Data with repeated events are known as multivariate failure time data. With this type of data individuals may experience the event, such as a recurrent infection, multiple times. A number of different models can be fit using PROC PHREG. The COVSANDWICH|COVS option on the PROC PHREG statement can be used with the AGGREGATE suboption and the ID statement to identify clusters of related observations. This use provides a robust covariance matrix when the observations are clustered (not independent) and is often employed in the analysis of multivariate failure time data.
A good source for information is the PROC PHREG documentation. The "Overview" section and "Analysis of Multivariate Failure Time Data" in the Details section of the PHREG documentation provide a thorough discussion of this type of data and of the different types of models that can be fit. See also the example titled "Analysis of Recurrent Events Data" in the PHREG documentation.
The course Survival Analysis Using the Proportional Hazards Model discusses a number of these models including the Andersen-Gill (1982) or AG counting-process model, the Prentice, Williams and Prentice (1981) or PWP conditional models, and the Wei, Lin and Weissfeld (1989) or WLW marginal model. The course notes are available for purchase, are written at an approachable level, and give some advice on which model might be appropriate for a given situation. Much of the information in the course notes can also be found in the "Recurrent Event Models" section of Hosmer and Lemeshow (1999). This text and other references are listed below.
Andersen, P. K., Borgan, Ø., Gill, R. D., and Keiding, N. (1992), Statistical Models Based on Counting Processes, Springer-Verlag, New York, NY.
Andersen, P. K. and Gill, R. D. (1982), "Cox's Regression Model Counting Process: A Large Sample Study," Annals of Statistics, 10, 1100 - 1120.
Hosmer, D.W., Jr. and Lemeshow, S. (1999) Applied Survival Analysis: Regression Modeling of Time to Event Data, John Wiley & Sons, New York, NY.
Prentice, R. L., Williams, B. J., and Peterson, A. V. (1981), "On the Regression Analysis of Multivariate Failure Time Data," Biometrika, 68, 373 - 379.
Wei, L. J., Lin, D. Y., and Weissfeld, L. (1989), "Regression Analysis of Multivariate Incomplete Failure Time Data by Modeling Marginal Distribution," Journal of the American Statistical Association, 84, 1065 - 1073.