there are often low levels of support and encouragement from community members for women to seek mental health treatment actively.In a patriarchal society, women have limited decision-making abilities and typically need the approval of their husbands and elders before seeking medical attention.In turn, the family's decisions are strongly influenced by community standards, perceptions, and beliefs, making it extremely difficult for a woman to seek treatment when needed.
Myths and misconceptions: Participants highlighted that communities often have grave misunderstandings about how common mental health disorders present themselves and when treatment or counselling should be sought.In many cases, women hesitate or fail to seek treatment since they believe their condition is not severe enough to warrant further attention.
Myths and misconceptions also arise from false belief systems that the community associates itself with.For example, in rural Karnataka, delivering a baby through C-section is looked down upon and attributed to the mother being 'abnormal.'Others misconceptions include the birthing mother being possessed by ghosts if complications were to arise during the intrapartum period.
Additionally, while there is a general acceptance of the notion that maternal mental health impacts the baby, there are no active steps taken by the community to promote mental health, probably due to the misconception that the mother can automatically exert control over her negative emotions for the sake of her baby without any active interventions.

Family and individual level barriers
Family and individual-level factors include the financial burden of availing mental health services, time constraints due to competing interests and commitments, lack of partner and family support, which adversely affects how empowered the woman feels about seeking mental healthcare, as well as her motivation to attend to her own mental and physical health needs.
Financial burden related to availing treatment services: Mental health services are perceived to be expensive and expected to impose a heavy financial burden.The perceived burden includes finances for direct payment for services, cost of travel and accommodation in distant urban locations where services are typically concentrated, and the loss of income arising from being absent from work.Lack of awareness about low-cost services offered by the Manochaitanya program adds to this faulty perception.
Lack of support from the partner or family: In a patriarchal society, women's health is often considered to be less important compared to other members in the family.Therefore, they fear a backlash from family members if they seek support for their mental health.This aspect is further exacerbated in women who experience domestic violence since poor interpersonal relationships further reduce the likelihood of being supported to seek treatment.Also, familial conflicts reduce how empowered women may feel to seek out services on their own.In many cases, when perinatal women perceive their families not to support them seeking mental healthcare, they discourage any help they would have otherwise received from supportive frontline workers for fear of abandonment or aggressive family reactions.

Lack of time due to multiple competing commitments:
During the perinatal period, women face the double burden of household chores and demanding child-care.They often do not have the necessary support at home to take over their duties when they need to dedicate themselves to attending mental health services.This prevents them from utilizing services, even if they are available.

Lack of motivation for self-care:
The participants perceived women to be less motivated about seeking services for their health.The other family members' needs and conflict avoidance within the family were almost always prioritized over their own.

Risk factors for poor maternal mental health
Participants cited several risk factors that contributed to poor maternal mental health in Karnataka.Stressors related to domestic violence, husband's alcohol dependence, and financial insecurity were most commonly cited.The woman's lack of empowerment and social support in the household and her community, particularly her inability to make important decisions for herself or her baby, was also highlighted as being common causes for feelings of anxiety and helplessness that many mothers face in day-to-day lives.Poor economic conditions that necessitated mothers to supplement the family's income while simultaneously taking care of an infant or food insecurity that impacted lactation sufficiency were cited as common risk factors for impoverished families.Another important area of concern relates to a preference for a boy child, especially in cases where previous pregnancies led to a girl child being born.This puts undue pressure and is a cause for anxiety during the entire antenatal period.The birth of a girl child often leads to reduced social support within the family, affecting the mother's mental health and reducing her feelings of empowerment within the family.
In some cases, poor maternal mental health outcomes result from gross misconduct and illtreatment of the mother by health professionals while delivering the child, especially in public health facilities.At a time when institutional delivery is encouraged, the experience in some mothers leaves them scarred.Such harsh behavior further exacerbates their vulnerability.

Interactions between barriers to service utilization and risk factors of poor maternal mental health
Our analysis revealed several factors that pose barriers to utilizing mental health services are also strongly related to the risk factors perceived to contribute to poor maternal mental health.In essence, risk factors of poor mental health increase the barriers to service utilization that operate at the family and the individual level.For example, women who are not empowered to make their own choices about seeking mental healthcare typically also experience domestic violence at home, particularly from in-laws and alcoholic husbands.Alcoholism in the husband was, in turn, perceived to be related to financial insecurity in the household, which in turn adversely affected the family's decision to seek mental health services, which they perceived as expensive.Women who needed to take on additional work to supplement household income were further disadvantaged since it imposed heavily on their time available to seek services.
Women were expected to continue with their regular household activities during the perinatal period, which is more demanding at this time due to pregnancy or child-rearing related exhaustion.This constant exhaustion state further limited the time and motivation required to seek mental health support when required.This was often compounded by unreasonable expectations from family members who did not support their own mental health during their perinatal period.Therefore, they saw no reason why any specialized treatment or counselling would be required in the first place.

