[Skip to Content]
[Skip to Content Landing]
December 21, 1994

Race and Access to Erythropoietin by Patients on Hemodialysis-Reply

Author Affiliations

University of Michigan Ann Arbor
University of Maryland Baltimore

JAMA. 1994;272(23):1819. doi:10.1001/jama.1994.03520230028021

In Reply.  —We agree with Dr Sadler on the importance of measuring the quality of Medicare data when using this database for scientific analyses. Our estimates published in this article are similar to our findings from a large data validation study.1 However, objectivity requires us to be constantly critical, especially for untested data items such as rHuEPO.Dr Sadler takes our minor discussion point out of context and thereby generalizes that we have "challenged the validity of the PMMIS." This is not our position. We basically agree with his four technical points (listed herein); however, our analyses show that our data quality estimates remain virtually unchanged.First, the point that we "used incomplete PMMIS files": PMMIS files are always incomplete. We used a file that was over 15 months old since US Renal Data System analyses suggest that by 15 months these files can be considered complete for most

First Page Preview View Large
First page PDF preview
First page PDF preview