OK: 1. Research on treatments should follow a pathway. It’s meant to begin with efficacy “can it work?” along with mechanistic studies on “how does it work?”. Once efficacy has been established, it’s time to move on to effectiveness “does it work for in the intended population”? 
https://twitter.com/DanBockmannDC/status/1624512949388824576
2. Efficacy studies are your classic placebo-controlled trial: ideal conditions, high internal validity. I think people misunderstand the role of efficacy studies. We don’t need many of these to answer “can it work?”.
3. Exercise in OA had shown efficacy, and the field moved on to effectiveness ages ago. That ship has sailed. That’s how the pathway works. Was it too early? Not good enough data? Maybe. But the ship sailed long ago, and how effectiveness is established too.
4. But when efficacy is weak or inconsistent between studies with different intervention protocols, as is the case in Ex for OA knee OA, we need more “how does it work?” mechanistic studies to understand and improve the intervention.
5. Ex for OA has very different types of exercise, with different mechanisms of action, and has different effects in terms of pain, functioning, coping, quality of life. We need to know more from mechanistic studies on “how does it work?” for each of these.
6. But effectiveness is clearly proven in lots of populations and settings, all kinds of exercise type, >100 trials for knee OA. doi.org/10.1016/j.ocar…
7. We don’t need more confusingly-designed efficacy studies that create more questions than answers. We don't even need more effectiveness studies that replicate existing knowledge. We need to improve the treatment for better effect - that will require some hard clear thinking...
8. And meanwhile we need to continue on with pragmatic implementation, because Ex is highly cost-effective for the healthcare system while helping many (not all) people with #osteoarthritis, saving $ in other healthcare consumption and societal costs. oarsijournal.com/article/S1063-…
• • •
Missing some Tweet in this thread? You can try to
force a refresh
