Background

There is a well-reported, very sizeable global gap between the need and provision of services to prevent, identify and treat mental, neurological and substance use (MNS) disorders. To expedite care as efficiently as possible, the WHO’s World Mental Health Report1 recommended the assessment and management of MNS disorders in primary healthcare (PHC) and community settings, as outlined in the Mental Health Action Plan.2 To deliver community mental healthcare, PHC staff require training in the assessment, diagnosis and management of MNS disorders, alongside other key components, including regular supervision and support. Central to this agenda is the WHO Mental Health Gap Action Programme3 (mhGAP) and intervention guide (IG).4

The mhGAP-IG is an evidence-based tool for the assessment and treatment of priority MNS disorders, featuring guidelines for clinical decision making. Aimed at non-specialist PHC staff in low-income and middle-income countries (LMICs), the mhGAP-IG is also used by governments, non-governmental organisations and researchers. An mhGAP Humanitarian Intervention Guide (HIG),5 for settings characterised by widespread psychological trauma and even less access to mental health specialists, has also been published,5 alongside an mhGAP-IG mobile device app.6

Our previous systematic review7 identified 33 published peer-reviewed studies and protocols which used the mhGAP-IG. Given that another review found only six published studies of experimental non-communicable disease clinical guideline implementation in LMICs,8 the mhGAP-IG literature was relatively substantial in 2017. The mhGAP-IG had been adopted by clinicians, government ministries, trainers, educators and academics in diverse LMICs. Applications ranged from local adaptation, training and clinical practice, to mobile applications for isolated PHC staff, tablet-based avatar-assisted family training, economic modelling, novel rating scales and comparison interventions in randomised controlled trials (RCTs). The literature was, however, dominated by studies from a subset of countries, suggesting that much implementation was not evaluated, or that evaluations were not widely disseminated. Reliance on relatively limited pre-mhGAP and post-mhGAP training knowledge assessments meant that opportunities may have been missed at times, to describe real-world contextual challenges to widespread uptake and scale-up.

Objective

In this study, we sought to identify evidence generated since the publication of our previous review for the practical implementation of the WHO mhGAP-IG and associated tools, in terms of how they have been used, evaluated and reported.

Study selection and analysis

We included any type of study design, review or report of evidence, experience or evaluation of using the mhGAP-IG in LMICs, from any country, in any language. In order to identify as many potentially relevant records as possible, we searched the following databases on 24 November 2020: 3ie, Cochrane Library, CINAHL, EMBASE, LILACS, Medline, PsycINFO, PubMed, SciELO, Scopus, Web of Knowledge. The only search terms used were ‘mental health gap action programme’ OR ‘mental health gap action program’ OR ‘mhGAP’. The term ‘intervention guide’ was not included, due to its variable use in the literature and in practice. Searches were conducted in English, but studies written in other languages were eligible for inclusion. In addition to database searches, the reference lists of relevant excluded papers were searched for eligible studies. Grey literature, including book chapters and conference presentations, were identified by repeating the search on Google Scholar. We used forward citation tracking of our 2017 systematic review to identify additional records.7 This work is registered on the International Prospective Register of Systematic Reviews (registration number: CRD42017068459).

Findings

Figure 1 shows the flow of studies from identification to screening, eligibility and inclusion. The titles and abstracts of the 2002 non-duplicated papers were screened by RK excluding 1807, which did not review or report evidence, experience or evaluation the mhGAP-IG, app or HIG. The details of a subset of excluded studies reporting non-mhGAP-IG-related integrations of mental healthcare into PHC were recorded for a separate review. The remaining 195 full-text articles were screened again for eligibility by JS and categorised into study types. Where studies could be categorised into more than one type (e.g., both clinical application and research), a judgement was made about the study’s primary focus. No papers were excluded based on language, and no relevant papers from high-income settings were identified.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

PRISMA flow diagram. mhGAP-IG, Mental Health Gap Action Programme Intervention Guide; PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-Analyses.

Table 1 shows the distribution of study types across geographical regions. Most studies (40%, n=62) were conducted in the African region and South-East Asia (25%, n=38). Notably few studies were conducted in the Eastern Mediterranean (7%, n=10), region of the Americas (4%, n=6) and the European region (1%, n=2). Ten per cent of studies (n=15) reported results from at least two countries on different continents. Most included studies described uses of the mhGAP-IG, app or HIG for training (38%, n=58), clinical use (30%, n=44) and research (16%, n=25).

Table 2 shows the study design methods used across different study types. RCTs comprised 15% (n=23) of studies and RCT protocols or trial registrations, 12% (n=19). Uncontrolled studies comprised 23% of studies (n=36), descriptive accounts 18% of studies (n=28), qualitative studies 8% (n=13) and other non-randomised designs 7% (n=10).

Of the 193 eligible studies, data were extracted by JS and RK from the 162 papers (describing 155 studies) not included in the 2017 systematic review (n=33). The heterogeneity of mhGAP-IG, app and HIG uses, outcome measures, and evaluations precluded meta-analysis. Due to the heterogeneity of designs, including a large proportion of non-randomised designs and variable reporting methods, we did not assess risk of bias across included studies.9 Extracted data included country involved, participants, sample, nature of use, evaluation conducted and summary of findings, where relevant. Below, we summarise the findings reported by studies categorised into six different types: use in training (online supplemental material table 4), use in clinical practice (online supplemental material table 5), use in research (online supplemental material table 6), local adaptation or contextualisation (online supplemental material table 7), economic evaluation (online supplemental material table 8) and other educational purposes (online supplemental material table 9).

