Free Trial

Safari Books Online is a digital library providing on-demand subscription access to thousands of learning resources.

  • Create BookmarkCreate Bookmark
  • Create Note or TagCreate Note or Tag
  • PrintPrint

CONCLUSION

As we have seen, competing risks analysis is easily accomplished with conventional software by doing a separate analysis for each event type, treating other events as censoring. The biggest drawback of competing risks analysis is the requirement that times for different event types be independent or at least that each event be noninformative for the others. This requirement is exactly equivalent to the requirement for random censoring discussed in Chapter 2. In either case, violations can lead to biased coefficient estimates.

The seriousness of this problem depends greatly on the particular application. For the LEADERS data set, I argued that death due to natural causes is likely to be noninformative for the risk of either a constitutional or a nonconstitutional exit. For the latter two types, however, the presumption of noninformativeness may be unreasonable. In thinking about this issue, it is helpful to ask the question, "Are there unmeasured variables that may affect more than one event type?" If the answer is yes, then there should be a presumption of dependence.


  

You are currently reading a PREVIEW of this book.

                                                                                                                    

Get instant access to over $1 million worth of books and videos.

  

Start a Free Trial


  
  • Safari Books Online
  • Create BookmarkCreate Bookmark
  • Create Note or TagCreate Note or Tag
  • PrintPrint