Introduction

About 200 million persons have some form of mental illness in India.(Sagar et al., 2020). Task-shifting (a process whereby specific tasks are moved, where appropriate, to health workers with shorter training and fewer qualifications) approaches have been advocated to address this huge public health challenge (Gururaj G et al., 2016; Hofmann-Broussard et al., 2017; World Health Organisation et al., 2008). In India and other parts of the globe, there are elegant task-shifting approaches (Chatterjee et al., 2009, 2014; Kumar et al., 2017; Nadkarni et al., 2019; Shidhaye et al., 2017).

In a South African study, the impact of a training program for community health workers (CHWs) from 4 health care settings (total n = 56) in mental health, demonstrated a gain in knowledge scores (n = 46; p ≤ 0.001) sustained over 3 months after the eight-sessions training program along with significant gain in their confidence and attitude after the training (Sibeko et al., 2018). Another study demonstrated not only the feasibility of HIV counselling but also the perceived need of community health workers (CHWs) for training in mental health. 67% of patients were willing to receive counselling by CHWs. Also, > 80% of individuals remained in contact with CHWs. (Myers et al., 2019). In an Indonesian study 3 days, modular training sessions in mental health for CHWs (N = 65) (2 day classroom training and 1 day fieldwork) showed a statistically significant improvement in knowledge, and attitudes in mental health 4 months after completion. (pre-test mean score: 11.9; post-test score: 13.1 and 15.3 4 months post-training) (Marastuti et al., 2020).

The MANAS trial in Goa, India studied, a community sample of 2776 patients and demonstrated that lay heath workers' led psychological interventions and collaboration with mental health professionals resulted in a 30% reduction in the prevalence of Common Mental disorders (CMDs), and a 37% reduction in suicide attempts over 12 months. The study further showed a reduction in the number of days lost to psychiatric morbidity and demonstrated that psychosocial interventions for CMDs can be effectively delivered by lay health care providers and general practitioners alike (Patel et al., 2011). A systematic review of CHWs-led community-based mental health interventions showed that the average duration of training of CHWs ranged from 2 days to 3 months. Formats included lectures, role-plays along with proficiency testing that included a variable duration of supervision and periodic reviews. Additionally, CHWs led mental health interventions and monitoring resulted in meaningful impact with 19 of the 27 trials showing positive outcome in the mental health condition (Barnett et al., 2018).

The use of technology in mental health capacity building is an emerging area that carries with it a huge potential for an exponential increase in the number and quality of human resources to cater to the burgeoning needs of the country. Another unmet need is the paucity of ‘effectiveness’ studies as opposed to the ‘efficacy’ studies, that suffer from the ‘generalizability’ point-of-view.

Technology and Training in Mental Health

Accessible digital technology has now made it relatively easy to reach out to community health workers and carries the potential to exponentially transform the training landscape. Several capacity-building programs in mental health are being carried out in India including non-specialist doctors, Community Health Officers (CHOs) (Ibrahim et al., 2020, 2021; Pahuja et al., 2020) nurses, and counsellors in Addiction medicine.(Bairy et al., 2019; Mehrotra et al., 2018; Sagi et al., 2017)An important research question is to see if technology can be seamlessly utilised in training CHWs.

ASHA and Her Role in Integrating Mental Health into Primary Care

The cadre of Accredited Social Health Activists (ASHAs), a quintessential CHW in the Indian public healthcare scenario, was created in 2006 by the union health ministry's flagship program, the National Rural Health Mission (Rahul et al., 2021). They act as an interface between the community and the health services. ASHA is a woman who resides in the community she would be working with. In addition to maternal and child health, ASHAs are involved in the identification of various communicable diseases (Tuberculosis, Leprosy, Malaria, sexually transmitted diseases) and non-communicable diseases. ASHAs’ role in mental health too is coming to the forefront in recent times (Colizzi et al., 2020; Malathesh et al., 2021; Thomas et al., 2008).

Need for the Study

Establishing the effectiveness of empowering ASHAs in mental health can potentially pave the way to meaningfully integrate mental health into general primary healthcare service delivery. In this pragmatic implementation research, we attempted to examine the issue by comparing the traditional training method vs digitally-driven longitudinal training.

