Moore et al.1 studied postoperative supplementary (‘rescue’) analgesic consumption and described its skewed distribution among patients. Figure 1 histogram by Moore et al. is very similar to my 1994 article's Figure 1 cumulative distribution curve.2 On the basis of my Monte Carlo simulations,2 I share the perspective by Moore et al. that trials using cut-points of analgesic usage provide for easy interpretation with little loss of statistical power, but for reasons additional to those that they discuss.1 Wilcoxon–Mann–Whitney tests have slightly higher statistical power to differentiate between groups than do tests using a cut-point.2 In addition, limitations to the Wilcoxon–Mann–Whitney tests’ corresponding power analyses and confidence intervals, from the many patients with zero analgesic consumption, have been addressed.3,4 Yet, such statistical advances do not compensate for the clinical issue that covariates, such as patient age, can have such large influences on analgesic consumption that the covariates often should be included in trial designs and interpretations.5,6 That can easily be done with cut-points.
Assistance with the letter: none.
Sources of funding: none.
Conflict of interest: none.
Comment from the Editor: Professor Moore did not wish to respond to this letter.
1. Moore RA, Mhuircheartaigh RJ, Derry S, McQuay HJ. Mean analgesic consumption is inappropriate for testing analgesic efficacy in postoperative pain: analysis and alternative suggestion. Eur J Anaesthesiol
2. Dexter F. Analysis of statistical tests to compare doses of analgesics among groups. Anesthesiology
3. O’Brien RG, Castelloe J. Exploiting the link between the Wilcoxon-Mann-Whitney test and a simple odds statistic. In: American Statistical Association 2005 Proceedings of the Section on Statistical Computing
; 2006; Alexandria, Virginia, USA. pp. 2192–2198.
4. Divine G, Kapke A, Havstad S, Joseph CLM. Exemplary data set sample size calculation for Wilcoxon–Mann–Whitney tests. Stat Med
5. Ip HY, Abrishami A, Peng PW, et al. Predictors of postoperative pain and analgesic consumption: a qualitative systematic review. Anesthesiology
6. Reeves M. The influence of age on sample size calculation in acute pain trials using morphine consumption as an end point. Anesth Analg