Research highlights – 8 July 2011

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.

Accuracy of blood pressure measurements
The problems of blood pressure monitoring are currently in the spotlight. Only last week, our Shortcuts section (2011;342:d4015) covered a study of data from a clinical trial that concluded that at least five repeat blood pressure readings were necessary to achieve ≥80% confidence in the accuracy of the recording (Annals of Internal Medicine 2011;154:781-8). Now the BMJ has published a systematic review by J Hodgkinson and colleagues evaluating the performance of different methods of diagnosing hypertension in primary care.

They conclude that compared with ambulatory monitoring, neither clinic nor home measurements (even repeated measurements) have sufficient sensitivity or specificity to be recommended as a single diagnostic test. More widespread use of ambulatory blood pressure for the diagnosis of hypertension would result in more appropriately targeted treatment. However, they acknowledge that their results should be treated with caution: the wide variety of different diagnostic thresholds used meant that fewer than half of the 20 eligible studies could be used in the meta-analysis.

Adding grist to the mill, a prospective cohort study published this week in Archives of Internal Medicine (2011;171:1090-8) investigated the prognostic value of ambulatory blood pressure measurement in comparison with office measurements in 436 patients with chronic kidney disease. They conclude that ambulatory measurement, particularly at night time, predicts renal and cardiovascular risk, whereas office measurement “does not predict any outcome.”

Painkillers and atrial fibrillation
NSAID use and atrial fibrillation are both common—especially in older people—so the existence of a link between the two would have major public health implications. Some studies have suggested that such an association exists but the findings were not clear cut.

Morten Schmidt and colleagues looked at the risk of atrial fibrillation or flutter associated with use of non-selective NSAIDs or selective COX 2 inhibitors (which previous studies did not investigate). The case-control study was based on population data from Denmark, and showed that both types of agent were associated with an increased risk, especially among new users (contrasting with a previous study in which long term users were most at risk).

Editorialist Jerry Gurwitz says that although the results don’t conclusively prove a link—and certainly not a causal one—caution in prescribing NSAIDs is advisable in any case for “older patients with a history of hypertension or heart failure, who are already at high risk for adverse effects of these drugs.”

Quality of clinical activities covered (and not covered) by the QOF pay for performance scheme in UK primary care
In 2006 a report from the US Institute of Medicine concluded that “pay for performance should be introduced as a stimulus to foster comprehensive and system-wide improvements in the quality of healthcare” ( This piqued interest in the UK, where general practitioners had already been grappling with such a scheme in the form of the Quality and Outcomes Framework (QOF). Ruth McDonald and colleagues’ qualitative study in two English general practices suggested that implementing the QOF scheme had not demotivated general practitioners: “Although I hate it, I do, you know, its very paradoxical but I actually think it’s a good idea and I think it makes things tangible” admitted one participant in the study (BMJ 2007;334:1357).

Now, along with other colleagues, two authors from that earlier BMJ study—Martin Roland and Stephen Campbell—have asked whether, in 148 English practices with datasets from the years 2000-07, pay for performance was associated with neglect of clinical activities that were not covered by QOF. They interrogated the General Practice Research Database to look at trends in quality of care for 42 activities (23 incentivised under the QOF, and 19 not incentivised) selected from 428 indicators of quality of care. Quality improved initially for incentivised activities but then plateaued. At first, the scheme had little impact either way on non-incentivised activities, but after three years, quality of care for some fell significantly below levels predicted from the period before the introduction of QOF.

Association between provision of mental illness beds and rate of involuntary admissions in the NHS in England 1988-2008
Patrick Keown and colleagues examined the rise in the rate of involuntary admissions for mental illness in England that has occurred as community alternatives to hospital admission have been introduced (doi:10.1136/bmj.d3736).