The Get Data Out programme routinely publishes cancer statistics produced by NHS E (previously PHE) in a consistent table, called the Get Data Out (GDO) table. This table collects patients into groups with common characteristics, and then publishes information such as incidence, treatment rates, survival and Routes to Diagnosis for these groups.
This document sets out the definitions of the cohort and groups for Get Data Out tables for the 2023 release of data on pancreatic cancer between 2013 and 2020.
The cohort of pancreatic tumours used for Get Data Out is all tumours coded in ICD-10 to C25.
Tumours were classified as Carcinoma, Neuroendocrine and Other by ICD-O-3 morphology code as documented in the Appendix. The Other group was very small (usually fewer than 10 tumours per year, less than 0.1% of all tumours) and hence was combined with the Carcinoma group.
Classifications were created with Melissa Matz at LSHTM and Dr Brian Rous at NCRAS.
The large group of malignant tumours, “Carcinoma and Other”, are then split by stage, into:
Please note that the staging system for these pancreatic cancers changed from TNM 7 to TNM 8 between 2017 and 2018 diagnoses. We believe that this changing definition reduces the number of stage 2 tumours and increases the number of stage 3 tumours, and so care must be taken if analysing the GDO data as a time series - the decrease in stage 1-2 tumours is being driven at least in part by the coding artefact of the changing stage system.
The registration of 2019 tumours were being completed during the COVID-19 pandemic. This led to reduced access to the usual data sources, and despite the registry’s best efforts a noticeable decrease in data quality in some fields. This is most commonly seen in an increase in ‘stage unknown’ tumours, and a corresponding decrease in other stage groups. This should be noted when undertaking time-series analysis on the data.
The vast majority of cases are staged in either the UICC 7, UICC 8 or ENETS 2007 system, where the recommended staging system changed from UICC 7 to UICC 8 between 2017 and 2018 diagnoses. A very small percentage of cancers (less than 1%, apart from 2013, where 1.085% of Pancreas cancers were staged using UICC 6), were staged in other systems, as reported in the known limitations page. The following table shows the percentage of cases staged in each of UICC7, UICC8 and ENETS 2007 over time.
The ideal is to split into:
This is possible for the larger groups. Where a full partition is not possible, groups are combined to make the group size large enough (e.g. ending with 70+, not 80+, or <60 instead of <50)
Where numbers allow (i.e. in patients aged over 60 in carcinomas of pancreas) the cohort is partitioned into 4 regions, North of England, South of England, Midlands and East of England, and London. Patients are assigned to a cohort by their postcode at diagnosis and the National Statistics Postcode Lookup. Regions are defined based on the NHS regions in 2017, which were current when this partition was devised, and have been kept steady to provide a timeseries.
Where numbers allow the data is then split into male and female.