When our network of hospitals had the 5 registrars reporting to the
same manager, we performed peer review quarterly on abstracts
selected randomly that would total at the end of the year to be 10%
of the annual analytic caseload. We reviewed the demographic info,
diagnosis info (text and codes), treatment info, follow-up and text
fields that substantiated the codes. The registrars were required
to maintain a 90% or better accuracy rate.

Now that we're separate facilities with different managers, we only
perform peer review for our non-CTR's in the group of hospitals.
Those registrars have every case reviewed versus the 10% that we
reviewed for our registries.

We still are expected to have a 90% or better accuracy rate for the
abstracts reviewed by our physicians. This rate seemed to be one
that was acceptable by our Quality Management department as well as
accuracy rates set by the ACOS for various standards.

Hope this helps.
Charla Carter, CTR
Cincinnati, OH