Quality control of automatically defined cancer cases by the automated registration system of the Venetian Tumour Registry

Abstract

The overall proportion of incorrect decisions is not high and similar to those reported by other registries, but errors are correlated to the diagnostic evidence pattern. As a further check, we decided to revise clinical cases for the three sites mentioned manually, in order to reduce the numbers proportion of both prevalent cases, and all cytology-based diagnoses, so as to reduce the number of 'false positives'. Coverage of hospital discharge source has been extended in order to decrease the proportion of cases based only on pathology records.

Consistency and accuracy of diagnostic cancer codes generated by automated registration: comparison with manual registration

“…Although the need for quality control of cancer registration data is recognized [ 14 ] few data exist on quality of automated cancer registration procedures. Assessments of accuracy using mainly automated methods are available for Arhus County, Denmark [ 7 ]; Veneto [ 8 ] and Ontario, Canada [ 7 ]. The approach in Arhus County was to compare the automatic data with those gathered manually by the Danish national cancer registry.…”

“…In a recently published comparison between manual and automatically generated data for all cancer sites in the Venetian Cancer registry (Italy) [ 8 ] concordance was 93.3%. This registry was the first in Europe to use an automatic procedure for cancer incidence generation, and it is reassuring that our concordance level was similar to that obtained by this pioneering approach.…”

Automated Selection of Relevant Information for Notification of Incident Cancer Cases within a Multisource Cancer Registry

“…In Italy, the Varese Cancer Registry estimated the proportion of wrongly-coded topographies from their data sources to be 1.6% on a full year registration [8,11]. Another evaluation performed on 1539 files in the Venetian Tumour Registry showed 85.7% concordance on IARC topography and IARC morphology and only 39 false positives (2.5%) [12]. These evaluations concerned cases that had several concordant sources leading to describe only one single tumour for a given patient and were then eligible to be registered automatically in these registries.…”

“…This finding underlines the importance of using diverse sources for the notification of new incident cases. Deterministic [8,11,12] or probabilistic algorithms using artificial learning techniques [13] have been developed and evaluated. These works were mainly based on interrelating and aggregating data from different sources, but without any selection of relevant information according to its source.…”

Indicators of guideline-concordant care in lung cancer defined with a modified Delphi method and piloted in a cohort of over 5,800 cases

“…Also, indicators have to be timely, reliable and economically computable on a large scale to inform quality improvement activities [10,11]. In the oncologic setting, this is impractical if indicators are retrospectively computed using data abstracted ad hoc from clinical records, while it is expected to be feasible using current health database linked to modern population cancer registers [12][13][14].…”