The Use of Mixed Models for the Analysis of Mediated Data with Time-Dependent Predictors

Linear mixed models (LMMs) are frequently used to analyze longitudinal data. Although these models can be used to evaluate mediation, they do not directly model causal pathways. Structural equation models (SEMs) are an alternative technique that allows explicit modeling of mediation. The goal of this paper is to evaluate the performance of LMMs relative to SEMs in the analysis of mediated longitudinal data with time-dependent predictors and mediators. We simulated mediated longitudinal data from an SEM and specified delayed effects of the predictor. A variety of model specifications were assessed, and the LMMs and SEMs were evaluated with respect to bias, coverage probability, power, and Type I error. Models evaluated in the simulation were also applied to data from an observational cohort of HIV-infected individuals. We found that when carefully constructed, the LMM adequately models mediated exposure effects that change over time in the presence of mediation, even when the data arise from an SEM.

• Publication year

  2011

• Venue

  Journal of Environmental and Public Health

• Publication date

  2011-05-14

• Fields of study

  Medicine, Psychology

• Identifiers
  DOI 10.1155/2011/435078PMID 21647351PMCID 3103895
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar, PubMed

• No claims are published for this paper.

• No concepts are published for this paper.

• Analysis of Longitudinal Data Using Latent Variable Models with Varying Parameters

  2011cited by this paper
• Performance of mixed effects models in the analysis of mediated longitudinal data

  2010cited by this paper
• Introduction to Statistical Mediation Analysis

  2008cited by this paper
• Bias in cross-sectional analyses of longitudinal mediation.

  2007cited by this paper
• Alcohol Consumption and HIV Disease Progression

  2007cited by this paper
• Conceptualizing and testing random indirect effects and moderated mediation in multilevel models: new procedures and recommendations.

  2006cited by this paper
• Latent curve models: A structural equation perspective

  2005cited by this paper
• Applied Longitudinal Analysis

  2005cited by this paper
• People are variables too: multilevel structural equations modeling.

  2005cited by this paper
• Structural Equation Models

  2005cited by this paper
• Lower level mediation in multilevel models.

  2003cited by this paper
• Have Multilevel Models Been Structural Equation Models All Along?

  2003cited by this paper
• Investigation of Mediational Processes Using Parallel Process Latent Growth Curve Modeling

  2003cited by this paper
• Testing mediational models with longitudinal data: questions and tips in the use of structural equation modeling.

  2003cited by this paper
• Estimating Multilevel Linear Models as Structural Equation Models

  2003cited by this paper
• A structural equations modeling approach to the general linear mixed model

  2001cited by this paper
• Multilevel Modeling of Individual and Group Level Mediated Effects

  2001cited by this paper
• A Structural Modeling Approach to a Multilevel Random Coefficients Model

  2000cited by this paper
• Putting the individual back into individual growth curves.

  2000cited by this paper
• Multilevel Mediation Modeling in Group-Based Intervention Studies

  1999cited by this paper
• General Longitudinal Modeling of Individual Differences in Experimental Designs: A Latent Variable Framework for Analysis and Power Estimation

  1997cited by this paper
• Latent Variable Modeling of Longitudinal and Multilevel Data. Project 2.4, Quantitative Models To Monitor the Status and Progress of Learning and Performance and Their Antecedents.

  1994cited by this paper
• Using covariance structure analysis to detect correlates and predictors of individual change over time

  1994cited by this paper
• Best Methods for the Analysis of Change: Recent Advances, Unanswered Questions, Future Directions

  1991cited by this paper
• Patterns of Change within Latent Variable Structural Equation Models

  1990cited by this paper
• Latent curve analysis

  1990cited by this paper
• Latent growth curves within developmental structural equation models.

  1987cited by this paper
• The moderator-mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations.

  1986cited by this paper
• Random-effects models for longitudinal data.

  1982cited by this paper
• Lower Level Mediation in Multilevel Models

  year unknowncited by this paper

• Do Age and Episodic Memory Task Performance Differentially Relate to Tract-specific White Matter Microstructure? Cross-sectional and Longitudinal Analyses in a Healthy Adult Sample.

  2026cites this paper
• Sibling psychopathology as a mediator between disability in youth with chronic physical illness and sibling Health-Related quality of life

  2026cites this paper
• Impact of Physical Activity Modalities on Text Processing and Comprehension in Adolescents

  2025cites this paper
• Mediation analysis in longitudinal study with high-dimensional methylation mediators

  2024cites this paper
• Exercise-Induced Hypoalgesia in Patients with Chronic Whiplash-Associated Disorders: Differences between Subgroups Based on the Central Sensitization Inventory

  2024cites this paper
• Bayesian Inference of Dynamic Mediation Models for Longitudinal Data

  2023cites this paper
• Time lags and time interactions in mixed-effects models impacted longitudinal mediation effect estimates.

  2022cites this paper
• Statistical analysis considerations within longitudinal studies of physical qualities in youth athletes: A qualitative systematic methodological review

  2022cites this paper
• The role of social isolation in physical and emotional outcomes among patients with chronic pain.

  2021cites this paper
• Effects of technology-based physical activity interventions for women after bariatric surgery: study protocol for a three-arm randomised controlled trial

  2021cites this paper
• Bayesian model selection for multilevel mediation models

  2021cites this paper
• Collective orientation and its implications for coordination and team performance in interdependent work contexts

  2020cites this paper
• Continuous‐time causal mediation analysis

  2019influential citation
• Legal Socialization and Self-Reported Criminal Offending: The Role of Procedural Justice and Legal Orientations

  2019cites this paper
• Mindfulness-based Cognitive Therapy for Obsessive-Compulsive Disorder: A Pilot Study

  2019cites this paper
• Legal Socialization and Self-Reported Criminal Offending: The Role of Procedural Justice and Legal Orientations

  2017cites this paper
• Bayesian dynamic mediation analysis.

  2017cites this paper
• Procedural Justice and Legal Socialization Among Serious Adolescent Offenders: A Longitudinal Examination

  2016cites this paper
• [Design of a genomic, environmental, microbial and metabolomic study on celiac disease: an approach to the future of personalized prevention of celiac disease].

  2016cites this paper
• Celiac Disease Genomic, Environmental, Microbiome, and Metabolomic (CDGEMM) Study Design: Approach to the Future of Personalized Prevention of Celiac Disease

  2015cites this paper
• Assessment of the dimensions, construct validity, and utility for rheumatoid arthritis screening of the COPCORD instrument

  2014cites this paper
• Pharmacodynamic evaluation of a reference and a generic toltrazuril preparation in broilers experimentally infected with Eimeria tenella or E. acervulina

  2014cites this paper
• Perception de la personnalité du locuteur dysphonique sur base de sa qualité vocale

  2013cites this paper
• Non-linear mixed models in the analysis of mediated longitudinal data with binary outcomes

  2012cites this paper