Continued Improvement in Hypertension Management in England
Results From the Health Survey for England 2006
This study evaluate whether blood pressure management has improved in England between 2003 and 2006, using cross-sectional, nationally representative, random samples of 8834 (in 2003) and 7478 (in 2006) noninstitutionalized adults (aged ≥16 years) of mean age 46 (in 2003) and 47 (in 2006) years. Overall mean blood pressure levels in 2006 were 130.8/74.2 mm Hg in men and 124.0/72.4 mm Hg in women. Awareness of hypertension increased significantly in the overall population (from 62% in 2003% to 66% in 2006; P<0.001), the increase being significant in women (from 64% in 2003% to 71% in 2006; P<0.001) but not in men (from 60% to 62%; P=0.26). Similarly, the proportion treated had risen significantly overall (from 48% to 54%; P<0.001) and in women (from 52% to 62%; P<0.001) but not in men (from 43% to 47%; P=0.05). Control rates (<140/90 mm Hg) were higher in 2006 than in 2003 (from 22% to 28%; P<0.001) and had increased more among women than men: from 23% to 32% in women (P<0.001) and from 21% to 24% in men (P=0.02). Among those on treatment, control rates increased from 46% to 52% (P<0.001; from 44% to 53% in women, P<0.001; and from 48% to 52% in men, P=0.18). The most common agents used for monotherapy have changed since 2003 and were angiotensin-converting enzyme inhibitors or angiotensin receptor blockers. Where ≥2 drugs were used, the most common antihypertensive class used varied by age and ethnicity. Awareness, treatment, and control of hypertension increased between 2003 and 2006, particularly in women, but opportunities for further improvement remain.
Recent population-based surveys of hypertension management throughout the world have been consistent in showing variably inadequate levels of blood pressure control.1–3 Data arising from the annual, nationally representative Health Survey for England (HSE) have shown improving levels of blood pressure control between 1994 and 19983,4 and subsequently in 2003,5 but despite this improvement, control rates (defined as systolic blood pressure <140 mm Hg and diastolic blood pressure <90 mm Hg) of those deemed hypertensive (ie, either systolic ≥140 mm Hg or diastolic ≥90 mm Hg or on treatment for hypertension) were only 22% in 2003.
Since April 2004, the Quality and Outcomes Framework was introduced in the new General Medical Services contract in the United Kingdom. This is a pay-for-performance system with which general practitioners receive remuneration for achieving various clinical targets, with points and payments awarded according to the level of achievement. It is a voluntary part of the new General Medical Services contract; general practices can aspire to achieve all, part, or none of the points available in the Quality and Outcomes Framework (although, to date, most practices with a General Medical Services contract have participated fully). One of the measured achievements is lowering blood pressure levels to <150 mm Hg and <90 mm Hg.6 The potential impact of this aspect of the practitioners’ new contract on blood pressure control in the population has not been evaluated in any rigorous way using a population-based sample in England. However, a recent study on data from general practices from April 2004 to March 2007 attributed improved blood pressure monitoring and control in the United Kingdom for the first time to the physician pay-for-performance incentives.7 Meanwhile, British guidelines continue to recommend optimal blood pressure targets to be <140 mm Hg systolic and <90 mm Hg diastolic and lower for those with diabetes mellitus or established vascular disease.8
In 2006, the focus of the HSE was cardiovascular disease and associated risk factors: these data allow an evaluation of whether blood pressure management, in terms of awareness, treatment, and control rates, has continued to improve since 2003.
The HSE is an annual, nationally representative sample of the noninstitutionalized English population of all ages, randomly selected in a stratified 2-stage process using the small-users Postcode Address File. The primary focus in 2003 and 2006 was on cardiovascular disease. The sampling and data collection methods have been described in detail elsewhere.9
Data collection, which took place throughout the year, was identical in 2006 and 2003. It involved an interview followed by a visit from a specially trained nurse who measured blood pressure, took a blood sample, and recorded use of prescribed medicines. Using the Omron HEM 90710 and an appropriately sized cuff, 3 sitting blood pressure readings were taken on the right arm after 5 minutes of rest. Blood pressure data used in this study are based on the means of the second and third measurements. Participants who had exercised, eaten, drunk alcohol, or smoked in the 30 minutes before measurements were excluded from analyses.
