Modeling risk using generalized linear models

被引:333
作者
Blough, DK
Madden, CW [1 ]
Hornbrook, MC
机构
[1] Univ Washington, Sch Publ Hlth & Community Med, Dept Hlth Serv, Seattle, WA 98195 USA
[2] Kaiser Permanente Ctr Hlth Res, Portland, OR USA
关键词
medical risk; two-part model; generalized linear model; quasi-likelihood; model calibration;
D O I
10.1016/S0167-6296(98)00032-0
中图分类号
F [经济];
学科分类号
02 ;
摘要
Traditionally, linear regression has been the technique of choice for predicting medical risk. This paper presents a new approach to modeling the second part of two-part models utilizing extensions of the generalized linear model. The primary method of estimation for this model is maximum Likelihood. This method as well as the generalizations quasi-likelihood and extended quasi-likelihood are discussed. An example using medical expense data from Washington State employees is used to illustrate the methods. The model includes demographic variables as well as an Ambulatory Care Group variable to account for prior health status. (C) 1999 Elsevier Science B.V. All rights reserved. JEL classification: I11.
引用
收藏
页码:153 / 171
页数:19
相关论文
共 18 条
  • [1] Ash A, 1989, Health Care Financ Rev, V10, P17
  • [2] A CHRONIC DISEASE SCORE WITH EMPIRICALLY DERIVED WEIGHTS
    CLARK, DO
    VONKORFF, M
    SAUNDERS, K
    BALUCH, WM
    SIMON, GE
    [J]. MEDICAL CARE, 1995, 33 (08) : 783 - 795
  • [3] COPAS JB, 1987, J ROY STAT SOC B MET, V49, P175
  • [4] COPAS JB, 1983, J R STAT SOC B, V45, P311
  • [6] Duan N., 1983, J EC BUSINESS STATIS, V1, P115, DOI [DOI 10.1080/07350015.1983.10509330, DOI 10.2307/1391852]
  • [7] Duan N., 1984, J BUS ECON STAT, V2, P283, DOI DOI 10.1080/07350015.1984.10509396
  • [8] FOWLES JB, 1994, COMP ALTERNATIVE APP
  • [9] HARRELL FE, 1996, PREDICTING OUTCOMES
  • [10] Hay J., 1984, J EC BUSINESS STAT, V2, P279