Prevention Science

, Volume 20, Issue 3, pp 394–406 | Cite as

Inverse Propensity Score Weighting with a Latent Class Exposure: Estimating the Causal Effect of Reported Reasons for Alcohol Use on Problem Alcohol Use 16 Years Later

  • Bethany C. BrayEmail author
  • John J. Dziak
  • Megan E. Patrick
  • Stephanie T. Lanza


Latent class analysis (LCA) has proven to be a useful tool for identifying qualitatively different population subgroups who may be at varying levels of risk for negative outcomes. Recent methodological work has improved techniques for linking latent class membership to distal outcomes; however, these techniques do not adjust for potential confounding variables that may provide alternative explanations for observed relations. Inverse propensity score weighting provides a way to account for many confounders simultaneously, thereby strengthening causal inference of the effects of predictors on outcomes. Although propensity score weighting has been adapted to LCA with covariates, there has been limited work adapting it to LCA with distal outcomes. The current study proposes a step-by-step approach for using inverse propensity score weighting together with the “Bolck, Croon, and Hagenaars” approach to LCA with distal outcomes (i.e., the BCH approach), in order to estimate the causal effects of reasons for alcohol use latent class membership during the year after high school (at age 19) on later problem alcohol use (at age 35) with data from the longitudinal sample in the Monitoring the Future study. A supplementary appendix provides evidence for the accuracy of the proposed approach via a small-scale simulation study, as well as sample programming code to conduct the step-by-step approach.


Latent class analysis Causal inference Propensity scores Alcohol use Motives Reasons for drinking 



The authors wish to thank Deborah D. Kloska for help with management of the Monitoring the Future data sets and Donna L. Coffman for early discussions that helped inform our thinking about causal latent class exposures.


This research was conducted at The Pennsylvania State University and The University of Michigan, and was supported by a seed grant from the National Center for Responsible Gaming (NCRG) and awards P50-DA039838, P50-DA010075, and R01-DA037902 from the National Institute on Drug Abuse (NIDA); data collection was supported by awards R01-DA001411 and R01-DA016575 from NIDA.

Compliance with Ethical Standards

Conflict of Interest

The authors declare that they have no conflicts of interest.

Ethical Approval

All procedures performed in this study involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

Informed Consent

Where appropriate, informed consent and assent were obtained from all individual participants included in this study.


The content is solely the responsibility of the authors and does not necessarily represent the official views of NCRG, NIDA, or the National Institutes of Health.

Supplementary material

11121_2018_883_MOESM1_ESM.docx (30 kb)
ESM 1 (DOCX 30 kb)