Conclusions and clinical implications

This updated systematic review of the literature on the mhGAP-IG, app and HIG demonstrates a dramatic increase in published studies of its use and evaluation since 2017.7 Most of the literature comes from the African region and South-East Asia, with notable evidence gaps including the region of the Americas and the Eastern Mediterranean. Countries participating in the programme for improving mental healthcare (PRIME) study (Ethiopia, India, Nepal, South Africa and Uganda) are well represented, demonstrating the benefits to mental health research and evaluation of investment, shared training and methodologies. PRIME identified barriers to integrating mental health into PHC, including limited funding, insufficient specialists to supervise non-specialist workers, inadequate health system structures to support roll-out of task-shared interventions, low community awareness of mental health and high levels of stigma.14 Building on PRIME, the AFFIRM study is conducting two RCTs in South Africa and Ethiopia of task-shared psychological interventions, building capacity through fellowships and mentorships for early career researchers from Ethiopia, Ghana, Malawi, Uganda and Zimbabwe, and building a network for interdisciplinary collaboration.15 Both projects demonstrate ways of enhancing the impact and sustainability of mental health research, through long-term collaborations between LMIC and HIC institutions, funded to design and evaluate new interventions from piloting to scale-up, and develop the next generation of global mental health researchers in LMICs.

Similar study types to those identified in our 2017 review7 have been published since, with the highest proportions of studies reporting on mhGAP-IG training courses, clinical and research applications. The surge in literature also includes contextual adaptations, economic analyses and novel applications to health education. Studies demonstrated the benefits of the mhGAP-IG’s use in training for improving the knowledge, attitudes and confidence of health workers in relation to mental health disorders. Clinical studies identified improvements in patient symptoms, quality of life, access to and engagement with mental healthcare associated with mhGAP-IG implementation.

The published evidence comprises a breadth of research methods, with high proportions of RCTs and protocols for RCTs, and smaller proportions of related feasibility studies, implementation studies and process evaluations. The publication of uncontrolled, non-randomised, descriptive and case studies contributes a valuable literature on the mhGAP-IG from settings where more extensive research studies may not be feasible. Adaptation studies recommended a range of contextual considerations to optimise the success of mhGAP-IG implementation.

Many research uses of the mhGAP-IG aimed to adapt and evaluate brief psychological interventions for settings where specialist mental healthcare is limited or unavailable. Often, such interventions were implemented alongside mhGAP-IG training (as ‘enhanced usual care’) or as a next step after successful mhGAP-IG training for clinical use. Literature on the Thinking Healthy Programme, Healthy Activity Programme, Counselling for Alcohol Problems and PM+, largely from South Asia, demonstrates the broader role of the mhGAP-IG in encouraging the development and rigorous evaluation of acceptable, feasible and scalable talking therapies in LMICs. Sharing of intervention manuals via the WHO website should facilitate the adaptation and evaluation of these interventions for other countries, cultures and contexts. SMIs such as schizophrenia, severe depression and bipolar affective disorder in PHC were less frequently targeted by intervention studies, the majority of which focused on mild to moderate depression. Lessons learned from the use of the mhGAP-IG for common mental disorder research could now be applied to research and intervention development for people with SMI.

Recognition has grown of the need to improve the quality of inpatient and community mental healthcare, attend to the human rights of people with mental disorders and promote recovery.16 The fact that most mhGAP-IG, app and HIG training combined lectures and small-group tasks with role play interactions suggests that scaling up the mhGAP-IG could have wider benefits, improving health worker communication skills and the compassion and respect with which care is delivered. Twenty-one included studies addressed or measured stigma, or planned to. Of 15 completed studies, nine reported reductions in one or more measures of stigma, even where health worker knowledge or skills were unchanged.17–25 Economic studies identified the scale of the mental health gap, its financial26 and disability impacts,27 and calculated the workforce requirements28–30 and costs of treatment.

The principal limitation of this study is the likelihood of publication bias, with studies finding no significant or negative results less likely to have been published. A substantial number of reports is likely to remain in the ‘grey literature’ outside peer reviewed journals. A further limitation is that we conducted single-author screening of potentially eligible studies, raising the risk that some eligible studies could have been missed.

Since their release, the WHO mhGAP-IG, app and HIG have made a remarkable impact on global mental health education and training, clinical practice, research and policy. Included studies used a range of methods: a testament to practitioners’ commitment to mhGAP-IG monitoring, evaluation and implementation research. Promising approaches to strengthen the evidence base include consortia bringing researchers together from a range of LMICs and HICs for collaborative, interdisciplinary study, benefiting from shared learning and pooled resources. The mhGAP-IG has created a dynamic field in which practitioners from diverse international contexts learn from each other’s experiences and adapt study designs and intervention models for their own context. The next step must be to fill gaps in the evidence base for under-studied regions and disorders, and to investigate the scale-up and sustainable integration of isolated interventions into long-term clinical practice.