Implementation of first episode psychosis intervention in India - A case study in a low-and middle-income country.

First Episode Psychosis (FEP) is a serious mental illness affecting adolescents and young persons. While many effective interventions are available, there has not been much research to understand the implementation of such interventions in India and other low- and middle-income countries (LMIC). We studied the implementation of an FEP intervention program in a specialist mental health facility in Chennai, India, using a well-established framework for doing so, the Consolidated Framework for Implementation Research (CFIR). We conducted 27 in-depth interviews with the service users (15 persons with FEP and 12 family caregivers of persons with FEP). We also conducted a focus group discussion with 8 service providers and in-depth interviews with 7 other service providers including those in the service management. A thematic analysis approach was used to identify emerging themes. First, we found CFIR effectively accommodated implementation challenges evident in LMICs; that is, it is transferable to LMIC settings. Second, we highlight barriers to implementation that include cost, limited human resources, cultural and professional hierarchy, divergence from evidence-based guidelines, and lack of awareness and stigma in the wider community. Third, we highlight facilitators for implementation such as, leadership engagement, the need for change that was recognized within the service, cosmopolitan perspectives derived from clinicians' local and international collaborative experiences and expertise, compatibility of the intervention with the existing systems within the organization, accommodating the needs of the service users, and rapport developed by the service with the service users. Fourth, we propose a model of service delivery incorporating a task-sharing approach for first episode psychosis in resource restricted settings based on the feedback from the stakeholders.

Background

First Episode Psychosis (FEP) affects adolescents and youth with an estimated incidence of about 34 per 100,000 person-years (Kirkbrideet al., 2017), having a significant impact on their functional ability and productivity (Austinet al., 2015; O'Keeffeet al., 2019; Raghavan et al., 2017). It is also associated with a higher risk for suicide (Hawton et al., 2005). It has a significant impact on family caregivers (Kumar et al., 2019). Providing interventions at the earliest is shown to improve outcomes (Ammingeret al., 2011) and the duration of untreated illness can adversely affect outcomes (Penttilä et al., 2014). While effective interventions for First Episode Psychosis (FEP) are available (Abidiet al., 2017), and are being used in India and other low- and middle-income countries (LMIC), they are available only in small pockets in these resource-restricted settings (Iyer et al., 2010; Raghavan et al., 2019; Farooq, 2013, Rangaswamy et al., 2012). A recent situational analysis of FEP service delivery in a large city in India showed that specialist multi-disciplinary intervention was available in only one setting (Dhandapaniet al., 2020). In the face of challenges and needs, the World Psychiatric Association has made the early intervention in FEP in LMIC a global strategic priority (Singh & Javed, 2020).

Interventions for mental disorders developed in high income settings may be as effective in LMIC settings (Patel & Thornicroft, 2009). However, many interventions transferred from HIC to LMIC settings fail to be implemented successfully due to factors such as lack of human resources, failure to build capacity, inadequate information systems, poor health financing, and fractured service delivery (Saracenoet al., 2007). The Programme for Improving Mental Health Care (PRIME) study from India reports challenges such as poor support from the service providers, lack of availability of medicines, low importance given to mental disorders by the policymakers, administrative issues, and lack of focused data on mental disorders (Shidhaye, 2015). Similar issues have also been reported in the implementation of national mental health policy in South Africa (Draperet al., 2009). In short, there are significant barriers to the implementation of evidence-based interventions in mental health in LMICs. Our study seeks to provide insight into the implementation challenges through the application of a well-regarded implementation science framework, the Consolidated Framework for Implementation Research (CFIR).

