Communications for Statistical Applications and Methods

The Korean Statistical Society (한국통계학회)
권호 표지
DOI QR코드

DOI QR Code

Weighted zero-inflated Poisson mixed model with an application to Medicaid utilization data

  • Lee, Sang Mee (Department of Public Health Sciences, University of Chicago) ;
  • Karrison, Theodore (Department of Public Health Sciences, University of Chicago) ;
  • Nocon, Robert S. (Department of Medicine, University of Chicago) ;
  • Huang, Elbert (Department of Medicine, University of Chicago)
  • Received : 2017.09.28
  • Accepted : 2017.12.24
  • Published : 2018.03.31
Download PDF
⟨ Previous Next ⟩

Abstract

In medical or public health research, it is common to encounter clustered or longitudinal count data that exhibit excess zeros. For example, health care utilization data often have a multi-modal distribution with excess zeroes as well as a multilevel structure where patients are nested within physicians and hospitals. To analyze this type of data, zero-inflated count models with mixed effects have been developed where a count response variable is assumed to be distributed as a mixture of a Poisson or negative binomial and a distribution with a point mass of zeros that include random effects. However, no study has considered a situation where data are also censored due to the finite nature of the observation period or follow-up. In this paper, we present a weighted version of zero-inflated Poisson model with random effects accounting for variable individual follow-up times. We suggested two different types of weight function. The performance of the proposed model is evaluated and compared to a standard zero-inflated mixed model through simulation studies. This approach is then applied to Medicaid data analysis.

Keywords

References

  1. Emerson SS, McGee DL, Fennerty B, Hixson L, Garewal H, and Alberts D (1993). Design and analysis of studies to reduce the incidence of colon polyps. Statistics in Medicine, 12, 339-351. https://doi.org/10.1002/sim.4780120316
  2. Goodman DC, Mick SS, Bott D, et al. (2003). Primary care service areas: a new tool for the evaluation of primary care services. Health Services Research, 38, 287-309. https://doi.org/10.1111/1475-6773.00116
  3. Hall DB (2000). Zero-inflated and binomial regression with random effects: a case study, Biometrics, 56, 1030-1039.
  4. Hsu CH (2005). Joint modelling of recurrence and progression of adenomas: a latent variable approach. Statistical Modelling, 5, 201-215. https://doi.org/10.1191/1471082X05st094oa
  5. Hsu CH (2007). A weighted zero-inflated Poisson model for estimation of recurrence of adenomas. Statistical Methods in Medical Research, 16, 155-166. https://doi.org/10.1177/0962280206071929
  6. Lambert D (1992). Zero-inflated Poisson regression, with an application to defects in manufacturing. Technometrics, 34, 1-14. https://doi.org/10.2307/1269547
  7. Lee AH, Wang K, Scott JA, Yau KK, and McLachlan GJ (2006). Multi-level zero-inflated Poisson regression modelling of correlated count data with excess zeros. Statistical Methods in Medical Research, 15, 47-61. https://doi.org/10.1191/0962280206sm429oa
  8. Lee AH, Wang K, and Yau KK (2001). Analysis of zero-inflated Poisson incorporating extent of exposure. Biometrical Journal, 43, 963-975. https://doi.org/10.1002/1521-4036(200112)43:8<963::AID-BIMJ963>3.0.CO;2-K
  9. McGilchrist C and Yau K (1995). The derivation of BLUP, ML, REML estimation methods for generalised linear mixed models. Communications in Statistics-Theory and Methods, 24, 2963-2980.
  10. Min Y and Agresti A (2005). Random effect models for repeated measures of zero-inflated count data. Statistical Modelling, 5, 1-19. https://doi.org/10.1191/1471082X05st084oa
  11. Moghimbeigi A, Eshraghian MR, Mohammad K, and McArdle B (2008). Multilevel zero-inflated negative binomial regression modeling for over-dispersed count data with extra zeros. Journal of Applied Statistics, 35, 1193-1202. https://doi.org/10.1080/02664760802273203
  12. Monod A (2014). Random effects modeling and the zero-inflated Poisson distribution. Communications in Statistics, 43, 664-680. https://doi.org/10.1080/03610926.2013.814782
  13. Mullahy J (1986). Specification and testing of some modified count data models. Journal of Econometrics, 33, 341-365.
  14. Ridout M, Demetrio CG, and Hinde J (1998). Models for count data with many zeros. In Proceedings of the XIXth International Biometric Conference, 19, 179-192.
  15. Yau KK and Lee AH (2001). Zero-inflated Poisson regression with random effects to evaluate an occupational injury prevention programme. Statistics in Medicine, 20, 2907-2920. https://doi.org/10.1002/sim.860