“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.

• Does treatment with GLP-1R agonists lead to weight loss in patients who are overweight or obese?
• Can vitamin A supplementation lower the risk of hearing loss in undernourished children?
• How effective is an adjuvanted monovalent vaccine against pandemic influenza A/H1N1 in people with underlying chronic diseases?
• Can a new algorithm estimate the absolute risk of having ovarian cancer in women with and without symptoms?

Improving diagnosis of ovarian cancer in primary care
Early diagnosis of ovarian cancer improves prognosis: five year survival among women diagnosed with stage I ovarian cancer is 90%, compared with only 20% for stage III cancer and 6% for stage IV. Currently, however, less than 30% of women are diagnosed with stage I cancer, as the disease has few established risk factors and a range of non-specific symptoms such as loss of appetite, weight loss, and abdominal pain that also occur with less serious and more common conditions.

Using routinely collected data from the QResearch database of general practices in England and Wales, Julia Hippisley-Cox and Carol Coupland have now developed and validated an algorithm to identify women at highest risk of ovarian cancer to facilitate early referral and investigation. The algorithm was based on simple clinical variables—such as age, family history of ovarian cancer, anaemia, abdominal pain and distension, rectal and postmenopausal bleeding, and loss of appetite and weight—which a patient is likely to be aware of, or which are routinely recorded in general practice. It was successful in predicting 340 (63%) of the 538 new cases of ovarian cancer in the validation cohort over the next two years, which were in the 10% of women with the highest predicted risks. The authors conclude that the algorithm had good discrimination and calibration, and could potentially be integrated into general practice clinical computer systems, to assess risk in women presenting with and without symptoms.

Of course, the development of this algorithm is just the first stage in achieving a routine test for ovarian cancer risk, and in his accompanying editorial. William Hamilton considers the further steps that will be needed before such computerised support for cancer diagnosis can become a reality.

How much weight loss is associated with use of GLP-1R agonists?
When first and second line treatments such as metformin and gliclazide have failed, a trial of exanatide is an option for patients with a body mass index over 35, for those with co-morbidities likely to improve with weight loss, or for people (such as lorry drivers) who for occupational reasons would rather avoid insulin, according to NICE guidelines.

The amount of weight loss associated with glucagon-like peptide-1receptor (GLP-1R) agonists may seem a tangential matter, because the key role of these drugs is to improve glycaemic control. But the question is clearly intriguing for readers of bmj.com; Tina Vilsbøll and colleagues’ meta-analysis is the third most read article online this week. However, beneath promising headlines of a mean 3.2 kg weight loss associated with the drugs, lies complexity and caution—highlighted by Raj Padwal in an accompanying editorial.

Trials included in the study were not powered to detect differences in weight. And the situation is complicated because many trials compared the GLP-1R agonists with active control drugs, such as insulin, which are known to cause weight gain. Weight loss in trials of GLP-1R agonists compared with placebo was less (1.9 kg) than that seen overall. By comparison, Vilsbøll and colleagues write, non-pharmacological measures achieve 1-5 kg weight loss, and sibutramine and orlistat 3-5 kg weight loss. Given that the safety profile of GLP-1R agonists has yet to be established, Padwal writes that their use cannot be recommended for weight reduction alone until their benefits and risks are clarified.

Consumption of fried foods and risk of coronary heart disease
In Spain, where olive or sunflower oil is used for frying, the consumption of fried foods was not associated with coronary heart disease or with all cause mortality, according to Pilar Guallar-Castillón and colleagues, reporting results from the European Prospective Investigation into Cancer and Nutrition. The Spanish cohort included 40, 757 adults aged 29-69, who were free of coronary heart disease at baseline in 1994-96 and were followed up until 2004.

The decline in coronary heart disease
Three studies examine the decline in mortality from coronary heart disease and its determinants in England, Poland, and Denmark (doi:10.1136/bmj.d8059, doi:10.1136/bmj.d8136, doi:10.1136/bmj.e356). Hugh Tunstall-Pedoe draws the findings together in an editorial, saying that the decrease seems to be associated with the effects of evidence based treatments in primary prevention, coronary care, and secondary prevention.