Theoretical framework

Implementation science represents “the scientific study of methods to promote the systematic uptake of research findings and other evidence-based practices into routine practice, and in so doing, supports improvement in the quality and effectiveness of health services and care” (Eccles & Mittman, 2006). Many theories and frameworks are available to guide the implementation of health care interventions (Tabak et al., 2012; Nilsen, 2015; Reed et al., 2019). The Consolidated Framework for Implementation Research (CFIR) is a popular framework due to its comprehensiveness and flexibility (Damschroder et al., 2009). It has been meaningfully applied across a wide range of health intervention implementation research across settings (Kirk et al., 2016; Stoner et al., 2020). Though most application has been in the high-income settings, a recent review included use of the CFIR in LMIC settings (Meanset al., 2020), around which we extend insight, as discussed following our empirical analysis. The CFIR was developed as a meta-theoretical framework including different pre-existing theories in implementation science literature and provides a consistent taxonomy of contextual determinants (barriers and facilitators) of implementation processes (Damschroder et al., 2009). It considers five domains related to implementation, within each of which lie several constructs: intervention characteristics (e.g. its core non-adaptable and periphery adaptable components, evidence strength and quality as viewed by stakeholders); outer setting (e.g. economic, social and political context, patients’ needs and resources); inner setting (e.g. organizational reward system, climate, culture, leadership, networks, knowledge management); characteristics of individuals involved (e.g. professional and individual mindsets about evidence and receptiveness to change); the process of implementation (e.g. processes of implementation planning, engagement, reflection, and evaluation).

We aimed to study the implementation of an FEP intervention program in a specialist mental health facility in India. Our objectives were to understand the experiences of service users and service providers of this FEP intervention service. We also planned to highlight the challenges faced in the implementation and how socio-cultural variables in the Indian context contributed to these. We used CFIR as a framework to structure our findings, but also to reflect upon the applicability of CFIR to LMIC settings, such as India. Our findings can potentially help to implement the FEP interventions across other settings, to scale up, and to sustain over long periods.

Methods

Setting

Schizophrenia Research Foundation (SCARF) is a not-for-profit non-governmental organization based in Chennai, India. It is a specialist mental health facility offering out-patient, in-patient, day centre and rehabilitation services for persons with mental illnesses including psychosis. The FEP service was initiated in the service in 2003 in collaboration with the Prevention and Early Intervention Program for Psychosis (PEPP) in Montreal, Canada (Anonymous, 2012). The PEPP model from Montreal, Canada, consisted of a 2-year integrated programme with pharmacological and psychosocial interventions overseen by a case manager. Patients with FEP met the case managers weekly and a psychiatrist biweekly until they reached clinical stability (Iyer et al., 2010), (Compton et al., 2008). This formed the basis of the specialist FEP service in Chennai, but as our empirical study reveals PEPP was adapted as it was adopted to fit with context, and we consider the implementation challenges and solutions in the process of adaption.

In (Name of the organization is anonymized for double-blind reviewing), the specialist FEP service is delivered by a dedicated team of psychiatrists and case managers. Psychiatrists confirm the diagnosis and provide medical interventions while the case managers coordinate care for individual clients under their care and offer non-pharmacological interventions including psychoeducation, supportive counselling, crisis management, stress management, caregivers support, medication adherence, relapse prevention, and monitoring for the side effects of medicines. The psychosocial interventions delivered were adapted to suit the Indian cultural context and resource constraints (Anonymous, 2020), which we discuss in our results. Simultaneously, the provider also caters to those with FEP in a generic manner as some persons with FEP are not referred to the specialist FEP service but are managed by the generic (non-FEP specialist) consultant psychiatrists. The pathway for FEP service users is described in Fig. 1 below.

Fig. 1

Existing FEP service user pathway.

Participants

Service users with FEP, their primary caregivers (family member living with the person with FEP), and service providers from the FEP service in Schizophrenia Research Foundation (SCARF), Chennai, India were invited to participate.

Service users

Of the 115 persons with FEP registered with the service, 25 individuals with FEP and their caregivers, selected using a purposive sampling approach to include people from different socio-economic backgrounds, gender, duration of illness, duration of treatment and engagement with the service, were approached initially for participation. Household monthly income was used to determine the socio-economic status of participants. Five persons with FEP and their caregivers did not wish to participate due to various reasons including lack of time (3 persons), being far away from the hospital (1 person) and no specific reason (1 person). Fifteen persons with FEP and 12 caregivers participated in a total of 27 in-depth interviews when further data collection from the service users was terminated as data saturation was achieved (Guest et al., 2006). The mean age of the service users was 25.6 years (SD 7.5) and the age ranged from 18 years to 43 years.

