Research highlights – 27 August 2010

Research questions “Research highlights” is a weekly round-up of research papers appearing in the print BMJ. We start off with this week’s research questions, before providing more detail on some individual research papers and accompanying articles.

Educating patients about diabetes
In 2008, M J Davies and colleagues tested the efficacy of a six hour group education programme for people with newly diagnosed type 2 diabetes—the diabetes education and self management for ongoing and newly diagnosed (DESMOND) programme (BMJ 2008;336:491-5). Compared with usual care, the intervention resulted in small improvements in weight loss, smoking cessation, and beliefs about illness at 12 months, but no difference in haemoglobin A1c levels. In a response to the article on, Roger Fisken, consultant physician and diabetologist at Friarage Hospital in Northallerton, points out: “The difference of 1.1kg in weight, though statistically significant, is not, frankly, terribly impressive.”

Now Davies and colleagues have done another study of the DESMOND intervention, this time looking at cost effectiveness, firstly on the basis of costs in the previous trial of the intervention and secondly on the basis of “real world” costs for a hypothetical primary care trust. Again the results weren’t huge, but this time in a good way: the estimated mean incremental cost per quality adjusted life year was £5387 using trial based intervention costs and £2092 using real world intervention costs, both values well under the cost effectiveness acceptability threshold of £20000 per QALY routinely used by NICE.

So although the intervention only produces small benefits in patients with diabetes, this one off programme is so cheap that trusts might as well give it a shot. Or, as the authors write more eloquently, “The collective mean estimated benefit of the changes arising from the DESMOND intervention is sufficient to outweigh the low intervention cost per patient.”

Stopping maintenance treatment for psychosis

This double blind trial by Eric Y H Chen and colleagues studies 178 young people who had received antipsychotic treatment for at least a year for first episode psychosis. Participants were randomised to receive placebo or maintenance treatment with quetiapine for a further year, and the treated group had a significantly lower relapse rate. This finding may seem unsurprising, but the study fills an important evidence gap and should improve prescribers’ and patients’ decisions. As Jonathan C Craig and colleagues explain in a linked editorial, trials of maintenance treatment for chronic conditions rarely evaluate what really matters: how long people should be treated.

Incidentally, the editorialists are nephrologists, not psychiatrists, and their editorial addresses the broad question of trial design in chronic diseases.

Pregnancy after miscarriage

“How soon can we try again?” was the question addressed by Eleanor Love and colleagues, who sought to define the ideal interval between a miscarriage and subsequent pregnancy. Their retrospective analysis included data from Scottish hospitals for over 30000 women who had a miscarriage in their first recorded pregnancy but later became pregnant again. They found that women who conceived within six months of an initial miscarriage had the best reproductive outcomes and lowest rates of complication in a subsequent pregnancy. A shortage of good evidence has made this question difficult for doctors to answer and, in an editorial, Julia Shelley explains why it’s been such a challenge to researchers.

Research online: For these and other new research articles see

Migraine and mortality risk

Two papers published online this week add to the evidence that people who have migraine with aura are at increased risk of cardiac and cerebrovascular disease. In Iceland, Larus Gudmundsson and colleagues examined whether migraine with aura in mid-life was associated with increased cardiovascular and other mortality (doi:10.1136/bmj.c3966), and in the United States, Kurth and colleagues assessed the relation in women between migraine with aura and haemorrhagic stroke (doi:10.1136/bmj.c3659). Both cohort studies found significantly increased risks, but editorialists Klaus Berger and Stefan Evers interpret the findings with caution and advise that discussions with patients about risk should be tailored to the individual (doi:10.1136/bmj.c4410).