Collapsing Consecutive Dates

[jdukes42]

Hi All,

I have a new question regarding collapsing consecutive dates.

For my study I am looking at antibiotic prescriptions. I have already determined the patients who would have been given a single primary antibiotic. However, now I have to address those individuals who would fall into the mixed therapy category. For these individuals I want to collapse the start and stop dates for antibiotics given on consecutive dates. In addition to collapsing to a single date, I need to retain the antibiotic given on those dates (see example below).

ID	RX_1	RX_2	Start_1	Start_2	Stop_1	Stop_2	New_Start	New_Stop	New_RX_1	New_RX_2
1	1	2	5/2/2001	5/3/2001	5/2/2001	5/3/2001	5/2/2001	5/3/2001	1	2
2	1	3	5/2/2001	5/2/2001	5/2/2001	5/2/2001	5/2/2001	5/2/2001	1	3

However, not all individuals will have multiple treaments that can be collapsed into one set of consecutive dates. In those cases I need a data set that will look something like this

ID

RX_1

RX_2

Start_1

Start_2

Stop_1

Stop_2

New_Start1

New_Stop1

New_Start2

New_Stop2

New_RX1_1

New_RX1_2

New_RX2_1

New_RX2_2

3

1

2

5/2/2001

5/10/2001

5/2/2001

5/19/2001

5/2/2001

5/2/2001

5/10/2001

5/19/2001

1

2

The overall data set has up to 8 different prescription (RX), start, and stop dates that have to be collapsed.

Thanks,

Jonathan

[art297]

Jonathan,

All of the dates in your example are from 2001.  What is the actual range of possible dates?

Art

.

[jdukes42]

Art-

The dates range from 2005 to 2010. Thanks again for your interest and help.

Jonathan

[art297]

Jonathan,

Take a look at this thread from SAS-L last year.  Two approaches are discussed/provided for capturing date ranges:

http://www.listserv.uga.edu/cgi-bin/wa?A2=ind1010E&L=sas-l&D=1&H=0&O=D&T=1&m=307534&P=4731

I would think that either approach could be modified to meet your needs.

Art

[jdukes42]

Art-

Thanks. I will take a look at this.

Jonathan

[Ksharp]

Not quit sure.

What does your dataset look like and what output do you want?

Ksharp