The inclusion criteria for service users and their caregivers were:

The person with FEP must have a DSM-IV-TR (American Psychiatric Association, 2000) diagnosis of either schizophrenia spectrum psychotic disorder or affective psychosis, irrespective of the time since the onset of psychosis, and not having received antipsychotic medication for more than 30 days since the onset of psychosis.

The person with FEP and their family member had visited the service at least once.

The family member must be staying with the person with FEP.

The family member must be aged 18 and above.

Service providers

All the service providers, including the service managers, involved in the FEP service, participated in either the focus group (FGD) or the in-depth interview (IDI). Eight service providers consisting of 2 consultant psychiatrists, 5 case managers (1 psychologist and 4 social workers) delivering specialist FEP service, and 1 service manager participated in the FGD. In addition, 5 generic psychiatrists in the hospital who also managed persons with FEP participated in in-depth interviews. Two senior managers in the hospital who oversaw the FEP service also participated in in-depth interviews. Details of participants are provided in Table 1 below.

Table 1

Number & detail of participants.

S No	Participant	In-depth Interview	Focus Group Discussion	Total
1	Service user – persons with FEP	15		15
2	Service user – caregivers of persons with FEP	12		12
3	Service providers – delivering specialist FEP service		8 (2 consultant psychiatrists, 4 social workers, 1 psychologist, 1 service manager)	8
4	Service providers – Consultants psychiatrists delivering generic service	5		5
5	Service managers	2		2
Total participants				42

Ethics

The Institutional Ethics Committee of Schizophrenia Research Foundation (SCARF) approved this study. The participants were required to provide voluntary informed consent.

Procedure

The service users and their caregivers were selected from the clinic register. They were contacted by the researcher by telephone to arrange an appointment at a mutually convenient time to participate in in-depth interviews. The service providers were invited to participate in a focus group discussion. Those who could not join the focus group were invited to participate in in-depth interviews.

The focus group and the in-depth interviews were conducted in Tamil and English. Topic guides (available on request) to explore the experiences of all participants was used. For the service users, this also included questions about their expectations and understanding of their condition and treatment. For the service providers and the service managers, the topic guides used for the in-depth interviews and the focus group discussions were developed based on the domains and constructs of the CFIR that were relevant to our research. The authors discussed each domain and construct to develop appropriate questions for the topic guides. The topic guides explored the nature of intervention being provided, the challenges and facilitators to implementation, sustenance, and scaling up. All Tamil interviews were translated and transcribed in English. The context and meaning of the responses were carefully retained during the translation. The two researchers who conducted the interviews and analysed the data were not part of the FEP service and were not involved in the clinical management of the service users. They were also not directly supervising or line managing the service providers who participated in the study. Two researchers independently coded the transcripts into themes manually using the phases of analysis recommended by Braun and Clarke (Braun & Clarke, 2006). The CFIR was used to guide the organization of the emergent themes. For clinicians and managers from service providers, CFIR framed interview questions in a deductive way. Meanwhile, for service users, questions were more open, with CFIR applied to organize themes in a more inductive way. The two authors read the transcripts completely to familiarize themselves with the data to achieve full immersion. While the inter-coder reliability number was not calculated, the codes derived by the two authors were largely comparable. The authors discussed the codes and themes, a few discrepancies were examined carefully, and the authors arrived at a consensus.

Results

Facilitators and barriers to the implementation of FEP intervention

The themes from the focus group and in-depth interviews are described below. The facilitators and barriers identified are listed below under the different CFIR domains in Table 2, Table 3 respectively.

Table 2

Implementation facilitators.

