A multi-level two-part random effects model, with application to an alcohol-dependence study

Division of Biostatistics and Epidemiology, Department of Public Health Sciences, University of Virginia, Charlottesville, VA 22908-0717, USA.
Statistics in Medicine (Impact Factor: 2.04). 08/2008; 27(18):3528-39. DOI: 10.1002/sim.3205
Source: PubMed

ABSTRACT Two-part random effects models (J. Am. Statist. Assoc. 2001; 96:730-745; Statist. Methods Med. Res. 2002; 11:341-355) have been applied to longitudinal studies for semi-continuous outcomes, characterized by a large portion of zero values and continuous non-zero (positive) values. Examples include repeated measures of daily drinking records, monthly medical costs, and annual claims of car insurance. However, the question of how to apply such models to multi-level data settings remains. In this paper, we propose a novel multi-level two-part random effects model. Distinct random effects are used to characterize heterogeneity at different levels. Maximum likelihood estimation and inference are carried out through Gaussian quadrature technique, which can be implemented conveniently in freely available software-aML. The model is applied to the analysis of repeated measures of the daily drinking record in a randomized controlled trial of topiramate for alcohol-dependence treatment.

  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The decline of fisheries over recent decades and a growing human population has coincided with an increase in aquaculture production. As farmed fish densities increase, so have their rates of infectious diseases, as predicted by the theory of density-dependent disease transmission. One of the pathogen that has increased with the growth of salmon farming is sea lice. Effective management of this pathogen requires an understanding of the spatial scale of transmission. We used a two-part multi-scale model to account for the zero-inflated data observed in weekly sea lice abundance levels on rainbow trout and Atlantic salmon farms in Chile, and to assess internal (farm) and external (regional) sources of sea lice infection. We observed that the level of juvenile sea lice was higher on farms that were closer to processing plants with fish holding facilities. Further, evidence for sea lice exposure from the surrounding area was supported by a strong positive correlation between the level of juvenile sea lice on a farm and the number of gravid females on neighboring farms within 30km two weeks prior. The relationship between external sources of sea lice from neighboring farms and juvenile sea lice on a farm was one of the strongest detected in our multivariable model. Our findings suggest that the management of sea lice should be coordinated between farms and should include all farms and processing plants with holding facilities within a relatively large geographic area. Understanding the contribution of pathogens on a farm from different sources is an important step in developing effective control strategies.
    Preventive Veterinary Medicine 04/2013; DOI:10.1016/j.prevetmed.2013.03.015 · 2.51 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: We present an architecture that enables developers to build applications that can flexibly control downloaded executable content. The architecture includes an access control model for representing security requirements and a browser service for deriving application requirements from signed content messages and executing content in limited domains.
    Object-Orientation in Operating Systems, 1996., Proceedings of the Fifth International Workshop on; 11/1996
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: In this paper we propose a four-part random effects model, with application to correlated medical cost data. Four joint equations are used to model respectively: (1) the probability of seeking medical treatment, (2) the probability of being hospitalized (conditional on seeking medical treatment), and the actual amount of (3) outpatient and (4) inpatient costs. Our model simultaneously takes account of the inter-temporal (or within-cluster) correlation of each patient and the cross-equation correlation of the four equations, by means of joint linear mixed models and generalized linear mixed models. The estimation is accomplished by the high-order Laplace approximation technique in Raudenbush et al. [Raudenbush, S.W., Yang, M., Yosef, M., 2000. Maximum likelihood for generalized linear models with nested random effects via high-order, multivariate Laplace approximation. Journal of Computational and Graphical Statistics 9, 141-157] and Olsen and Schafer [Olsen, M.K., Schafer, J.L., 2001. A two-part random effects model for semicontinuous longitudinal data. Journal of the American Statistical Association 96, 730-745]. Our model is used to analyze monthly medical costs of 1397 chronic heart failure patients from the clinical data repository (CDR) at the University of Virginia.
    Computational Statistics & Data Analysis 05/2008; 52(9):4458-4473. DOI:10.1016/j.csda.2008.02.034 · 1.15 Impact Factor