Detecting Specific Patterns of Change for Two Outcome Scores in a Mental Health Study by Means of Linear Mixture Models
Abstract
Common practice evaluates the success of any form of (medical) treatment by evaluating the change and the relation of change for particular outcome criteria. However, we must assume that the expected change over time can only be observed for—perhaps only small—subgroups. The purpose of this article, therefore, is to explain and demonstrate a method on how to measure the relationships of change for important outcomes in a longitudinal study of schizophrenia treatment and to identify latent subgroups of patients characterized by different shapes of trajectories and relations of trajectories. We will focus on the change over time for the Social and Occupational F unctioning A ssessment S core (SOFAS) (American Psychiatric Association, 1987) and the subjective health-related quality of life assessed by SF-36(Ware, Gandek, & The IQOLA Project Group, 1994). The sample employed for this study consists of 307 patients with a diagnosis of schizophrenia (ICD-10 F20.0) treated by psychiatric services in Leipzig. The method adopted to portray these changes is the growth mixture model, which combines latent class analysis with the approach of modeling individual trajectories. Results show that a positive relation between the outcomes of interest can only be observed for about 15% of the sample.
References
1987). Diagnostic and Statistical Manual of Mental Disorders (DSM-III-R) . (3rd ed.). Washington, DC: American Psychological Association
(Anderson, J. , Gerbing, D. W. (1988). Structural equation modeling in practice: A review and recommended two-step approach. Psychological Bulletin, 103, 411– 423Anderson, J. , Gerbing, D. W. (1992). Assumptions and comparative strength of the two-step approach. Sociological Methods & Research, 20, 321– 333Bauer, D. J. , Curran, P. J. (2003a). Distributional assumptions of growth mixture models: Implications for overextraction of latent trajectory classes. Psychological Methods, 8, 338– 363Bauer, D. J. , Curran, P. J. (2003b). Overextraction of latent trajectory classes: Much ado about nothing ? Reply to Rindskopf (2003), Muthén (2003) and Cudeck & Henly (2003). Psychological Methods, 8, 384– 393Cudeck, R. , Henly, S. J. (2003). A realistic perspective on pattern representation in growth data: Comment on Bauer and Curran (2003). Psychological Methods, 8, 378– 383Curran, P. J. , Hussong, A. M. (2002). Structural equation modelling of repeated measures data: Latent curve analysis. In D. S. Moskowitz, S. L. Hershberger (Eds.), Modelling intraindividual variability with repeated measures data: methods and applications. (pp. 59-85). London: Lawrence Erlbaum AssociatesFornell, C. , Yi, Y. (1992a). Assumption of the two-step approach: Reply to Anderson and Gerbing. Sociological Methods & Research, 20, 334– 339Fornell, C. , Yi, Y. (1992b). Assumptions of the two-step approach to latent variable modeling. Sociological Methods & Research, 20, 291– 320Jedidi, K. , Jagpal, H. S. , DeSarbo, W. S. (1997a). Finite mixture structural equation models for response-based segmentation and unobserved heterogeneity. Marketing Science, 16, 39– 59Jedidi, K. , Jagpal, H. S. , DeSarbo, W. S. (1997b). STEMM: A general finite mixture structural equation model. Journal of Classification, 14, 23– 50Jöreskog, K. G. (1973). A general method for estimating a linear structural equation system. In A. S. Goldberger, O. D. Duncan (Eds.), Structural equation models in the social sciences (pp. 85-112). New York: Seminar PressKass, R. E. , Raftery, A. E. (1995). Bayes factor. Journal of the American Statistical Association, 90, 773– 795Kass, R. E. , Wasserman, L. (1995). A reference Bayesian test for nested hypotheses and its relationship to the Schwartz criterion. Journal of the American Statistical Association, 90, 928– 934Kilian, R. , Matschinger, H. , Becker, T. , Angermeyer, M. C. (2003). A longitudinal analysis of the impact of social and clinical characteristics on the costs of schizophrenia treatment. Acta Psychiatrica Scandinavica, 107, 351– 360Kilian, R. , Roick, C. , Matschinger, H. , Bernert, S. , Mory, C. , Becker, T. , Angermeyer, M. C. (2001). Instruments