Intended for healthcare professionals

Editor's Choice US editor's choice

US Highlights

BMJ 2006; 333 doi: (Published 30 November 2006) Cite this as: BMJ 2006;333:0-g
  1. Douglas Kamerow, US editor (dkamerow{at}

Everyone uses Google, including doctors and patients. Should we rely on it to help make diagnoses? Hangwi Tang and Jennifer Ng evaluated (doi: 10.1136/bmj.39003.640567.AE) Google's diagnostic skills with 26 recent case studies from the New England Journal of Medicine. They found that in 15 of the cases they were able to find the ultimate diagnosis by Googling major search terms from the case record. While this doesn't prove anything, and it certainly depends on the knowledge and skills of the Googlers, it does confirm that there is a lot of relevant information out there on the web. In a related editorial (doi: 10.1136/bmj.39044.369745.BE), Martin Gardner points out that all that Google is doing is searching—very quickly and widely—for medical terms and returning results that have a high number of matches, regardless of whether the term is a diagnosis, treatment, anatomic part, or anything else. Only with a future web that includes context and meaning—called the semantic web—will we have something approaching an expert system.

Should doctors be prohibited from having sexual relations with their patients? Current guidelines in the US and UK prohibit such relationships with current patients, and the UK's General Medical Council has just issued a warning—but not a prohibition—against sex with previous patients as well. Julian Sheather, in an editorial (doi: 10.1136/bmj.39042.529641.BE), feels that on balance the GMC is right. While there are certainly times when such a prohibition seems silly—when a doctor has treated a colleague in the emergency room, for example—or unnecessarily harsh—in rural areas or small towns—in general doctors should think long and hard before commencing a relationship with a former patient.

One big item on the agenda of the new Democratic Congress is decreasing the number of Americans without health insurance. Princeton economist Uwe Reinhardt asks (doi: 10.1136/bmj.39042.375544.BE) if Democratic leadership is likely to result in meaningful change. He thinks not, because of low interest in raising taxes as well as a lack of consensus about how to go about it.

Finally, should we be giving statins to everyone with an increased heart disease risk? The Heart Protection Study Collaborative Group did a cost effectiveness modeling study (doi: 10.1136/bmj.38993.731725.BE) of the lifetime use of 40 mg of simvastatin across a range of age and risk groups. They found that the medicine was cost effective for those with at least a 1% annual risk of a major vascular event—a much lower threshold than current prophylactic recommendations.

View Abstract