It happens all the time: The medical literature fails to support the effectiveness of an intervention, even though practitioners know for a fact the intervention actually does have a positive effect in some patients. But a new study suggests some researchers are now starting to look beyond the averages at factors that predict those positive responses. And that’s good news for practitioners who treat patients with knee osteoarthritis (OA).
It’s not unusual for studies of knee OA interventions to report that some patients have a positive response, some have a negative response, and some have no response. In terms of statistical significance, the inevitable conclusion is that the intervention being studied has no effect. In terms of clinical relevance, this type of study underscores the heterogeneous nature of the knee OA population and suggests that any intervention is probably worth a try if nothing else seems to work.
Researchers from the University of Salford in the UK found a similar range of outcomes in their recent study of lateral wedge orthoses in patients with medial tibiofemoral OA, which was epublished in March by Osteoarthritis & Cartilage. Sure enough, they found that, compared with a shoe-only condition, walking with lateral wedge orthoses was associated with a decreased peak external knee adduction moment (EKAM)—suggesting a positive response—in some patients, but an increased peak EKAM in other patients.
For the group overall, the peak EKAM did decrease significantly, but only by 5.85%. By comparison, when only those who experienced a decrease in peak EKAM were analyzed, that mean decrease was a more impressive 11.39%.
But the Salford investigators didn’t stop there. They analyzed the participants’ gait mechanics more closely, and found that those with a greater peak ankle eversion angle or a greater ankle angle at peak EKAM were more likely to respond positively to the intervention (see “Inclination insights: Ankle motion predicts wedge insole effects,” page 13).
Of course, these findings are still a bit removed from clinical application. They need to be replicated in a larger population, and it has yet to be determined whether the same ankle angle variables would also be predictive of longer-term effects of lateral wedge orthoses. And, given that most clinicians do not have regular access to sophisticated gait analysis equipment, it will also be important to identify variables that correlate with the predictors identified in the study—and can be easily measured in the clinic.
But the implications of this study go beyond its specific findings. Instead of asking simply whether an intervention was or was not effective for a population, the Salford researchers asked which patients were most likely to respond to an intervention, and why. That’s a fundamentally different approach to this problem, and one that I hope will be applied to studies of offloading braces, footwear, gait retraining, strength exercises, and other potential interventions for knee OA.
Knowing the knee OA population is heterogeneous isn’t nearly as helpful, clinically, as knowing which patients will respond to which interventions. The Salford study represents an important step toward achieving that goal.