The WHO mhGAP Guideline in Context: Read the Recommendation and Its Conditions

Edited with AI

Review & Opinion

A recommendation can fit a diagnosis and still be a poor match for the service available. The World Health Organization (WHO) Mental Health Gap Action Programme (mhGAP) guideline helps learners examine that tension: what care is supported, how certain the evidence is, and what competent delivery requires.

The 2023 third edition concerns mental, neurological and substance use conditions, principally in non-specialist health services in low- and middle-income countries. Its audience includes doctors, nurses, other health workers, planners and programme managers. It belongs on the reading list of mental health and psychosocial support (MHPSS) learners because it connects broad service ambitions with specific care decisions. It does not turn every psychosocial helper into a clinical provider. [1]

Our judgment is that its greatest educational value lies in the reasoning around each recommendation. Reading only the recommendation box loses much of what makes the document useful.

Know which document you have

The 2023 guideline and the mhGAP Intervention Guide, version 2.0, published in 2016, are distinct publications. The guideline reviews evidence and makes recommendations. The Intervention Guide translates recommendations into clinical assessment and management protocols. The 2023 guideline itself explains this distinction in section 4.2 and identifies practical details that belong in derivative tools. [1, 2]

This matters when someone says a programme is “based on mhGAP.” That description leaves several questions unanswered. Which edition informs its clinical decisions? Which tool guides delivery? What training and supervision accompany it? A citation to the 2023 guideline cannot establish that an older training package incorporates its recommendations.

For learners, the useful habit is to name the publication and the decision it supports. A guideline, an implementation resource and an intervention manual contribute different kinds of knowledge. Our review of WHO's implementation manual follows the service-planning decisions that surround an intervention. Reading these resources informs different responsibilities; it cannot demonstrate mastery of them.

A recommendation has more than one strength

The guideline distinguishes the strength of a recommendation from certainty in its supporting evidence. Its methods also consider benefits and harms, preferences, resources, equity, feasibility and human rights. A recommendation is therefore a judgment about action under specified conditions, rather than an evidence score in disguise. [3]

Two examples make this concrete. For adults with generalized anxiety disorder or panic disorder, recommendation ANX2 supports brief, structured interventions based on cognitive behavioural therapy (CBT) principles, with a strong recommendation and moderate evidence certainty. ANX3 considers delivery formats and carries a conditional recommendation with low certainty. The first statement does not make every format equally suitable. [4]

For adults with post-traumatic stress disorder (PTSD), recommendation STR1 considers several psychological approaches, including CBT with a trauma focus, eye movement desensitization and reprocessing (EMDR), and stress management. The recommendation is conditional and evidence certainty is low. Its surrounding text discusses comparative findings, access and preferences rather than presenting an interchangeable menu. [4]

The reading lesson is simple: preserve the population, comparison and qualifications when taking a recommendation into a meeting. “Recommended by WHO” is too broad to settle a choice between programmes. Nor does a recommendation for a method family validate every local adaptation that shares one of its techniques.

Competence is part of the recommendation

The PTSD section explicitly addresses training, demonstrated competence and supervision. It also acknowledges the complexity of trauma-focused techniques and the risks of inadequately supported delivery by lay providers. The anxiety section likewise addresses provider time, training and supervision. [4]

These requirements deserve operational weight. In our view, a proposal that names a recommended intervention but leaves its supervision arrangements unspecified has not completed the implementation argument. This is a question to investigate, not proof that the proposed intervention will fail.

The same distinction helps when reading Problem Management Plus in context. A structured intervention package gives a learner a more defined delivery model to examine. The mhGAP guideline helps locate such choices within a wider clinical evidence and care framework. Neither document substitutes for training in the actual intervention or for the responsibilities attached to a professional role.

Where the guideline reaches its limits

The guideline's breadth is a strength for health services. It is less suited to answering every question about housing, protection, livelihoods, community participation or humanitarian coordination. A service planner needs a wider view of the response, including resources such as the MHPSS Minimum Service Package.

Another limit is transferability. The PTSD section identifies the small number of studies conducted in low- and middle-income countries as a constraint on generalization. [4] Learners should therefore separate evidence that informed a recommendation from evidence that their own service can deliver it acceptably and effectively. Local monitoring can inform that second question; a well-designed service description alone cannot answer it.

Our criticism is directed at shortcuts in using the guideline. Its own text is more careful than a slide containing a list of approved techniques.

A useful way to study it

Choose one care question within your role. Read the relevant recommendation, then write down the population, evidence certainty, implementation conditions and unresolved questions. Discuss where your service meets those conditions and where further preparation is needed. This is a reading exercise, not an assessment or treatment protocol.

Keep the person's route through support in view. Psychological First Aid addresses initial help; the stabilization-groups discussion examines another proposed function within a broader pathway. Their relevance depends on purpose, fit and boundaries. They should not become automatic substitutes for indicated clinical care.

Use mhGAP to make the question more precise: what care is justified here, for this population, with these delivery conditions? That is a stronger foundation for learning than collecting recommendations as endorsements.

Sources

1. WHO. Mental Health Gap Action Programme guideline, third edition, 2023. Introduction, sections 1.1–1.4, printed pp. 2–3.
2. WHO, 2023. Section 4.2, Derivative products, printed p. 112.
3. WHO, 2023. Methods, sections 2.3–2.4, printed pp. 5–10.
4. WHO, 2023. Recommendations ANX2–ANX3, printed pp. 20–22; STR1, printed pp. 46–47.

Professional scope

Deep Emotional Work Academy materials are non-clinical professional education by default. They are not medical or psychological advice, diagnosis, treatment, psychotherapy, cure, or a guaranteed mental-health outcome. Any clinical use, patient-facing application, or integration into institutional standards of practice requires review, adaptation, approval, and supervision by qualified authorities within the receiving institution. Those decisions remain within the responsibility, qualifications, governance structures, and legal scope of the participating professionals and host institution. Matthias Behrends contributes as an educator and instructor. All activities in instructor-led institutional training require the involvement of a licensed clinician and a designated supervisor.