Perceived impact of poor maternal mental health on the mother, child, and the family
Majority of the participants were of the opinion that maternal depression is a public health concern and that it adversely affects the mother, child, and family.Specifically, poor maternal mental health was perceived to adversely impact a mother's self-care practices, mother-child bonding, and her relationship with the rest of the family.It was also broadly acknowledged that these adverse impacts would ultimately lead to poorer child development outcomes.Overall, our framework (Fig. 1) suggests that multiple levels of barriers to service utilization decrease access to available services and impedes our ability to understand the true burden of perinatal mental health conditions in the state.Therefore, all participants unanimously voiced their concerns over the poor state of affairs in dealing with poor maternal mental health in Karnataka, specifically -limited mental health services available to them, the lack of awareness about those that existed, and the several barriers that women faced in being able to access those services.
On the flip side, the framework also suggests that a multi-pronged strategy to address these barriers would improve the uptake of services and have several positive downstream effects.
Better utilization of services would reduce the adverse impacts of poor maternal mental health on the mother, child, and family relationships and prioritize integrating perinatal mental health services in primary care settings and increasing demand for more high-quality services.Over time, this would help overcome the dearth of data that impedes the effective implementation, monitoring and evaluation of perinatal mental health programs in the state of Karnataka.

Recommended model for universal identification, referral, and treatment of perinatal mental health conditions in India
Based on this evidence, and drawing from the literature describing successful perinatal mental health programs for low resource settings, we recommend the following model for integrating perinatal mental healthcare in primary care settings in Karnataka (Fig. 2).First, we propose a robust integration of perinatal mental health services with maternal and child health programs, operationalized through screening for common mental health disorders during each antenatal visit, during the postnatal period when the mother brings the child for immunizations, and by ASHA workers during scheduled home visits as part of the Home Based Neonatal Care (HBNC) program (35).Screening could be led by frontline workers such as ASHA's, Anganwadi workers, ANMs, and staff nurses in district and taluk hospitals, Community Health Centers (CHCs), and Primary Health Centers (PHCs).Two different screening approaches could be piloted to test acceptability, feasibility, and efficacyone questionnaire-based.The other using a structured interview formatboth probed into the levels of stress, anxiety, and depressive symptoms experienced by the mother.As outlined in a World Health Organization (WHO) report on maternal mental health (27), these interactions between frontline workers and mothers should be sensitive, respectful, courteous and non-judgmental, to build a relationship of trust and openness.
If the symptoms are mild, the same frontline workers will provide basic counselling and support, which could include active assistance with social and marital problems, connecting women to appropriate agencies that focus on specific areas such as domestic violence or financial aid, and introduce them to peer, support groups, when available.Records would be maintained to ensure follow-up counselling and check-ins during subsequent check-ups.Women with moderate symptoms would be referred to the PHC medical officer for follow-up assessment and referrals as needed.Women who report persistent moderate-to-severe symptoms would be referred directly to the Manochaitanya clinic or the district hospital for evaluation and treatment by a specialist.
A novel initiative named the Digital Nerve Center (DiNC) is currently being piloted in two districts of Karnataka.By harnessing technology, it aims to improve access by providing personalized information about available healthcare services and coordinating visits (32).Screening and referrals from the Digital Nerve Centre (DiNC) would also feed into this pipeline following the same logic model (see Fig. 2).Regular refresher training and supervision of all frontline staff would be undertaken to ensure continued delivery of high-quality services and to provide psychological support to staff when needed.
Finally, as a preventive and promotive measure, and to improve awareness and reduce stigma, topics on maternal mental health and its common determinants would be discussed in existing platforms related to maternal and child health, integrated with themes known to cause stress anxiety perinatal period.For example, discussions related to breastfeeding could also include topics on how lactation failure could potentially cause stress, along with a list of action points to follow in such scenarios.Some examples of relevant platforms that could host these discussions include mother's support group meetings led by ASHA's in the pregnancy and postpartum periods, village health and nutrition days (VHNDs), and village health, nutrition, and sanitation committee (VHNC) meetings.