  1. Asparouhov, T., & Muthén, B. (2014). Auxiliary variables in mixture modeling: Three-step approaches using Mplus. Structural Equation Modeling, 21, 329–341.CrossRefGoogle Scholar
  2. Austin, P. C. (2011). An introduction to propensity score methods for reducing the effects of confounding in observational studies. Multivariate Behavioral Research, 46, 399–424.CrossRefGoogle Scholar
  3. Bakk, Z., & Vermunt, J. K. (2016). Robustness of stepwise latent class modeling with continuous distal outcomes. Structural Equation Modeling, 23, 20–31.CrossRefGoogle Scholar
  4. Bang, H., & Robins, J. M. (2005). Doubly robust estimation in missing data and causal inference models. Biometrics, 61, 962–972.CrossRefGoogle Scholar
  5. Bolck, A., Croon, M., & Hagenaars, J. (2004). Estimating latent structure models with categorical variables: One-step versus three-step estimators. Political Analysis, 12, 3–27.CrossRefGoogle Scholar
  6. Bray, B. C., Almirall, D., Zimmerman, R. S., Lynam, D., & Murphy, S. A. (2006). Assessing the total effect of time-varying predictors in prevention research. Prevention Science, 7, 1–17.CrossRefGoogle Scholar
  7. Bray, B. C., Lanza, S. T., & Tan, X. (2015). Eliminating bias in classify-analyze approaches for latent class analysis. Structural Equation Modeling, 22, 1–11.CrossRefGoogle Scholar
  8. Cardoso, J. B., Goldbach, J. T., Cervantes, R. C., & Swank, P. (2016). Stress and multiple substance use behaviors among Hispanic adolescents. Prevention Science, 17, 208–217.CrossRefGoogle Scholar
  9. Coffman, D. L., & Zhong, W. (2012). Assessing mediation using marginal structural models in the presence of confounding and moderation. Psychological Methods, 17, 642–664.CrossRefGoogle Scholar
  10. Coffman, D., Patrick, M. E., Palen, L., Rhoades, B. L., & Ventura, A. (2007). Why do high school seniors drink? Implications for a targeted approach. Prevention Science, 8, 241–248.Google Scholar
  11. Coffman, D. L., Caldwell, L. L., & Smith, E. A. (2012). Introducing the at-risk average causal effect with application to HealthWise South Africa. Prevention Science, 13, 437–447.CrossRefGoogle Scholar
  12. Cohen, J. (1988). Statistical power analysis for the behavioral sciences. Hillsdale: Laurence Erlbaum.Google Scholar
  13. Collins, L. M., & Lanza, S. T. (2010). Latent class and latent transition analysis: With applications in the social, behavioral, and health sciences. New York: Wiley.Google Scholar
  14. Dehejia, R. H., & Wahba, S. (2002). Propensity score matching methods for non-experimental causal studies. Review of Economics and Statistics, 84, 151–161.CrossRefGoogle Scholar
  15. Dziak, J. J., Bray, B. C., Zhang, J., Zhang, M., & Lanza, S. T. (2016). Comparing the performance of improved classify-analyze approaches for distal outcomes in latent profile analysis. Methodology, 12, 107–116.CrossRefGoogle Scholar
  16. Evans-Polce, R. J., Patrick, M. E., & Miech, R. (2017). Patterns of reasons for vaping in a national sample of adolescent vapers. Paper presented at the Society for Prevention Research 25th Annual Meeting: “Prevention and Public Systems of Care: Research, Policy and Practice,” Washington.Google Scholar
  17. Gilreath, T. D., Astor, R. A., Estrada Jr, J. N., Benbenishty, R., & Unger, J. B. (2014). School victimization and substance use among adolescents in California. Prevention Science, 15, 897–906.Google Scholar
  18. Green, K. M., & Stuart, E. A. (2014). Examining moderation analyses in propensity score methods: Application to depression and substance use. Journal of Consulting and Clinical Psychology, 82, 773–783.CrossRefGoogle Scholar
  19. Héroux, M., Janssen, I., Lee, D. C., Sui, X., Hebert, J. R., & Blair, S. N. (2012). Clustering of unhealthy behaviors in the aerobics center longitudinal study. Prevention Science, 13, 183–195.CrossRefGoogle Scholar
  20. Imbens, G. (1999). The role of the propensity score in estimating dose-response functions (Tech. Work. Paper No. 237). Cambridge: National Bureau of Economic Research. Retreived from
  21. Jiang, L., Beals, J., Zhang, L., Mitchell, C. M., Manson, S. M., Acton, K. J., ... & Special Diabetes Program for Indians Demonstration Projects. (2012). Latent class analysis of stages of change for multiple health behaviors: Results from the Special Diabetes Program for Indians diabetes prevention program. Prevention Science, 13, 449–461.Google Scholar
  22. Johnston, L. D., O’Malley, P. M., Bachman, J. G., Schulenberg, J. E., & Miech, R. A. (2016). Monitoring the Future national survey results on drug use, 1975–2015: Volume 2, college students and adults ages 19–55. Ann Arbor: Institute for Social Research, The University of Michigan.Google Scholar
  23. Kang, J. D. Y., & Schafer, J. L. (2007). Demystifying double robustness: A comparison of alternative strategies for estimating a population mean from incomplete data (with discussion and rejoinder). Statistical Science, 22, 523–539.CrossRefGoogle Scholar
