General relative risk functions for case-control studies.

While multiplicative (log-linear and logistic) models have a firmly established place in epidemiologic methodology, additive and other more general model structures are needed also. The authors propose a parametric family of relative risk functions ranging from subadditive to supramultiplicative that is generated by varying the exponent in a power transform for the log relative risk. The choice of model is facilitated by graphic analysis of goodness-of-fit statistics computed for various values of the exponent. Intermediate quantities available as by-products of the fit are useful for checking the influence of particular observations on the estimated regression coefficients. Three examples illustrate the applications of these methods to random, stratified, and matched samples of cases and controls. Computer software is available for each of these situations. Even though different relative risk models may have markedly different implications for the multifactorial nature of the disease process, it may be difficult to distinguish between them unless the data are quite extensive.

[1]  S. Greenland Limitations of the logistic analysis of epidemiologic data. , 1979, American journal of epidemiology.

[2]  David R. Cox,et al.  Regression models and life tables (with discussion , 1972 .

[3]  N. Breslow,et al.  Cohort Analysis in Epidemiology , 1985 .

[4]  D. Pregibon Logistic Regression Diagnostics , 1981 .

[5]  N Breslow,et al.  Are there two logistic regressions for retrospective studies? , 1978, Biometrics.

[6]  A. G. Munford,et al.  The Combined Effect of Two Factors on Disease in a Case‐Control Study , 1980 .

[7]  H. Dayal Additive excess risk model for epidemiologic interaction in retrospective studies. , 1980, Journal of chronic diseases.

[8]  P. Holland,et al.  Discrete Multivariate Analysis. , 1976 .

[9]  Víctor M. Guerrero,et al.  Use of the Box-Cox transformation with binary response models , 1982 .

[10]  Duncan C. Thomas General relative-risk models for survival time and matched case-control analysis , 1981 .

[11]  I. Bross,et al.  A study of etiological factors in cancer of the esophagus , 1961, Cancer.

[12]  Robin Thompson,et al.  Composite Link Functions in Generalized Linear Models , 1981 .

[13]  Norman E. Breslow,et al.  Odds ratio estimators when the data are sparse , 1981 .

[14]  J. Nelder,et al.  The GLIM System Release 3. , 1979 .

[15]  Norman E. Breslow,et al.  Maximum Likelihood Fitting of General Risk Models to Stratified Data , 1983 .

[16]  G. Berry Dose-response in case-control studies. , 1980, Journal of Epidemiology and Community Health.

[17]  N. Breslow,et al.  Statistical methods in cancer research. Vol. 1. The analysis of case-control studies. , 1981 .

[18]  K J Rothman,et al.  The estimation of synergy or antagonism. , 1976, American journal of epidemiology.

[19]  K. Rothman,et al.  The effect of joint exposure to alcohol and tobacco on risk of cancer of the mouth and pharynx. , 1972, Journal of chronic diseases.

[20]  R. Saracci Asbestos and lung cancer: An analysis of the epidemiological evidence on the asbestos—smoking interaction , 1977, International journal of cancer.

[21]  R. Pyke,et al.  Logistic disease incidence models and case-control studies , 1979 .

[22]  P. McCullagh,et al.  Generalized Linear Models , 1984 .

[23]  J. Anderson Separate sample logistic discrimination , 1972 .

[24]  D. Cox,et al.  An Analysis of Transformations , 1964 .

[25]  K. Rothman,et al.  Models of varying parametric form in case-referent studies. , 1982, American journal of epidemiology.