ANALYSIS OF STATISTICAL TESTS TO COMPARE VISUAL ANALOG SCALE MEASUREMENTS AMONG GROUPS

Background: A common type of study performed by anesthesiologists dete rmines the effect of an intervention on pain reported by groups of pat ients. The goal of this study was to evaluate the effectiveness of t, analysis of variance (ANOVA), Mann-Whitney, and Kruskal-Wallis tests t o compare visual analog scale (VAS) measurements between two or among three groups of patients. These results may be particularly helpful du ring the design of studies that measure pain with a VAS. Methods: One VAS measurement was obtained from each of 480 nulliparous women in lab or who were receiving oxytocin (149), nalbuphine (159), or epidural bu pivacaine (172). Multiple simulated samples were then drawn from these data. These simulated samples were used in computer simulations of cl inical trials comparing VAS measurements among groups. t and ANOVA tes ts were performed before and after an arcsin transformation was used, to make the data closer to a normal distribution. VAS measurements wer e also compared after they were divided into five ranked categories. R esults: The statistical distributions of VAS measurements were not nor mal (P < 10(-7)). Arcsin transformation made the distributions closer to normal distributions. Nevertheless, no statistical test incorrectly suggested that a difference existed among groups, when there was no d ifference, more often than the expected rate. tor ANOVA tests had a sl ightly greater statistical power than the other tests to detect differ ences among groups. Because arcsin transformation both decreased diffe rences among means and reduced the variance to a lesser extent, it dec reased power to detect differences among groups. Statistical power to detect differences among groups was not less for a five-category VAS t han for a continuous VAS. Conclusions: We conclude that t and ANOVA, w ithout an accompanying arcsin transformation, are good tests to find d ifferences in VAS measurements among groups.