Digital Versus In-Person Physical Therapy in Adults With Musculoskeletal Conditions: Retrospective Matched-Cohort Analysis of Surgery and Low-Value Surgical Rates

Background: Musculoskeletal (MSK) disorders are leading causes of disability worldwide, with clinical guidelines recommending physical therapy–based interventions. Digital MSK programs offer an alternative to address logistical and socioeconomic barriers to regular in-person care. However, evidence comparing surgical use between digital and in-person physical therapy remains limited, particularly for low-value procedures. Objective: This study aimed to evaluate the 12-month incidence of MSK surgery and low-value surgical procedures among participants initiating a multimodal Digital Care Program (DCP) versus a matched-cohort initiating in-person physical therapy. Methods: Retrospective, matched-cohort study, using exact and propensity matching, with a Health Insurance Portability and Accountability Act (HIPAA)–deidentified US nationwide merged claims dataset (July 2022-February 2025). Eligible adults had spine, knee, hip, or shoulder conditions, ≥24 months uninterrupted health insurance coverage to an employer-sponsored DCP, and no MSK surgery in the prior year. The intervention group (IG) participated in a DCP combining exercise, education, and cognitive behavioral therapy, with real-time biofeedback and remote physical therapist oversight. The comparator group (CG) initiated in-person physical therapy, identified from a third-party claims database, using relevant MSK ICD-10 (International Statistical Classification of Diseases, Tenth Revision) codes as primary diagnosis. The primary outcome was the incidence of any MSK surgery within 12 months; the secondary outcome was the incidence of low-value surgery based on Choosing Wisely–aligned definitions. Cohort characteristics were compared using t test and chi-square test. Risk ratios (RRs) were calculated overall and by pain site, age group, and Social Deprivation Index. Results: In a matched cohort of 4190 individuals, predominantly middle-aged (~52 years old) women (1335/2095, 63.7%) with spinal pain (1123/2095, 53.6%), participation in the digital program was linked to a 58% (95% CI 49-66) lower relative risk of surgery at 12 months compared to those initiating in-person physical therapy (RR 0.42, 95% CI 0.34-0.52; E-value=4.19 [lower CI 3.29]). For surgeries categorized as low-value, IG was associated with 82% (95% CI 68-90) lower relative risk (RR 0.17, 95% CI 0.09-0.31; E-value=11.24 [lower CI 5.91]). Overall MSK surgical trends were consistent across pain sites, with greatest relative differences for knee (IG: 40/414 9.7% vs CG: 122/414, 29.5%; RR 0.26; 95% CI 0.17-0.38) followed by hip (19/203, 9.4% vs 42/203, 20.7%; RR 0.40; 95% CI 0.22-0.71). Lower surgery incidences in the IG (overall and low-value) were found across all socioeconomic and age strata. Conclusions: This real-world study demonstrated, for the first time, that participation in a digital MSK program was associated with substantially lower incidences of both overall and low-value surgeries compared to those who opted for in-person physical therapy among commercially-insured adults. These findings suggest that digital MSK programs can mitigate access barriers, promote adherence to guideline-concordant care, and reduce unnecessary procedures, including among underserved populations.