[Previous] [Contents] [Next]

Scottish Health Survey 1995: Volume 1

6 BLOOD PRESSURE

Wei Dong, Paola Primatesta and Lualhati Bost

6.1 Introduction

In Scotland’s Health: A Challenge To Us All, The Scottish Office identified coronary heart disease as a top priority for improving the health of the Scottish population and set a target of reducing mortality among adults aged under 65 by 40% between 1990 and 2000.¹ One of the major risk factors for heart disease and stroke is raised blood pressure. A reduction in blood pressure through primary and secondary prevention should therefore be targeted in order to achieve a decline in cardiovascular disease (CVD) mortality and morbidity.^2,3,

Blood pressure is determined by genetic and environmental factors, known and unknown. It is associated with many factors such as age, sex, socioeconomic status, body weight, salt intake, excessive alcohol consumption and physical activity.⁴ In addition, many medications may affect blood pressure levels.

It is widely accepted that the relationship between blood pressure and CVD is continuous and there is no threshold effect in risk for CVD. However, in clinical settings it is common practice to classify individuals as being hypertensive or normotensive using various definitions and to report prevalence of hypertension at a population level.

The Scottish Health Survey is the first survey to provide data on blood pressure based on a nationally representative sample of the Scottish population (aged 16-64). This chapter describes the Health Survey results for: mean systolic and diastolic blood pressures; the prevalence of high blood pressure; and the proportion of those with high blood pressure who were aware of their condition (detected), and whose condition was treated and controlled. An important aim of this chapter is to provide baseline data which may be useful in assessing trends over time. Variations in blood pressure between regions and between social classes are also presented, and the relationships between blood pressure and various life-style factors are examined.

Finally, the Health Survey results are compared with data from the 1984-86 Scottish Heart Health Study (SHHS) in order to look at changes in blood pressure over the last decade, and with results from the 1994 Health Survey for England (1994 being the most recent year in which the English survey focused on cardiovascular disease).

6.2 METHODS

The detailed protocol for blood pressure measurement is contained in Appendix B in the Technical Report. Briefly, blood pressure was measured using an automated device, the Dinamap 8100 monitor. This device has been validated against the mercury sphygmomanometer and was found to give systolic readings and diastolic readings that are slightly higher in older age groups but lower in younger age groups.⁵

Using an appropriately sized cuff, three blood pressure readings were taken on the right arm with the informant in a seated position after five minutes rest. Blood pressure of pregnant women was not measured. The data presented here are based on the mean of the second and third readings from informants in whom all three recordings were completed.

Blood pressure readings were obtained from 6851 informants, which is 98% of all those visited by a nurse. However, informants who had eaten, drunk alcohol or smoked in the 30 minutes prior to the measurement were excluded from the analysis (14% of those measured), as were informants from whom less than three readings were obtained (1%). Thus, the analysis includes valid blood pressure readings from 5768 informants (2580 men and 3188 women). Table 6.1

Of those seen by a nurse, women aged 16-44 years had a slightly lower response rate than men due to the exclusion of pregnant women; women aged 45-64 years had a slightly higher response rate than men of that age, as fewer women had eaten, drunk or smoked prior to the measurement (a full discussion of response rates may be found in the Technical Report).

Among the informants who had three valid blood pressure readings, 528 were taking one or more anti-hypertensive agents which may affect blood pressure readings. Of these, 40% were on diuretics, 45% were on beta-blockers, 17% were taking ACE inhibitors and 30% were on calcium antagonists. (It should be noted that these categories of medicine are not mutually exclusive.) In line with the reports on the Health Survey for England, informants who were taking these medications were included in the analysis (unless otherwise stated). Table 6.2

6.3 classification of blood pressure levels

Informants were classified into one of four groups on the basis of their systolic (SBP) and diastolic (DBP) blood pressure and current use of anti-hypertensive medications, as follows:

The term high blood pressure is used to refer to those who were normotensive treated, hypertensive treated or hypertensive untreated.

6.4 blood pressure by age and sex

6.4.1 Systolic blood pressure (SBP)

Mean SBP for all informants was 128 mmHg. Mean SBP was 131 mmHg for men and 124 mmHg for women.

