This Article Figures Only Full Text Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Blackwell, E. Articles by Miller, G. E. Search for Related Content PubMed PubMed Citation Articles by Blackwell, E. Articles by Miller, G. E. Related Collections Statistical Corner

Applying Mixed Regression Models to the Analysis of Repeated-Measures Data in Psychosomatic Medicine

From the Department of Psychology (E.B., G.E.M.), University of British Columbia, Vancouver, British Columbia, Canada; and the Rush Institute for Aging (C.F.M.d.L.), Rush-Presbyterian–St. Luke's Medical Center, Chicago, Illinois.

Address correspondence and reprint requests to Ekin Blackwell, MA, UBC Department of Psychology, 2136 West Mall, Vancouver, BC V6T 1Z4, Canada. E-mail: ekinb{at}interchange.ubc.ca

Objective: Although repeated-measures designs are increasingly common in research on psychosomatic medicine, they are not well suited to the conventional statistical techniques that scientists often apply to them. The goal of this article is to introduce readers to mixed regression models, which provide a more flexible and accurate framework for managing repeated-measures data.

Methods and Results: We begin with a summary of the advantages that mixed regression models have over conventional statistical techniques in the context of repeated-measures designs. Next, we outline the conceptual and mathematical underpinnings of mixed regression models for a nonstatistical audience. The article ends with two examples of how these models can be applied in psychosomatic research; one deals with a prospective investigation of depressive symptoms and change in body mass index in older adults and the other with a diary study of social interactions and cortisol secretion.

Conclusions: Mixed regression models offer a flexible and powerful approach to analyzing repeated-measures data. They possess important advantages over more traditional strategies, and more widespread application of these models is likely to enhance the overall quality of psychosomatic research.

Key Words: mixed regression models • analysis of change • repeated measures • nested designs • random effects

Abbreviations: HLM = hierarchical linear model; OLS = ordinary least squares; ANOVA = analysis of variance; BP = blood pressure; CES-D = Center for Epidemiologic Studies–Depression; BMI = body mass index; D = dominance; EPAQ = Extended Version of the Personality Attributes Questionnaire.

This article has been cited by other articles:

				I. Korstjens, A. M. May, E. van Weert, I. Mesters, F. Tan, W. J. G. Ros, J. E. H. M. Hoekstra-Weebers, C. P. van der Schans, and B. van den Borne Quality of Life After Self-Management Cancer Rehabilitation: A Randomized Controlled Trial Comparing Physical and Cognitive-Behavioral Training Versus Physical Training Psychosom Med, May 1, 2008; 70(4): 422 - 429. [Abstract] [Full Text] [PDF]