Social support best improves task-shared care

Task-shared psychosocial programs worked best when they strengthened social support, managed problems, and increased activity.
*Individual participant data component network meta-analysis of randomized trials; Level 1a.

Citation

Papola D, Tedeschi F, Efthimiou O, et al. Optimising and personalising task-shared psychosocial interventions for common mental disorders. Lancet Psychiatry. 2026;13:747-760. doi:10.1016/S2215-0366(26)00193-8

Background

Psychosocial treatments delivered by trained non-specialists can expand care for depression and anxiety, especially where specialists are scarce. This study examined which treatment parts add the most benefit.

Patients

Adults with depression, anxiety, or psychological distress in 34 randomized trials; total 11,963 participants. Exclusions included severe mental illness, substance use disorders, trauma-related disorders, pregnancy-related conditions, and cognitive impairment.

Intervention

Task-shared psychosocial interventions delivered by trained non-specialist providers, separated into key treatment components.

Control

Usual care, enhanced usual care, or another task-shared psychosocial intervention.

Outcome

Change in depression and anxiety symptoms at study endpoint, rescaled from 0 to 100, with higher scores meaning worse symptoms.

Follow-up Period

Study endpoint; timing varied across trials.

Results

Thirty trials provided individual participant data from 10,612 participants; four trials contributed summary data. Negative values mean lower symptom scores and greater benefit.

Component Symptom score effect, percentage points (primary outcome) Direction
Strengthening social support −9.48 (95% credible interval −13.29 to −6.60) Benefit
Behavioral activation −4.15 (95% credible interval −7.48 to −0.05) Benefit
Problem management −4.08 (95% credible interval −5.37 to −2.83) Benefit
Relaxation +7.97 (95% credible interval 3.35 to 12.11) Worse symptoms
Cognitive reframing +5.17 (95% credible interval 0.78 to 9.15) Worse symptoms

Incremental mean difference estimates the added effect of including one component within a treatment package.

Controls generally represented locally available usual or enhanced usual care. Component estimates are exploratory because components were not independently randomized.

Limitations

Nearly one-third of trials had high risk of bias, mostly from participant dropout. Findings apply to bundled, task-shared interventions, not stand-alone components or specialist psychotherapy. Ethnicity, culture, and local context were not available for personalization. Depression and anxiety were analyzed together.

Funding

European Commission; funder had no reported role.

Clinical Application

Prioritize social support, problem management, and activity planning in task-shared counseling; avoid routinely adding relaxation or cognitive reframing.