The interviewers collected sociodemographic information, including self-assigned ethnicity, using Census 2001 categories. During the interview, in the cardiovascular module of questions, participants were asked if they had been told by a doctor or a nurse that they had high blood pressure; information about diabetes mellitus or history of cardiovascular disease (angina, heart attack, or stroke) was also collected. In 2006, to avoid an overlong interview, this module was administered to only half of the sample aged ≥65 years, selected at random. This was to avoid a decrease in the response rate for old people. On the other hand, the information on nonrespondents was used to compute the specific weight used in the analysis to adjust for potential selection bias. Research ethics approval was obtained from the appropriate committees before each survey.
Hypertension was defined as systolic blood pressure ≥140 mm Hg, diastolic blood pressure ≥90 mm Hg, or being on treatment for blood pressure. Isolated systolic hypertension in individuals aged ≥30 years was defined as follows: stage 1 was defined as systolic blood pressure 140 to 159 mm Hg and diastolic blood pressure <90 mm Hg; and stage 2 was defined as systolic blood pressure ≥160 mm Hg and diastolic blood pressure <90 mm Hg.
Details of which, if any, antihypertensive agents were being taken were recorded by the nurse. In accordance with previous analyses, respondents who were not sure whether a blood pressure–lowering drug that they were taking had been prescribed to treat hypertension were considered as a treated hypertensive individual if they also reported a history of hypertension. We examined the use of antihypertensive drugs by class and compared this with the current guidelines11 by age and ethnicity (<55 and not black versus ≥55 years of age or black).
We limited analyses to participants aged ≥16 years with no missing data. Both 2003 and 2006 samples were weighted to allow for nonresponse differences both to the interview and then to the nurse visit, as described previously.5 We computed awareness, treatment, and control rates among hypertensive men and women from HSE 2006 and compared these with 2003 subjects using the 2-sided z test. We defined awareness as a self-report of having been diagnosed as hypertensive by a doctor or nurse, excluding women diagnosed during pregnancy. For control rates, we considered 2 blood pressure target levels: <140/90 mm Hg (the target recommended in most hypertension guidelines) and <150/90 mm Hg (an “audit standard,” which is also included in recent British guidance8 to acknowledge the difficulties of achieving the “optimal” target in some patients and also because this is the level that attracts payment in the general practitioners’ contract).6
We calculated the Framingham-based equation12 to estimate the 10-year cardiovascular disease risk in individuals aged ≥30 years without self-reported coronary heart disease or stroke. We included age, sex, smoking, systolic blood pressure, total and high-density lipoprotein cholesterol, and self-reported doctor-diagnosed diabetes mellitus in the equation.
In 2006, 14 142 adults aged ≥16 years were interviewed, among whom 10 489 had a nurse visit. Of these, 8762 had valid blood pressure readings (1336 had eaten, drunk alcohol, or smoked in the 30 minutes before blood pressure recording; 3 valid measurements were unavailable on 192 people: 89 refused or had invalid results; and 110 were pregnant). A total of 1067 informants aged ≥65 years were excluded because, by design, they did not undergo the cardiovascular module of questions. A further 217 individuals were excluded because they reported taking medication that lowered blood pressure but did not report a history of hypertension. We included the remaining 7478 (3314 men and 4164 women; corresponding with 4019 men and 4374 women, when analyses were weighted) in the analysis; the mean age was 47 years in both sexes.
Mean systolic blood pressure rose across the whole age range in both men and women but was higher in men than women until the age of 70 years (Table 1 and Figure 1). Diastolic blood pressures also rose with age in both sexes but only until the age of 60 years, above which blood pressures fell systematically. Diastolic blood pressures were generally higher in men than women, except among those aged 16 to 19 years and aged ≥70 years. Overall mean blood pressure levels were 130.8/74.2 mm Hg in men and 124.0/72.4 mm Hg in women. Hypertension rates increased with age in both sexes and were more prevalent in men than women, except in the age range from 70 to 79 years.
