I am conducting a study on a cohort of people with a follow-up period of 7 years. I wish to use Cox Proportional Hazard model to estimate HR between an exposure and the length of time of an event. One missing information is the date of birth for the all subjects, but month and year are available.This prevents the calculation of exact age at the time of the study.

Any suggestions will be much appreciated? Any sensitivity analysis should be conducted?


  • $\begingroup$ How old are the subjects at the start of the study? $\endgroup$ – onestop Dec 5 '10 at 10:48

It is hard to imagine a situation when the effect of age with a precision of a month is not sufficient - even for babies after the first few months of life nobody uses weeks. For adults, even rounding to years should be just fine.

| cite | improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy