Recommendation Rates for Physical Therapy, Lifestyle Counseling, and Pain Medications for Managing Knee Osteoarthritis in Ambulatory Care Settings: A Cross-Sectional Analysis of the National Ambulatory Care Survey (2007-2015)

Objective: To describe and compare triennial rates of physicians' recommendations for physical therapy (PT), lifestyle counseling, and pain medication for knee osteoarthritis (OA) and to identify patient, physician, and practice factors associated with each treatment recommendation.

Methods: We conducted a cross-sectional analysis examining data between 2007 and 2015 from the National Ambulatory Medical Care Survey. Visits to orthopedists and primary care physicians for knee OA were identified and assessed for the following: PT referral, lifestyle counseling, nonsteroidal antiinflammatory drug (NSAID) prescriptions, and narcotics prescriptions. Triennial rates for each treatment were calculated. We examined associations between patient (e.g., race, insurance), physician, and practice factors (e.g., ownership, location) and treatments prescribed using multivariate logistic regression that accounted for complex sampling design.

Results: A total of 2,297 physician visits related to knee OA (~67 [±4] million weighted visits) were identified. For visits to orthopedists, PT and lifestyle recommendation rates declined (158 to 88 of 1,000 visits and 184 to 86 of 1,000 visits, respectively), while NSAID and narcotics prescriptions increased (132 to 278 of 1,000 visits and 77 to 236 of 1,000 visits, respectively) over time (P < 0.05). For visits to primary care physicians, there were no significant changes in rates of PT, lifestyle counseling, and narcotics prescriptions over time, while NSAIDs prescriptions increased (221 to 498 of 1,000 visits; P < 0.05). Treatment recommendations were associated with nonclinical factors, including practice type, location, and type of provider.

Conclusion: In patients with knee OA, PT and lifestyle counseling seem underutilized, while pain medication prescriptions increased during the investigated timeframe. Variation in treatment choices were associated with nonclinical factors. Future research is necessary to examine ways to improve PT and lifestyle utilization and reduce variation in care for knee OA.