Editor's Choice US editor's choice

Alarming results

BMJ 2007; 334 doi: http://dx.doi.org/10.1136/bmj.39218.433391.3A (Published 17 May 2007) Cite this as: BMJ 2007;334:0-a
  1. Douglas Kamerow, US editor
  1. dkamerow{at}bmj.com

    In the UK they are labelled “alarm symptoms,” and in the US we call them “red flags”: signs or symptoms in clinical care that are thought to predict serious disease and often lead to specialist referral. But what do we really know about the seriousness of dysphagia, hematuria, hemoptysis, or rectal bleeding? Not much, until now. Roger Jones and colleagues examined these symptoms and subsequent diagnoses in over 750,000 primary care patients, using the UK General Practice Research Database (doi: 10.1136/bmj.39171.637106.AE). They found that the new onset of alarm symptoms is associated with an increased likelihood of cancer, especially in men and the elderly. For example, 7.5% of men with new hemoptysis had a subsequent diagnosis of a respiratory tract cancer, many times the “background rate.”

    In a related editorial, Robin Fox and John Fletcher point out that new diagnoses of cancer are rare in primary care, on average less than one a year per doctor for each of the cancers in this study (doi: 10.1136/bmj.39212.467037.BE). Even though likelihood ratios in the hundreds sound impressive, the vast majority of patients with these symptoms won't, on further investigation, have cancer. And studies based on routine diagnostic coding can reveal more about gaming the system to get an urgent referral than about diagnostic accuracy. Still, Fox and Fletcher conclude that this study shows that alarm symptoms should be taken seriously.

    Also alarming is the amount of preventable morbidity and mortality attributed to venous thromboembolism in hospitalized patients. David Fitzmaurice and Ellen Murray briefly review the numbers: perhaps 60 000 deaths a year in the UK from venous thromboembolism; excess thromboembolism rates of up to 60% because of the lack of pharmacological thromboprophylaxis; and more than 25 000 potentially preventable deaths a year in the UK (doi: 10.1136/bmj.39210.496505.BE).

    These data perfectly introduce an overview of a recent surgical thromboprophylaxis guideline prepared by the UK National Institute for Health and Clinical Excellence (NICE) (doi: 10.1136/bmj.39174.678032.AD). Although there is some residual debate about whether mechanical prophylaxis (elastic support stockings) is as good as pharmacoprophylaxis with low molecular weight heparin, there should be no question that prophylaxis of some sort is needed for every surgical patient.

    View Abstract