Content area
Full Text
Health Serv Outcomes Res Method (2013) 13:1838 DOI 10.1007/s10742-012-0086-x
Alex A. Bohl David K. Blough Paul A. Fishman Jeffery R. Harris
Elizabeth A. Phelan
Received: 29 November 2011 / Revised: 16 April 2012 / Accepted: 1 May 2012 / Published online: 15 May 2012 Springer Science+Business Media, LLC 2012
Abstract Generalized additive models for location, scale, and shape (GAMLSS) are a class of semi-parametric models with potential applicability to health care cost data. We compared the bias, accuracy, and coverage of GAMLSS estimators with two distributions [gamma and generalized inverse gaussian (GIG)] using a log link to the generalized linear model (GLM) with log link and gamma family and the log-transformed OLS. The evaluation using simulated gamma data showed that the GAMLSS and GLM gamma model had similar bias, accuracy, and coverage and outperformed the GAMLSS GIG. When applied to simulated GIG data, the GLM gamma was similar or improved in bias, accuracy, and coverage compared to the GAMLSS GIG and gamma; furthermore, the GAMLSS estimators produced wildly inaccurate or overly-precise results in certain circumstances. Applying all models to empirical data on health care costs after a fall-related injury, all estimators produced similar coefcient estimates, but GAMLSS estimators produced spuriously smaller standard errors. Although no single alternative was best for all simulations, the GLM gamma was the most consistent, so we recommend against using
A. A. Bohl (&) J. R. Harris E. A. Phelan
Health Promotion Research Center, Department of Health Services, School of Public Health, University of Washington, 1107 NE 45th St., Suite 200, Seattle, WA 98105, USAe-mail: [email protected]
A. A. Bohl P. A. Fishman J. R. Harris E. A. Phelan
Department of Health Services, University of Washington, Seattle, WA, USA
A. A. Bohl
Mathematica Policy Research, Inc., Cambridge, MA, USA
D. K. Blough
Department of Pharmacy, University of Washington, Seattle, WA, USA
P. A. Fishman
Group Health Research Institute, Seattle, WA, USA
E. A. Phelan
Division of Gerontology and Geriatric Medicine, Department of Medicine, University of Washington, Seattle, WA, USA
Are generalized additive models for location, scale, and shape an improvement on existing modelsfor estimating skewed and heteroskedastic cost data?
123
Health Serv Outcomes Res Method (2013) 13:1838 19
GAMLSS estimators using GIG or gamma to test for differences in mean health...