This Article Full Text (PDF) References Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Saved Citations Download to citation manager Request Permissions Request Reprints Add to My Marked Citations Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Citing Articles via Scopus Google Scholar Articles by Lipscomb, J. Articles by Matchar, D. B. Search for Related Content PubMed PubMed Citation Articles by Lipscomb, J. Articles by Matchar, D. B. Social Bookmarking                         What's this?

Predicting the Cost of Illness

A Comparison of Alternative Models Applied to Stroke

Marek Ancukiewicz

Giovanni Parmigiani

Vic Hasselblad

Greg Samsa

David B. Matchar

Predictions of cost over well-defined time horizons are frequently required in the anal ysis of clinical trials and social experiments, for decision models investigating the cost-effectiveness of interventions, and for macro-level estimates of the resource im pact of disease. With rare exceptions, cost predictions used in such applications con tinue to take the form of deterministic point estimates. However, the growing availability of large administrative and clinical data sets offers new opportunities for a more general approach to disease cost forecasting: the estimation of multivariable cost functions that yeld predictions at the individual level, conditional on intervention(s), patient charac teristics, and other factors. This raises the fundamental question of how to choose the "best" cost model for a given application. The central purpose of this paper is to demonstrate how to evaluate competing models on the basis of predictive validity. This concept is operationalized according to three alternative criteria: 1) root mean square error (RMSE), for evaluating predicted mean cost; 2) mean absolute error (MAE), for evaluating predicted median cost; and 3) a logarithmic scoring rule (log score), an information-theoretic index for evaluating the entire predictive distribution of cost. To illustrate these concepts, the authors conducted a split-sample analysis of data from a national sample of Medicare-covered patients hospitalized for ischemic stroke in 1991 and followed to the end of 1993. Using test and training samples of about 500,000 observations each, they investigated five models: single-equation linear models, with and without log transform of cost; two-part (mixture) models, with and without log transform, to directly address the problem of zero-cost observations; and a Cox pro portional-hazards model stratified by time interval. For deriving the predictive distri bution of cost, the log transformed two-part and proportional-hazards models are su perior. For deriving the predicted mean or median cost, these two models and the commonly used log-transformed linear model all perform about the same. The untrans formed models are dominated in every instance. The approaches to model selection illustrated here can be applied across a wide range of settings. Key words: cost anal ysis ; cost of illness; statistical models; econometric models; stroke; cerebrovascular disease. (Med Decis Making 1998;18 suppl:S39-S56)

Medical Decision Making, Vol. 18, No. 2, S39-S56 (1998)
DOI: 10.1177/0272989X9801800207


CiteULike    Complore    Connotea    Del.icio.us    Digg    Reddit    Technorati    Twitter    What's this?

This article has been cited by other articles:

				R. Luengo-Fernandez, A. M. Gray, and P. M. Rothwell Costs of Stroke Using Patient-Level Data: A Critical Review of the Literature Stroke, February 1, 2009; 40(2): e18 - e23. [Abstract] [Full Text] [PDF]

				W. E. Pelham, E. M. Foster, and J. A. Robb The Economic Impact of Attention-Deficit/Hyperactivity Disorder in Children and Adolescents J. Pediatr. Psychol., July 1, 2007; 32(6): 711 - 727. [Abstract] [Full Text] [PDF]

				J. Gundgaard Income-related inequality in utilization of health services in Denmark: Evidence from Funen County Scand J Public Health, October 1, 2006; 34(5): 462 - 471. [Abstract] [PDF]

				J. Lipscomb, M. S. Donaldson, N. K. Arora, M. L. Brown, S. B. Clauser, A. L. Potosky, B. B. Reeve, J. H. Rowland, C. F. Snyder, and S. H. Taplin Cancer Outcomes Research J Natl Cancer Inst Monographs, October 1, 2004; 2004(33): 178 - 197. [Full Text] [PDF]

				J. Cook, M. Drummond, and J. F Heyse Economic endpoints in clinical trials Statistical Methods in Medical Research, April 1, 2004; 13(2): 157 - 176. [Abstract] [PDF]

				T. H. Wagner, S. Chen, and P. G. Barnett Using Average Cost Methods to Estimate Encounter-Level Costs for Medical-Surgical Stays in the VA Med Care Res Rev, September 1, 2003; 60(3_suppl): 15S - 36S. [Abstract] [PDF]

				N. J. Cooper, A. J. Sutton, M. Mugford, and K. R. Abrams Use of Bayesian Markov Chain Monte Carlo Methods to Model Cost-of-Illness Data Med Decis Making, January 1, 2003; 23(1): 38 - 53. [Abstract] [PDF]

				S. D. Reed, D. K. Blough, K. Meyer, and J. G. Jarvik Inpatient costs, length of stay, and mortality for cerebrovascular events in community hospitals Neurology, July 24, 2001; 57(2): 305 - 314. [Abstract] [Full Text] [PDF]

				J. L. Wagner, S. R. Alberts, J. A. Sloan, S. Cha, J. Killian, M. J. O'Connell, P. Van Grevenhof, J. Lindman, and C. G. Chute Incremental Costs of Enrolling Cancer Patients in Clinical Trials: a Population-Based Study J Natl Cancer Inst, May 19, 1999; 91(10): 847 - 853. [Abstract] [Full Text] [PDF]