JMIR Publications

ABSTRACT

Background:

Musculoskeletal (MSK) disorders are leading causes of disability and healthcare spending worldwide. Clinical guidelines consistently recommend early, guideline-concordant physical therapy to reduce the need for surgical intervention. However, systemic, logistical, and socioeconomic barriers limit access to in-person care. Digital MSK programs offer an alternative delivery model that may improve access, adherence, and outcomes. Despite their growing adoption, evidence directly comparing surgical utilization between digital and in-person physical therapy remains limited, particularly regarding the reduction of low-value procedures.

Objective:

This study evaluated 12-month incidence of MSK surgery, including procedures classified as low-value, between participants in a multimodal digital care program (DCP) and a matched cohort initiating in-person physical therapy.

Methods:

We conducted a retrospective, matched cohort study, using exact and propensity matching, with claims data from July 2022 to February 2025. Eligible adults (≥18 years) had spine, knee, hip, or shoulder conditions, ≥12 months of continuous medical eligibility before and after the index date, and no MSK surgery in the prior year. The intervention group (IG) participated in an employer-sponsored DCP combining exercise, education, and cognitive behavioral therapy, delivered via a class II medical device with real-time biofeedback and remote physical therapist oversight. The comparator group (CG) included patients receiving in-person physical therapy, identified from a third-party U.S. nationwide claims database, using relevant MSK ICD-10 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. Odds ratios (ORs) were calculated overall and by pain site, age group, and Social Deprivation Index (SDI).

Results:

In a matched cohort of 4,236 individuals, participation in the digital program was linked to a 59% lower 12-month surgery incidence compared with in-person physical therapy (5.4% vs. 13.2%; OR 0.38, 95%CI 0.30; 0.47), and 77% lower incidence of surgeries categorized as low-value based on clinical guidelines (0.9% vs. 3.9%; OR 0.22 95%CI 0.14; 0.37). Overall MSK surgical trends were consistent across pain sites, with greatest relative differences for knee (9.5% vs 29.4%; OR 0.25; 95%CI 0.17;0.37) and hip (9.2% vs 20.9%; OR 0.39; 95% CI 0.22; 0.69), favoring IG. Lower surgery incidences in the IG (both overall and low-value) were found across all socioeconomic and age strata. Patients in the IG had higher engagement and reported clinically meaningful improvements in pain and function.

Conclusions:

Participation in a multimodal digital MSK program was associated with substantially lower incidences of both overall and low-value surgery compared with in-person physical therapy in a large, real-world, matched cohort of commercially-insured adults. These findings suggest that digital MSK programs can help address access barriers, promote adherence to guideline-concordant care, and reduce unnecessary procedures, including among underserved populations.