Mental Health Services Research

, Volume 6, Issue 4, pp 239–246 | Cite as

A Method for Analyzing Longitudinal Outcomes with Many Zeros

  • Haiyi Xie
  • Gregory McHugo
  • Anjana Sengupta
  • Robin Clark
  • Robert Drake

Abstract

Health care utilization and cost data have challenged analysts because they are often correlated over time, highly skewed, and clumped at 0. Traditional approaches do not address all these problems, and evaluators of mental health and substance abuse interventions often grapple with the problem of how to analyze these data in a way that accurately represents program impact. Recently, the traditional 2-part model has been extended to mixed-effects mixed-distribution model with correlated random effects to deal simultaneously with excess zeros, skewness, and correlated observations. We introduce and demonstrate this new method to mental health services researchers and evaluators by analyzing the data from a study of assertive community treatment (ACT). The response variable is the number of days of hospitalization, collected every 6 months over 3 years. The explanatory variable is group: ACT vs. standard case management. Diagnosis (schizophrenia vs. bipolar disorder), time, and the baseline values of hospital days are covariates. Results indicate that clients in the ACT group have a higher probability of hospital admission, but tend to have shorter lengths of stay. The mixed-distribution model provides greater specification of a model to fit these data and leads to more refined interpretation of the results.

mixed distribution excess zeros two-part model repeated measures assertive community treatment 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

REFERENCES

  1. Drake, R., McHugo, G., Clark, R., Teague, G., Xie, H., & Miles, K. (1998). Assertive community treatment for patients with co-occurring severe mental illness and substance abuse disorder: A clinical trial. American Journal of Orthopsychiatry, 68(2), 201–215.Google Scholar
  2. Duan, N., Manning, W. G., Morris, C. N., & Newhouse, J. P. (1983). A comparison of alternative models for the demand for medical care. Journal of Economic and Business Statistics, 1, 115–126.Google Scholar
  3. Green, W. (1994). Accounting for excess zeros and sample selection in Poisson and negative binomial regression models.(Working paper EC-94-10). New York: New York University, Department of Economics.Google Scholar
  4. Grunwald, G. K., & Jones, R. H. (2000). Markov models for time series with mixed distribution. Environmetrics, 11, 7–339.Google Scholar
  5. Hajivassiliou, V. A. (1994). Asimulation estimation analysis of the external debt crises of developing countries. Journal of Applied Econometrics, 9,1–131.Google Scholar
  6. Hall, D. B. (2000). Zero-inflated Poisson and binomial regression with random effects: A case study. Biometrics, 56, 1030–1039.Google Scholar
  7. Heckman, J. (1974). Shadow prices, market wages, and labor supply. Econometrica, 42, 674–679.Google Scholar
  8. Heckman, J. (1976). The common structure of statistical models of truncation, sample selection, and limited dependent varibales, and a sample estimator for such models. The Annals of Economic and Social Measurement, 5, 475–592.Google Scholar
  9. Hur, K. (1999). A random-effects zero-inflated Poisson regression model for clustered extra-zero counts. Unpublished PhD Dessertation, University of Illinois at Chicago.Google Scholar
  10. Lachenbruch, P. (1976). Analysis of data with clumping at zero. Biometrische Zeitschrift, 18, 351–356.Google Scholar
  11. Lachenbruch, P. (1992). Utility of regression analysis in epidemiologic studies of the elderly. In W. R. Wallace (Ed.), The epidemiologic study of the elderly (pp. 371–381). Oxford, UK: Oxford University Press.Google Scholar
  12. Laird, N. M., & Ware, H. (1982). Random-effect models for longitudinal data. Biometrics, 38, 963–974.Google Scholar
  13. Lambert, D. (1992). Zero-inflated Poisson regression, with an application to detects in manufacturing. Technometrics, 34, 1–14.Google Scholar
  14. Manning, W. G., Duan, N., & Rogers, W. H. (1987). Monte Carlo evidence on the choice between sample selection and two-part models. Journal of Econometrics, 35, 59–82.Google Scholar
  15. Muthen, L. K., & Muthen, B. O. (2004). Mplus User's Guide (Version Three). Los Angeles, CA: Muthen & Muthen.Google Scholar
  16. Olsen, M., & Schafer, J. (2001). A two-part random-effects model for semicontinuous longitudinal data. Journal of American Statistical Association, 96, 730–745.Google Scholar
  17. Tobin, J. (1958). Estimation of relationships for limited dependent variables. Econometrica, 26, 24–36.Google Scholar
  18. Tooze, A. J., Grunwald, G. K., & Jones, R. H. (2002). Analysis of repeated measures data with clumping at zero. Statistical Methods in Medical Research, 11, 341–355.Google Scholar
  19. Xie, H., Hur, K., & McHugo, G. (2000). Using random-effects zero-inflated Poisson model to analyze longitudinal count data with extra zeros. Paper presented at the 53rd Session of the Inter-national Statistical Institute, Seoul, South Korea.Google Scholar
  20. Yau, K., & Lee, A. (2001). Zero-inflated Poisson regression with random effects to evaluate an occupational injury prevention program. Statistics in Medicine, 20, 2907–2920.Google Scholar
  21. Zeger, S., Liang, K., & Albert, P. (1988). Models for longitudinal data: A generalized estimating equation approach. Biometrics, 44, 1049–1060.Google Scholar

Copyright information

© Springer Science+Business Media, Inc. 2004

Authors and Affiliations

  • Haiyi Xie
    • 1
    • 2
  • Gregory McHugo
    • 1
    • 2
  • Anjana Sengupta
    • 1
    • 2
  • Robin Clark
    • 3
  • Robert Drake
    • 1
    • 2
  1. 1.Dartmouth Medical SchoolLebanon
  2. 2.New Hampshire-Dartmouth Psychiatric Research CenterLebanon
  3. 3.Center for Health Policy and ResearchUniversity of Massachusetts Medical SchoolUSA

Personalised recommendations