Discussion
The framework developed in this study (Fig. 1) highlights three key points regarding perinatal mental health service in the state of Karnataka, India, from the perspective of a wide range of local stakeholders who are involved in maternal and child health in various capacities.First, it comprehensively lays out the common barriers to service utilization that operate at multiple levelsthe health system, community, family, and the individual.Second, it highlights the interrelatedness between the above barriers and the risk factors of poor maternal mental health.
Finally, the framework posits that addressing these barriers, particularly those related to the family and individual factors, would have several positive downstream effects (highlighted as blue boxes in the framework).It would help counter the risk factors of poor perinatal mental health and its adverse impacts on the mother, child, and family, create awareness about and demand for high-quality mental health services, and enable building a sizeable database to help quantify the true burden of poor maternal mental health in the state.These points are elaborated below.
Identification of those who need interventions is the first step to efficient utilization of any health service.Therefore, concerted efforts should be directed towards universal and systematic screening of perinatal women for mental health conditions.A systematic review of screening instruments to assess depression during the perinatal period identified seven commonly used tools in low-resource settings (36).Among them, the Edinburgh Postnatal Depression Scale (EPDS) emerged as the most accurate and sensitive across settings.
The institution of a universal screening mechanism needs to go hand-in-hand with training and sensitization of healthcare staff on the use of these instruments.Most health professionals, especially in remote areas, have limited training and time to identify and refer women to appropriate channels for diagnosis and treatment or provide basic mental health counselling in the absence of available specialist services.These barriers are not unique to India since LMICs, in general, are plagued by inadequate screening and referral mechanisms and shortages of trained health staff, which eventually lead to a huge treatment gap for all mental health conditions, including common perinatal mental health disorders such as depression and anxiety (37,38).The dearth of trained workforce and heavy workload of existing specialist staff may be addressed by combining task-sharing and a stepped-care approach.Several countries that are aiming to integrate mental health in primary care have broadly adopted this strategy with local adaptations (27) such as the National Prenatal Depression Initiative (NPDI) in Australia (39), the Perinatal Mental Health Project in South Africa (37) as well as smaller scale district-level pilot projects in South America, Middle-east and Africa (27).Task-sharing entails training and supporting frontline workers on basic screening and counselling techniques and establishing a clear responsibility and referral pathway for severe and complex cases.This serves the dual purpose of building capacity for mental health treatment at different levels within the health system while freeing up specialist time to focus exclusively on difficult cases (37,40,41).
The dedicated perinatal mental health program poised to be launched by the Karnataka government (Mathruchaitanya) will follow a task-sharing, stepped-care approach, thereby promising to address the health system level barriers described above.However, the additional burden of implementing these new mental health programs on already overburdened frontline workers and primary physicians (42) will further impinge on their time and availability.A potential solution may be the continued incentivizing of private partners to provide mental healthcare services in public settings as initiated through the public-private partnership model in the Manochaitanya program (30), or the training of community volunteers who could take on some of the responsibilities of frontline health workers such as identifying, referring and monitoring individuals with mental health issues in the community.There is a large body of evidence demonstrating the feasibility and efficacy of providing quality mental health services for common mental health disorders by non-specialist community health workers and volunteers.
An excellent example is the Thinking Healthy Program and its various local adaptations that were adopted by the WHO as an evidence-based, cost-effective, non-specialist led, communitybased intervention for perinatal mental health (43)(44)(45)(46).
While it is possible to address health system-level barriers through strategic local adaptations of the stepped-care model, the specific barriers that operate at the level of the community, family, and the individual will need more contextual and innovative solutions since biases against mental health disorders are long-standing and deep-seated.Stigma, often arising from significant gaps in knowledge about mental health conditions in the community and extant misconceptions regarding how they present, emerged as an important community-level barrier for women to access services.Stigma also resulted in community and family members failing to provide the necessary social support desperately needed by perinatal women to seek services.Recent systematic reviews list stigma as an important barrier to accessing services even in developed nations like the UK (26,47), and a meta-analytic review of 56 articles (48) reported a moderate negative association (-0.27) between stigma and help-seeking behavior, highlighting it as one of the key deterrents to service utilization that needs addressing across the board.