Overall, hypertension was observed in 30% of informants (32% of men and 29% of women), of which in those aged ≥30 years, almost half, 14% (16% of men and 12% of women), was because of stage 1 isolated systolic hypertension. The mean blood pressure levels and prevalences of hypertension in 2006 compared favorably with those reported in 2003 when mean blood pressures were 131.4/74.5 mm Hg in men and 125.7/73.3 mm Hg in women and overall hypertension rates were 33% and 30%, respectively.
Two thirds of those classified as hypertensive in 2006 were aware of their diagnosis, awareness being more common among women than men in all of the age groups (Table 2 and Figure 2). More than 60% of the female hypertensive population were on treatment for hypertension, but fewer than half the hypertensive men were treated. Similarly, control rates to <140/90 mm Hg were higher overall among women than men, with approximately one third and one quarter of the respective hypertensive populations having reached “controlled” status.
Of those on treatment for hypertension in 2006, 52% of informants (52% of men and 53% of women) had controlled blood pressure (Table 2). This compared with 46% (48% and 44%, respectively) for the equivalent populations in 2003. Similarly, the rates of awareness, treatment, and control observed in 2006 were all consistently greater than the equivalent figure reported in 2003, particularly among women. Using the audit standard for control (<150/90 mm Hg), overall rates of control were significantly higher in 2006 compared with 2003, with 37% of informants (33% of men and 42% of women) having blood pressures <150/90 mm Hg (Table 2). Of those on treatment for high blood pressure in 2006, 69% (70% of men and 69% of women) were controlled to this audit level.
About three quarters of patients with a self-reported history of cardiovascular disease or diabetes mellitus or an estimated 10-year risk of cardiovascular disease of ≥20% were hypertensive (Table 3). Of those hypertensive patients who had cardiovascular disease or diabetes mellitus, ≈85% were treated, and ≈44% had their blood pressures controlled to <140/90 mm Hg. However, for those hypertensives who did not have CHD or stroke but whose estimated CV risk was ≥20%, only 55% were treated and 17% controlled. These rates are all greater than the equivalent figure in 2003, particularly treatment rates in the estimated risk group, which rose from 45%.
More than 60% of patients on treatment for hypertension were receiving ≥2 antihypertensive drugs (Table 4), which compares with 56% in 2003. For those receiving monotherapy, the most common agents used were blockers of the renin-angiotensin system (RAS; either angiotensin-converting enzyme inhibitors or angiotensin receptor blockers), irrespective of age or ethnicity. Overall diuretics, β-blockers, and calcium channel blockers (CCBs) were the second, third, and fourth most commonly used agents but with similar levels of usage. However, this order changed when stratified by age and ethnicity: β-blockers were clearly the second most commonly used agents for those <55 years of age, but diuretics and CCBs were more frequently used than β-blockers among older patients or those of African origin.
The most common combination of drugs among those taking 2 agents was RAS blockers plus diuretics, with 3 combinations (diuretic plus CCB, RAS blockers plus β-blockers, and RAS-blockers plus CCBs) having similar levels of usage (Table 4). Again, when stratified by age and ethnicity, the order of preference changed for those aged <55 years, with greater use of RAS blockade plus β-blockers, whereas diuretics plus CCBs were used together relatively less often in this age group. When 3 agents were used, the most common combination of agents was RAS blockers, diuretics, and CCBs, with RAS blockers, diuretics, and β-blockers a close second.
These survey data from 2006 show that, for the first time, the majority of those on treatment for hypertension in England had their blood pressures controlled to the target of <140/90 mm Hg. Important and significant increases in rates of awareness, treatment, and control are apparent since 2003, particularly among women. As reported previously, the rates in 2003 were superior to those observed in 1998,5 which, in turn, improved on the 1994 rates.3,4 Our findings confirm, on the basis of a population-based survey, that further improvements have occurred in recent years, as was also shown in an earlier general practitioner–based analysis of Quality and Outcomes Framework data.7 Among those on treatment, the improved control rates to targets of both <140/90 and <150/90 mm Hg were associated with an increased use of ≥2 agents. The agents used to lower blood pressure are not strictly in keeping with current British guidance,11 particularly among those on monotherapy.