Methodology

Settings and Design

Details of the methods is described elsewhere (Kumar et al., 2020). To summarize, the Ramanagara district belongs to the Karnataka state of South India. It has 61 Primary Health Centres (PHCs). Of these, as a first step, we randomly (simple randomization generated through a computer) selected 6 (six) PHCs to be included in the study. Further, again using simple randomisation method, 3 (three) were chosen to be the ‘study group’ PHCs. The remaining 3 (three) PHCs formed the ‘control group’ (CG). All consenting ASHAs in each of 6 (six) PHCs were recruited for the study. Accordingly, 71 ASHAs of the 6 PHCs were recruited: 35 in SG and 36 CG. The adult Population served by the Study Group-ASHAs (SG-ASHAs) was 22,623 and 12,400 by the Control Group-ASHAs (CG-ASHAs) (Office of the Registrar General & Census Commissioner, 2011).

SG-ASHAs received mental health training through hybrid mode (vide below) and Continued regular mentoring and monitoring while the CG-ASHAs got only the one-time onsite training and continued to provide DMHP services as before. The objective was to compare the effectiveness of two modes of training ASHAs i.e.longitudinal hybrid mode vs One-time Classroom Training. Outcome measures were to compare (a) the number of individuals with possible mental illness identified by the ASHAs of both groups and (b) pre and Post KAP scores of both groups.

Components of the Training

Onsite Training

A one-day (6 hours) in-person classroom teaching consisting of orientation to various mental illnesses, their presentation and identification, and referral using a simple screening tool, ‘Symptoms-in-Others’, a simple tool to screen for severe mental illnesses (Kapur & Isaac, 1978; K. S. Reddy et al., 2013). This onsite training was given to both SG-ASHAs and CG-ASHAs.

Online Training (ECHO model)

(Arora et al., 2011; Bairy et al., 2021): Curriculum content was sourced from the ‘Manual for community health workers’ that was earlier developed by the National Institute of Mental Health and Neurosciences (NIMHANS), Bengaluru at the request of the World Health Organisation (WHO), India Office and the Ministry of Health and Family Welfare, Govt. of India (submitted to the WHO Office and the Ministry for approval). The manual is in the point-of-care format containing simple descriptions of various clinical scenarios and ways of identifying problem behaviours/mental states and instructions for further management. Local language i.e. Kannada was the medium of instruction. The session content was vetted by the authors CNK, PLN, and BCM, and imparted by NH and NK. SG-ASHAs completed the training over 18 months. Each session lasted from 60 to 90 min. The time gap between two sessions ranged from 4 to 6 weeks intervals. Each training session began with an ASHA discussing details of the case seen by her, subsequently, NH/NK discussed a topic based on the curriculum through case vignettes followed by a brief didactic lecture. At the end of each session take-home points were discussed. The sessions covered the following topics: (a) sad, worried, and nervous persons (b) violent, fearful and disorganized persons (c) suicide and how to manage persons with suicidal ideas (d) behavioural problems due to substance abuse (e) Myths surrounding the mental illnesses (f) two sessions on cases encountered by them in the community focusing on rehabilitation and disability aspects. The online component was delivered only to the SG-PHC ASHAs. Also, the online training occurred in a group format wherein all ASHAs of each of the PHC gathered in their respective PHCs.

Technology for the Online Training Sessions

Each SG-ASHA was provided a smartphone with a 4G internet Data plan. Hub and Spoke model of online training was adopted. Zoom digital platform was used for conducting training sessions from the Hub i.e., NIMHANS. Spokes (SG-ASHAs) logged in from their respective primary health centers (PHCs) in groups as described above. Sessions were conducted at a mutually convenient date and time.

Mentoring and Monitoring

ASHAs of both groups were contacted at least once in 2 weeks (apart from the on-demand calls originating from ASHAs) to check on the progress of case-finding and for offering guidance in handling various mental health-related issues they encountered in their communities (through telephone calls), which was carried out primarily by the research nurse and a research social worker. On-demand basis, ASHAs also contacted the research team for assistance and guidance in handling mental health issues encountered in the community.

