Single-item screens identified patients with elevated levels of depressive and somatization symptoms in outpatient physical therapy

Objective Develop efficient and accurate screening tools to identify elevated levels of depressive or somatization symptoms, which can adversely affect functional status outcomes. Methods We conducted a secondary analysis of prospectively collected depressive and somatization symptoms (Symptom Checklist 90-Revised) data from 10,920 patients receiving outpatient physical therapy for a variety of neuromusculoskeletal diagnoses. Item response theory methods were used to analyze data, with particular emphasis on differential item functioning among groups of patients, and to identify potential screening items. Screening item accuracy for identifying patients with elevated symptoms was assessed with receiver-operating characteristic analyses. Results Seven items for depressive and 10 items for somatization symptoms represented unidimensional scales. Differential item functioning was negligible for demographic and clinical variables known to affect functional status outcomes. Items providing maximum information at the 88th percentile for depressive and 77th percentile for somatization scales accurately dichotomized patients into elevated versus not elevated symptom levels. Conclusions Lack of differential item functioning suggested depressive and somatization screening could be useful in routine clinical practice and allowed the development of single-item screens that accurately identified patients with elevated depressive or somatization symptoms. Item response theory-based single-item screens may facilitate evaluation and management of heterogeneous populations receiving outpatient physical therapy.