CFIR domain	Facilitator	Selected Quotes
Intervention characteristics	Intervention source	“When we wanted to start providing interventions for our FEP clients we chose the one from Montreal, Canada. We knew we could trust the program and McGill University is quite reputed.” (Service manager 3)
	Adaptability; adapted to meet the local needs of service users	“This package for FEP is adapted through the experiences that we have gained over the period of years. What's good about this is that it is fit for purpose here.” (Service provider 2)
	Cost to the service user	“Everything is provided free to us and that's why we can get this treatment. We are very thankful.” (Caregiver of P9)
Outer setting	Families facilitating adherence to treatment	“I don't think I should take medicines or come to the hospital. But my parents insist, and they say it is helping me. So, I come. Left to myself, I will not take the medicines. Nothing is wrong with me.” (Person with FEP P11)
	Cosmopolitanism (local and international collaboration experience)	“A few members of the team had been to visit NIMHANS (National Institute of Mental Health and Neurosciences, Bengaluru, India), Canada, Singapore and the UK over the years for training. This was possible because of the links we have made and the relationships we have developed with other teams.” (Service manager 2)
Inner setting	Need for change	“We knew that FEP in younger people should be viewed differently from those with chronic psychosis. There was a need to identify resources. The clinicians were involved in dealing with both the FEP and chronic psychosis clients.” (Service manager 2)
	Compatibility of the FEP intervention with the existing system	“The skills that we have learnt as case managers in FEP can be used for clients with other diagnoses also.” (Service provider – Case manager 1)
	Leadership engagement	“The Director and all senior people have been very enthusiastic about what we do. They have provided us with guidance and support all along.” (Service provider – Case manager 2)
Characteristics of individuals	Rapport with the young service users	“They (case managers) were all young and they could understand our issues. I felt comfortable speaking to them. I could share my concerns with them that I could not speak to others about.” (Person with FEP P3)

Table 3

Implementation barriers.

CFIR domain	Barrier	Selected Quotes
Intervention characteristics	Lack of standardized protocols	“Different consultant psychiatrists have different preferences for choosing medicines. Lack of an SOP (standard operating procedure) is a problem. Continuing Medical Education (CME) programs can also help to discuss guidelines about prescribing for young people.” (Service manager 1)
	Cost to the service provider	“Right now, we have been paying staff from the research budget. It costs to pay for all the staff including the case managers. When the funding runs out it is going to be a challenge. We need to explore alternative sources of funding.” (Service manager 2)
	Relative advantage of FEP intervention over the generic service provision	“I have a busy clinic. I sometimes forget to refer them. Even when I ask the service users to go and find this person in the FEP service, they just come back to me next time saying that they do not want to see anyone else.” (Consultant A)
Outer setting	Perceived importance of medical intervention over non-pharmacological interventions	“Was expecting that she will be given medicines and that she will sleep better. I did not know they can also help with improving her self-confidence.” (Caregiver of P6)
	Lack of awareness about FEP & treatment	“My friend who knew someone with mental illness referred me to come here. I did not know anything about this. I went to the local temple and had been asked to complete some rituals.” (Caregiver of P4)
	Stigma	“I don't want my friends and teachers to know I have been coming to this hospital and taking medicines. They will tease me.” (Person with FEP P3)
Inner setting	Poor network & communication within organization	“Everyone in the organization should buy-in to the FEP model and agree to refer. That's the challenge. Regular updates and continuing medical education (CME) programs can help.” (Service manager 1)
	Hierarchy in the organization limiting the role of managers	“I feel I cannot make independent decisions about non-pharmacological treatment for service users without running them past the consultant. They may think that I crossed a line if I did that.” (Service provider – case manager 3)
	Role of CM not being recognized by the service users	“It is true that most patients/families are not familiar with the term ‘case manager’. Most often case managers are viewed as the doctor's ​assistant who collects the detailed report and accompanies them to meet the consultant to summarise the progress. Case managers understanding these cultural dynamics, usually take the doctors' help to emphasize the need for therapy, etc.” (Service provider – Case manager 2)
	Difficulty in reporting all issues for the service users with the consultants	“The doctor gives me tablets that helps me to sleep but I still hear voices. She (the Psychiatrist) is terribly busy, and I only get a few minutes to speak to her. She is very experienced and will know that I still have problems. I did not want to tell her that I am not better as she may feel offended. But I spoke about this to the sister (case manager) and she spoke to the doctor on my behalf. She also explained that it may take some time for my problems to get better.” (Person with FEP P1)
Characteristics of individuals	Lack of formal training in FEP for staff	“Many case managers are fresh graduates with little clinical experience. Some of them may have worked with persons with chronic mental illnesses. Also, as psychiatric training in India does not have dedicated FEP training, many consultants again come with experience only with chronic psychosis. We need to provide dedicated FEP training for students and future work force.” (Service manager 3)
Implementation process	Lack of dedicated funding to support sustenance and scaling up	“We don't have separate funds for running an FEP service. Some of the existing staff working have a permanent contract of employment so they can continue to provide the service after the research funding stops. We will need funds to run the program.” (Service manager 1)