Assessments

House to House Surveys

ASHAs completed two rounds of house-to-house surveys covering the area under their purview during the study period (October 2018 till March 2021). For the first round, they used ‘Symptoms in Others’ (Kapur & Isaac, 1978), a simple tool (that takes 2 min to complete) to identify severe mental disorders in the community. The first round of the survey was conducted in August 2019. During this activity, it was noted that in addition to the severe mental illnesses (Symptoms in Others is designed for the former purpose), only a few people having symptoms with anxiety and substance use spectrum were reported by ASHAs. The research team then conceptualised and designed a tool that could capture a wider spectrum of symptoms. The team members reviewed and deliberated upon the existing tools and funneled them down to a set of simple questions. This was named ‘Mental Health Screening and Counselling Toll for Field Level Workers of India (MERIT)’, MERIT takes about 5 min to apply and psychometric properties showed favourable results with good inter-rater reliability between ASHAs with kappa statistic value k = 0.792; p < 0.001 and excellent concurrent validity (a measure of agreement between ASHA and Mental health professional) kappa statistic k = 0.744; p < 0.001. Face and validity exercise showed that 14/17 (82.35%) mental health professionals (who rated the tool) strongly agreed that questions were appropriate for screening out various mental health issues in the adult population. It may be noted that we are in the process of publishing the psychometric properties of MERIT. The second survey was then carried out using MERIT between November 2020 and March 2021.

22,623 people were surveyed in the SG and 12,400 were covered in the CG. Extrapolating the results from the National Mental Health Survey 2016 (Murthy, 2017), 279, 3712, and 1610 cases of SMDs, CMDs, and SUDs respectively were expected to be prevalent, in the area covered by all 6 PHCs. For this study, each screen positive (by ASHAs) was considered to be a potential ‘case’.

Scores on Knowledge, Attitude, and Practice

All ASHAs were requested to take up an assessment of their knowledge attitude, and practices (KAP) twice during the study. First, before the start of the training and the second at the end of completion of training. KAP questionnaire consisted of 28 questions, prepared by the researchers based on their experience working in the field of community psychiatry. The questions were of objective type and, the correct choice carried one mark (total maximum score: 28). The same Questionnaire was applied for pre- Assessment and Post Assessment (after 18 months). Figure 1 gives details of enrolment, randomization, and the numbers included in the final analysis.

Fig. 1
figure 1

Flow diagram of the study and control group recruited for the final analysis of Knowledge Attitude and Practice scores in Mental Health

Full size image

Statistical Analysis

Discrete variables were analysed using Chi-square test. KAP scores were compared using Repeated Measures ANOVA (RM ANOVA) to study the gain in KAP scores, both in terms of time effect and the group X time interaction effect. Analysis of Covariance (ANCOVA) was used to study the sociodemographic (Age, education and years of experience as ASHAs) variables’ impact on the post-test KAP scores. Analysis was carried out using the software Statistical Package for Social Sciences (SPSS), version 27 (IBM Corp, 2020).

Results

The sociodemographic characteristics of ASHAs are as shown in Table 1.

Table 1 Socio-demographic details of the Accredited social health activists (ASHAs)
Full size table

Mentoring and Monitoring

The research team contacted ASHAs a total of 1648 times at an average of 48 calls per week (SG-ASHAs got 840 calls, whereas CG-ASHAs, 808 times). On the other hand, SG-ASHAs contacted the research team 430 times, while CG-ASHAs contacted 340 times (Chi-square = 5.0; p = 0.02).

Identification of Persons with Possible Mental Illness

Prevalence figures from the National Mental health Survey 2016 (Gururaj G et al., 2016) were used as a benchmark to calculate the estimated prevalence of mental illnesses in the community. After counting the number of cumulative screen positives (from both surveys), we calculated the fraction (of the actual prevalence) of individuals with probable mental illnesses. Chi-square test was carried out to analyse the difference between the two groups (Table 2).

Table 2 Proportions of Individuals with Likely mental illnesses identified Total adult population
Full size table

KAP Scores

The mean total KAP score increased from 16·76 to 18·57 (p < 0·01) in SG-ASHAs while it increased from 18·65 to 18.84 (p = 0·76) in CG-ASHAs (paired sample t-test). Further, as the baseline pre-test KAP scores of CG-ASHAs were more, we used them as covariates for the RM ANOVA. While the time effect favoured the SG-ASHAs, the same did not hold true for the time X group interaction effect (Table 3). Finally, we carried out ANCOVA to study the impact of sociodemographic characteristics of ASHAs on the post-test KAP scores controlling for the pre-test KAP scores. Not only the overall model was not significant (F = 0.400 and p = 0.84), but none of the independent variables [Education status (F = 0.37, p = 0.543); age (F = 0.17, p = 0.68); years of experience (F = 0.44, p = 0.48)] emerged as a predictor.

