Journal
COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE
Volume 129, Issue -, Pages 63-70Publisher
ELSEVIER IRELAND LTD
DOI: 10.1016/j.cmpb.2016.03.008
Keywords
Competing risks; Multiple-outcome data; Cumulative incidence function; Sub distribution function; Inverse probability of treatment weighting; Propensity score
Ask authors/readers for more resources
In observational studies without random assignment of the treatment, the unadjusted comparison between treatment groups may be misleading due to confounding. One method to adjust for measured confounders is inverse probability of treatment weighting. This method can also be used in the analysis of time to event data with competing risks. Competing risks arise if for some individuals the event of interest is precluded by a different type of event occurring before, or if only the earliest of several times to event, corresponding to different event types, is observed or is of interest. In the presence of competing risks, time to event data are often characterized by cumulative incidence functions, one for each event type of interest. We describe the use of inverse probability of treatment weighting to create adjusted cumulative incidence functions. This method is equivalent to direct standardization when the weight model is saturated. No assumptions about the form of the cumulative incidence functions are required. The method allows studying associations between treatment and the different types of event under study, while focusing on the earliest event only. We present a SAS macro implementing this method and we provide a worked example. (C) 2016 Elsevier Ireland Ltd. All rights reserved.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available