As can be seen in the above tables, the CFIR has been used to systematically explore the barriers and facilitators to the implementation of FEP interventions in a service in the LMIC context. Barriers such as lack of resources to provide services beyond the duration of the research study, dearth of capacity building programs to facilitate sustenance and scaling up, inherent hierarchy within the organizational culture limiting the work of non-medical work force, lack of awareness about FEP and the interventions among service users, the expectation of service users to only consult the doctors, and the service user preference for pharmacological interventions are more prominent in this LMIC setting.

On the other hand, specific facilitators that appear to operate in the setting are the need for change within the organization to implement a special intervention for FEP, the cosmopolitanism of the organization as reflected in clinicians’ local and international collaboration experience, the reputation of the intervention source, the adaptability and the compatibility of the intervention within the organization, the leadership engagement, the rapport the service providers had with the young service users, the adaptation of the intervention to the service user needs, and the availability of the family members to facilitate adherence to treatment.

Needs of service users

The primary need for the persons with FEP was recovery from symptoms such as poor sleep and hearing voices.

“I did not want to hear those terrible voices threatening me. I could not sleep at all.” (Person with FEP P8)

Some service users and many relatives wanted their family members with FEP to return to studies or work at the earliest.

“I could not work for the past few months and my family is struggling. I need to get back to work as soon as possible.” (Person with FEP P4)

“I just wanted my son to be well again to get back to college. He had already lost about a year.” (Caregiver of P1)

Many service users wanted appointments on weekends and shorter wait time in the clinic so that they do not miss their studies and work.

“As I need to go to college, appointments on Saturdays are very helpful.” (Person with FEP P1)

“Waiting in the clinic to see the doctor takes a lot of time. I miss my day at work due to this.” (P7)

The service users did not want to be followed-up for longer periods of time. They wanted a quick resolution of their problems and when they were better, they did not want to come to the clinic. They were, however, willing to be contacted by telephone and mentioned that they would come back if they needed in the future.

“I am better now. I am going to college regularly. I don't want to keep coming to the clinic.” (P8)

“Now that I am well, I don't need to come to the clinic. I have the phone numbers for the counsellors (case managers), and they can speak to me, if necessary.” (P5)

The family members too did not want longer follow-up periods. They were willing to watch out for signs of relapse and bring the person with FEP to the clinic if necessary.

“Coming to hospital takes a lot of time. He is better. She (case manager) told us about what to look out for if he were to become unwell again. We can bring him back if necessary, in the future.” (Caregiver of P8)

Scaling up

The service providers, given the resource restrictions and challenges with capacity building, were concerned about potential scaling up of FEP services in India.