Mean SBP increased continuously with age in men, from 126 mmHg in the 16-24 age group to 141 mmHg in those aged 55-64 years. In women, mean SBP remained fairly constant until age group 25-34 after which it increased continuously with age: mean SBP rose from 117 mmHg in women aged 16-24 years to 139 mmHg in those aged 55-64 years. Men had higher SBP than women in each of the five age groups.

Systolic blood pressure was approximately normally distributed in men but was slightly skewed to the right in women. The variability of SBP increased with age in men and women after the 35-44 age group. Table 6.3, Figures 6A, 6B

Figure 6A: Distribution of systolic blood pressure (mmHg), by sex

Figure 6B: Systolic blood pressure (SBP), by age and sex

6.4.2 Diastolic blood pressure (DBP)

Mean DBP was 71 mmHg for all informants. It was 73 mmHg in men and 68 mmHg in women.

Among men, mean DBP rose with age until the 45-54 age group and then remained unchanged. It increased from 61 mmHg in those aged 16-24 years to 80 mmHg in those aged 45-54 and 55-64 years. Among women, mean DBP increased continuously with age, from 62 mmHg in those aged 16-24 years to 74 mmHg in those aged 55-64 years. Men had higher mean DBP than women in all five age groups.

Diastolic blood pressure was normally distributed in men and women. The variability of DBP remained fairly constant with age in men but increased slightly with age in women. Table 6.4, Figures 6C, 6D

Figure 6C: Distribution of diastolic blood pressure (DBP), by sex.

Figure 6D: Diastolic blood pressure (DBP), by age and sex

6.4.3 Pulse pressure

Pulse pressure is the difference between SBP and DBP. Among men, mean pulse pressure decreased continuously between the 16-24 (64 mmHg) and 45-54 (54 mmHg) age groups, and then increased for men aged 55-64 years (61 mmHg).

Among women, mean pulse pressure decreased between the 16-24 and 25-34 age groups, and thereafter increased with age, with women aged 55-64 years having the highest mean pulse pressure (65 mmHg). This is consistent with the results that, after age 25-34 years, mean SBP in women increased more than DPB. Table 6.5

6.4.4 Mean arterial pressure (MAP)

Mean arterial pressure is defined as the sum of the level of DBP plus one third of the difference between SBP and DBP. MAP increased with age in both sexes. For men, MAP rose from 87 mmHg in those aged 16-24 years to 105 mmHg in those aged 55-64 years. For women, MAP rose from 84 mmHg in those aged 16-24 years to 100 mmHg in those aged 55-64 years. Men had a higher MAP than women across all five age groups. Table 6.6, Figure 6E

6.4.5 High blood pressure

The survey definition of high blood pressure consists of three categories, namely normotensive treated, hypertensive treated and hypertensive untreated. The prevalence of high blood pressure was 11.9% among all informants (11.0% among men and 12.9% among women). The prevalence increased with age in both sexes: in men, from 0.4% in those aged 16-24 years to 33.6% in those aged 55-64 years; in women, it increased from 0.2% in those aged 16-24 years, to 40.6% in those aged 55-64 years.

The prevalence of high blood pressure was very similar among men and women for all age groups but the oldest (55-64), with normotensive treated being more common in women and hypertensive untreated being more common in men (see also Section 6.4.6). For those aged 55-64 years, high blood pressure was more common among women than men, with the main difference being more women in the normotensive treated category than men. Table 6.7, Figure 6F

Figure 6E: Mean arterial pressure (MAP), by age and sex

Figure 6F: Prevalence of high blood pressure, by age and sex

6.4.6 Detection, treatment and control of high blood pressure

While there is general recognition of the importance of reducing blood pressure, the levels of detection, treatment and control of high blood pressure are unsatisfactory in many countries.^6,7,8, In this report, the ‘detection rate’ for high blood pressure was estimated as the proportion of those with survey-defined high blood pressure who also reported a history of hypertension. ‘Treatment rate’ was defined as the proportion of all those with high blood pressure who were currently taking anti-hypertensive medications; and ‘control rate’ was defined as the proportion of those taking anti-hypertensive medication who had a SBP < 160 mmHg and DBP < 95 mmHg. It should be noted that the number of informants with high blood pressure was small in both men and women (particularly those aged under 45), so the results should be interpreted with caution.

