Introduction

The Ovarian Cancer Audit Feasibility Pilot is a collaboration between the gynaecological oncology clinical community, the charity sector and Public Health England, with the aim of performing meaningful analyses of routinely collected data for the purpose of improving treatment and outcomes for women diagnosed with ovarian cancer in England. The Ovarian Cancer Audit Feasibility Pilot is jointly funded by the British Gynaecological Cancer Society, Target Ovarian Cancer and Ovarian Cancer Action, and is being delivered by analysts at the National Cancer Registration and Analysis Service (NCRAS), which is part of Public Health England. The pilot will run for two years from 2019 and publish a range of outputs on ovarian cancer, including a final report on the audit and its findings, bringing all of the analyses into one place. Outputs can be found on the project page on this website.

Building on the disease profile report and this report looking at geographic variation, future analyses will include a closer examination of surgery and an exploration of the factors associated with short-term mortality. This will bring together an impressive new body of evidence in relation to ovarian cancer treatment and outcomes in England and provide the baseline of data and structures needed for a continuing ovarian cancer audit.

The first publication from the Ovarian Cancer Audit Feasibility Pilot was the Disease Profile in England report¹, which describes incidence, mortality, stage and survival for ovary, fallopian tube and primary peritoneal carcinomas (‘ovarian cancer’) diagnosed in England.

Amongst its many findings, the report showed marked geographic variation in cancer survival across England at a Cancer Alliance level. Excluding borderline diagnoses, five-year net survival ranged between 29% and 50% across the 19 Cancer Alliances in England for the period 2013 to 2017.

One possible reason for such disparity was variation in the local clinical management of disease. To explore this hypothesis, this geographic variation report describes differences in treatment between Cancer Alliances in England, and the extent to which these might be explained by variation in tumour and patient characteristics.

Beginning with a description of how treatment differed according to stage at diagnosis, patient age, tumour morphology and comorbidities, the report moves on to look at regional variation in treatment at Cancer Alliance level. In addition to unadjusted (crude) results, models are presented that control for confounding variables (including age, stage and morphology) that may differ between regions, helping isolate any variation that may be attributable to clinical decision making.

Results from these analyses indicate that the probability of accessing surgery and chemotherapy varies between regions within England, even after accounting for differences in patient and tumour characteristics. These findings indicate an opportunity for identifying examples of best practice that could be disseminated to Cancer Alliances where treatment probabilities are lower, leading to improvements in therapy and better outcomes for women with ovarian cancer. Methods underlying these results are described at the end of the report.