“Given lack of specific training for staff in FEP, this will be an issue for future work force if we need to expand the service.” (Service manager 3)

“Now we have staff who can screen intensely for persons with FEP in the outpatient department and ask other consultants to refer patients. Also, we have dedicated staff who can proactively call service users to ensure adherence to treatment and follow-up. Eventually, we will need to stop doing some of these activities.” (Service provider – Case manager 1)

They mentioned during the FGD and in-depth interviews that while specialist FEP services are needed and important, having such resource intensive FEP services across all settings will be exceedingly difficult in a resource restricted setting like India. The service providers suggested that where feasible a specialist FEP service (as shown in Fig. 1) should exist. However, when such resources are scant, they propose a model that integrated specialist FEP service to a generic service (see Fig. 2) that in their opinion was more sustainable in a resource restricted setting like India.

Fig. 2

FEP Service integrated in a generic service user pathway.

“Unless we have specialist FEP services, we will not be able to do high quality work and address the special needs of those with FEP. But resource is an issue. We need to adapt and innovate.” (Service manager 1)

In this model, each generic service could have a smaller number of specialist FEP case managers trained in the FEP interventions. The consultant psychiatrist decides whether someone with FEP should be referred for psychosocial management by a generic case manager. There are more generic case managers in a service compared with specialist FEP case manager. The generic case managers provide and coordinate non-pharmacological interventions and offer active follow-up by telemedicine while being trained and supervised by the specialist FEP case manager. The specialist FEP case manager in addition to training and supervising generic case managers also deals with persons with FEP with special needs that cannot be provided by a generic case manager.

Discussion

We explored the experiences of the service users and the service providers to identify the barriers and facilitators in implementing an FEP intervention in a specialist mental health service in India. Implementation studies tend to privilege the views of those providing, rather than in receipt of, the service. Our study seeks to provide a balance to our understanding of implementation, particularly important given the cultural and socio-economic dimensions of the setting, which service users are well-positioned to comment upon regarding their implications for implementation. We highlight that the CFIR model is deficient in this respect, and might accommodate how the demands of service users influence implementation of clinical interventions more explicitly as one its dimensions, as discussed further below. We do not suggest that service user demands, such as rapid control of symptoms to aid return to work or study can be easily met, and agency of service users to meet their demands is likely limited in the face of resource constraints in LMICs.

Our analysis of the implementation of FEP interventions in India, applying a widely accepted implementation research framework, the CFIR, reveals a range of issues that appear more pronounced in this setting. Regarding the applicability of this framework to an LMIC context such as India, given that CFIR has been derived from high-income nation contexts, our study highlights CFIR as a useful framework to delineate implementation challenges. However, that the economic and cultural specificities of India, emphasize some of the implementation challenges as more pronounced. In essence, we do not need to re-invent an implementation science framework, but we need to take a more nuanced view of its application, as discussed below. The utility of CFIR in LMIC settings with some adaptation has also been demonstrated in a recent review (Meanset al., 2020) and this confirms our assertion. We do however note that we did not interrogate the valence or strength of each CFIR theme in our study, further research may seek to do so in LMIC settings.

Various factors such as the tension for change within the organization to provide specialist interventions for FEP, the cosmopolitanism resulting in a relationship with the McGill University in Montreal, and the reputation of the intervention source appear to have played a role in the selection of intervention model for FEP in Schizophrenia Research Foundation (SCARF), Chennai. Also, the adaptability and the compatibility of the intervention with the existing systems, leadership engagement, and the ability of the service providers to establish a good rapport with the young service users all appear to have influenced the continued use of the intervention. These factors appear to influence the choice and sustained use of interventions in other high-income settings as well (Greenhalgh et al., 2004), (Innis et al., 2015).