Table 3 RM ANOVA for KAP Scores after controlling for baseline scores
Full size table

Discussion

This study demonstrates that compared to the traditional classroom training, the longitudinal hybrid mode is a more effective way to empower ASHAs to carry out mental health work. SG-ASHAs not only identified more number of potential cases but showed meaningful improvement in KAP Scores. A corollary finding is a feasibility of utilising digital technology to engage grassroot health workers for mental health, notwithstanding the challenges (Gajera et al., 2021a). To the best of our knowledge, this is the first study to examine the longitudinal impact of mental health training in India. The absence of the group-time interaction effect could be due to the ceiling effect. It was also observed that the knowledge scores of CG dropped from the baseline thus indicating that extended longitudinal training did benefit in terms of improving KAP Scores of SG.

Another confounder could have been the baseline education status of SG-ASHAs. Though they outnumbered CG-ASHAs with regard to completing high school education, the former had lesser baseline KAP Scores. Moreover, on ANCOVA, educational status did not emerge as a predictor of the post-test KAP scores, demonstrating that just higher basic formal education may not have impacted the KAP Scores of ASHAs.

Periodic collaborative engagement (in terms of video and audio consultations) could have contributed to better outcomes (Pahuja et al., 2020). Identification of potential common mental disorders and substance use problems is a relatively novel and important finding from the public health point of view. The fact that CHWs can work towards their identification augurs well for the future. Here again, the finding was reversed with regards to severe mental disorders. This could be since the identification of severe mental disorders in the community is a no-brainer. Essentially, persons with severe mental disorders are the face of psychiatry in the community with their obvious symptoms and disability. Previous studies have shown that identifying them can easily be done by field-level workers such as ASHAs (S. Reddy et al., 2014). In contrast, persons with CMDs commonly present with physical complaints and these are unlikely to be identified as symptoms associated with mental health issues (Gautam & Jain, 2010). Also, subjective individual-level factors such as active engagement, inquisitiveness, etc. do play a role in better identification of cases in the community. Anecdotally, the latter was observed to be better in one CG-PHC. Though randomization is supposed to account for all these factors, the sample sizes for the study could have limited their applicability.

These efforts can be easily scaled up towards involving these cadres in mental health work at the grassroot level. A couple of such initiatives may be noteworthy wherein the remotest healthcare professionals are reached out in innovative ways. (Bairy et al., 2021; Gajera et al., 2021b; Gowda et al., 2018; Ibrahim et al., 2021; Sukumar et al., 2020).

Methodological Issues

An important limitation is that significant differences did not emerge concerning the attitude scores. This could be because of a couple of factors (a) lesser number of questions testing attitude and (b) Our repeated anecdotal observations showed better overall motivation by a group of 'control' group ASHAs throughout the program. Another limitation is that the sample size was not calculated at the beginning itself and this may have led to bias. However, their number was chosen keeping in mind the budget and quantum of work the research team could handle. Whatever be the cause, we believe that the direction of the results points to only one direction. (c) another criticism could be the ‘unvalidated’ nature of the KAP questionnaire. These were chosen based on the extensive experience of the researchers and with the consensus of experts about the appropriateness and scoring pattern of the questionnaire. This is an eye-opener in the sense that future efforts should concentrate on the better spread of questions targeting different dimensions of knowledge and skill acquisition by the ASHAs. Detailed analysis on population-level outcomes (in terms of outcomes of treatment received by persons with psychiatric disorders and its impact on reducing the treatment gap for psychiatric disorders) is currently underway. (d) challenges related to the use of digital technology are already discussed separately (Malathesh et al., 2020).

Mentoring of Community Health Workers (CHWs)

At the first level, the CHWs (inclusive of ASHAs) can be mentored on promotive and preventive aspects of mental health. Additionally, routine follows ups can be carried out by them. They can learn to identify common side effects of psychotropic medications and alert family members. They can impart information on the importance of 'treatment adherence. They can act as the link between families and the health system by alerting them about early signs of relapses and recurrences, in addition, mentoring would benefit them to handle difficult scenarios, crisis intervention, psychoeducation about side-effects of psychotropics, giving information on or outpatient clinics that are nearest to patients’ houses, providing information on disability benefits, etc. Mentoring will also enable CHWs to address other issues such as stigma and generate awareness regarding mental health in the community. Another point to note is that these things will not take much time, fortnightly sessions can be scheduled for a batch of about 50 ASHAs. Finally, other higher centers including the departments of psychiatry in medical colleges and mental health institutes can come forward with such initiatives. These can act as referral centers for complex cases.

