Although initially more expensive, a physical therapy course for knee osteoarthritis is more cost-effective after 1 year than a course of glucocorticoid injections, according to data published in JAMA Network Open.

“A steroid injection is a common treatment for OA, as well as referrals to physical therapy for exercise. All of the clinical practice guidelines from major organizations advise exercise as a core treatment recommendation. However, most of the data regarding actual care patterns that we have available from large health systems show that most people don’t actually get both.

“The real-world challenge is that an injection is easy to get and usually much cheaper,” he added. “Physical therapy requires more time — eight to 12 visits — and often with co-pays for each visit. In reality, that may push patients towards considering an injection.”

— Daniel I. Rhon, DPT, DSc, PhD, of the Brooke Army Medical Center, JBSA Fort Sam Houston, in Texas

Secondary Analysis

To compare the incremental cost-effectiveness between physical therapy and intra-articular glucocorticoid injection, as initial treatment strategies, Rhon and colleagues conducted a secondary analysis of data from a previous 1-year, randomized clinical trial from 2017. According to the researchers, that trial reported health care system costs, including knee-specific care costs, and quality-adjusted life-years. The health economist and team of statisticians working with Rhon and colleagues were blinded to the primary study’s results.

In the primary study, researchers recruited and randomized 156 patients with radiographic OA, from two large US military hospitals, 1:1 to receive either intra-articular glucocorticoid injections or physical therapy. These participants were followed for 1 year.

For the secondary analysis, Rhon and colleagues examined incremental cost-effectiveness between the two groups as the primary outcome. The researchers used acceptability curves of bootstrapped incremental cost-effectiveness ratios (ICERs) to identify the proportion of ICERs under the specific willingness-to-pay level, defined as $50,000 to $100,000. Rhon and colleagues reported health care system costs — both total and knee related — and health-related QALYs.

According to the researchers, mean 1-year knee-related medical costs were $2,113 (standard deviation = $4,224) among patients who received glucocorticoid injections, and $2,131 (SD = $1,015) in the physical therapy group. Meanwhile, the mean difference in QALYs significantly favored physical therapy at 1 year (0.076; 95% CI, 0.02-0.126), the researchers wrote. Physical therapy was more cost-effective than glucocorticoids, with an ICER of $8,103 for knee-related medical costs and a 99.2% probability that results settle below the willingness-to-pay threshold of $100,000.

In addition, four patients who received glucocorticoids ultimately underwent surgery, compared with zero in the physical therapy group.

“Our study showed that when you look at the quality of life gained over the following year — the measure commonly used to establish cost-effectiveness — even though physical therapy was likely the more expensive option initially, it was cost-effective at 1 year,” Rhon says. “Patients that had physical therapy were less likely to have surgery, and most patients ended up having more than 1 injection.

“All of this should be considered when making the decision to have an injection instead of engaging with physical therapy,” he adds. “The idea of a quick injection and then you’re done may not necessarily pan out that way in every case — or even most cases. Maybe the point here is to consider the long game when you are considering what treatment is best.”

[Source: Healio Rheumatology]