“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.

• Is vitamin A supplementation associated with reduced death and illness in children?
• Is immediate adenoidectomy more effective than initial watchful waiting in children with recurrent upper respiratory tract infections?
• What are the prevalence, stability, and impact of subtypes of headache?
• Do changes in heart disease mortality after age 45 years differ between women and men?

Adenoidectomy in children with recurrent upper respiratory tract infections
“The adenoidectomy rate is more than three times higher in the Netherlands than in the US, and the proportion of children operated on for infections varies fivefold across these two countries, suggesting that there is no international consensus as to which children benefit from the operation.” So say Maaike van den Aardweg and colleagues in the introduction to their paper on bmj.com.

With that in mind, the authors compared two strategies for children aged 1-6 who had been referred for adenoidectomy, in a randomised controlled trial with number of upper respiratory tract infections per person year as the primary outcome. The children were randomised either to adenoidectomy within six weeks, with or without myringotomy, or to a strategy of initial watchful waiting.

Research is often as messy as real life, and this study was hampered by 40% of the children in the initial watchful waiting group having adenoidectomy anyway during the trial’s two year follow-up period. But that means 60% didn’t have surgery, and the per protocol and as treated analyses showed the same results as the intention to treat analyses: the prevalence of infections decreased equally over time in both groups. So the authors were able to conclude that immediate adenoidectomy was no better than initial watchful waiting.

Changes in headache over time
What can a patient with headache expect over time—if they are troubled by migraines now, will they remain so in the future, or will they be bothered by different types of headaches? There are good data to describe the prevalence of headache subtypes, but we know little about how headaches change over time. As headache is such a common problem, this seemed like a substantial gap in doctors’ knowledge.

Kathleen Merikangas and colleagues looked at data from a Swiss cohort study with 30 years’ worth of follow-up. At the start of the study the participants were around 20 years old. They were interviewed every few years and each time asked whether they had suffered with headache or migraine in the past 12 months. If they answered yes, the doctor or psychologist conducting the interview documented the clinical characteristics of their predominant headache. The authors used this information to describe how participants’ primary headaches changed over time, and how they affected their lives.

Although the information was carefully gathered, there is a big limitation in how the headaches were classified. Each participant could be assigned only one type of headache at any one time point; in reality some patients might have experienced more than one type at once. Nevertheless, these data make a valuable contribution to scarce literature and the paper should make interesting reading for doctors who encounter headache regularly, or perhaps even suffer with them.

Mortality from heart disease after menopause
The BMJ gives priority to research that will improve the decision making of doctors, whether they’re clinicians, policymakers, or researchers. Of course, the question “will this help doctors make better decisions?” is subjective, and we rely on authors, reviewers, and the manuscript committee—and post-publication feedback from readers and evidence appraisal services (such as EvidenceUpdates and Journal Watch)—to hone these judgments.

 The hardest papers to judge in this way are often those reporting what we’d consider to be pure epidemiology, of which Dhananjay Vaidya and colleagues’ modelling study is a good example.  The authors looked at three birth cohorts in England and Wales and the United States (born 1916-25, 1926-35, and 1936-45) and noted their all cause mortality and disease specific mortality rates for heart disease (in men and women) and breast cancer (only in women) between 1950 and 2000. 

Using regression analyses the authors confirmed and extended the findings of previous cross sectional studies. They did not see a sudden acceleration in female heart disease mortality around menopausal age; rather, the reduction in the mortality gap between the sexes seemed to be due to an increase in mortality in men at around the same age. And in the full, open access, version of this paper on bmj.com the authors conclude that “efforts to improve cardiac health in women should focus on their lifetime risk rather than only after menopause.” We thought that was a pretty useful bottom line that could improve decision making. Were we right?

Changes in severity of 2009 pandemic H1N1/A influenza in England
A study by A M Presanis and colleagues provides a complete overview of the severity of the first two waves of the 2009 pandemic H1N1/A influenza outbreak in England, synthesising all available surveillance data. The findings suggest a mild pandemic, with the case severity ratio (the probability that an infected individual develops severe disease) and the infection severity ratio (the probabilities of all infections leading to severe events) highest in children and older adults and increasing over time. Infection attack rates were highest among school age children. The results suggest underascertainment of severe cases and highlight the limits of current surveillance systems.