I have the following data:
Account Number Age Age Group Medical/Surgical Patient Type 8AM pass
through Before1/After0 8AM PET Before0/After1 8AM ET Date 1 Date 2 Date
3
123456789 0 <18 MED I 3 1 1 10/21/2011 10/22/2011 10/23/2011
234567890 0 <18 MED I 3 1 1 10/09/2011 10/10/2011 10/11/2011
GIVEN the number of dates in each row (date 1 -3) in this case, and
there could be up to 28 date columns---,
I need exactly that many total rows (3, one for each date), with the
account number, age, age group, med/surg, patient type, 8AM pass
through, Before1/After0 8AM PET, Before1/After0 8AM ET columnar data's
info repeated, and the date being reduced to one column, like so--
Account Number Age Age Group Medical/Surgical Patient Type 8AM pass
through Before1/After0 8AM PET Before0/After1 8AM ET Date 1
123456789 0 <18 MED I 3 1 1 10/21/2011
123456789 0 <18 MED I 3 1 1 10/22/2011
123456789 0 <18 MED I 3 1 1 10/23/2011
234567890 0 <18 MED I 3 1 1 10/09/2011
234567890 0 <18 MED I 3 1 1 10/10/2011
234567890 0 <18 MED I 3 1 1 10/11/2011
Exactly how do I do this? I cannot figure this out.
Any expertise would be greatly appreciated.
I have to run a census report based off certain statistics, and those repeat/unchanging columns are the certain stats. The specific report I am trying to run is a census report, and that is why each unique date/account number combo needs its own row. The database I pull this info from only contains the beginning and ending time stamps as to when the patient is in a given location of the hospital. When you wish to do a 11AM census for example, there would be two ways to do it that I could think of,
1. a snapshot in time, calculating all patients that cross or enter a location on a certain very second, say exactly 11AM, or 3PM.
2. or say a true hourly census, for example, 11AM meaning, all patients that enter or hit a location on or across any time point within 11:00:00 AM through 11:59:99.9999/100ths of a second.
As you can imagine in the first method example, what if the patient entered a location at 11:04 am and left at 11:58 am? this patient would not get counted in the 11AM "hourly census" nor the noon census.
In any event, the data base only gives me time stamp end points, beginning time patient entered a location, and ending time they left and maybe were transferred elsewhere within the hospital, but not in that location anymore. I figured out how to parse the multiple dates for the given hourly time stamps that it crosses each day, since it will be a correct hourly/daily census count now.
IT is the department that is using method #1 right now, and I can see no value to undercounting patients, especially since a census is all about counting correctly. So far, I have figured out the correct answer, I just need to parse the date column into a single column rather than multiple and copy the rest of the stat info down dependent upon how many many unique dates I have given the unique patient and visit combo. In other words, if a patient stayed in a given location from 4/2/2012 at 6:45 am to 4/11/2012 at 5:45am, if I were doing an 8AM census, then that patient crossed that time stamp 10 times, 4/2-4/10/2012.
Eric Lutz
[Non-text portions of this message have been removed]
Tidak ada komentar:
Posting Komentar