The detection rate was 60% for men and women. Informants aged 45-64 years had a higher detection rate than those aged 16-44 in both sexes. Table 6.8

Among those with high blood pressure, the treatment rate was about 68% for all informants. It was lower among men than women (60% vs. 74%). This sex difference was greater among those aged under 45 years, possibly because younger women tend to have more contact with medical services than men of a similar age (because of pregnancy or contraception-related consultations). Tables 6.9

Among those with high blood pressure who were treated, the control rate was 83% for all informants, 77% for men and 88% for women. The control rate was lower in men than women in both the 16-44 (based on small numbers) and 45-64 age groups. Tables 6.10

6.5 BLOOD PRESSURE AND REGION

6.5.1 Region and mean systolic and diastolic blood pressure

Table 6.11 shows mean SBP and mean DBP by region. Regional differences in mean SBP were very small. Among men, mean SBP was either 131 or 132 mmHg, except for the Highland & Islands where it was 130 mmHg. Among women, mean SBP varied between 122 mmHg and 126 mmHg across regions. (Note that the estimates in Table 6.11 for some of the age-sex groups within regions are based on small sample sizes.) Table 6.11

After adjustment for age using linear regression, significant variations in mean SBP between regions were found among women (p=0.008) but not among men (p=0.82). Among women, mean SBP was lowest in the Forth Valley, Argyll & Clyde region, being 1.9 mmHg lower than the overall mean; it was highest in the two regions of Borders, Dumfries & Galloway and Lanarkshire, Ayrshire & Arran, being 1.7 mmHg higher than the overall mean. Table 6.12

Similarly, the variations in unadjusted mean DBP were very small across the regions in both men (ranging from 72 mmHg to 74 mmHg) and women (ranging from 66 mmHg to 70 mmHg).

After adjustment for age using linear regression, significant variations in mean DBP between regions were found among women (p<0.001) but not among men (p=0.087). Among women, mean DBP was lowest in the Forth Valley, Argyll & Clyde region, being 2.3 mmHg lower than the overall mean. Mean DBP in other regions did not differ significantly from the overall mean.

6.5.2 Region and high blood pressure

The prevalence of high blood pressure by region, age and sex is shown in Tables 6.13 and 6.14. Age-standardised prevalence varied from 7.4% in the Highland & Islands and Lothian & Fife to 12.6% in Greater Glasgow among men; and from 8.1% in Lothian & Fife to 15.8% in Borders, Dumfries & Galloway among women. Table 6.13,Table 6.14

After adjustment for age using logistic regression analysis, there was no statistically significant difference in the prevalence of high blood pressure between regions in either men or women (p=0.25 for men and p=0.46 for women).

6.6 BLOOD PRESSURE and social class

Coronary heart disease and stroke in Scotland, as in many other countries, have shown significant social class variations in both sexes, with the more socio-economically disadvantaged being at higher risk.^9,10,11,12, Some of these differences may be due to social class differentials in known CVD risk factors such as blood pressure which, in turn, may be attributed to social class differentials in obesity.¹³

This section examines variations in mean SBP, mean DBP and the prevalence of high blood pressure by social class. Social class is defined as the occupation of the chief income earner within the informant’s household (see the Glossary - Appendix E in the Technical Report - for details).

6.6.1 Systolic and diastolic blood pressure and social class

For both sexes, social class variations in mean SBP were small, ranging from 131 mmHg to 132 mmHg in men, and from 124 mmHg to 127 mmHg in women.

Similarly the social class variations in mean DBP were very small, ranging from 72 mmHg to 74 mmHg in men, and from 68 mmHg to 70 mmHg in women.

After adjustment for age using linear regression, no statistically significant association between mean blood pressure (systolic and diastolic) and social class was found (p=0.89 in men and p=0.48 in women for SBP, and p=0.08 in men and p=0.86 in women for DBP). Table 6.15, Table 6.16

6.6.2 High blood pressure and social class

The prevalence of high blood pressure by social class, age and sex is shown in Tables 6.17 and 6.18. Among men, age-standardised prevalence varied from 6.7% in Social Classes I to 10.5% in Social Class IIIM, with no clear gradient between manual and non-manual social classes. Among women, the age-standardised prevalence varied from 7.9% in Social Class II to 14.2% in Social Class IV, with manual social classes showing a higher prevalence of high blood pressure than non-manual social classes. Table 6.17, Table 6.18

After adjustment for age using logistic regression, social class difference in the prevalence of high blood pressure was significant in women (p=0.019) but not in men (p=0.18). Women in manual social classes, particularly Social Classes IIIM and IV, had a higher prevalence than those in non-manual social classes.

