Letters Cancer referrals

Cancer cases diagnosed as emergencies may be fewer than reported

BMJ 2013; 347 doi: https://doi.org/10.1136/bmj.f4773 (Published 30 July 2013) Cite this as: BMJ 2013;347:f4773
  1. Laurence Leaver, general practitioner1,
  2. Barnabas J Gilbert, medical student2,
  3. Jenny Jones, former project manager3,
  4. Martin Allaby, consultant in public health4
  1. 1Jericho Health Centre, Oxford OX2 6NW, UK
  2. 2Green Templeton College, Oxford OX2 6HG, UK
  3. 3Thames Valley Cancer Network, Oxford OX3 7HE, UK
  4. 4Solutions for Public Health, Oxford Business Park South, Oxford OX4 2GX, UK
  1. laurenceleaver{at}nhs.net

The news item highlighting survival after elective surgery for colorectal cancer reiterates dismay at the “stubbornly high proportion of patients who present as an emergency” (21% of cancers diagnosed between 1 April 2011 and 31 March 2012).1 2 Using data on cancers diagnosed during 2006-08,3 the current “Be Clear on Lung Cancer” campaign emphasises that “39% of lung cancer patients present as emergencies, compared to 24% across all cancers.”4 We should treat these hospital based figures with caution because it may be difficult for a hospital to be certain that this is the first presentation or first referral for each cancer, and some of the data are from as early as 2006.

Thames Valley Cancer Network facilitated a GP audit of all 809 cancers diagnosed from 1 August 2010 to 31 July 2011 across 33 practices.5 Cases were identified from GP databases and individually reviewed by a GP, including a detailed audit of the routes to diagnosis. Only 13% (95% confidence interval 11% to15%) of all cancers, 18% (11% to 25%) of colorectal cancers, and 29% (18% to 39%) of lung cancers were diagnosed as emergencies compared with 24%, 26%, and 39% respectively in the national data for 2006-08.

GP review of all 809 patients found that 19 out of 125 who were referred routinely had had reasons to be referred urgently. This shows that there is some scope for earlier diagnosis, but in most cases there was no reason to have managed patients differently.

These data suggest that, where improvement is possible, some of the “headline” figures have already improved, at least in the Thames Valley region. Future gains in early diagnosis may well be highly dependent on screening programmes or campaigns to encourage patients to present earlier.


Cite this as: BMJ 2013;347:f4773



View Abstract

Log in

Log in through your institution


* For online subscription