One way to counter barriers related to stigma is by increasing awareness about mental health conditions in the community, the potential for their cost-effective treatment, training community leaders on communication strategies that build sympathetic narratives about mental health conditions, and avoiding misconceptions about their presentation (49).Additionally, increased awareness about the risk factors for poor maternal mental health such as domestic violence and stressors related to voicing a preference for a boy child may further help the family to avoid stressful behaviors that are known to impact the mother's mental health adversely, thereby acting as a preventive measure.The Indian government's schemes that support the girl child such as the "Beti bachao, beti padhao" (Save the girl child, educate the girl child) and "Sukanya Samriddhi Yojana" (Girl child prosperity scheme) will further bolster such efforts.
Several experts thought that increased visibility of the Manochaitanya program, as is evident from the dramatic increase in its uptake in the past six years (31), would greatly help generate the necessary awareness about mental health disorders in the community.Continued sensitization and capacity building of frontline workers on mental health disorders would build credibility for the program within the community as they get exposed to more knowledge about different mental health conditions and their determinants on the one hand, and the support they receive for affordable and accessible specialist healthcare on the other.Finally, the provision of universal screening for maternal mental health with well-planned and efficient care pathways may go a long way to counter stigma-related barriers to accessing services.Together, these may lead to shifts in community perspectives about mental health and increase mental health-seeking behaviours.These measures will also help address some of the family and individual level barriers, including misconceptions regarding mental health treatment being prohibitively expensive and time-intensive, and empower women to voice their need for mental health treatment in a more accepting society.
An interesting outcome of our analysis was the interaction between factors that were perceived as barriers to service utilization and the causes of poor maternal mental health.Based on this observed interaction, we propose that solving for these common underlying factors may reap the dual benefit of improving service utilization and reduce the risk of poor mental health.There is evidence to show that improving mental health literacy at the community and family levels improves mental health outcomes, partly by promoting help-seeking behavior (50).Specifically, in the Karnataka context, women are expected to continue with usual household duties through the perinatal period with little support from the family.They may also experience domestic violence and have poor decision-making abilities in the family.This lack of social support and increased stress are both risk factors for poor mental health and barriers to seeking treatment.Therefore, solving for these common underlying factors through increased awareness about the negative consequences of maternal stress on child health and development may simultaneously spur improved attitudes towards providing social support to seek mental healthcare when needed, as well as provide increased support to help reduce maternal stress in the immediate postpartum period.Similarly, interventions to address physical, verbal, or emotional abuse, lack of partner or family support, and financial insecurity may have spin-off effects on both fronts.
Based on our findings and available evidence in the literature, we have proposed a model that combines task-sharing and a stepped-care approach for perinatal mental health in Karnataka, which can unify the several mental health programs that have been recently been launched or are in the pipeline.We believe that this unified model holds promise since similar models have successfully been implemented at varying levels of scale in other low-resource settings such as South Africa, Uganda, and Kerala, India (29,37,38,51,52).An important component is integrating perinatal mental health delivery in existing maternal and child health programs and largely led by trained non-specialist frontline workers.Successful implementation of this model may lead to increased uptake of mental health services, thereby providing the necessary demanddriven impetus to strengthen the Indian mental health infrastructure further, ensure increased accessibility, a broader range of available services, and the appointment of dedicated care providers for the provision of mental health services in the future.

