Effect of progabide on frequency of seizures.

SUMMARY:

Epileptic Seizure data come from a clinical trial of 59 epileptic patients, randomized to either the anti-epileptic drug progabide, or to a placebo. Before treatment, the number of seizures in an eight-week period were counted for each patient as a baseline measurement. Then the number of seizures in four consecutive two-week periods under treatment were counted. The objective is to determine whether the new treatment is effective at reducing the rate of seizures among epileptic patients. The data frame seizure has six columns. y , time, Subject, group.

DATA DESCRIPTION:

ARGUMENTS:

y
the number of seizures in the previous time period (two or eight weeks).
Subject
an identifier for each patient.
Time
the time in weeks since the beginning of the study.
group
the group that the patient belonged to, placebo or progabide.

SOURCE:

Thall and Vail (1990); Breslow and Clayton (1993)

EXAMPLES:

seizure.gee<- gee(y~Time+group,cluster=Subject,family="poisson",link="log", 
         correlation=list(type="exchangeable"),variance="glm.scale", data=Seizure)