## Creating a new date when two diseases are first present

Solved
Occasional Contributor
Posts: 5

# Creating a new date when two diseases are first present

If I had a dataset that contained a Diabetes_Date and a Hypertension_date like the following:

ID      Diabetes_date      Hypertension_date

111     4/5/05                         7/10/06

222     3/4/14                         9/6/10

I need a new variable date that would list the date at which they had both Diabetes and Hypertension.  So for the above example it would be 7/10/06 for 111 and the 222 ID would be 3/4/14.  I know I could do a few if statements to accomplish this, but wondering if there are other easy methods.

Thanks so much!

Accepted Solutions
Solution
‎04-15-2014 03:26 PM
Super Contributor
Posts: 578

## Re: Creating a new date when two diseases are first present

Wouldn't it just be max(diabetes_date,hypertension_date)?  I think the max function propagates null values.

All Replies
Solution
‎04-15-2014 03:26 PM
Super Contributor
Posts: 578

## Re: Creating a new date when two diseases are first present

Wouldn't it just be max(diabetes_date,hypertension_date)?  I think the max function propagates null values.

Super Contributor
Posts: 644

## Re: Creating a new date when two diseases are first present

No, I do not think max() propagates nulls unless all the parameters are null.  In fact I use max(value, 0) to set missing values to zero.  You need the N() function as well to count the number of non null values.

If n(Diabetes_date, hypertension_date) = 2

then combined Date = max(diabetes_date,hypertension_date) ;

Richard

🔒 This topic is solved and locked.