How Spacing of Data Collection May Impact Estimates of Substance Use Trajectories

Abstract

The goal of this study is to provide an empirical example using longitudinal cigarette smoking data that compares results of growth mixture trajectory models on the basis of contiguous and snapshot measurements. Data were drawn from an intensive longitudinal study of college freshman (N = 905) with a previous history of smoking. Participants provided weekly smoking reports for 35 consecutive weeks. We found that using contiguous weekly data (35 waves) or 6-wave or 4-wave snapshot data provided similar trajectory curves and proportions. However, there were notable differences in individual trajectory assignments on the basis of contiguous and snapshot measurements.

RÉSUMÉ

Comment l’espacement des collections de données peut influencer les estimations de trajectoires d’utilisation de substance

Le but de cette étude est de fournir un exemple empirique qui, en utilisant les données longitudinales de la consommation de cigarettes, compare les résultats des modèles de trajectoires de mélange de développement basées sur des mesures contigues et des mesures figées. Les données furent tirées d’une étude longitudinale intensive d’étudiants universitaires en première année (N = 905) ayant un antécédent de consommation de cigarettes. Les participants fournirent un rapport hebdomadaire pendant 35 semaines consécutives. Nous avons découvert que l’utilisation de données hebdomadaires contigues (35 vagues), ou des données figées de 6 vagues ou 4 vagues, fournirent des courbes et proportions de trajectoire similaires. Cependant, il y avaient des différences notables dans les assignements de trajectoires individuelles basés sur des mesures adjacentes et figées.

RESUMEN

Cómo el espaciamiento de colección de datos puede impactar aproximaciones de trayectorias del consumo de sustancias

El objetivo de este estudio es el proporcionar un ejemplo empírico utilizando datos longitudinales de consumo de cigarrillos que compara los resultados de modelos de trayectorias del crecimiento con mixturas basados en mediciones continunas vs mediciones puntuales. Se utilizaton datos de estudiantes universitarios (N = 905) fumadores. Los particpantes proporcionaron informes de su consumo de cigarrillos durante 35 semanas consecutivas. Se encontró que los modelos obtenidos utilizando todas las 35 mediciones contiguas proporcionaron trayectorias medias y proporciones de sujetos en cada mixtura parecidas a las obtenidas cuando únicamente se utilizaron mediciones puntuales (se utilizaron 6 puntos temporales y también 4 puntos únicamente). Sin embargo, las trayectorias individuales estimadas con ambos métodos (medidas contiguas vs. medidas puntuales) fueron notablemente diferentes.

Keywords :

• growth mixture trajectory modeling
• UpTERN
• contiguous measurements
• snapshot measurements
• design of intensive longitudinal study

THE AUTHORS

Xianming Tan, Ph.D., is a Research Associate in the Methodology Center, Pennsylvania State University, University Park, Pennsylvania, USA. Dr. Tan's work focuses on the development and application of finite mixture models. He has a particular interest in the design and analysis of intensive longitudinal studies.

Lisa Dierker, Ph.D., is Professor of Psychology and Chair of the Quantitative Analysis Center at the Wesleyan University, Middletown, Connecticut, USA. Dr. Dierker's research focuses on the application of state-of-the-art statistical methods in understanding the natural history of nicotine dependence and other addictive behaviors. Her research is currently funded by the National Institute on Drug Abuse (NIDA), the National Science Foundation, and the McManus Charitable Trust.

Jennifer Rose, Ph.D., is a Research Associate Professor at the Wesleyan University, Quantitative Analysis Center, Middletown, Connecticut, USA. Dr. Rose's research focuses on the application of advanced statistical methods to the study of nicotine dependence symptoms in adolescent smokers. She is Co-editor and Contributor to the book, Multivariate Applications in Substance Use Research: New Methods for New Questions published by Lawrence Erlbaum Associates, a book that has become a reference for several state-of-the-art statistical methods.

Runze Li, Ph.D., is Professor of Statistics and Professor of Public Health Sciences, a member of the Methodology Center, Pennsylvania State University, University Park, Pennsylvania, USA. Dr. Li's research interest includes analysis of intensive longitudinal data, variable selection for high-dimensional data, and statistical methodology applications to substance use research, life science research, and engineering research. His work has been funded by the National Institute on Drug Abuse (NIDA) and the National Science Foundation. He has published in a broad assortment of methodological and substantive journals. Dr. Li has been Associate Editor of Annals of Statistics, Journal of American Statistical Association, and Statistica Sinica. Awards include NSF Career award and Fellow of Institute of Mathematical Statistics.