Overview of Fit Cox Model in a Counting Process Form

Use Fit Cox Model in a Counting Process Form to describe the relationship between predictors and survival. The predictors can be fixed or time-dependent, which means the predictor values can change during the study period. You can include interaction and polynomial terms and perform stepwise selection of terms.

The data must be in the counting process style of input. In the counting process input form, multiple rows represent each subject. Each row describes a time interval when the values of all the variables are constant. Time-dependent predictors change between rows. The intervals begin just after the start time and include the end time.

The counting process form of data is a flexible format that allows you to analyze data that contains both fixed dependent predictors and time-dependent predictors. In addition, this form of input handles recurrent events data. For example, a research study examines the effect of a drug on a tumor. Subjects take a higher dose of a medicine in certain treatment phases, so the medicine dose is not the same over the course of the study. This characteristic makes the medicine dose a time-dependent predictor. Also, a subject can have a tumor go into remission but not leave the study. Because the subject can undergo treatment, remission, and recurrence multiple times over the course of the study, the study has to account for recurrent events.

Key results of comparative studies that use Cox regression often report relative risks for predictors under different treatments. For example, a study on a cancer treatment concludes that the relative risk for two groups is 4, which means that the patients in one group are cancer-free at 4 times the rate of patients in the other group over the study period. Minitab displays relative risks for each variable so that you can easily assess their effect on the frequency of the event.