Marginalized two-part models for semicontinuous data with application to medical costs
Public DepositedAdd to collection
You do not have access to any existing collections. You may create a new collection.
Downloadable Content
Download PDFCitation
MLA
Smith, Valerie. Marginalized Two-part Models for Semicontinuous Data with Application to Medical Costs. Chapel Hill, NC: University of North Carolina at Chapel Hill Graduate School, 2015. https://doi.org/10.17615/4r0v-wk40APA
Smith, V. (2015). Marginalized two-part models for semicontinuous data with application to medical costs. Chapel Hill, NC: University of North Carolina at Chapel Hill Graduate School. https://doi.org/10.17615/4r0v-wk40Chicago
Smith, Valerie. 2015. Marginalized Two-Part Models for Semicontinuous Data with Application to Medical Costs. Chapel Hill, NC: University of North Carolina at Chapel Hill Graduate School. https://doi.org/10.17615/4r0v-wk40- Last Modified
- March 19, 2019
- Creator
-
Smith, Valerie
- Affiliation: Gillings School of Global Public Health, Department of Biostatistics
- Abstract
- In health services research, it is common to encounter semicontinuous data characterized by a point mass at zero followed by a right-skewed continuous distribution with positive support. Examples include health expenditures, in which the zeros represent a subpopulation of patients who do not use health services, while the continuous distribution describes the level of expenditures among health services users. Semicontinuous data are typically analyzed using two-part mixture models that separately model the probability of health services use and the distribution of positive expenditures among users. However, because the second part conditions on a nonzero response, conventional two-part models do not provide a marginal interpretation of covariate effects on the overall population of health service users and non-users, even though this is often of greatest interest to investigators. Here, we propose a marginalized two-part model that yields more interpretable effect estimates in two-part models by parameterizing the model in terms of the marginal mean. This model maintains many of the important features of conventional two-part models, such as capturing zero-inflation and skewness, but allows investigators to examine covariate effects on the overall marginal mean, a target of primary interest in many applications. Using a simulation study, we examine properties of the maximum likelihood estimators from this model. We illustrate the approach by evaluating the effect of a behavioral weight loss intervention on health care expenditures in the Veterans Affairs (VA) health care system. We then extend this marginalized two-part model to clustered or longitudinal data structures by incorporating random effects. This longitudinal marginalized two-part model is fit following a fully Bayesian approach with non-informative or weakly informative prior distributions, and we illustrate it by analyzing the effect of a copayment increase in the VA health system. Finally, using simulation studies, we compare the performance of the marginalized two-part model to commonly used one-part generalized linear models (GLMs) fit via quasi-likelihood estimation over a range of simulated data scenarios with varying percentages of zero-valued observations.
- Date of publication
- May 2015
- Keyword
- Subject
- DOI
- Identifier
- Resource type
- Rights statement
- In Copyright
- Advisor
- Koch, Gary
- Herring, Amy
- Preisser, John
- Maciejewski, Matthew
- Neelon, Brian
- Degree
- Doctor of Public Health
- Degree granting institution
- University of North Carolina at Chapel Hill Graduate School
- Graduation year
- 2015
- Language
- Publisher
- Place of publication
- Chapel Hill, NC
- Access right
- There are no restrictions to this item.
- Date uploaded
- June 23, 2015
Relations
- Parents:
This work has no parents.
Items
Thumbnail | Title | Date Uploaded | Visibility | Actions |
---|---|---|---|---|
Smith_unc_0153D_15272.pdf | 2019-04-09 | Public | Download |