Shrier Ian, Christensen Robin, Juhl Carsten, Beyene Joseph
Centre for Clinical Epidemiology, Lady Davis Institute, Jewish General Hospital, McGill University, 3755 Cote Ste-Catherine Road, Montreal, Quebec H3T 1E2, Canada.
The Parker Institute, Bispebjerg and Frederiksberg Hospital, Nordre Fasanvej 57, DK-2000 Copenhagen F, Denmark.
J Clin Epidemiol. 2016 Dec;80:57-67. doi: 10.1016/j.jclinepi.2016.07.012. Epub 2016 Jul 30.
To compare results from meta-analyses for mean differences in minimal important difference (MID) units (MD), when MID is treated as a random variable vs. a constant.
Meta-analyses of published data. We calculated the variance of MD as a random variable using the delta method and as a constant. We assessed performance under different assumptions. We compare meta-analysis results from data originally used to present the MD and data from osteoarthritis studies using different domain instruments.
Depending on the data set and depending on the values of rho and coefficient of variation of the MID (CoV), estimates of treatment effect and P-values between an approach considering the MID as a constant vs. as a random variable may differ appreciably. Using our data sets, we provide examples of the potential magnitude. When rho = 0.5 and CoV = 0.8, considering MID as a constant overestimated the treatment effect by 33-110% and decreased the P-value for heterogeneity from above 0.95 to below 0.08. When rho = 0.8 and CoV = 0.5, the magnitude of the effects was similar.
Considering MID as a random variable avoids unrealistic assumptions and provides more appropriate treatment effect estimates.
比较当最小重要差异（MID）被视为随机变量与常数时，关于MID单位均值差异的荟萃分析结果。
已发表数据的荟萃分析。我们使用德尔塔法计算MD作为随机变量时的方差以及作为常数时的方差。我们评估了不同假设下的性能。我们比较了最初用于呈现MD的数据的荟萃分析结果，以及使用不同领域工具的骨关节炎研究数据的荟萃分析结果。
根据数据集以及MID的相关系数（rho）和变异系数（CoV）的值，将MID视为常数与视为随机变量的方法之间，治疗效果估计值和P值可能存在显著差异。利用我们的数据集，我们给出了潜在差异幅度的示例。当rho = 0.5且CoV = 0.8时，将MID视为常数会使治疗效果高估33%至110%，并使异质性的P值从高于0.95降至低于0.08。当rho = 0.8且CoV = 0.5时，效果幅度相似。
将MID视为随机变量可避免不切实际的假设，并提供更合适的治疗效果估计。
