Repeated Measures Regression in Laboratory, Clinical and Environmental Research: Common Misconceptions in the Matter of Different Within- and Between-Subject Slopes

• Published in: International journal of environmental research and public health Vol. 16; no. 3; p. 504
• Main Authors: Hoover, Donald, Shi, Qiuhu, Burstyn, Igor, Anastos, Kathryn
• Format: Journal Article
• Language: English
• Published: Switzerland MDPI AG 11.02.2019; MDPI
• Subjects: Adult; Air pollution; Bias; Body fat; Cholesterol; Clinical Laboratory Services; Cross-Sectional Studies; Data Interpretation, Statistical; Epidemiology; Estimates; Health care policy; HIV; Human immunodeficiency virus; Humans; Laboratories; Linear Models; Nitrogen dioxide; Outdoor air quality; Regression analysis; Within-subjects design; New York; United States > US; within-/between-subject associations; cross-sectional regression; mixed models; working correlation structure; generalized estimating equations; repeated measures

When using repeated measures linear regression models to make causal inference in laboratory, clinical and environmental research, it is typically assumed that the within-subject association of differences (or changes) in predictor variable values across replicates is the same as the between-subject...