Expect the unexpected

Medicine is beset with traps for the unwary: incidental findings, unintended consequences, perverse incentives, symptoms that are easily missed. This week’s journal has its fair share of examples.

Zoe Morris and colleagues ask how often MRI scans of the brain find things they weren’t looking for (doi:10.1136/bmj.b3016). Their meta-analysis concludes that one in every 37 scans over the past 20 years had an incidental finding. In his linked editorial, Aad van der Lugt warns that rates are probably even higher with modern scanners and imaging protocols (doi:10.1136/bmj.b3107). Patients and research participants should be warned beforehand and carefully counselled afterwards.

At the other end of the spectrum are patients whose crucial symptom goes unrecognised. Subarachnoid haemorrhage is missed in 20-50% of patients at first presentation (doi:10.1136/bmj.b2874), unsurprising perhaps when a full time GP with a list of 2000 patients is likely to see only one case every seven years. One in 10 patients who present to general practice with sudden severe headache turns out to have subarachnoid haemorrhage, and early diagnosis and referral can improve the outcome

What of “payment by results?” Has it delivered greater productivity from England’s hospitals without damaging quality of care? Shelley Farrar and colleagues find that it has (doi:10.1136/bmj.b3047), but they can’t tell us about the consequences for primary care. In his linked editorial (doi:10.1136/bmj.b3081) Gerard Anderson warns that a similar initiative running for 25 years within the United States’ Medicare programme led to what came to be known as “discharge sicker and quicker,” putting additional burden on home and community care. On the other hand, the expected increase in the number of readmissions didn’t materialise.

Anderson revels in the unpredictability of it all. The UK’s health select committee took a rather different view of the lack of a reliable evidence base for health policy, recently passing damning criticism of the persistent lack of evaluation of new initiatives to tackle health inequalities. Ian Forde and Dagmar Zeuner (doi:10.1136/bmj.b3219) look at the evidence for one such initiative, conditional cash transfer, which gives money to disadvantaged families as long as they take up services aimed at improving the health and wellbeing of their children. Similar schemes have worked in Latin America and the United States, although not without some unintended consequences. In Brazil, for example, rates of weight gain among children fell because mothers thought they would lose the cash if their children were no longer underweight.

Will such schemes work in Britain where, despite the Labour government’s commitments to reduce childhood poverty, most recent figures show a slight increase in the number of children living below the poverty line and a decline in social mobility? The authors conclude that, to succeed, the services will need to be of high quality, the incentives will have to be sufficient (which means more than the one of £200 currently proposed), targeting of areas and families will have to be carefully thought through, and the whole thing will have to be robustly evaluated.