Item-level depression and anxiety change during residential substance use disorder treatment
Objective Aggregate scores on the Patient Health Questionnaire-9 (PHQ-9) and Generalized Anxiety Disorder-7 (GAD-7) may conceal differences in symptom-level change during residential substance use disorder (SUD) treatment. This study examined whether improvements in total scores concealed differential symptom change and residual burden.Method Routine item-level data from 98 adult treatment episodes were assessed a median of 20 days apart. After descriptive reconnaissance, a timestamped analysis plan was frozen before definitive modelling. Cumulative-logit generalised estimating equations tested Time × Item interactions; secondary analyses examined residual endorsement, complete resolution, threshold dependence and prespecified sensitivities.Results Change differed across PHQ-9 items, Wald χ2(8) = 27.32, Holm-adjusted p = .001, and GAD-7 items, Wald χ2(6) = 18.42, Holm-adjusted p = .005; both findings remained significant across all executable prespecified sensitivity analyses. PHQ-9 ordinal shifts ranged from OR = 10.37 for concentration difficulty to 37.37 for anhedonia, and GAD-7 shifts from 6.40 for irritability to 19.01 for trouble relaxing. Residual fatigue (40.8%), sleep disturbance (37.8%), nervousness (42.9%), uncontrollable worry (38.8%) and irritability (37.8%) remained common at discharge.Conclusions Aggregate improvement concealed non-uniform symptom change and residual burden. Item-level review may add clinically specific information to validated totals without treating individual items as diagnostic measures.
Authors
- Christopher Lomas (ORCID: https://orcid.org/0000-0003-2137-3124)
- Sam Lythgoe
Publication Details
- Journal
- Journal of Substance Use
- Published
- 2026-11-18
- DOI
- https://doi.org/10.1080/14659891.2026.2764871
- Primary Topic
- Anxiety, Depression, Psychometrics, Treatment, Cognitive Processes
- Type
- article
- Field-Weighted Citation Impact
- 0.00