Understanding improvements in cancer diagnosis in primary care: Time trends analysis (2006-2015) of diagnostic routes and emergency presentation sub-type in England

Talk Code: 
6C.1
Presenter: 
Lucy Elliss-Brookes
Co-authors: 
Annie Herbert, Gary Abel, Sam Winters, Sean McPhail, Georgios Lyratzopoulos
Author institutions: 
University College London, University of Exeter Medical School (Primary Care), Public Health England (National Cancer Registration and Analysis Service)

Problem

Primary care has a crucial, though often misrepresented, role in the diagnosis of cancer. Many patients continue to be diagnosed as emergencies, which have a complex aetiology and are associated with worse survival. Examining time trends in diagnostic routes can help to appreciate the evolving role of general practice in diagnostic pathways for cancer patients.

Approach

We examined time trends in diagnostic routes overall, and in emergency presentations specifically. We paid particular attention to a specific sub-type of emergency presentation, i.e. those generated after emergency GP referrals (which we term GP-EP). We used the Routes to Diagnosis dataset, including data on patients diagnosed with cancer in England during 2006-2015. We described trends in the numbers of cancer cases diagnosed through any route (i.e. including non-emergency presentation). Further, we described these trends after adjusting for patient characteristics and cancer site case-mix, using multivariable logistic regression.

Findings

The number of all cancer cases in England increased during 2006-2015 increased by 20%, while the number of emergency presentations decreased by 4%, with progressive reductions in related proportions. The percentage of patients diagnosed via two-week wait referral increased from 25% to 38% (19% to 30% adjusted). Reductions in the proportion of cancer patients diagnosed as emergencies was accompanied by a changing composition of emergency presentation sub-types. Among the 554,621 emergency presenters, the GP-EP share declined continually during the study period (adjusted percentages down from 31% in 2006 to 17% in 2015). Across the study years, emergency presenters were more likely to be GP-referred if they were diagnosed with pancreatic, gallbladder, ovarian, or acute leukaemia, and if they lived in areas of lower level of deprivation (least vs. most deprived quintile: 25% vs. 19%).

Consequences

We describe continued reduction in the proportion of patients diagnosed with cancer as an emergency, and particularly emergency presentations generated by an emergency GP referral. Reductions in overall emergency presentations seem to chiefly reflect downward trends in GP-EP referrals, which in turn likely reflect continually increasing use of the two-week-wait referral pathways during the last decade, through the implementation of NICE referral guidelines.

Submitted by: 
Lucy Elliss-Brookes
Funding acknowledgement: 
AH and GL are supported by a Cancer Research UK Advanced Clinician Scientist Fellowship to GL (award C18081/A18180). GL and GAA are associate directors (co-investigators) of the multi-institutional CanTest Research Collaborative funded by a Cancer Research UK Population Research Catalyst award (C8640/A23385). This project involves data derived from patient-level information collected by the NHS, as part of the care and support of patients with cancer. The data are collated, maintained and quality assured by the National Cancer Registration and Analysis Service, which is part of Public Health England (PHE).