Strengths of this study include the large nationally representative random sample, which is more generalizable than data arising from patients attending for healthcare. In addition, the use of identical questions and protocols for objective measurements in the 2003 and 2006 surveys allowed direct valid comparisons between the 2 years. Our use of repeated independent cross-sectional surveys ensured that there was no “Hawthorne effect,”13 whereby reporting abnormal results to participants or their general practitioners (which is ethically required) acts as an intervention in its own right, affecting future awareness, treatment, and control of the abnormality (raised blood pressure, in this instance).
The main weakness of these data is the reduced proportion of interviewed participants who received a nurse visit. However, bias was minimized by the use of nonresponse weighting that combined adjustment to ensure that those interviewed reflected the national and regional population, based on sociodemographic profiles, with further adjustment, using variables from the interview, to minimize differences between those interviewed who did and did not have a nurse visit.9 The further reduction in the sample size is attributable to those who did not undergo the cardiovascular module of questions. Had we included the total of those who had valid blood pressure readings (8762), the response rate would have been 84% of those with a nurse visit. Because blood pressure measurements in all of these subjects were available, we compared them with the subsample selected for this article using the relative nonresponse weight. Mean systolic and diastolic blood pressures were not significantly different in the 2 samples, even after selecting those aged ≥65 years, which, in our study, was half of the total sample.
The prevalence of hypertension, however defined, is probably exaggerated as a result of making the diagnosis on the basis of a single set of readings, contrary to optimal practice.8,14 Some compensation for the possible exaggeration in the prevalence of hypertension was the use of the last 2 of 3 readings, measured at home, by a nurse under careful conservative conditions,9 all of which are likely to generate lower blood pressures than those produced in routine clinical practice. In addition, it should be pointed out that control rates among those who definitely should be treated at 140/90 mm Hg according to guidelines that prevailed in 20068 are clearly unsatisfactory (Table 3). These findings are particularly worrisome given that guidelines current in 2006 recommended a lower target of <130/80 mm Hg for such patients.8
Current and earlier guidance8,11 in the United Kingdom recommend CCBs or diuretics for those of African origin and for those aged ≥55 years (the majority of hypertensives). However, angiotensin-converting enzyme inhibitors or angiotensin receptor blockers (RAS blockers) were the most common “single” agents used overall and when stratified as per National Institute for Health and Clinical Excellence/British Hypertension Society recommendations.11 When blood pressure is inadequately controlled by monotherapy, the National Institute for Health and Clinical Excellence/British Hypertension Society guidance currently recommend the use of either “A+C” (where A stands for angiotensin-converting enzyme inhibitors or angiotensin receptor blockers and C for CCBs) or A+D (where D stands for diuretics). Overall in 2006, A+D was the most common combination used, but A+C was not used more commonly than C+D or A+B (where B stands for β-blockers).
The use of β-blockers as monotherapy has fallen from 29% in 20035 to 21% in keeping with latest guidance, although, in younger patients, 33% of those on monotherapy used β-blockers, and among those on 2 drugs, >35% used β-blockers, conflicting with current guidelines. The most common combinations of 3 and 4 drugs (A+C+D and A+C+D+B or A+C+D+α-blockers) used are in keeping with current British guidelines.11 Apparent deviation from the “A/CD” treatment algorithm may in part reflect the coexistence of other conditions that are considered compelling indications (eg, β-blockers for active coronary heart disease).
According to British Guidance before 20068 and since11, a systolic pressure threshold for treatment of ≥140 mm Hg only applies to those with diabetes mellitus, established cardiovascular disease, or a 10-year estimated risk of cardiovascular disease of ≥20%. Of the 38% of women and 53% of men considered to be in the untreated hypertensive population (see Table 2) ≈4% were diabetic or had established cardiovascular disease, 3% had a systolic blood pressure >160 mm Hg or diastolic >100 mm Hg, and more than one third had a high estimated cardiovascular risk. Nevertheless, the remaining majority of this untreated hypertensive population did not, according to current British guidance,8,11 merit intervention.
