Reply to Misinterpretation of the Longitudinal Results in the Randomized, Double-Blind, Placebo-Controlled Trial of Infrared-Laser Moxibustion for Cancer-Related Fatigue
Reply to Misinterpretation of the Longitudinal Results in the Randomized, Double-Blind, Placebo-Controlled Trial of Infrared-Laser Moxibustion for Cancer-Related Fatigue
复制标题
对红外激光艾灸治疗癌症相关疲劳的随机、双盲、安慰剂对照试验中纵向结果的误解的答复
作者：
Huijuan Mao;Ke Cheng;Jun J. Mao;Xuefen Chen;Xueyong Shen
谢谢你给我机会回应莫兰等人。关于我们的论文的“致编辑的信”。我们使用混合效应线性模型对我们的纵向数据进行了重复测量分析，但由于期刊的字数限制，我们没有在手稿中提供统计输出。混合线性模型显示主效应和交互作用检验结果：组因素（P= 0.013）和时间因素（P< 0.0001）均具有统计学意义。组因素和时间因素之间的相互作用显示出趋势，但未达到统计学显著性（P= 0.129）。正如我们在结局测量部分的手稿中所报告的，试验的主要终点是在治疗结束时（即第4周）。我们的分析（和Moran et al.的分析结果一致，表明两组疲劳评分差异有统计学意义，有利于真激光灸。Cohen的标准差为1.141，表明具有较大的相关临床效应。正如我们在手稿中所报告的，与随机对照试验的设计一致，两组之间基线时的疲劳评分相似（P= 0.407）。因此，我们在第2周、第4周和第8周对组间疲劳评分进行的比较是适当的。第2周的差异证据弱于第4周和第8周;然而，由于激光灸通常需要四周的完整疗程才能有效治疗，我们并不期望在各个时间点的效应量相似。
Thank you for the opportunity to respond to Moran et al.’s “Letter to the Editor” regarding our paper1. We performed repeated measures of analysis of our longitudinal data using the mixed effects linear model, but did not present the statistical output in our manuscript due to the journal’s word limits. The mixed linear model showed the results of the main effect and interaction test: the group factor (P= 0.013) and time factor (P< 0.0001) were both statistically significant. The interaction between the group factor and time factor showed a trend but did not reach statistical significance (P= 0.129).As we reported in our manuscript in the Outcome Measures section, the primary endpoint of the trial was at the end of treatment (ie, Week 4). Both our analysis (and Moran et al.’s analysis were consistent and indicated that the fatigue scores were statistically different between the two groups favoring true laser moxibustion. The Cohen’sd was 1.141, indicating a large and relevant clinical effect. As we reported in the manuscript, and consistent with the design of a randomized controlled trial, the fatigue scores were similar (P= 0.407) at baseline between the two groups. Thus our comparisons of the fatigue scores between groups at weeks 2, 4, and 8 were appropriate. Evidence of the difference at Week 2 was weaker than it was at Week 4 and Week 8; however, as laser moxibustion typically requires a full course of four weeks to be therapeutically effective, we did not expect the effect size to be similar across the time points.