JMIR Publications

ABSTRACT

Background:

A 24-week self-directed digitally-delivered intervention, was found to improve pain and function in people with knee OA. However, it is possible that this intervention may be better suited to certain sub-groups of people with knee OA compared to others.

Objective:

To explore whether certain individual baseline characteristics moderate the effects of this intervention on changes in pain and/or function over 24-weeks in people with knee OA.

Methods:

An exploratory analysis was conducted on data from a randomized controlled trial involving 206 people with a clinical diagnosis of knee OA. The trial compared a self-directed digitally-delivered intervention comprising of web-based education, exercise and physical activity program supported by automated exercise behaviour change mobile phone text messages to web-based education alone (control). The primary outcomes were changes in overall knee pain (11-point numerical rating scale) and physical function (Western Ontario and McMaster Universities Osteoarthritis Index function subscale) at 24-weeks. Five baseline patient characteristics were selected as potential moderators: i) number of comorbidities; ii) number of other painful joints; iii) pain self-efficacy; iv) exercise self-efficacy; and v) self-perceived importance of exercise. Separate linear regression models for each primary outcome and each potential moderator were fit including treatment group, the moderator and an interaction between treatment group and moderator, adjusting for the outcome at baseline.

Results:

Participants with higher pain self-efficacy at baseline demonstrated smaller improvements in function at 24-weeks with the intervention compared to control. Specifically, at 24-weeks, within the intervention group, each 1-unit increase in baseline pain self-efficacy was associated with a reduction of 0.62 (95% CI (-1.93, 0.68)) units in the mean change in WOMAC function (i.e. less improvement), whereas in the control group each 1-unit increase in baseline pain self-efficacy was associated with a 1.52 (0.27, 2.78) unit improvement in mean change in WOMAC function (interaction p=0.02). There was only weak evidence that pain self-efficacy moderated the effect of the intervention on pain and that number of comorbidities, number of other painful joints, exercise self-efficacy or exercise importance moderated the effect of the intervention on pain or function.

Conclusions:

With the exception of pain self-efficacy, which moderated change in function but not pain, we found limited evidence that our selected baseline patient characteristics moderated intervention outcomes. This indicates people with a range of baseline characteristics respond similarly to the unsupervised, digitally-delivered exercise intervention. As these findings are exploratory in nature, they require confirmation in future studies. Clinical Trial: N/A