ObjectiveThe purpose of this study was to evaluate whether physical therapy use influenced subsequent use of musculoskeletal-related surgeries, injections, magnetic resonance imaging (MRI), and other imaging. MethodsWe conducted a retrospective cohort study of patients aged 18 to 64 years who had an ambulatory care visit at the University of Utah system, after implementation of the 10th revision of the International Statistical Classification of Diseases and Related Health Problems with adequate data collection in the system at the time of the data pull, between October 1, 2015, and September 30, 2018. We identified patients (n = 85 186) who received care for a musculoskeletal condition (lower back pain, cervical, knee, shoulder, hip, elbow, ankle, wrist/hand, thoracic, and arthritis diagnoses). Regression analyses were used to evaluate the association between physical therapy use and medical care use while controlling for relevant factors. ResultsIn patients referred to physical therapy (n = 15 870), physical therapy use (n = 3812) was associated with increased MRI use (incidence rate ratio, 1.24; 95% confidence interval, 1.15-1.33; P < .001) and surgery use (incidence rate ratio, 1.11; 95% confidence interval, 1.00-1.23; P < .001). Several other factors were also associated with increased health care use, including being referred by an orthopedic provider, obesity, non–lower back pain diagnoses, and having 1 or more comorbidities. ConclusionOutpatient physical therapy use for musculoskeletal conditions in adult patients younger than 65 years at the University of Utah system, a mountain west United States academic health care system, was associated with increased rates of MRI and surgery. This finding is contrary to prior research suggesting that physical therapy improves outcomes in some diagnosis groups. A referral from an orthopedic provider, non–lower back pain diagnoses, and obesity were also associated with increased medical care utilization.