TY - JOUR
T1 - Tutorial
T2 - The Practical Application of Longitudinal Structural Equation Mediation Models in Clinical Trials
AU - Goldsmith, Kimberley A.
AU - MacKinnon, David P.
AU - Chalder, Trudie
AU - White, Peter D.
AU - Sharpe, Michael
AU - Pickles, Andrew
PY - 2018/6
Y1 - 2018/6
N2 - The study of mediation of treatment effects, or how treatments work, is important to understanding and improving psychological and behavioral treatments, but applications have often focused on mediators and outcomes measured at a single time point. Such crosssectional analyses do not respect the implied temporal ordering that mediation suggests. Clinical trials of treatments often provide repeated measures of outcomes and, increasingly, of mediators as well. A trial with repeated measurements allows for the application of various types of longitudinal structural equation mediation models. These provide for flexibility in modeling, including the incorporation of some types of measurement error and unmeasured confounding that can strengthen the robustness of findings. The usual approach is to identify the most theoretically plausible model and apply that model. In the absence of clear theory, we put forward the option of fitting a few theoretically plausible models, providing a type of sensitivity analysis for the mediation hypothesis. In this tutorial, we outline how to fit several longitudinal mediation models. This will allow readers to learn about one type of model that is of interest, or about several alternative models so that they can take this sensitivity approach. We use the “Pacing, Graded Activity, and Cognitive Behavioral Therapy: A Randomized Evaluation” (PACE) trial of rehabilitative treatments for chronic fatigue syndrome (ISRCTN 54285094) as a motivating example and describe how to fit and interpret various longitudinal mediation models using simulated data similar to those in the PACE trial. The simulated dataset and Mplus code and output are provided.
AB - The study of mediation of treatment effects, or how treatments work, is important to understanding and improving psychological and behavioral treatments, but applications have often focused on mediators and outcomes measured at a single time point. Such crosssectional analyses do not respect the implied temporal ordering that mediation suggests. Clinical trials of treatments often provide repeated measures of outcomes and, increasingly, of mediators as well. A trial with repeated measurements allows for the application of various types of longitudinal structural equation mediation models. These provide for flexibility in modeling, including the incorporation of some types of measurement error and unmeasured confounding that can strengthen the robustness of findings. The usual approach is to identify the most theoretically plausible model and apply that model. In the absence of clear theory, we put forward the option of fitting a few theoretically plausible models, providing a type of sensitivity analysis for the mediation hypothesis. In this tutorial, we outline how to fit several longitudinal mediation models. This will allow readers to learn about one type of model that is of interest, or about several alternative models so that they can take this sensitivity approach. We use the “Pacing, Graded Activity, and Cognitive Behavioral Therapy: A Randomized Evaluation” (PACE) trial of rehabilitative treatments for chronic fatigue syndrome (ISRCTN 54285094) as a motivating example and describe how to fit and interpret various longitudinal mediation models using simulated data similar to those in the PACE trial. The simulated dataset and Mplus code and output are provided.
U2 - 10.1037/met0000154
DO - 10.1037/met0000154
M3 - Article
SN - 1082-989X
VL - 23
SP - 191
EP - 207
JO - PSYCHOLOGICAL METHODS
JF - PSYCHOLOGICAL METHODS
IS - 2
ER -