{"title":"评估锚定法中最小临床重要差异的偏差:模拟方法。","authors":"Greg Hather, Polyna Khudyakov","doi":"10.1080/10543406.2025.2547586","DOIUrl":null,"url":null,"abstract":"<p><p>Anchor based methods have been used in clinical studies to determine minimal clinically important differences (MCID) for clinical outcome assessments. However, the theoretical properties and robustness of the methodology are not fully understood. We conducted a simulation study to explore the performance of anchor-based methods across a range of values for outcome variance, placebo effects, anchor measurement noise, and confounding. Our results demonstrate that considerable placebo effects, anchor measurement error, and confounders may introduce a substantial bias into the estimated MCID. We also discuss strategies to identify and mitigate these biases.</p>","PeriodicalId":54870,"journal":{"name":"Journal of Biopharmaceutical Statistics","volume":" ","pages":"1-8"},"PeriodicalIF":1.2000,"publicationDate":"2025-08-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Evaluating bias in the anchor method for the minimal clinically important difference: a simulation approach.\",\"authors\":\"Greg Hather, Polyna Khudyakov\",\"doi\":\"10.1080/10543406.2025.2547586\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><p>Anchor based methods have been used in clinical studies to determine minimal clinically important differences (MCID) for clinical outcome assessments. However, the theoretical properties and robustness of the methodology are not fully understood. We conducted a simulation study to explore the performance of anchor-based methods across a range of values for outcome variance, placebo effects, anchor measurement noise, and confounding. Our results demonstrate that considerable placebo effects, anchor measurement error, and confounders may introduce a substantial bias into the estimated MCID. We also discuss strategies to identify and mitigate these biases.</p>\",\"PeriodicalId\":54870,\"journal\":{\"name\":\"Journal of Biopharmaceutical Statistics\",\"volume\":\" \",\"pages\":\"1-8\"},\"PeriodicalIF\":1.2000,\"publicationDate\":\"2025-08-19\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Journal of Biopharmaceutical Statistics\",\"FirstCategoryId\":\"3\",\"ListUrlMain\":\"https://doi.org/10.1080/10543406.2025.2547586\",\"RegionNum\":4,\"RegionCategory\":\"医学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q4\",\"JCRName\":\"PHARMACOLOGY & PHARMACY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Biopharmaceutical Statistics","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1080/10543406.2025.2547586","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"PHARMACOLOGY & PHARMACY","Score":null,"Total":0}
Evaluating bias in the anchor method for the minimal clinically important difference: a simulation approach.
Anchor based methods have been used in clinical studies to determine minimal clinically important differences (MCID) for clinical outcome assessments. However, the theoretical properties and robustness of the methodology are not fully understood. We conducted a simulation study to explore the performance of anchor-based methods across a range of values for outcome variance, placebo effects, anchor measurement noise, and confounding. Our results demonstrate that considerable placebo effects, anchor measurement error, and confounders may introduce a substantial bias into the estimated MCID. We also discuss strategies to identify and mitigate these biases.
期刊介绍:
The Journal of Biopharmaceutical Statistics, a rapid publication journal, discusses quality applications of statistics in biopharmaceutical research and development. Now publishing six times per year, it includes expositions of statistical methodology with immediate applicability to biopharmaceutical research in the form of full-length and short manuscripts, review articles, selected/invited conference papers, short articles, and letters to the editor. Addressing timely and provocative topics important to the biostatistical profession, the journal covers:
Drug, device, and biological research and development;
Drug screening and drug design;
Assessment of pharmacological activity;
Pharmaceutical formulation and scale-up;
Preclinical safety assessment;
Bioavailability, bioequivalence, and pharmacokinetics;
Phase, I, II, and III clinical development including complex innovative designs;
Premarket approval assessment of clinical safety;
Postmarketing surveillance;
Big data and artificial intelligence and applications.