Curriculum for Mentoring ASHAs

Topics and the nature of interactions were chosen from the CHW manual prepared for the WHO and the Ministry of Health and Family Welfare, Govt. of India. A bottom-up approach was taken by way of holding consultations with multiple stakeholders. A 2 day workshop was conducted at NIMHANS in 2018 to deliberate what kind of work is expected from CHWs and the kinds of cases they come across in the community. Learning methods included case vignettes and role play videos depicting persons with psychiatric disorders. A symptomatic trans-diagnostic approach was taken to devise chapters that also included the basic interventions a CHW could carry out. Another aspect of the curriculum was the brief screening tool that was devised as a point-of-care guide. The tool named Mental Health Screening and Counselling tool for field level workers of India (MERIT), helped CHWs apply the same to all families in a door-to-door survey. The tool takes about four minutes to apply in each household and also contains a very brief counselling guide.

Indeed, the use of non-specialist health workers in delivering community-based care is not a new phenomenon in India and excellent models have demonstrated their feasibility, utility, and effectiveness in all kinds of psychiatric disorders (Nadkarni et al., 2019; Patel et al., 2011; Shidhaye et al., 2017). However, this study is unique in the sense that the public health machinery was involved in testing the effectiveness, and hence, this could be more generalisable. This aspect cannot however take away the fact the prerequisites for the success of the program including the inherent motivation and accountability of CHWs for mental health work. Administrative mechanisms to monitor the progress and periodic review of the program at the PHC/Block and District levels are necessary.

Incentivising Mental Health Work of ASHAs

The role of ASHAs in maintaining “Continuity of Care” during the challenging period of the first wave of Corona Virus Disease Pandemic-19 (COVID 19) Pandemic in India is highlighted (Rahul et al., 2021), and would like to reiterate that this would be an essential and important aspect of public mental healthcare delivery and would no doubt go a long way in reducing the treatment gap. Not only the out-of-pocket expenditure of patients reduce by making treatment available nearer to doorsteps, but also, the motivation and drive of CHWs will increase by having dedicated incentives for mental health work. In this context, it may be noted that the XII 5 year plan document for DMHPs (that is currently active) makes provision for 2 CHWs per primary health center (Policy Group DMHP, 2012). Supportive policies are already there for such a cause and these need to be harnessed appropriately and flexibly for reaching out to more people. One of the challenges faced during the study was the fact that ASHAs salaries were delayed which led to disruption in service provisions by them, therefore if incentivisation of CHWs is initiated under DMHP, consistency in payments needs to be ensured, as any discontinuity in care in early phases of treatment could negatively impact the care/outcome of the patient. Also, Caution needs to be exercised while incentivising the service wherein clear guidelines need to be set to ensure the quality of care is maintained.

Qualitative Perspectives from ASHAs

ASHAs felt that mental health work is important and felt that the training program helped broaden their horizon and they also believed they can integrate mental health work with other responsibilities provided that is made easy and less time-consuming. This is the spirit with which the above modules were prepared. This in itself was one of the secondary objective of the larger study, with only a few important points highlighted here.

ASHAs are already being given brief classroom training in states such as Karnataka and Gujarat (Armstrong et al., 2011; Shah et al., 2019). Outcomes need to be monitored against the indicators and technology can be better utilised to amplify the impact of engaging them. Lastly, there are multiple other cadres of CHWs. They belong to other sectors such as the women and child development department (Anganwadi workers) (Sandhyarani & Rao, 2013) and Disability Department (village rehabilitation workers) (Peterson, 1999). They should also be involved in mental health work in whatever way possible. Anganwadi workers can concentrate on women and children while the village rehabilitation workers can concentrate on those with severe mental illnesses in ensuring follow-up and maintaining continuity of care.

Conclusion

Compared to the traditional one-time classroom training, the longitudinal hybrid model is associated with better skills retention and better identification of those with possible mental illness. Task shifting and task sharing in mental health can and should incorporate services of ASHAs or CHWs.