Representation of racial and ethnic minorities in U.S. depression trials comparing psychotherapy vs. controls: A meta-analysis using state population benchmarks.

INTRODUCTION

Psychotherapies for depression are effective. To generalize to the diverse U.S. population, clinical trials must adequately represent racial/ethnic minority groups. Prior reviews of diversity of clinical trial participants are limited by incomplete reporting of race/ethnicity, use of national rather than state-level comparators, and pooling practices where a few diverse trials skew estimates.

METHODS

We evaluated racial/ethnic representation in U.S. randomized trials comparing psychotherapy with an inactive control in depressed adults using meta-analysis with state-specific population benchmarks. We assessed reporting of race/ethnicity and, using meta-analysis, evaluated minority representation relative to state population benchmarks. Analyses included sensitivity and subgroup tests (through 2001 and 2002 onward) and meta-regressions assessing links between minority representation and treatment effectiveness. Seventy-six trials (98 comparisons; 8819 participants) were included.

RESULTS

Race/ethnicity was not reported in 29.1% of studies. The "total minority" estimate suggested slight overrepresentation (+2%) but was inconsistent across sensitivity analyses and obscured subgroup differences in representation and temporal patterns. Subgroup analyses showed uneven representation, with Black participants overrepresented and Hispanic/Latino, Asian, and multiracial participants persistently underrepresented. Psychotherapy showed a moderate effect (g = 0.62). The apparent association between greater minority representation and lower treatment effectiveness was no longer significant after adjustment for trial characteristics or exclusion of high-risk-of-bias trials.

DISCUSSION: Incomplete reporting limits evaluation of representativeness. Aggregate "total minority" estimates obscures important subgroup differences, including persistent underrepresentation of some racial/ethnic groups. More complete reporting and subgroup-specific analyses are needed to evaluate racial/ethnic representation in depression treatment clinical trials.