Effects of a 12-Week Digital Care Program for Chronic Knee Pain on Pain, Mobility, and Surgery Risk: Randomized Controlled Trial.

ABSTRACT

BACKGROUND: Chronic knee pain, most commonly caused by knee osteoarthritis, is a prevalent condition which in most cases can be effectively treated through conservative, non-surgical care involving exercise therapy, education, psychosocial support, and weight loss. However, most people living with chronic knee pain do not receive adequate care, leading to unnecessary use of opiates and surgical procedures. OBJECTIVE: Assess the efficacy of a remotely delivered digital care program for chronic knee pain. METHODS: We enrolled 162 participants into a randomized controlled trial between January and March 2017. Participants were recruited from participating employers using questionnaires for self-assessment of their knee pain, and randomized into treatment (n=101) and control (n=61) groups. Participants in the treatment group were enrolled in the Hinge Health digital care program for chronic knee pain. This is a remotely delivered, home-based 12-week intervention that includes sensor-guided exercise therapy, education, cognitive behavioral therapy, weight loss, and psychosocial support through a personal coach and team-based interactions. The control group received three education pieces regarding self-care for chronic knee pain. Both groups had access to treatment-as-usual. The primary outcome was the Knee Injury and Osteoarthritis Outcome Score (KOOS) Pain subscale and KOOS Physical Function Shortform (KOOS-PS). Secondary outcomes were visual analog scales (VAS) for pain and stiffness respectively, surgery intent, and self-reported understanding of the condition and treatment options. Outcome measures were analyzed by intention to treat (excluding 7 control participants who received the digital care program due to administrative error) and per protocol. RESULTS: In an intent-to-treat analysis the digital care program group had a significantly greater reduction in KOOS Pain compared to the control group at the end of the program (greater reduction of 7.7, 95% CI 3.0 to 12.3, P=.002), as well as a significantly greater improvement in physical function (7.2, 95% CI 3.0 to 11.5, P=.001). This was also reflected in the secondary outcomes VAS pain (12.3, 95% CI 5.4 to 19.1, P<.001) and VAS stiffness (13.4, 95% CI 5.6 to 21.1, P=.001). Participants' self-reported likelihood (from 0% to 100%) of having surgery also reduced more strongly in the digital care program group compared to the control group over the next 1 year (-9.4 percentage points, pp, 95% CI -16.6 to -2.2, P=.01), 2 years (-11.3 pp, 95% CI -20.1 to -2.5, P=.01), and 5 years (-14.6 pp, 95% CI -23.6 to -5.5, P=.002). Interest in surgery (from 0 to 10) also reduced more so in the digital care program compared to control group (-1.0, 95% CI -1.7 to -0.2, P=.01). Participants' understanding of the condition and treatment options (on a scale from 0 to 4) increased more substantially for participants in the digital care program than those in the control group (0.9, 95% CI 0.6 to 1.3, P<.001). In an analysis on participants that completed the intervention (per protocol analysis) all primary and secondary outcomes remained significant at greater effect magnitudes compared to intention to treat, with those completing the program showing a 61% (95% CI 48 to 74) reduction in VAS pain compared to 21% (95% CI 5 to 38) in the control group (P<.001). Accounting for the cost of administering the program, we estimate net cost savings on surgery alone of US $4340 over 1 year and $7900 over 5 years for those participants completing the digital care program compared to those in the control group receiving treatment-as-usual. In an exploratory subgroup analysis including only participants exhibiting clinical symptoms of osteoarthritis the program proved equally effective. CONCLUSIONS: This trial provides strong evidence that a comprehensive 12-week digital care program for chronic knee pain, including osteoarthritis, yields significantly improved outcomes for pain, physical function, stiffness, surgery risk, and understanding of the condition, compared to a control group. TRIAL REGISTRATION International Standard Randomized Controlled Trial Number (ISRCTN) 13307390; http//www.isrctn.com/ISRCTN13307390 (Archived by WebCite at http//www.webcitation.org/6ycwjGL73).