As interventions move from one setting to another they are adapted as they are adopted. In our study, we see that the FEP intervention being implemented locally is in a different form than its original version from Montreal, Canada. The adaptation process did not use a framework such as the Cultural Treatment Adaptation Framework (Chu & Leino, 2017) or follow a structured process (Movsisyanet al., 2019) but was made over time, based on the experiential knowledge of the service providers and the need of the service users. However, despite this limitation of lack of use of a formal framework, a recent study comparing FEP intervention in Montreal, where the intervention was originally developed, and Chennai, where it is adapted and delivered, showed comparable for some, and even better for other, outcome measures in Chennai (Anonymous, 2020). The adaptation processes used in this setting may need a more careful exploration as this may provide clues to the success of culture and setting specific adaptations in LMIC settings.

The main driver for adaptation has been the needs of local service users. The service users in our study needed rapid control of symptoms, early return to work or studies, convenient appointment times, shorter follow-up periods and use of telephone calls for follow-up. Any planned intervention for FEP in LMIC settings should take these identified needs into consideration. The stress upon early return to work by FEP service users and their families is possibly due to the economic impact of mental illness. The cost of treatment for a person with mental illness in India is mostly met by the families as out-of-pocket expenditure and is perceived to be a significant burden (Gururaj et al., 2016). A recent study from India highlights this issue (Chavan et al., 2018), and even though it did not focus on FEP, there is no reason to believe that the situation for FEP will be any different to other mental illness. In our study we identified that from the service users’ point of view provision of subsidized intervention is an incentive to adhere to the treatment. Shorter follow-up periods for persons with FEP as suggested by the service users in our study should, however, be considered with caution. Current evidence suggests judicious use of antipsychotics in combination with psychosocial interventions for FEP (Alvarez-Jimenezet al., 2016).

The availability of the families of persons with FEP willing to work with the service providers can contribute to the reported better outcomes of FEP in some resource-poor settings, possibly by ensuring better adherence to treatment (Mallaet al., 2020). A recent systematic review emphasised the importance of providing tailor-made interventions to engage families (Selick et al., 2017). However, working with families requires two further considerations: cultural adaptation of family therapies developed in the West (Singh et al., 2013; Singh, 2017) and services to have necessary resources to deliver these (Eassom et al., 2014).

The participants in our study were willing to be contacted by telephone for follow-up. Evidence for using telephones and other technology-based interventions to improve adherence and outcomes for persons with mental illness has been well established (Krzystanek et al., 2019; Schulzeet al., 2019, Singh, 2017; Steinkampet al., 2019). There is also a greater understanding among service providers of the importance of person-centred service delivery models especially in FEP services to improve acceptance and adherence by the service users (Dixon et al., 2016; Stewart, 2013).

We identified many barriers that affected the implementation of the FEP intervention in the local setting. The cost of providing and sustaining the specialist FEP intervention is a cause for concern for the service providers. In addition, the dearth of organized capacity-building programs plays a limiting role in the sustenance and scaling-up of the service. Material and human resource restrictions are quite pronounced in LMIC settings and pose a significant challenge in the implementation of interventions (World Health Organization, 2015). The implementation research frameworks, especially in LMIC settings, should especially consider resource constraints. We agree with the suggested new domain by Means et al. (Meanset al., 2020) – Characteristics of systems – within the CFIR and we also recommend including resource issues, as highlighted in our study, as a construct within this domain. To address the economic factors influencing the financing of health care workforce in LMIC settings the World Health Organization has provided a framework for the policy makers and international agencies (WHO and Global Health Workforce Alliance., 2008). Involving the local staff to identify solutions to the problems in implementation by creating a sense of belonging and respect, improving staff empowerment, and increasing knowledge and skills of staff as human resource management solutions have been suggested in LMIC settings (Dieleman et al., 2009).

A feature of the service structure that plays into the FEP intervention implementation in India is the inherent hierarchy in professional organizations, which leads to dominance of doctors in developing and delivering the FEP service. Such hierarchical structure within the healthcare workforce is not unique to India or to mental health (Braithwaiteet al., 2016; O'Leary et al., 2010; Waubenet al., 2011) but appear to be accentuated in India (Nair & Healey, 2006). The leadership of an FEP service needs to reflect the cultural contributions to the hierarchical structures within healthcare services in a LMIC context, such as India. In addition, efforts to study the team characteristics and methods to improve the collective efficacy of the team to take into account the inherent hierarchy within the health care services in LMIC settings have been proposed (Meanset al., 2020).

