Cochrane Kelsey M, Williams Brock A, Fischer Jordie A J, Samson Kaitlyn L I, Pei Lulu X, Karakochuk Crystal D
Food, Nutrition, and Health, Faculty of Land and Food Systems, The University of British Columbia, Vancouver, British Columbia, Canada.
Curr Dev Nutr. 2020 Sep 24;4(10):nzaa152. doi: 10.1093/cdn/nzaa152. eCollection 2020 Oct.
Regression to the mean (RTM) is a statistical phenomenon where second measurements are more likely to be closer to the mean. This is particularly observed in those with baseline values further from the mean. Anemic individuals (hemoglobin <120 g/L) are often recruited when evaluating iron supplementation programs, as they are more likely to elicit a greater hemoglobin response; however, they are also at greater risk for RTM as their baseline values are lower than the overall population mean.
The aim was to calculate and apply RTM to a previously conducted iron supplementation trial of women in Cambodia at increasingly severe baseline anemia cutoffs (hemoglobin <120 g/L, <115 g/L, and <110 g/L).
Women received either 60 mg/d iron (= 191) or placebo (= 185) for 12 wk. Hemoglobin was measured at baseline and at 12 wk (endline), and change in hemoglobin was calculated in each group for each cutoff. RTM was calculated in the placebo group at each cutoff and applied to the change observed at each cutoff in the iron group to obtain the RTM-free effect.
In the placebo group, mean change in hemoglobin increased as cutoffs became more extreme (0.9 g/L to 1.9 g/L in those with baseline hemoglobin <120 g/L and <110 g/L, respectively). RTM estimates similarly increased: 1.0 g/L (<120 g/L), 1.3 g/L (<115 g/L), and 1.8 g/L (<110g/L). When applying RTM to the iron group, we found that ∼10% of the "treatment effect" could be attributable to RTM at each cutoff. However, iron supplementation was still effective in increasing hemoglobin, with an increased effect in those with lower baseline values, as proven by the RTM-free effect at each cutoff: 8.7 g/L (<120 g/L), 10.9 g/L (<115 g/L), and 13.6g/L (<110 g/L).
RTM may have accounted for ∼10% of the observed change in hemoglobin following iron supplementation; however, appropriate use of a placebo group in the statistical analyses of the trial controls for this potential RTM effect.
均值回归（RTM）是一种统计现象，即第二次测量值更有可能接近均值。在基线值离均值较远的人群中尤其明显。在评估铁补充计划时，通常会招募贫血个体（血红蛋白<120 g/L），因为他们更有可能引发更大的血红蛋白反应；然而，由于他们的基线值低于总体人群均值，他们也面临更大的均值回归风险。
目的是计算均值回归并将其应用于柬埔寨女性先前进行的铁补充试验中，采用越来越严格的基线贫血临界值（血红蛋白<120 g/L、<115 g/L和<110 g/L）。
女性接受60 mg/d铁剂（=191人）或安慰剂（=185人），为期12周。在基线和12周（终点）时测量血红蛋白，并计算每组在每个临界值下血红蛋白的变化。在安慰剂组中计算每个临界值下的均值回归，并将其应用于铁剂组在每个临界值下观察到的变化，以获得无均值回归效应。
在安慰剂组中，随着临界值变得更严格，血红蛋白的平均变化增加（基线血红蛋白<120 g/L和<110 g/L的人群中分别为0.9 g/L至1.9 g/L）。均值回归估计值也类似增加：1.0 g/L（<120 g/L）、1.3 g/L（<115 g/L）和1.8 g/L（<110 g/L）。将均值回归应用于铁剂组时，我们发现在每个临界值下，约10%的“治疗效果”可归因于均值回归。然而，铁补充剂在增加血红蛋白方面仍然有效，在基线值较低的人群中效果增强，这在每个临界值下的无均值回归效应中得到证实：8.7 g/L（<120 g/L）、10.9 g/L（<115 g/L）和13.6 g/L（<110 g/L）。
均值回归可能占铁补充后观察到的血红蛋白变化的约10%；然而，在试验的统计分析中适当使用安慰剂组可控制这种潜在的均值回归效应。
