Objectives To determine the diagnostic accuracy of the Whooley questions in the identification of depression; and, to examine the effect of an additional ‘help’ question.

Design Systematic review with random effects bivariate diagnostic meta-analysis. Search strategies included electronic databases, examination of reference lists, and forward citation searches.

Inclusion criteria Studies were included that provided sufficient data to calculate the diagnostic accuracy of the Whooley questions against a gold standard diagnosis of major depression.

Data extraction Descriptive information, methodological quality criteria, and 2×2 contingency tables were extracted.

Results Ten studies met inclusion criteria. Pooled sensitivity was 0.95 (95% CI 0.88 to 0.97) and pooled specificity was 0.65 (95% CI 0.56 to 0.74). Heterogeneity was low (I²=24.1%). Primary care subgroup analysis gave broadly similar results. Four of the ten studies provided information on the effect of an additional help question. The addition of this question did not consistently improve specificity while retaining high sensitivity as reported in the original validation study.

Conclusions The two-item Whooley questions have high sensitivity and modest specificity in the detection of depression. The current evidence for the use of an additional help question is not consistent and there is, as yet, insufficient data to recommend its use for screening or case finding.

Trial registration number CRD42014009695.