NLCA senior clinical lead Paul Beckett examines why case ascertainment has been such a key feature of our reports in the past, but is now no longer so crucial.
From the early days of the National Lung Cancer Audit (NLCA), ensuring that the data are as complete as possible has always been crucial to its success. As well as completeness of the data for individual patients, this also means making sure that all appropriate patients are included in the audit.
I remember attending a meeting on the planned audit as a new consultant in 2003, and asking Mick Peake how we could ensure that trusts did not cherry-pick the cases that they submitted. The answer was to publish the ‘case ascertainment rate’, a comparison of the number of cases submitted each year compared with what would be expected based on historic data from the cancer registries.
How could we ensure that trusts did not cherry-pick the cases they submitted?
We featured the case ascertainment rate prominently as a key performance metric in our early reports, and it was a useful way of encouraging trusts to make sure that they were capturing and submitting data on all their cases of lung cancer. Indeed, we saw the number of cases captured by the audit rise from around 40% of expected cases in 2005 to very close to 100% by the end of the decade.
As time went by, trust mergers and changes in referral pathways meant that it was sometimes hard to keep the expected number of cases accurate, but we tried to update it regularly.
The case ascertainment rate is now de facto 100%
However, in 2017, the need for a case ascertainment rate seems to have passed. No longer can trusts ‘choose’ which cases to submit. The NLCA now uses processed cancer registration data linked to a number of other data sources, and the inclusion of cases that were not previously known to multidisciplinary teams (MDTs) has led to several thousand extra patients being included per year. As a result, the case ascertainment rate is now de facto 100% (at least for English trusts). We have therefore decided to drop this rate as a key performance indicator, although we will continue to report the actual number of annual cases both by trust and nationally.
Paul Beckett, NLCA senior clinical lead