  24. Lanza, S. T., & Rhoades, B. L. (2013). Latent class analysis: An alternative perspective on subgroup analysis in prevention and treatment. Prevention Science, 14, 157–168.CrossRefGoogle Scholar
  25. Lanza, S. T., Bray, B. C., & Collins, L. M. (2013a). An introduction to latent class and latent transition analysis. In J. A. Schinka, W. F. Velicer, & I. B. Weiner (Eds.), Handbook of psychology (Vol. 2, 2nd ed., pp. 691–716). Hoboken: Wiley.Google Scholar
  26. Lanza, S. T., Moore, J. E., & Butera, N. M. (2013b). Drawing causal inferences using propensity scores: A practical guide for community psychologists. American Journal of Community Psychology, 52, 380–392.CrossRefGoogle Scholar
  27. Lanza, S. T., Coffman, D. L., & Xu, S. (2013c). Causal inference in latent class analysis. Structural Equation Modeling, 20, 361–383.CrossRefGoogle Scholar
  28. Lanza, S. T., Tan, X., & Bray, B. C. (2013d). Latent class analysis with distal outcomes: A flexible model-based approach. Structural Equation Modeling, 20, 1–26.CrossRefGoogle Scholar
  29. Lanza, S. T., Dziak, J. J., Huang, L., Wagner, A., & Collins, L. M. (2015). PROC LCA & PROC LTA users’ guide (Version 1.3.2). University Park: The Methodology Center, Penn State. Retrieved from
  30. Lanza, S. T., Schuler, M. S., & Bray, B. C. (2016). Latent class analysis with causal inference: The effect of adolescent depression on young adult substance use profile (Chp. 16, pp. 385–404). In W. Wiedermann & A. von Eye (Eds.), Causality and statistics. Hoboken: Wiley.Google Scholar
  31. Li, F., Lock Morgan, K., & Zaslavsky, A. M. (2016). Balancing covariates via propensity score weighting. Journal of the American Statistical Association. Advance online publication.
  32. Low, S., Smolkowski, K., & Cook, C. (2016). What constitutes high-quality implementation of SEL programs? A latent class analysis of second step® implementation. Prevention Science, 17, 981–991.CrossRefGoogle Scholar
  33. Lunceford, J. K., & Davidian, M. (2004). Stratification and weighting via the propensity score in estimation of causal treatment effects: A comparative study. Statistics in Medicine, 23, 2937–2960.CrossRefGoogle Scholar
  34. McCaffrey, D. F., Griffin, B. A., Almirall, D., Slaughter, M. E., Ramchand, R., & Burgette, L. F. (2013). A tutorial on propensity score estimation for multiple treatments using generalized boosted models. Statistics in Medicine, 32, 3388–3414.CrossRefGoogle Scholar
  35. Merrill, J. E., Wardell, J. D., & Read, J. P. (2014). Drinking motives in the prospective prediction of unique alcohol-related consequences in college students. Journal of Studies on Alcohol and Drugs, 75, 93–102.CrossRefGoogle Scholar
  36. Miech, R. A., Johnston, L. D., O’Malley, P. M., Bachman, J. G., Schulenberg, J. E., & Patrick, M. E. (2017). Monitoring the Future national survey results on drug use, 1975–2016: Volume I, secondary school students. Ann Arbor, MI: Institute for Social Research, The University of Michigan.Google Scholar
  37. Muthén, L.K. and Muthén, B.O. (2015). Mplus User’s guide (7th ed.) Los Angeles, CA: Muthén & Muthén.Google Scholar
  38. Patrick, M. E., & Schulenberg, J. E. (2011). How trajectories of reasons for alcohol use relate to trajectories of binge drinking: National panel data spanning late adolescence to early adulthood. Developmental Psychology, 47, 311–317.CrossRefGoogle Scholar
  39. Patrick, M. E., Schulenberg, J. E., O’Malley, P. M., Johnston, L., & Bachman, J. (2011). Adolescents’ reported reasons for alcohol and marijuana use as predictors of substance use and problems in adulthood. Journal of Studies on Alcohol and Drugs, 72, 106–116.CrossRefGoogle Scholar
  40. Patrick, M. E., Bray, B. C., & Berglund, P. (2016). Reasons for marijuana use among young adults and long-term associations with marijuana use and problems. Journal on Studies of Alcohol and Drugs, 77, 881–888.CrossRefGoogle Scholar
  41. Patrick, M. E., Evans-Polce, R., Kloska, D. D., Maggs, J. L., & Lanza, S. T. (2018). Age-related changes in associations between reasons for alcohol use and high-intensity drinking across young adulthood. Journal of Studies on Alcohol and Drugs.Google Scholar
  42. R Core Team (2015). R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing. Retreived from
  43. Ridgeway, G., Kovalchik, S. A., Griffin, B. A., & Kabeto, M. U. (2015). Propensity score analysis with survey weighted data. Journal of Causal Inference, 3, 237–249.CrossRefGoogle Scholar
  44. Robins, J. M., Hérnan, M. A., & Brumback, B. (2000). Marginal structural models and causal inference in epidemiology. Epidemiology, 11, 550–560.CrossRefGoogle Scholar
  45. Rosenbaum, P. R., & Rubin, D. B. (1983). The central role of the propensity score in observational studies for causal effects. Biometrika, 70, 41–55.CrossRefGoogle Scholar
  46. Rubin, D. B. (1997). Estimating causal effects from large data sets using propensity scores. Annals of Internal Medicine, 127, 757–763.CrossRefGoogle Scholar