6.7 BLOOD PRESSURE and behavioural characteristics

A number of behavioural characteristics which are common in Scotland - such as excessive alcohol consumption, being overweight, and low levels of physical activity - are known to have adverse effects on blood pressure. This section examines the association between blood pressure and some of these behavioural factors.

It should be borne in mind that, given the cross-sectional nature of the Health Survey, the direction of an association cannot be presumed. This is especially true since it is likely that some informants may have altered their lifestyle because of a previous diagnosis of hypertension. The 528 informants (192 men and 336 women) who were taking anti-hypertensive agents were included in the following analysis, which might therefore underestimate any existing association.

The unadjusted mean blood pressures according to the level of each factor is shown in Tables 6.19-6.22. In order to examine the independent associations of these risk factors with blood pressures adjusting for age, linear regressions including all factors and age were used.

Obesity

Among both sexes, mean SBP and mean DBP increased as level of BMI increased for men and women of all ages. Table 6.19

Cigarette smoking

Among men and women of all ages, unadjusted mean SBP and mean DBP tended to be higher among ex-smokers and informants smoking at least 20 cigarettes per day, although the differences were small. Part of the reason for a higher blood pressure among ex-smokers may be because they were on average older, or they had stopped smoking as a result of a diagnosis of hypertension. Table 6.20

Alcohol consumption

Among men, unadjusted mean SBP and mean DBP were highest among those who drank over 21 units per week, and tended to be lower among ex-drinkers. Among women, unadjusted mean SBP and mean DBP were highest among ex-drinkers. There was very little variation in blood pressure among other categories of alcohol consumption level. (It should be noted that, as the number of ex-drinkers was small, the data should be interpreted with caution.) Table 6.21

Physical activity

For both men and women, mean SBP and mean DBP were negatively associated with levels of physical activity in most age groups. (See Chapter 2 for a description of how levels of physical activity were derived.) Table 6.22

In a linear regression analysis for SBP, adjusted for age, BMI, cigarette smoking, alcohol consumption and physical activity, significant association was only found between alcohol consumption and SBP in men: compared with non- or occasional drinkers, those who drank over 21 units of alcohol a week had significantly higher (2.4 mmHg) SBP, while ex-drinkers had significantly lower (5.5 mmHg) SBP. For DBP, significant associations were only found in men between alcohol consumption, physical activity level and DBP. Compared with non- or occasional drinkers, those who drank over 21 units of alcohol a week had significantly higher (1.8 mmHg) DBP, while ex-drinkers had significantly lower (3.8 mmHg) DBP. Those who took exercise more than twice a week (levels 3-5) had DBP 1.4 mmHg lower than those who did not take at least this amount of exercise. These results remained unchanged after further adjustment was made for the use of anti-hypertensive agents. Table 6.23

6.8 Comparison with the Scottish Heart health study

This section compares the blood pressure results from the Health Survey with similar data from the 1984-86 Scottish Heart Health Study (SHHS) in order to estimate changes in blood pressure over the last decade. Comparisons have been restricted to those aged between 40-59 years only, as this was the age range included in SHHS.

Because of a number of important differences in survey methodology between the Health Survey and SHHS (see Chapter 1), the results presented below should be interpreted with caution. While it is beyond the scope of this chapter to examine in detail all the possible reasons why there may be differences in blood pressure between the Health Survey and SHHS, some of the likely causes are discussed in Section 6.8.3.

6.8.1 Comparison of systolic and diastolic blood pressure

Among men, mean SBP was 134 mmHg in both surveys. Among women, mean SBP was lower in the Health Survey at 129 mmHg than in SHHS where it was 131 mmHg.

Mean DBP was lower in the Health Survey than in SHHS for both sexes and across all age groups. Among men, mean DBP was 84 mmHg in SHHS and 79 mmHg in the Health Survey. The corresponding figures for women were 81 mmHg and 72 mmHg respectively.

