An Exploration of Physical Therapy Outcomes and Psychometric Properties of the Patient-Specific Functional Scale After an Interdisciplinary Pain Management Program

Abstract

Objectives: The Patient-Specific Functional Scale (PSFS) has been used to assess physical functioning in various chronic pain populations. There is a lack of evidence for its use with diverse pain populations who have high-impact chronic pain (HICP). The primary aim of this study was to assess the psychometric properties of the PSFS with a heterogeneous sample of patients with HICP. Materials and Methods: A retrospective observational study of prospective functional outcomes collected before and after participation in an intensive pain management program: PSFS, Disabilities of the Arm, Shoulder, and Hand, Lower Extremity Functional Scale, Neck Disability Index, Oswestry Disability Index, multidimensional patient impression of change, pain intensity, pain limitation, and fear-avoidance from the charts of 509 intensive pain management program completers. The reliability, validity, and responsiveness of the PSFS were analyzed using Cronbach α, Pearson correlations, and receiver operating characteristics. Results: Statistically significant improvements were found for all outcomes after program participation (all P < 0.0001). Correlations of the PSFS with similar outcomes (convergent validity) were lower than expected (r = 0.16 to 0.33). A low correlation (r = −0.12) with an unrelated measure, fear avoidance, supported divergent validity. PSFS change score correlations with similar outcome measures ranged from 0.46 to 0.53. The area under the curve values for the PSFS ranged from 0.801 to 0.857, suggesting a moderate ability for the PSFS to detect improvement. Discussion: The psychometric properties of the PSFS showed mixed support for construct validity but good support for the responsiveness of the PSFS when used with a heterogeneous sample of patients with HICP.