  47. Rubin, D. B. (2001). Using propensity scores to help design observational studies: Application to the tobacco litigation. Health Services & Outcomes Research Methodology, 2, 169–188.CrossRefGoogle Scholar
  48. Schafer, J. L., & Graham, J. W. (2002). Missing data: Our view of the state of the art. Psychological Methods, 7, 147–177.CrossRefGoogle Scholar
  49. Schafer, J. L., & Kang, J. (2008). Average causal effects from nonrandomized studies: A practical guide and simulated example. Psychological Methods, 13, 279–313.CrossRefGoogle Scholar
  50. Schulenberg, J. E., Patrick, M. E., Kloska, D. D., Maslowsky, J., Maggs, J. L., & O’Malley, P. M. (2015). Substance use disorder in early midlife: A national prospective study on health and well-being correlates and long-term predictors. Substance Abuse: Research and Treatment, 9(Suppl 1), 41–57.Google Scholar
  51. Schulenberg, J. E., Johnston, L. D., O’Malley, P. M., Bachman, J. G., Miech, R. A., & Patrick, M. E. (2017). Monitoring the Future national survey results on drug use, 1975–2016: Volume II, college students and adults ages 19–55. Ann Arbor: Institute for Social Research, The University of Michigan.Google Scholar
  52. Schuler, M. S. (2013). Estimating the relative treatment effects of natural clusters of adolescent substance abuse treatment services: Combining latent class analysis and propensity score methods. Unpublished doctoral dissertation. Baltimore: Johns Hopkins University Retrieved from Scholar
  53. Schuler, M. S., Leoutsakos, J. S., & Stuart, E. A. (2014). Addressing confounding when estimating the effects of latent classes on a distal outcome. Health Services Outcomes and Research Methodology, 14, 232–254.CrossRefGoogle Scholar
  54. Spilt, J. L., Koot, J. M., & van Lier, P. A. (2013). For whom does it work? Subgroup differences in the effects of a school-based universal prevention program. Prevention Science, 14, 479–488.CrossRefGoogle Scholar
  55. Stapinski, L. A., Edwards, A. C., Hickman, M., Araya, R., Teesson, M., Newton, N. C., et al. (2016). Drinking to cope: A latent class analysis of coping motives for alcohol use in a large cohort of adolescents. Prevention Science, 17, 584–594.CrossRefGoogle Scholar
  56. Stapleton, J. L., Turrisi, R., Cleveland, M. J., Ray, A. E., & Lu, S. E. (2014). Pre-college matriculation risk profiles and alcohol consumption patterns during the first semesters of college. Prevention Science, 15, 705–715.CrossRefGoogle Scholar
  57. Tan, Z. (2010). Bounded, efficient and doubly robust estimation with inverse weighting. Biometrika, 97, 661–682.CrossRefGoogle Scholar
  58. Van Buuren, S. (2007). Multiple imputation of discrete and continuous data by fully conditional specification. Statistical Methods in Medical Research, 16, 219–242.CrossRefGoogle Scholar
  59. Varvil-Weld, L., Crowley, D. M., Turrisi, R., Greenberg, M. T., & Mallett, K. A. (2014). Hurting, helping, or neutral? The effects of parental permissiveness toward adolescent drinking on college student alcohol use and problems. Prevention Science, 15, 716–724.CrossRefGoogle Scholar
  60. Vermunt, J. K. (2010). Latent class modeling with covariates: Two improved three-step approaches. Political Analysis, 18, 450–469.CrossRefGoogle Scholar
  61. Vermunt, J. K., & Magidson, J. (2015). Upgrade manual for Latent GOLD 5.1. Belmont, MA: Statistical Innovations.Google Scholar
  62. White, I. R., & Thompson, S. G. (2005). Adjusting for partially missing baseline measurements in randomized trials. Statistics in Medicine, 24, 993–1007.CrossRefGoogle Scholar
  63. Yamaguchi, K. (2015). Extensions of Rubin’s causal model for a latent-class treatment variable: An analysis of the effects of employers’ work-life balance policies on women’s income attainment in Japan. Research Institute of Economy, Trade and Industry Discussion Paper Series (No. 15-E-090). Tokyo, Japan: The Research Institute of Economy, Trade and Industry. Retrieved from
  64. Zanutto, E. L. (2006). A comparison of propensity score and linear regression analysis of complex survey data. Journal of Data Science, 4, 67–91.Google Scholar
  65. Zhang, Z., Liu, W., Zhang, B., Tang, L., & Zhang, J. (2016). Causal inference with missing exposure information: Methods and applications to an obstetric study. Statistical Methods in Medical Research, 25, 2053–2066.CrossRefGoogle Scholar

Copyright information

© Society for Prevention Research 2018

Authors and Affiliations

  • Bethany C. Bray
    • 1
    • 2
    Email author
  • John J. Dziak
    • 1
  • Megan E. Patrick
    • 3
  • Stephanie T. Lanza
    • 1
    • 4
    • 5
  1. 1.The Methodology Center, Penn StateUniversity ParkUSA
  2. 2.College of Health and Human Development, Penn StateUniversity ParkUSA
  3. 3.Institute for Social ResearchUniversity of MichiganAnn ArborUSA
  4. 4.Edna Bennett Pierce Prevention Research Center, Penn StateUniversity ParkUSA
  5. 5.Department of Biobehavioral Health, Penn StateUniversity ParkUSA

Personalised recommendations