The use of highly structured care to achieve blood pressure targets

BMJ 2012; 345 doi: http://dx.doi.org/10.1136/bmj.e7777 (Published 20 November 2012) Cite this as: BMJ 2012;345:e7777
  1. Christopher E Clark, clinical academic fellow1,
  2. Richard McManus, professor of primary care2
  1. 1Primary Care Research Group, Institute of Health Services Research, University of Exeter Medical School, Exeter EX1 2LU, UK
  2. 2Department of Primary Care Health Sciences, University of Oxford, Oxford, UK
  1. christopher.clark{at}pms.ac.uk

Computer support helps, but lower targets may not be achievable or cost effective in primary care

Hypertension is a major risk factor for premature morbidity and mortality throughout the world.1 In 2010 the prevalence of hypertension in the health survey for England was 32% in men and 29% in women.2 Hypertension is a common reason for consultation in primary care, and blood pressure is a key indicator within the UK Quality and Outcomes Framework.3 Although the “rule of halves” (half of people with hypertension are undetected, half of those detected are not treated, and in half of those treated hypertension is not controlled) was published 40 years ago,4 the problem of suboptimal control of high blood pressure remains. Reviews have identified that a structured approach to care is a key component of effective strategies to reduce blood pressure.5 6

In a linked paper (doi:10.1136/bmj.e7156), Stewart and colleagues present the findings of a large trial of structured care that used a computerised algorithm to titrate antihypertensives to target blood pressure.7 The trial was conducted in primary care, where most patients with hypertension are treated,2 and recruited from a range of practices across almost the entire continent of Australia. Participants were adults with new and established hypertension whose blood pressure remained above target after stopping any current antihypertensive drugs and a 28 day run-in period during which they received 80 mg valsartan daily. Almost a third were excluded after the run-in period, mostly because their blood pressure was adequately controlled. This study therefore looked at patients with hypertension who were not controlled by treatment with an angiotensinogen receptor blocker (ARB).

Eligible participants were then set individualised blood pressure targets, using computer support, according to their cardiovascular risk assessment at entry: 125/75 mm Hg in the presence of proteinuria, 130/80 mm Hg in the presence of target organ damage, or 140/90 mm Hg if neither of these were detected. They were then randomised to usual care with treatment over and above ARB determined by their general practitioner or to an intervention. The intervention used an intensive algorithm based blood pressure management strategy with computer support, and it was designed to achieve the patient’s target blood pressure within six months. Within the intervention group participants were further randomised to monotherapy with ARB or dual therapy with the addition of a diuretic or calcium channel blocker.

Significant pooled reductions in blood pressure were seen in both groups (13.2/7.7 mm Hg for the intervention and 10.1/5.5 mm Hg for usual care), with the primary outcome measure of achievement of target in favour of the intervention (36.2% v 27.4%). This difference in primary outcome was significant only for those with blood pressure targets above 125/75 mm Hg. Prescription rates and drug side effects for combinations of treatments were greater in the treatment arm, which suggests some success in overcoming clinical inertia.8

Cautious interpretation of the findings is warranted. Participants were relatively young (mean age 58 years), but only 40% had uncomplicated hypertension. The authors carefully describe “usual care,” which was enhanced by clinical profiling and target setting at study entry.9 Outcome blood pressure measurements for participants were obtained for pragmatic reasons by the GPs or nurses delivering the intervention using an unblinded protocol. This is a source of bias in studies of this type, and independent assessment would have been preferable.10 “Aberrant” blood pressure readings were discarded and included those with a greater than 10 mm Hg difference between systolic blood pressure readings, which is less than the mean difference between first and third readings seen in a recent Canadian study using automated sphygmomanometers.11 Participants randomised to the intervention were seen twice as often as those receiving usual care; therefore, the omission of an associated cost effectiveness analysis is important if such a strategy is to be implemented on a wider scale.

The authors of this industry sponsored study describe the initial intervention (valsartan) as one of the most commonly prescribed antihypertensive agents globally. It is, however, an alternative first line agent only for people under 55 years of age in the UK.12 The more stringent blood pressure targets in this study are lower than those advised by the National Institute for Health and Clinical Excellence, and they are arguably lower than current evidence suggests other than for stroke.12 13 Despite 14% of participants in each group being assigned blood pressure targets higher than computer recommendations at the GP’s discretion, there was still evidence of deviation from the study protocol because GPs failed to up-titrate treatment for the more stringent blood pressure targets. More withdrawals and twice as many adverse events occurred in the intervention group. Given that performance is likely to be worse outside of trial conditions, achievement of aggressive blood pressure reduction may not be realistic, and may even be inadvisable, for many patients.

This study demonstrates the potential returns from computer aided structured care for the management of high blood pressure and how it could be delivered across a large geographical area. We now need to understand whether these effects depend on the choice of drug, how cost effective such interventions are, and which groups would benefit from the more stringent targets used in this study.


Cite this as: BMJ 2012;345:e7777


  • Research, doi:10.1136/bmj.e7156
  • Competing interests: Both authors have completed the ICMJE uniform disclosure form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare: no support from any organisation for the submitted work; no financial relationships with any organisations that might have an interest in the submitted work in the previous three years; no other relationships or activities that could appear to have influenced the submitted work.

  • Provenance and peer review: Commissioned; not externally peer reviewed.