Limitations of the study
First, certain barriers to mental health service utilization may have been overlooked, especially those specific to the most remote areas in Karnataka, since this manuscript documents the perceptions and opinions of participants largely based in urban areas.Second, data was only collected through in-depth interviews, which does not allow for the emergence of group consensus on topics of interest as is possible through focused-group discussions.Finally, a major limitation of the proposed model is the addition of yet another program to an already overburdened health system.However, the proposed solutions and the socio-economic benefits it will potentially accrue, would likely encourage the system to sustain and further strengthen mental healthcare delivery in primary care settings.

Conclusion
We identified multiple barriers to the effective utilization of mental health services in Karnataka, ranging across the health system, community, family, and individual levels.Based on our results and drawing examples from the literature, we have proposed a stepped-care model that integrates all the existing relevant programs in Karnataka into a universal perinatal mental health program to realize the goal of making mental health services available and accessible to all women during this critical period of their lives.In turn, this may reduce the risk of adverse neonatal, infant, and child health and development outcomes and have additional spillover benefits of increased awareness about mental health conditions in the community, reduced stigma, and eventually lead to improved healthcare-seeking behavior in the community.Needless to say, although contextualized for Karnataka, India, similar models would also be applicable to the rest of the country and other low-resource settings across the globe.However, important limitations related to heavy caseloads in remote locations with limited specialist support must be seriously considered during the iterative refinement of these programs in the future since time constraints of the existing overburdened healthcare staff would overturn any attempts to implement any of these programs effectively.Therefore, relevant stakeholders must continue to track implementation challenges across different contexts and keep innovating to address common barriers to help transform these ambitious but fledgling programs into effective well-oiled machinery.Once empirically tested and refined, components of the optimized mental health service delivery model can inform policy formulation in India and other LMICs."They feel awkward about everyone starting to discuss the issue, so they behave like all is good in their life when they are in a group.They do not come forward due to the fear of society.People fear society.""Whenever a female child is born the mother receives a sympathetic look.A woman is never given the place she deserves when she achieves something; we aren't ready to accept it."1.2.2Lack of awareness about the need for and potential of treatment "We don't speak about mental health.That is the main barrier.""Women ignore the problem and continue to stay in the misconception that some things are common in such situations.""We refer women to the psychiatrist, but they will not be ready to go anywhere.We tell her to go to the office meant for domestic violence, but she refuses to go and says this is the story of every household, and she always hesitates to move forward."1.2.3 Lack of awareness about service availability "We are blessed in Karnataka to have psychiatrists and psychologists in all districts in Karnataka, along with teams in district hospitals and medical colleges.""We have counselling facilities, the government has provided many, but we don't know how much people have understood it."1.2.4 Lack of social support "Just after six or seven months, they have to go to work and take care of kids.These are some physical challenges.Those below the poverty line and people who work at the grass root level face more of these issues as they have to handle many things at the Sr No.

Category
Theme Sub-theme Illustrative quote complaining and say that if we spend our time here, then who will cook for our husbands?Or if we sit here, our in-laws will scold us."1.3.3Lack of time due to other competing commitments "On some days they visit <the hospital> after the consultation time is over and say we were busy with our household chores so could not come earlier.They have to finish their routine work.These lags are because the family members do not allow the woman to visit hospitals.""Some women don't express <their need for help> even if they have some problem; they think that they will end up quarrelling with the husband in the presence of kids, so they quietly tolerate and go on.""Some women don't care and move on, thinking that it is their destiny." 2

Risk factors of poor maternal mental health
-Preference for a boy child -Lack of empowerment "Some people say they have a feeling of fear before delivery.When probed, they talk about the pressure from the family regarding having a baby boy and that she will not be allowed to go back home if she happens to deliver a girl.They even complain about the family not taking proper care, not allowing them to go outdoors, not even to their parental house."-Domestic violence -Lack of partner and family support -Financial hardships "Some husbands drink and don't bother at all about their wives.They never bring them to the hospital at any point of time.Never bring a single thing home.Mothers-in-law complain that she doesn't work and only breastfeeds.These women don't have enough food to eat and continuously worry that this will lead to low milk production.She has to spend sleepless nights, and on top of that they blame for her for the girl child."Framework to describe the various levels at which barriers to maternal mental health service utilization operate, their interaction with perceived causes of poor perinatal mental health, and its impact on the mother, child, and the family Figure 2 Recommended model to integrate perinatal mental health in primary care settings in Karnataka

Figure 1 :
Figure 1: Framework to describe the various levels at which barriers to maternal mental health service utilization operate, their

Figure 2 :
Figure 2: Recommended model to integrate perinatal mental health in primary care settings in Karnataka

Figures
Figures

Table 1 :
Types and number of health service providers interviewed for this study

"
Many times, the doctors are not available at the PHCs (primary health centers).There are facilities, but what about accessibility?We see women deliver at home because of the non-availability of ambulance services on time.When the condition is so, how can we expect support for mental health?"We have the satellite-based training for ASHA workers … around four thousand ASHA workers are trained every day.We add these mental health training modules in that.It may not occupy the same time and place as preventing hypertensive disorders or the correction of anaemia.However, we have initiated this to introduce the concept of mental disorders during pregnancy and the necessity to address"

"
The biggest challenge is that the gynaecologist has a huge number to cover.If he/she asks one question, how are you? the <perinatal> woman will keep talking.My husband is a problem, I have financial problems, my child has this, then I am worried whether I will have a baby boy or not.Here time is the constraint.I think that is the biggest barrier.Otherwise, every gynaecologist, every paediatrician, every mental It is difficult to help them <perinatal women> if they don't show any interest."