6.8.2 Comparison of the prevalence of raised blood pressure

The proportion of people with DBP over 100 mmHg (as reported in SHHS) showed a marked decline between SHHS and the Health Survey a decade later: for men it fell from 10% in SHHS to 3% in the Health Survey, and for women it fell from 7% in SHHS to 1% in the Health Survey. Table 6.24

6.8.3 Discussion

Changes in blood pressure results between SHHS and the Health Survey may be at least partly explained by differences in study design. For example, an important source of variation in measuring blood pressure is the type of equipment used, and the two studies used different equipment for taking their measurements. While SHHS used the Hawksley random zero sphygmomanometer, the Health Survey used the DINAMAP 8100 monitor. Evidence suggests that the Dinamap tends to give higher SBP readings (of about 6-8 mmHg) and lower DBP readings (of 1 mmHg and possibly 3-9 mmHg) than the random zero sphygmomanometer.^14,15,16,,

In other words, if the random zero sphygmomanometer had been used on the Health Survey, results of mean SBP would have been lower and mean DBP would have been higher than those obtained using the DINAMAP monitor. Therefore, it seems likely that some of the observed difference, particularly in mean DBP, between these two surveys is explained by the different types of equipment used.

The studies also differed in the way in which blood pressure was recorded and presented. In SHHS, results were based on the average of the only two blood pressure readings taken. In the Health Survey, on the other hand, the results were based on the mean of the final two of the three readings taken, a method which gives physiologically more representative results, as blood pressure tends to decrease between the first and subsequent readings.

In order to address this difference, data from the Health Survey were re-computed based on the mean of the first two readings taken. This resulted in slightly higher means for both SBP (+2 mmHg) and DBP (+1 mmHg). Comparing these revised results for the Health Survey with SHHS data, it can be seen that, in men mean SBP was slightly higher in the Health Survey than in SHHS (136 versus 134 mmHg), while in women mean SBP was 131 mmHg in both surveys. Using the revised mean DBP, it was still lower in the Health Survey than in SHHS for both sexes (80 versus 84 mmHg in men, 73 versus 81 mmHg in women respectively). The difference in the proportion of men and women with DBP >100 mmHg remained much lower in the Health Survey than in SHHS.

It may be that the results in Scotland mirror those in England, where blood pressure has shown an apparent decline between 1991 and 1995.¹⁷ Although the reasons for this are not well understood, it is thought that better diagnosis and treatment of hypertension may have partly contributed to this improvement. It is possible that there has been a similar reduction in blood pressure in Scotland in recent years, although this cannot be confirmed until data from future Health Surveys are available.

6.9 BLOOD PRESSURE in Scotland and england

This section compares blood pressure results in Scotland with those from the 1994 Health Survey for England. Comparisons are with England as a whole as well as with Northern England, which includes the two regional health authorities of Northern & Yorkshire and North West.

The Health Surveys for Scotland and England were similar in design, and blood pressure data were collected using exactly the same equipment and measurement protocol. Moreover, nurses working in Scotland and England were trained by the same personnel. Therefore, it is very unlikely that blood pressure measurements would systematically differ between these two surveys.

Response rates to the blood pressure measurements were also very similar between the two countries, although a higher proportion of informants in Scotland than in England smoked, ate or drank immediately before their blood pressure was measured. These informants from both surveys have been excluded from the results described below. (Further analysis showed that their exclusion makes very little difference to the results. This was expected, as the number of informants who smoked half an hour prior to the blood pressure measurement was relatively small and would not therefore be expected to have a significant impact on the level of mean blood pressure in the survey.)

6.9.1 Comparison of systolic and diastolic blood pressure

Mean SBP was lower in Scotland than in England across all age groups for both men (131 vs. 135 mmHg) and women (124 vs. 128 mmHg). Among all informants, mean SBP was 128 mmHg in Scotland compared with 131 mmHg in England. The difference between the countries in mean SBP was slightly larger than the difference between them in median SBP, which suggests that there were more people with very high SBP in England. Table 6.25, Figure 6G

After adjustment for age using linear regression, mean SBP was 3.0 mmHg lower among men, and 4.2 mmHg lower among women, in Scotland than in England (p<0.001 for both men and women). There was a significant interaction of the difference in mean SBP with treatment (i.e., taking any anti-hypertensive agents) in women, suggesting that the difference was not consistent between treated and untreated groups, with the greater difference found among those who were treated.

