Self-reported disability trajectories and their predictors among patients receiving care by physical therapists for musculoskeletal conditions: a retrospective analysis of registry data

Abstract

Objectives: To identify clustered trajectories of self-reported disability following the initiation of care by a physical therapist in outpatient orthopaedic settings and to determine baseline factors that distinguish between different trajectories.

Design: Retrospective cohort study using electronic health record and patient-reported outcome data.

Setting: Data were extracted from the ATI Patient Outcomes Registry, encompassing patient encounters from over 900 ATI outpatient physical therapy clinics in 26 states across the USA.

Participants: Patients receiving physical therapy after surgery were excluded. The final analytical sample included 597 245 unique patients initiating care between 1 January 2016 and 31 December 2021 for management of a spine, upper extremity or lower extremity musculoskeletal condition.

Interventions: Patients received treatments which could include strengthening and range of motion exercises, manual therapy, education, functional training and pain-relieving modalities.

Primary and secondary outcome measures: The primary outcome was trajectory of self-reported, joint-specific disability measure scores up to 6 months following initial evaluation.

Results: Three distinct disability trajectory clusters were identified (proportion of sample; beta coefficient (95% CI)): significant immediate worsening (3.4%; -1.31 (-1.33, -1.28)), significant gradual improvement (61.4%; (0.36 (0.35, 0.36)) and minimal change (35.2%; -0.20 (-0.21, -0.19)). Results were similar when stratified by primary diagnosis of upper extremity, lower extremity or spine conditions, with small differences in the relative proportion of trajectory class membership by body region. Predictive factors for less favourable disability trajectories included older age, lower physical and mental health scores, body region, higher social deprivation index, insurance type and certain comorbidities.

Conclusions: Most patients showed improvement in disability after exposure to treatment by a physical therapist, but a notable proportion experienced minimal change or worsening. Multiple demographic, physical, mental and social health factors differentiated trajectory class membership, highlighting opportunities to improve how and to whom this type of guideline-supported non-pharmacological care is delivered.

Keywords: Clinical Reasoning; Health Services; Musculoskeletal disorders; PAIN MANAGEMENT; Patient Reported Outcome Measures; REHABILITATION MEDICINE.

Figure 1

Study sample derivation from the ATI Patient Outcomes Registry. SI: Sacroiliac

Figure 2

Disability trajectories for overall (A), spine (B), upper extremity, (C) and lower extremity (D) cohorts.

Figure 3

Marginal effects plots for overall sample—significant immediate worsening versus minimal change (reference). LASSO, least absolute shrinkage and selection operator. API, Auto-personal injury. SI, sacroiliac.

Figure 4

Marginal effects plots for overall sample—significant gradual improvement versus minimal change (reference). LASSO, least absolute shrinkage and selection operator. API, Auto-personal injury. SI, sacroiliac.