Acupuncture for chronic knee pain: a randomised clinical trial. Authors' reply.

Drs White and Cummings state that we negatively reported findings of our ‘positive study’—namely, our trial evaluating acupuncture for chronic knee osteoarthritis (OA). Our study was not a ‘positive study’ as they suggest. Rather, our study showed no benefit of needle or laser acupuncture compared with sham laser acupuncture on either of our two primary outcomes (pain and function), and small clinically irrelevant benefits compared with no acupuncture for both laser and needle acupuncture. In accordance with best practice in the conduct and interpretation of clinical trials, we (1) a priori specified our primary outcome measures, including the minimum clinically important difference (MCID) on the scale of measurement for each outcome measure; (2) determined and recruited the sample size needed to show meaningful between-group differences in primary outcomes based on our chosen MCIDs and estimated SDs with adequate power; and (3) interpreted our findings based on both of these primary outcomes, their MCIDs and considering both control conditions included in our study design (ie, sham laser acupuncture and no acupuncture). Accordingly, the conclusion stating that our findings do not support acupuncture for these patients is appropriate and is the only conclusion that can (and should) be drawn from our study. We powered our study to detect the absolute MCID between groups in pain (1.8 units) and function (6 units), which have been recommended specifically for our chosen primary outcome measurement tools (numerical rating scale and Western Ontario and McMaster Universities OA Index, respectively). Estimated between-group differences did not reach these thresholds, nor were active acupuncture groups statistically different from sham laser. Drs White and Cummings do not like our choice of MCIDs for our primary outcomes. Their comments focus solely on MCIDs expressed as a dimensionless quantity (or, equivalently, in SD units). However, this approach confounds the absolute size of effect on the scale of the outcome measure with the variability (SD) of the outcome measure between patients. By their definition, two trials with exactly the same difference in mean outcome between treatments would result in a different MCID solely because of differing diversity of patient populations. Our approach of specifying an absolute MCID separates the definition of a clinically meaningful change from decisions regarding the patient population of interest, which clarifies the clinical interpretation of the trial results. Drs White and Cummings imply that our presentation of secondary results was unthoughtful because we did not quote the results from the supplemental content. The results in the main tables (tables 3 and 4) are based on analyses using multiple imputation of the nonnegligible amount of missing outcome data. The results in the supplemental content are based on complete case analyses, which require stricter assumptions about the missing outcome data than the main tables and, as such, are likely to be more prone to bias than the main analyses. We therefore concentrated interpretation on the most statistically sound analyses. Drs White and Cummings suggest our comparison of needle acupuncture with sham laser acupuncture is ‘clinically irrelevant’ because ‘sham laser is not an available therapy’. We are rather surprised by this statement. Of course sham laser is not an available therapy: sham treatments, by their very definitions, are fake treatments used for one reason only in clinical research—to isolate the active physiological effects of a given treatment from placebo contextual effects. If this assertion by the authors was true, then most pharmacological research needs to be discarded given that a sham treatment (placebo pill) is the gold standard control condition for drug clinical trials. An individual patient meta-analysis has shown that effect sizes for needle acupuncture compared with sham needle acupuncture are very small (0.16, 95% CI 0.07 to 0.25) and unlikely to be of any clinical relevance. It is unclear why Drs White and Cummings think we should have used a sample size large enough to detect such small and clinically meaningless differences between needle and sham laser acupuncture. They highlight that the National Institute for Health and Clinical Excellence (NICE) chose a threshold of 0.5 (standardised mean difference) to interpret findings of their meta-analysis when devising clinical practice guidelines for OA. However, Drs White and Cummings failed to state that NICE considered the comparison of acupuncture with sham acupuncture to be the most appropriate clinical comparison to assess the benefits and harms of acupuncture, which is why acupuncture is not recommended for OA by NICE in the 2014 guidelines. Drs White and Cummings state that our manual needle acupuncture was not optimal because we did not use electrical stimulation. However, the most up-to-date evidence from a 2013 meta-regression study that pooled data from 29 trials of acupuncture for chronic pain (including OA) showed that electrical stimulation did not significantly influence the effects of needle acupuncture on pain. Drs White and Cummings state that we should have concluded that ‘patients with OA knee should consider acupuncture as an option’. We agree with this statement when it is with reference to no treatment at all, as long as it is qualified with the appropriate evidence from the trial—that is, ‘patients with OA knee should consider needle, laser or sham laser acupuncture as an option’—because all were shown to be equally efficacious in comparison with no treatment. To omit the Letter