The randomised controlled trial (RCT) method provides conclusions about a treatment’s average effect, one might say a measure of average lives guiding policy decisions. Policies are aimed at the majority not you as an individual. When an RCT shows the average treatment effect is less than placebo the media headline might read treatment has no effect. Actually the methodology of an RCT does not negotiate all biases reported in the literature and does not address more complex scenarios such as those found in medicine. The simple individual level treatment effect is lost yet as an individual I’d like to know how treatments would effect me, not the crowd.
The largest trial for Chiropractic Care was criticised for having too small a sample size by something like 18 people (out of 1200 odd that were recruited) Yet the averaging out from the data of the 1200 participants were not accepted as ‘real’. The data that showed Chiropractic to be better than best current back pain services. Yet for the past 95 years in the UK people have been paying for chiropractic care, and since 1952, paying for chiropractic care in addition to the NHS. Surely the test of time would suggest that it is worth the organisational effort to winkle out the real effect for the most common ailment in the country, back pain. An effort to help sooth GP time demands and release them for more appropriate appointments. Chiropractic is a small minority profession of just 3300. The irony is that the healthcare profession that provides Care by hand needs a helping hand to be seen by policy makers.
In essence, it could take a few weeks to get a GP appointment and then another week or so of watchful waiting and lifestyle advice leads to a further 3 month wait for Physiotherapy, where the first two sessions are assessment and reassurance. Is waiting for treatment on the NHS really worth it when you can access Chiropractic Care, directly, gaining a treatment appointment within a few days?
You only get one body…
Krauss A (2018) Why all Randomised Controlled Trials Produce Biased Results. Annals of Medicine. 50 (4) 312-322