One of the diagnostic challenges facing GPs is to separate the majority of patients who consult them, and have relatively minor or self-limiting disease, from the minority who turn out to have serious disease, such as cancer. A large study involving over 400 GPs and 16,000 patient contacts conducted in Denmark has shown that a general practitioner's suspicion of the presence of serious disease at the end of the consultation is more strongly predictive of serious disease than many of the symptoms and signs conventionally used in schemes such as the two-week rule for urgent suspected cancer referrals.

Professor Peter Vedsted's research group demonstrated a positive predictive value for diagnosis of cancer or serious disease of almost 10% in the two months following consultation. Suspicions arose most frequently in patients with symptoms from blood, blood-forming organs, digestive organs and female genital organs, and were lowest in those with psychological problems or symptoms from the musculoskeletal, endocrine or cardiovascular systems.

The authors argue that patients in whom GPs have a suspicion of serious disease should have rapid, direct access to investigations such as imaging and endoscopy. They suggest that the two week wait referral systems in the UK and Denmark (which countries, interestingly, have the worst cancer survival figures in Europe) may actually prevent early diagnosis because they are based on symptom check lists, and do not include GPs' "gut feelings" about serious problems.

Research: Predictive values of GPs' suspicion of serious disease: a population-based follow-up study, Peter Hjertholm, Grete Moth, Mads Lind Ingeman and Peter Vedsted, British Journal of General Practice, DOI: 10.3399/bjgp14X680125, published 27 May 2014.