Re: st: Need help programming for dates in Stata

[Nick Cox <njcoxstata@gmail.com>]

I can't see what code you tried.

I'm assuming your date variables are Stata daily date variables. If
not, let that be so.

If I understand you correctly, you need to clone to fill in missings.

clonevar startdate2 = startdate
replace startdate2 = mdy(12, 31, 2008) if missing(startdate)

clonevar stopdate2 = stopdate
replace stopdate2 = mdy(12, 31, 2012) if missing(stopdate)


Then on a quick and perhaps hasty analysis

gen used_2009 = startdate2 <= mdy(12,31,2009)

gen used_2010 = (startdate2 <= mdy(12,31,2010)) & (stopdate >= mdy(1,1,2010))

and the definitions for 2011 and 2012 resemble those for 2010.

My first instinct was "It's a loop" but with just four years making it
general seems unimportant.

Code not tested. NB: one criterion; two criteria.

Nick

On Tue, Mar 5, 2013 at 6:43 PM, Michael Stewart
<michaelstewartresearch@gmail.com> wrote:

> I have a list of start and stop dates for medications
> I am trying to find if there is way to classify  them to specifically
> say if they are used in particular years
>
> Med  Start date    Stopdate
> xxx    11/12/2009    11/12/2012
> yyy      .               1/2/2010
> zzz    1/1/2009      .
>
>
> I want to create four  variables Used_2009   used_2010    used_2011
> used_2012
>
> The criteria is
> 1)If medication is  used even for a day in a year, it is considered to
> be used for that year
> 2)If the start date is blank, then it is deemed to be used from a much
> prior day
> 3)If ending date is blank/missing then  that medication is currently active
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/