Editor's Choice

Not always obvious

BMJ 2011; 342 doi: https://doi.org/10.1136/bmj.d3286 (Published 25 May 2011) Cite this as: BMJ 2011;342:d3286
  1. Jane Smith, deputy editor
  1. jsmith{at}bmj.com

The BMJ Publishing Group held its 3rd annual award ceremony last week. As Nigel Hawkes describes (doi:10.1136/bmj.d3183), it was a splendid affair, celebrating work that was variously innovative, caring, intelligent, and brave. A colleague commented afterwards that it was too easy to get cynical about healthcare and the many agendas that people pursue within it: the awards evening, he thought, was a complete contrast—a genuine celebration of good work. I thought so too.

Richard Peto, Oxford epidemiologist, got the lifetime achievement award for, as he said in accepting it, studying the “bleeding obvious”—the big things: smoking, alcohol, hypertension, cholesterol (hear him talk more about this on last week’s podcast, http://podcasts.bmj.com/bmj). In a way all the finalists for the lifetime achievement award had reached the shortlist because of dealing with big picture stuff (though not always with obvious answers): George Alleyne for his work with HIV and non-communicable diseases in the developing world, and Jack Wennberg for forcing the healthcare community to understand the extent to which supply determines demand (BMJ 2011;342:d1062).

Yet most of the time our authors are struggling with the less than obvious—as the research pages this week illustrate. A study by Matejka Rebolj and colleagues from Denmark sought to find out if it was possible to improve the specificity of the hybrid capture 2 test for human papillomavirus DNA, used in cervical screening, without reducing its sensitivity (doi:10.1136/bmj.d2757). They concluded that raising the cut-off level of the test would substantially improve the test’s specificity while maintaining its sensitivity at over 90%. But editorialists Peter Saseni and Alejandra Castanon warn that ignoring results between the old and the new cut-off values is questionable: they suggest that less intensive management (such as more frequent screening) may be more appropriate than simply concluding that this is a true negative result (doi:10.1136/bmj.d2941).

And even epidemiologists dealing with common conditions can come unstuck because things aren’t obvious. Our obituary this week, of David Sencer, a former head of the Centers for Disease Control in Atlanta, describes how he and his colleagues dealt with an outbreak of a swine flu virus in 1976 at Fort Dix army base in New Jersey (doi:10.1136/bmj.d3276). The virus resembled the strain that had caused the 1918-19 flu pandemic. Faced with the options of doing nothing; developing a vaccine and stockpiling it so it could be distributed swiftly if necessary; or developing a vaccine and immunising people as fast as they could, Dr Sencer and his colleagues opted for the last approach. No epidemic occurred, but some of those receiving the vaccine developed Guillain-Barré syndrome and over 20 died— all in the middle of a presidential election campaign. At the same time an outbreak of a mystery infection occurred among army veterans and 29 people died; the press complained that it was taking too long to find a cause. CDC staff traced the source to an American Legion convention in Philadelphia and to a new bacterium—Legionella, and the new president did the obvious thing for a politician— he sacked Dr Sencer.


Cite this as: BMJ 2011;342:d3286


View Abstract