Heterogeneity often exists among behavioral growth trajectories in a study population. In the evaluation of intervention effects for longitudinal randomized trials, it is informative to examine the impact of an intervention on subgroups characterized by different types of growth trajectories. This paper presents an application of growth mixture modeling to an ordinal-scale substance-use behavior outcome from the Aban Aya Youth Project (AAYP), a longitudinal preventive intervention trial targeting health-compromising behaviors among adolescents. Results suggested two classes of adolescent substance use growth trajectories: Class 1 (44.7% of the sample) started at a higher level and had a relatively shallow increase over time; Class 2 had a lower baseline level and a rapid increase over time. The intervention effectively reduced the rapid increase of substance use for the adolescents in the second class.
Get full access to this article
View all access options for this article.
References
1.
AgrestiA.1990Categorical data analysis. New York: Wiley.
2.
BauerD.J.2005A semiparametric approach to modeling nonlinear relations among latent variables. Structural Equation Modeling: A Multidisciplinary Journal, 4, 513–535.
3.
BauerD.J.CurranP.J.2003Distributional assumptions of growth mixture models: Implications for over extraction of latent trajectory classes. Psychological Methods, 8 (3), 338–363.
4.
BeetsM.W.FlayB.R.VuchinichS.SnyderF.J.AcockA.LiK. K.BurnsK.WashburnI.J.DurlakJ.2009Using a social and character development program to prevent substance use, violent behaviors, and sexual activity among elementary-school students in Hawai'i. American Journal of Public Health, 99 (8), 1438–1445.
5.
BrykA.S.RaudenbushS.W.1992Hierarchical linear models for social and behavioral research: Application and data analysis methods. Newbury Park, CA: Sage.
6.
ColderC.R.CampbellR.T.RuelE.RichardsonJ.L.FlayB.R.2002A finite mixture model of growth trajectories of adolescent alcohol use: Predictors and consequences. Journal of Consulting and Clinical Psychology, 70, 976–985.
7.
ColderC.R.MehtaP.BalandaK.CampbellR.T.MayhewK.P.StantonW.R.PentzM. A.FlayB.R., 2001Identifying trajectories of adolescent smoking: An application of latent growth mixture modeling. Health Psychology, 20, 127–135.
8.
FlahertyB.P.2002Assessing reliability of categorical substance use measures with latent class analysis. Drug and Alcohol Dependence, 68, S7–S20.
9.
FlayB.R.GraumlichS.SegawaE.BurnsJ.L.HollidayM.Y., & Aban Aya Investigators.2004Effects of two prevention programs on high-risk behaviors among African American youth: A randomized trial. Archives of Pediatric and Adolescent Medicine, 158, 377–384.
10.
HallD.2000Zero-inflated Poisson and binomial regression with random effects: A case study. Biometrics, 56, 1030–1039.
HedekerD.GibbonsR.1994A random-effects ordinal regression model for multilevel analysis. Biometrics, 50, 933–944.
13.
HurK.HedekerD.HendersonW.G.KhuriS.DaleyJ.2002Modeling clustered count data with excess zeros in health care outcomes research. Health Services and Outcomes Research Methodology, 3, 5–20.
14.
LairdN.M.WareJ.H.1982Random-effects models for longitudinal data. Biometrics, 38, 964–974.
15.
LambertD.1992Zero-inflated Poisson regression with an application to defects in manufacturing. Technometrics, 34, 1–14.
MuthénB.O.1989Latent variable modeling in heterogeneous populations. Psychometrika, 54, 557–585.
18.
MuthénB.O.1991Analysis of longitudinal data using latent variable models with varying parameters. In CollinsL. M.HornJ.L. (Eds.), Best methods for the analysis of change: Recent advances, unanswered questions, future directions (pp. 1–17). Washington DC: American Psychological Association.
19.
MuthénB. O.2001Latent variable mixture modeling. In MarcoulidesG.A.SchumackerR.E. (Eds.), New developments and techniques in structural equation modeling (pp. 1–33). Mahwah, NJ: Lawrence Erlbaum.
20.
MuthénB.O.2002Beyond SEM: General latent variable modeling. Behaviormetrika, 29, 81–117.
21.
MuthénB.O.2004Latent variable analysis: Growth mixture modeling and related techniques for longitudinal data. In KaplanD. (Ed.), Handbook of quantitative methodology for the social sciences. Newbury Park, CA: Sage Publications.
NaginD.S.LandK.C.1993Age, criminal careers, and population heterogeneity: Specification and estimation of a nonparametric, mixed Poisson model. Criminology, 31, 327–362.
27.
NaginD.S.TremblayR.E.2001Analyzing developmental trajectories of distinct but related behaviors: A group-based method. Psychological Methods, 6, 18–34.
28.
NylundK.L.AsparouhovT.MuthénB.O.2007Deciding on the number of classes in latent class analysis and growth mixture modeling: A Monte Carlo simulation study. Structural Equation Modeling: A Multidisciplinary Journal, 14, 535–569.
29.
SchwartzG.1978Estimating the dimension of a model. The Annals of Statistics, 6, 461–464.
30.
ScloveL.1987Application of model-selection criteria to some problems in multivariate analysis. Psychometrika, 52, 333–343.
31.
SegawaE.NgweJ. E.LiY.FlayB. R., & Aban Aya Coinvestigators. 2005Evaluation of the effects of the Aban Aya Youth Project in reducing violence among African American adolescent males using latent class growth mixture modeling techniques. Evaluation Review, 29 (2), 128–148.
32.
TofighiD.EndersC.K.2006Identifying the correct number of components in a finite mixture model. Computational Statistics, 9, 65–78.
33.
VerbekeG.LesaffreE.1996A linear mixed-effects model with heterogeneity in the random-effects population. Journal of the American Statistical Association, 91, 217–221.
34.
WitkiewitzK.van der MaasH.L.HuffordM.R.MarlattG.A.2007Nonnormality and divergence in posttreatment alcohol use: Reexamining the project MATCH data “another way.”Journal of Abnormal Psychology, 116 (2), 378–394.
35.
XuW.HedekerD.2001A random-effects mixture model for classifying treatment response in longitudinal clinical trials. Journal of Biopharmaceutical Statistics, 11 (4), 253–273.
