Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
| From | roland andersson <rolandersson@gmail.com> |
| To | statalist@hsphsun2.harvard.edu |
| Subject | Re: st: imputation of missing dates |
| Date | Tue, 10 May 2011 18:15:09 +0200 |
And here is another one: STATISTICS IN MEDICINE Statist. Med. 2004; 23:3803–3820 Choice of time-scale in Cox’s model analysis of epidemiologic cohort data: a simulation study Anne C. M. Thiebaut and Jacques Benichou Greetings Roland 2011/5/10 roland andersson <rolandersson@gmail.com>: > Yulia > > I wonder if you made any specific intervention at the examination and > want to see the effect of this intervention on survival. Otherwise you > may consider using attained age as analysis time for the survival > analysis, as there is no exact point in time when the persons came > under risk or change in risk. In that case you do not need an exact > date for the start of the analysis. > > See some of the references below (I can give you more exact refernce > if you contact me directly. > > Stocks 2010 on prostate cancer > Qiao 2009 on metabolic syndrome and stroke > Luo 2008 on pancreas cancer > Lamarca 1998 attained age survival analysis > Cheung 2003 survival analysis attained age > > Roland Andersson > > > 2011/5/9 Yulia Blomstedt <sungurik@gmail.com>: >> Dear Marteen and Carlo, thank you for the valuable advise! >> >> Dear Roland, thank you for your input. Unfortunately, we do not know >> the exact date of the invitation (only the year). One could have >> calculated it from the date of birth, but we know only the year of >> birth. >> We need EXAMDATE for non-participants in order to conduct a survival >> analysis (we know the exact date of death). >> My first choice was to calculate the survival from the YEAR when >> indviduals were invited to participate in the examination. Since, this >> would result in an inprecise person-time, I tried to explore the >> possiblity of creating an exact EXAMDATE. After all the received >> advice, this, however, does not seem as a very good alternative. >> >> Sincerely, >> Yulia >> * >> * For searches and help try: >> * http://www.stata.com/help.cgi?search >> * http://www.stata.com/support/statalist/faq >> * http://www.ats.ucla.edu/stat/stata/ >> > * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/