Hi All,
I am planning to conduct a time stratified case crossover analysis and struggling to arrange my data set for that purpose. In this design, an index hospital admission case will have several controls- all in the same weekday, month and year. As an example, if there is an admission to a hospital today Friday (25/0708), that will be my index case. My controls will be all preceding Fridays of July 2008. Thus I will have three controls for the index event. If on the other hand a patient is admitted on Wednesday 15/07/08 all other Wednesdays of July (1st, 8th, 22nd and 29th) will be serving as controls. In the second scenario, two controls are before the index case and two are after. I would appreciate any ideas and suggestions on how to realize this plan. I have copied a sample data set below.
Many thanks in advance!
date_admis date_control control_seq
01/07/2008 08/07/2008 1
01/07/2008 15/07/2008 2
01/07/2008 22/07/2008 3
01/07/2008 29/07/2008 4
02/07/2008 09/07/2008 1
02/07/2008 16/07/2008 2
02/07/2008 23/07/2008 3
02/07/2008 30/07/2008 4
04/07/2008 11/07/2008 1
04/07/2008 18/07/2008 2
04/07/2008 25/07/2008 3
08/07/2008 01/07/2008 1
08/07/2008 15/07/2008 2
08/07/2008 22/07/2008 3
08/07/2008 29/07/2008 4
Thanks
Kumsaa
_________________________________________________________________
Connect to the next generation of MSN Messenger�
http://imagine-msn.com/messenger/launch80/default.aspx?locale=en-us&source=wlmailtagline
*
* 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/
