Background/Purpose: Osteoarthritis (OA) is the most common form of arthritis. Physical therapy (PT) has previously been demonstrated to benefit patients with osteoarthritis, especially with early referral.  However, few studies have investigated factors affecting PT use and none have involved a Canadian population, where the mixed private and public payer system could affect usage rates. The objective of this study is to identify factors associated with PT use in a population-based cohort with OA-associated knee pain.

Methods: 255 subjects were recruited as a random population sample (MoDEKO study) in Greater Vancouver, Canada. Inclusion criteria were: 1) age 40–79 years, 2) pain, aching, or discomfort in/around the knee on most days of the month at any time in the past, and 3) any pain, aching, or discomfort in/around the knee in the past 12 months. Exclusions were: 1) inflammatory arthritis or fibromyalgia, 2) knee arthroplasty, 3) knee injury or surgery within the past 6 months, 4) referred pain from hips or back, and 5) inability to undergo MRI. PT use during the past 12 months (dependent variable) was ascertained by self-report. Independent variables included predisposing characteristics including age, sex, ethnicity, level of education (≤/> grade 12), and median household income using the 2006 Canadian Census data based on postal codes, as well as need characteristics including body mass index (BMI), self-reported knee swelling in the past 12 months, physician diagnosis of OA, radiographic severity (Kellgren–Lawrence grading (KL) ≥2/KL <2), and pain severity on flat walking, as assessed by the Western Ontario and McMaster Universities (WOMAC) Osteoarthritis Index.  Univariate logistic regression analysis was performed. Variables with p<0.20 were included in a multivariable logistic model to assess the association with PT use.

Results: 29 of 255 (weighted 12%) subjects reported PT use.  Mean age was 57, mean BMI 26.5, mean household median income $61,560, and mean WOMAC pain 17. 56% were female, 76% Caucasian, 77% attended post-secondary education, 39% were diagnosed with OA by a physician, 38% had KL ≥2, 51% reported knee swelling. In multivariable logistic regression analysis, female gender (OR 3.19, 95% CI 1.29-7.89) and self-reported knee swelling (OR 3.51, 95% CI 1.42-8.68) were significantly associated with PT use, while household income (as a factor of $10,000, OR 1.15, 95% CI 0.97-1.37), physician diagnosis of OA (OR 1.19, 95% CI 0.53-2.71), WOMAC pain on flat walking (as a factor of 10, OR 1.11, 95% CI 0.93-1.32) were included in the multivariable model, but were not statistically significant.

Conclusion: In this population-based study of subjects with knee pain, women and those with self-reported knee swelling were three times more likely to be current PT users. Interestingly, disease severity, physician diagnosis of OA, and household income were not significantly associated with current PT usage. Further longitudinal research will be important to identify other factors affecting PT use in order to target individuals at risk of underuse and/or overuse.

« Back to 2015 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/factors-associated-with-physical-therapy-use-in-osteoarthritis-of-the-knee-results-from-a-population-based-study/