

   SSmmookkiinngg,, AAllccoohhooll aanndd ((OO))eessoopphhaaggeeaall CCaanncceerr

        data(esoph)

   FFoorrmmaatt::

        data frame with records for 88 age/alcohol/tobacco com-
        binations.

         [,1]     "agegp"          Age group               1  25-34 years
                                                           2  35-44
                                                           3  45-54
                                                           4  55-64
                                                           5  65-74
                                                           6  75+
         [,2]     "alcgp"          Alcohol consumption     1   0-39 gm/day
                                                           2  40-79
                                                           3  80-119
                                                           4  120+
         [,3]     "tobgp"          Tobacco consumption     1   0- 9 gm/day
                                                           2  10-19
                                                           3  20-29
                                                           4  30+
         [,4]     "ncases"         Number of cases
         [,5]     "ncontrols"      Number of subjects

   DDeessccrriippttiioonn::

        Data from a case-control study of (o)esophageal cancer
        in Ile-et-Vilaine, France.

   AAuutthhoorr((ss))::

        Thomas Lumley

   SSoouurrccee::

        Breslow and Day (1980).  "Statistical Methods in Cancer
        Research.  1: The Analysis of Case-control studies";
        IARC Lyon.

   EExxaammpplleess::

        data(esoph)
        summary(esoph)
        ## effects of alcohol, tobacco and interaction, age-adjusted
        model1 <- glm(cbind(ncases, ncontrols) ~ agegp + tobgp * alcgp,
                      data = esoph, family = binomial())
        anova(model1)
        ## Try a linear effect of alcohol and tobacco
        model2 <- glm(cbind(ncases, ncontrols) ~ agegp + codes(tobgp) + codes(alcgp),
                      data = esoph, family = binomial())
        summary(model2)