The challenges experienced by non-medical health care workers within the health care systems in LMIC are exacerbated by service users' and their families’ expectations they see a doctor, and their preference for pharmacological rather than non-drug interventions to address their mental health condition. This expectation is due to the limited awareness about mental illness and therapeutic options among the service users. This again agrees with the recommendation by Means et al. (Meanset al., 2020) about including “community characteristics” as a construct within the domain of “outer setting” in the CFIR. While the expectations of the service users to meet the health care professionals of their choice can influence their perception about the quality of health care received by them (Paddisonet al., 2018), it is important that the services attempt to engage service users not only in direct care but also in organizational design and governance by adopting experience-based co-design methods to develop educational material for future service users, inform policy, enhance care processes, and governance (Bombardet al., 2018).

A practical challenge in implementing the intervention within the organization has been to ensure that all the health care providers follow standard protocols for managing persons with FEP. Highlighting the advantages of specialist FEP intervention and improving communications within the organization to increase the uptake of the intervention within the organization are suggested by the service providers and the managers to facilitate this. Many clinicians are reluctant to adhere to the guidelines as provision of healthcare is often influenced by many contextual factors that are not considered in the guidelines. There is growing evidence to suggest the importance of “mindlines” which are “guidelines-in-the-head” of the clinicians derived from their experience, informal discussions and networking within the organization (Gabbay et al., 2016). Measures to understand local “mindlines” of the clinicians and to address contextual issues may help in successful implementation.

Raising awareness about FEP among public and service users, to address stigma as well as specific religious and cultural issues influencing help-seeking and treatment adherence should be prioritized and could be modelled on a multi-component intervention to improve awareness and reduce stigma about mental illness in general that has been shown to be effective in India (Maulik et al., 2019). Other studies highlight similar cultural challenges in LMIC settings focused upon stigma (Gabbay et al., 2016). We also note lack of resources and challenges in training were identified as barriers in the implementation of a community based mental health intervention using volunteers in India (Maulik et al., 2019).

The model proposed by the service providers is derived from their overall experience considering the contextual realities in our setting. In other settings, the existing ground level psychiatric service providers could be trained in managing the common needs of FEP service users and more experienced and expensive medical staff could provide specialist FEP service for those with complex needs and supervision for junior staff. The task-shifting and task-sharing approach suggested has been a tried and tested model for enhancing service delivery in LMIC settings (Padmanathan & De Silva, 2013). It incorporates a locally adapted intervention, is supported by the leadership, needs fewer specialist workforce, which is potentially cost effective, and provides opportunity for developing capacity within the service in the longer run.

An important limitation of our study is that the information was gathered from one service in an urban setting and this limits generalization of the findings. However, not many services provided specialist FEP service in Chennai (Dhandapaniet al., 2020). This, to our knowledge, is the first systematic effort to study the implementation facilitators and barriers of an FEP intervention in India.

Conclusion

We studied the implementation of an FEP intervention program in a specialist mental health facility, using a well-established framework for doing so, the Consolidated Framework for Implementation Research (CFIR). First, we found CFIR effectively accommodated implementation challenges evident in LMICs; that is, it is transferable to LMIC settings. Second, we highlight barriers to implementation include: cost; limited human resources; cultural and professional hierarchy; divergence from evidence-based guidelines; and lack of awareness and stigma in the wider community. Third, we highlight facilitators for implementation: leadership engagement; the need for change was recognized; cosmopolitan perspectives derived from clinicians’ international experiences and expertise; compatibility of the intervention with the existing systems within the organization; accommodating the needs of the service users; and rapport developed with service users. Fourth, we propose a model of service delivery incorporating a task-sharing approach for first episode psychosis in resource restricted settings based on the feedback from the stakeholders.

Declaration of interests

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.