Aaron R Caldwell, David B Allison, Andrew W Brown, Gary L Gadbury, Thirupathi Reddy Mokalla, R Drew Sayer, Andrew D Vigotsky
{"title":"Concerns Regarding the Standard Deviation of Individual Responses for Assessing Treatment Response Heterogeneity.","authors":"Aaron R Caldwell, David B Allison, Andrew W Brown, Gary L Gadbury, Thirupathi Reddy Mokalla, R Drew Sayer, Andrew D Vigotsky","doi":"10.1152/japplphysiol.00485.2025","DOIUrl":null,"url":null,"abstract":"<p><p>The estimation of treatment response heterogeneity (TRH) is increasingly important as medicine moves toward personalized approaches. While various statistical methods have been proposed to quantify TRH in parallel-group trials, the standard deviation of individual responses (SD<sub>IR</sub>) has gained prominence within physiological research. This method is intended to quantify individual response variation by comparing standard deviations of change scores between intervention and control groups. We acknowledge that SD<sub>IR</sub> represents an improvement over many other flawed approaches that often involve responder counting. However, SD<sub>IR</sub> has critical limitations: 1) it cannot overcome the fundamental problem of causal inference because the correlation between potential outcomes remains unidentifiable, 2) it is incorrectly predicated on the assumption that TRH is present only when treatment group variance exceeds control group variance, and 3) it is statistically inefficient. We present an alternative framework, which involves assessing heteroskedasticity and estimating the bounds for the standard deviation of treatment effects (SD<sub>D</sub>). The presence of heteroskedasticity between treatment groups is a sufficient but not necessary condition for the presence of TRH. Further, SD<sub>D</sub> makes fewer assumptions than SD<sub>IR</sub> and, therefore, paints a more complete picture of potential TRH. Using data from a published exercise physiology study, we demonstrate how SD<sub>D</sub> can better characterize uncertainty in TRH estimation. We recommend researchers probe TRH by assessing heteroskedasticity, providing bounds for SD<sub>D</sub>, and estimating outcome distributions and probabilities while carefully crafting the theoretical rationale for the presence of TRH.</p>","PeriodicalId":15160,"journal":{"name":"Journal of applied physiology","volume":" ","pages":""},"PeriodicalIF":3.3000,"publicationDate":"2025-09-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of applied physiology","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1152/japplphysiol.00485.2025","RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"PHYSIOLOGY","Score":null,"Total":0}
引用次数: 0
Abstract
The estimation of treatment response heterogeneity (TRH) is increasingly important as medicine moves toward personalized approaches. While various statistical methods have been proposed to quantify TRH in parallel-group trials, the standard deviation of individual responses (SDIR) has gained prominence within physiological research. This method is intended to quantify individual response variation by comparing standard deviations of change scores between intervention and control groups. We acknowledge that SDIR represents an improvement over many other flawed approaches that often involve responder counting. However, SDIR has critical limitations: 1) it cannot overcome the fundamental problem of causal inference because the correlation between potential outcomes remains unidentifiable, 2) it is incorrectly predicated on the assumption that TRH is present only when treatment group variance exceeds control group variance, and 3) it is statistically inefficient. We present an alternative framework, which involves assessing heteroskedasticity and estimating the bounds for the standard deviation of treatment effects (SDD). The presence of heteroskedasticity between treatment groups is a sufficient but not necessary condition for the presence of TRH. Further, SDD makes fewer assumptions than SDIR and, therefore, paints a more complete picture of potential TRH. Using data from a published exercise physiology study, we demonstrate how SDD can better characterize uncertainty in TRH estimation. We recommend researchers probe TRH by assessing heteroskedasticity, providing bounds for SDD, and estimating outcome distributions and probabilities while carefully crafting the theoretical rationale for the presence of TRH.
期刊介绍:
The Journal of Applied Physiology publishes the highest quality original research and reviews that examine novel adaptive and integrative physiological mechanisms in humans and animals that advance the field. The journal encourages the submission of manuscripts that examine the acute and adaptive responses of various organs, tissues, cells and/or molecular pathways to environmental, physiological and/or pathophysiological stressors. As an applied physiology journal, topics of interest are not limited to a particular organ system. The journal, therefore, considers a wide array of integrative and translational research topics examining the mechanisms involved in disease processes and mitigation strategies, as well as the promotion of health and well-being throughout the lifespan. Priority is given to manuscripts that provide mechanistic insight deemed to exert an impact on the field.