For the CRDS team, as we started looking at the quality of the data as a part 
of our report to the CDM, after some tweaking for local oddities (no hispanic 
is NOT a race, and boy we handle smoking oddly here in WI), our data was coming 
up pretty clean. But one of the thing that tripped us up was the number of 
patients that had a race the equated to unknown (almost 50% instead of the goal 
less than 15%). The hypothesis is that the older imported records did not 
correctly have race correctly code seems to be born out by Debbie's analysis.

-TM

From: Yoshihara Deborah L
Sent: Tuesday, April 26, 2016 7:59 AM
To: Mish Thomas F; Wanta Keith M
Subject: race of active patients

So if I look at the active patients instead of all patients, the race ratios 
are more in line.
(2 visits within the last 3 years, separated by 30 days)
Keith - perhaps note in the ADD that we have historical data that probably 
didn't collect race.

count

race

percentage

414131

White

86.2%

24971

Black

5.2%

16000

Asian

3.3%

6983

Decline

1.5%

873

Hawaiian

0.2%

4828

A. Indian

1.0%

4513

Null

0.9%

7972

Unavailable

1.7%


Total  480271
_______________________________________________
Gpc-dev mailing list
[email protected]
http://listserv.kumc.edu/mailman/listinfo/gpc-dev

Reply via email to