Figure 6G: Mean systolic blood pressure for Scotland and England, by age and sex

Similarly, mean DBP was lower in Scotland than in England across all age groups for both men (73 vs. 75 mmHg) and women (68 vs. 71 mmHg). Among all informants, mean DBP was 71 mmHg in Scotland compared with 73 mmHg in England. The difference between the two countries in median DBP was similar to that in mean DBP. Table 6.26

After adjustment for age using linear regression, mean DBP was 1.7 mmHg lower among men, and 2.7 mmHg lower among women, in Scotland than in England (p<0.001 for both men and women). Again, there was an interaction of the difference in mean DBP with treatment (i.e., taking any anti-hypertensive agents) in women, suggesting a greater difference among those who were treated.

Comparing mean SBP and mean DBP between Scotland and Northern England showed even greater differences than the comparison with England as a whole.

6.9.2 Comparison of high blood pressure

Compared with England, the prevalence of high blood pressure (i.e., being normotensive treated, hypertensive treated or hypertensive untreated) among men was slightly lower in Scotland in most age groups, with a prevalence of 11.0% for men in Scotland and 12.9% for men in England.

Among women, in contrast to the significantly lower mean SBP and mean DBP in Scotland, the prevalence of high blood pressure (i.e., being normotensive treated, hypertensive treated or hypertensive untreated) was greater in Scotland than in England for women in three of the five age groups (35-44, 45-54 and 55-64). For all women informants, the prevalence of high blood pressure was 12.9% in Scotland and 11.2% in England. After adjustment for age using logistic regression, the prevalence of high blood pressure was significantly higher for women in Scotland than in England (p=0.02), but there was no difference for men (p=0.08).

Given that women in Scotland have lower mean blood pressure (both SBP and DBP) than women in England, it is somewhat surprising to find that women in Scotland are more likely than those in England to have high blood pressure. This could be explained by differences between the two countries in the normotensive treated and hypertensive treated categories. For example, looking at informants aged 45-64 (who were more likely to have high blood pressure than younger informants and so give more stable estimates), treatment was more common in Scotland than in England (64% versus 52% in men, and 74% versus 63% in women). In addition, among those being treated, blood pressure was more likely to be controlled (SBP <160 and DBP <95) in Scotland than in England: in Scotland, the control rates were 75% in men and 86% in women, while in England, they were 71% and 74% respectively.

These results were supported by further analysis which looked at the proportion of informants with raised blood pressure (SBP ³ 160 mmHg or DBP ³ 95 mmHg), regardless of treatment (i.e., with hypertensive treated and hypertensive untreated). This proportion was significantly lower in Scotland than in England for both men and women, adjusted for age. (Table not shown.)

The pattern was similar between Scotland and Northern England: i.e., compared with Northern England, the prevalence of high blood pressure in Scotland was lower in men and higher in women, although the magnitude of the differences were greater than that between Scotland and England as a whole. Table 6.27

6.9.3 Discussion

Given that Scotland has higher rates of cardiovascular disease mortality than England, it is somewhat surprising to find that Scotland had significantly lower blood pressure than England. The Health Survey findings are supported by evidence from the national Dietary and Nutritional Survey of British Adults (NDNS). This study, which was based on a nationally representative sample of 16-64 year olds, also reported lower systolic and diastolic blood pressure for people (especially men) living in Scotland than in England.¹⁸ As the NDNS results excluded people who were taking anti-hypertensive agents, the Health Survey data was re-analysed in order to compare mean blood pressure between Scotland and England after excluding those informants on treatment. The revised results tended to be 1 mmHg lower for both SBP and DBP for most age/sex groups in both Scotland and England. Mean blood pressure remained lower in Scotland than in England.

Finally, it should be noted that the Health Survey only included people aged 16-64 years, and that simple extrapolation of the survey results to the entire adult population would not necessarily be valid; further data would be needed to determine if this pattern applied to adults aged 65 and over.

[Previous] [Contents] [Next]