for the economic evaluation of psychiatric service systems: Methodological foundations of the European standardisation and the German adaption. Psychiatrische Praxis, 28, 74– 78Lo, Y. , Mendell, N. R. , Rubin, D. B. (2001). Testing the number of components in a normal mixture. Biometrika, 88, 767– 778McLachlan, G. J. , Peel, D. (2000). Finite mixture models. . New York: John WileyMehta, P. D. , West, S. G. (2000). Putting the individual back into individual growth curves. Psychological Methods, 5, 23– 43Meredith, W. , Tisak, J. (1990). Latent curve analysis. Psychometrika, 55, 107– 122Muthén, B. O. (1991). Analysis of longitudinal data using latent variable models with varying parameters. In L. M. Collins, J. Horn (Eds.), Best methods for the analysis of change. recent advances, unanswered questions, future directions (pp. 1-17). Washington, DC: American Psychological AssociationMuthén, B. O. (2001). Latent variable mixture modeling. In A. Marcoulides, R. E. Schumacker (Eds.), New developments and techniques in structural equation modeling (pp. 1-33). London: Lawrence ErlbaumMuthén, B. O. (2002). Beyond SEM: General latent variable modeling. Behaviormetrika, 29, 81– 117Muthén, B. O. (2003). Statistical and substantive checking in growth mixture modeling: Comment on Bauer & Curran (2003). Psychological Methods, 8, 369– 377Muthén, B. O. (2004). Latent variable analysis: Growth mixture modeling and related techniques for longitudinal data. In D. Kaplan (Ed.), Handbook of quantitative methodology for the social sciences (pp. 345-368). Newbury Park, CA: Sage PublicationsMuthén, B. O. , Curran, P. J. (1997). General longitudinal modelling of individual differences in experimental designs: A latent variable framework for analysis and power estimation. Psychological Methods, 2, 371– 402Muthén, B. O. , Muthén, L. K. (2000). Integrating person-centered and variable-centered analysis: Growth mixture modeling with latent trajectory classes. Alcoholism: Clinical and Experimental Research, 24, 882– 891Muthén, L. K. , Muthén, B. O. (2004). Mplus User’s Guide . (3rd ed.). Los Angeles: Muthén & MuthénNagin, D. S. (1999). Analyzing developmental trajectories: A semi-parametric, group-based approach. Psychological Methods, 4, 157–Nagin, D. S. , Tremblay, R. E. (2001). Analyzing developmental trajectories of distinct but related behaviors: A group based method. Psychological Methods, 6, 18– 34Reinecke, J. (2006). Longitudinal analysis of adolescents' deviant and delinquent behavior: Applications of latent class growth curves and growth mixture models. Methodology, 2, 100– 112Rindskopf, D. (2003). Mixture or homogenous? Comment on Bauer & Curran. Psychological Methods, 8, 364– 368Rogosa, D. R. (1988). Myths about longitudinal research. In K. W. Schaie, R. T. Campbell, W. Meredith, S. C. Rawlings (Eds.), Methodological issues in aging research (pp. 171-209). New York: SpringerSchwartz, G. (1978). Estimating the dimensions of a model. The Annals of Statistic, 6, 461– 4642003). Stata Statistical Software: Release 8 . College Station, TX: StataCorpLP
(Sörbom, D. (1974). A general method for studying differences in factor means and factor structure between groups. British Journal of Mathematical and Statistical Psychology, 27, 229– 239Vuong, Q. H. (1989). Likelihood ratio tests for model selection and non-nested hypotheses. Econometrica, 57, 307– 333Ware, J. E. , Gandek, B. (1994). The SF-36 Health Survey: Development and use on mental health research and the IQOLA project. International Journal of Mental Health, 23, 49– 73White, H. R. , Johnson, V. , Buyske, S. (2000). Parental modeling and parenting behavior effects on offspring alcohol and cigarette use: A growth curve analysis. Journal of Substance Abuse, 12, 287– 310Willet, J. B. (1988). Questions and answers in the measurement of change. In E. Z. Rothkopf (Ed.), Review of research in education: Vol 15, (pp. 345-422). Washington DC: American Educational Research AssociationWillett, J. B. , Sayer, A. G. (1994). Using covariance structure analysis to detect correlates and predictors of individual change over time. Psychological Bulletin, 116, 363– 381