Community-, facility-, and individual-level outcomes of a district mental healthcare plan in a low-resource setting in Nepal: A population-based evaluation
Mark J. D. Jordans, Nagendra P. Luitel, Brandon A. Kohrt, Sujit D. Rathod, Emily C. Garman, Mary De Silva, Ivan H. Komproe, Vikram Patel, Crick Lund
PLOS Medicine · 2019 · DOI: 10.1371/journal.pmed.1002748
Countries: Nepal
What Was Studied
This programme evaluation examined a district-level Mental Healthcare Plan (MHCP), developed under the multi-country PRIME research programme in partnership with Nepal's Ministry of Health, that integrated depression, alcohol use disorder (AUD), psychosis, and epilepsy care into 10 primary healthcare facilities in Chitwan district (population 579,984). The evaluation combined four linked components: (1) a community survey targeting 2,000 participants per round for adequate statistical power (3,482 participants actually surveyed across two independent-sample waves, before and 30 months after implementation began); (2) 12 months of routine health-facility service-utilisation records of clinically diagnosed cases, before versus during MHCP implementation (N=727 combined); (3) a repeated facility-based cross-sectional survey of outpatient attendees (3,627 participants across three waves -- baseline, 6 months, and 24 months after health-worker mhGAP training) with clinical consultation forms reviewed by a psychiatrist; and (4) four prospective treatment cohorts followed for 12 months (depression N=137, AUD N=175, psychosis N=95, epilepsy N=42). Together these assessed all four steps of the care cascade -- contact coverage, clinical detection, initiation of minimally adequate treatment, and clinical/functional outcomes. Data were collected between January 2013 and February 2017 using instruments validated or culturally adapted for Nepal (PHQ-9, AUDIT, WHODAS 2.0, PANSS). The design was observational and uncontrolled -- before-after comparisons without a control or comparison group.
What They Found
Using 12-month routine service-utilisation data, contact coverage rose across all four disorders: from 0% to 12.2% for depression, 0% to 7.5% for AUD, 3.2% to 53.4% for psychosis, and 1.3% to 13.0% for epilepsy. However, the independently-sampled community survey found no statistically significant change in population-level treatment-seeking (depression contact +3.3 percentage points, 95% CI -5.1 to 11.7; AUD contact among men +6.3 points, 95% CI -3.3 to 15.9). At facility level, health-worker detection of depression rose from 8.9% at baseline to 24.6% six months after mhGAP training (+15.7 points, Cohen's h=0.43) but fell back to 19.2% at 24 months; AUD detection rose from 1.1% to 60.0% at six months (+58.9 points, h=1.56) before dropping sharply to 12.1% at 24 months. Among those diagnosed, minimally adequate treatment was provided to 93.9% (depression) and 95.1% (AUD) of patients at 6 months, falling to 66.7% and 75.0% respectively at 24 months. In the 12-month treatment cohorts, patients showed small-to-moderate symptom improvements: depression PHQ-9 scores fell by 7.22 points (95% CI -9.54 to -4.89, d=-0.58), AUD AUDIT scores fell by 9.68 points (d=-0.34), and psychosis PANSS scores fell by 6.42 points (d=-0.43); the epilepsy cohort showed no significant reduction in seizures or functioning. Combining all cascade stages, the authors calculate the programme achieved overall 'effective coverage' of only 1 in 34 people with depression and 1 in 23 people with AUD -- underscoring that facility-level gains do not automatically translate into population-level treatment-gap closure.
What This Means for Nepal
This is the flagship district-level test of the mhGAP task-sharing model the knowledge base identifies as grounding Nepal's mental-health evidence base, conducted in Chitwan under the same tiered primary-care structure (health posts, primary healthcare centres, district hospital) that MoHP would need to replicate nationally. It demonstrates that brief training (5-9 days) plus quarterly supervision can rapidly build primary health workers' capacity to detect and adequately treat depression and AUD -- directly relevant given Nepal's psychiatrist density of only ~0.17-0.22 per 100,000, heavily concentrated around Kathmandu, and its still-unenacted Mental Health Act (only a 1996 Policy and 2020 Strategy exist). But the finding that only 1 in 34 people with depression and 1 in 23 with AUD achieved effective coverage overall, alongside AUD detection attrition from 60% (6 months) to 12% (24 months), gives two concrete lessons for any national mhGAP scale-up: (1) MoHP/DoHS should budget for sustained, ongoing supervision -- not front-loaded training alone -- to prevent the detection and treatment-quality attrition documented here between 6 and 24 months; (2) even with the MHCP's Community Informant Detection Tool (CIDT) already active at community level, population-level contact coverage did not significantly improve, so scale-up should pair facility training with adequately resourced, adequately powered community demand-generation -- through FCHVs and stigma-reduction efforts -- rather than assuming facility-side investment alone will close Nepal's >90% treatment gap.
Contextualisation
Nepal has no enacted Mental Health Act -- only a 1996 Mental Health Policy and a 2020 Strategy & Action Plan -- and carries a treatment gap of over 90% for common mental disorders, with psychiatrists concentrated overwhelmingly around Kathmandu (national density ~0.17-0.22 per 100,000). This PRIME programme evaluation, conducted in Chitwan district as the district-level mhGAP pilot the knowledge base identifies as grounding Nepal's mental-health evidence, is the most detailed real-world test of task-shared, primary-care mental healthcare in the country to date. Its four-stage evaluation of the care cascade -- contact, detection, treatment initiation, and outcomes -- is directly relevant to MoHP's stalled ambitions to scale mhGAP nationally, but its finding that only 1 in 34 people with depression and 1 in 23 with AUD achieved effective coverage overall is a critical caution against treating facility-level gains alone as evidence that Nepal's treatment gap is closing.