Previous surveys had clearly shown improving trends in management before 2004.3–5 The production and dissemination of a series of guidelines for the management of hypertension orchestrated by the British Hypertension Society seem likely to have influenced people and helped to produce this positive trend. Nevertheless, between 2003 and 2006, particularly large improvements in treatment and control rates were seen among women, which were similar in magnitude to those that occurred between 1998 and 2003. Although it is hard to evaluate its impact, the new General Medical Services Contract6 is likely to have contributed to these improvements. However, further improvements in detection, treatment, and control are required to reduce cardiovascular complications, particularly in those at high risk.
Despite the shortcomings of any cross-sectional survey to evaluate hypertension management, the data from this large nationally representative sample from England represent more evidence of an ever-improving approach to hypertension management, which, had it not occurred between 2003 and 2006, is likely to have cost between 4000 and 8000 fatal or nonfatal cardiovascular events.15
We thank Paola Primatesta for her advice, colleagues in the Joint Health Surveys Unit of University College London and NatCen (the National Centre for Social Research), the interviewers and nurses who collected the data, and the participants in the surveys.
Sources of Funding
Health Survey for England 2003 was funded by the Department of Health and Health Survey for England 2006 by the National Health Service Information Centre for health and social care. The funders played a substantial role in determining the content of the surveys but played no part in the design, analysis, or interpretation of this study, nor in drafting this article or the decision to submit it for publication. N.P. is grateful for support from the National Institute for Health Research Biomedical Research Centre Funding Scheme. The authors’ work is independent of the funders. The views expressed in this article are those of the authors and do not necessarily reflect those of the funders.
N.P. has received honoraria for speaking at conferences and attending ad hoc advisory boards from most of the major pharmaceutical companies producing antihypertensive and lipid-lowering agents and from several companies for consultancy and organizing educational meetings. He has received funds for several staff members to carry out research from Pfizer, Novartis, Servier, Menarini, and Mars, Inc. The remaining authors report no conflicts.
- Received November 3, 2008.
- Revision received November 13, 2008.
- Accepted January 5, 2009.
Wolf-Maier K, Cooper RS, Kramer H, Banegas JR, Giampaoli S, Joffres MR, Poulter N, Primatesta P, Stegmayr B, Thamm M. Hypertension treatment and control in five European countries, Canada, and the United States. Hypertension. 2004; 43: 10–17.
Primatesta P, Brookes M, Poulter NR. Improved hypertension management and control. Results from the Health Survey for England 1998. Hypertension. 2001; 38: 827–832.
Department of Health. Delivering Investment in General Practice: Implementing the New GMS Contract. London, United Kingdom: Department of Health; 2003.
Ashworth M, Medina J, Morgan M. Effect of social deprivation on blood pressure monitoring and control in England: a survey of data from the quality and outcomes framework. BMJ. 2008; 337: a2030.
Williams B, Poulter NR, Brown MJ, Davis M, McInnes GT, Potter JF, Sever PS, Thom SM. British Hypertension Society guidelines for hypertension management 2004 (BHS-IV): summary. BMJ. 2004; 328: 634–640.
Craig R, Mindell J, eds. Health Survey for England 2006. London, United Kingdom: The Information Centre; 2008.
National Collaborating Centre for Chronic Conditions. Hypertension: Management of Hypertension in Adults in Primary Care: Partial Update. NICE Clinical Guideline. London, United Kingdom: Royal College of Physicians; 2006.
Mayo E. The Human Problems of an Industrial Civilization. New York, NY: MacMillan; 1933: Chapter 3.
Mancia G, De Backer G, Dominiczak A Cifkova R, Fagard R, Germano G, Grassi G, Heagerty AM, Kjeldsen SE, Laurent S, Narkiewicz K, Ruilope L, Rynkiewicz A, Schmieder RE, Boudier HA, Zanchetti A. 2007 ESH-ESC practice guidelines for the management of arterial hypertension. ESH-ESC Task Force on the Management of Arterial Hypertension. J Hypertens. 2007; 25: 1751–1762.
Hansson L, Zanchetti A, Carruthers SG, Dahlof B, Elmfeld D, Julius S, Ménard J, Rahn KH, Wedel H, Westerling S, for the HOT Study Group. Effects of intensive blood-pressure lowering and low-dose aspirin in patients with hypertension: principal results of the Hypertension Optimal Treatment (HOT) randomised trial. Lancet. 1998; 351: 1755–1762.