Benefits of more granular measurement of depression: Deconstructing the PHQ-9 into the PHQ-22

Background: Depression is a leading contributor to global disease burden, yet its measurement is complicated by substantial heterogeneity in how symptoms are assessed. The widely used Patient Health Questionnaire-9 (PHQ-9) contains several double-barreled items that merge distinct symptom experiences within single questions, potentially hiding important clinical differences. However, the empirical impact of disaggregating these items remains largely unexplored. Thus, to address this gap, the present study aimed to develop and psychometrically evaluate the PHQ-22, an expanded instrument that decomposes all double-barreled PHQ-9 items into their constituent symptoms.Methods: Using data from a Czech community sample (N \= 6,390), we employed network psychometric methods including Exploratory Graph Analysis, bootstrap stability testing, and measurement invariance analysis to evaluate the PHQ-22's psychometric properties, and compared its criterion validity against the PHQ-9 using elastic net regression. Results: The PHQ-22 network exhibited a four-community structure with adequate stability. Measurement invariance across gender groups was supported. Criterion validity analyses demonstrated that PHQ-22 items collectively showed improved prediction of depression related impairment, stress and loneliness outcomes compared to PHQ-9 items while accounting for model complexity. In 34.2% of all comparisons, items derived from the same PHQ-9 parent item showed significantly different associations with stress and loneliness outcomes after False Discovery Rate correction.Conclusion: These findings provide evidence that separating double-barreled depression items enhances criterion validity and reveal symptom-specific patterns obscured by composite items. However, further validation in clinical samples is needed.