Factors Associated With Treatment and Control of Hypertension in a Healthy Elderly Population Free of Cardiovascular Disease: A Cross-sectional Study

Abstract

BACKGROUND

Despite readily available treatments, control of blood pressure (BP) with population aging remains suboptimal. Further, there are gaps in the understanding of the management of high BP in the aged. We explored antihypertensive treatment and control among elderly hypertensive participants free from overt cardiovascular disease (CVD), and identified factors related to both “untreated” and “treated but uncontrolled” high BP.

METHODS

We analyzed baseline data from 19,114 individuals aged ≥65 years enrolled from Australia and United States (US) in the ASPirin in Reducing Events in the Elderly study. Hypertension was defined as an average systolic/diastolic BP ≥140/90 mm Hg and/or the use of any BP lowering medication. “Controlled hypertension” was defined if participants were receiving antihypertensive medication and BP <140 and 90 mm Hg. Descriptive analyses were used to summarize hypertension control rates; logistic regression was used to investigate relationships with treatment and BP control.

RESULTS

Overall, 74% (14,213/19,114) of participants were hypertensive; and of these 29% (4,151/14,213) were untreated. Among those treated participants, 53% (5,330/10,062) had BP ≥140/90 mm Hg. Participants who were untreated were more likely to be men, have higher educational status, and be in good physical health, and less likely to have significant comorbidities. The factors related to “treated but uncontrolled” BP included older age, male, Black race (vs. White), using antihypertensive monotherapy (vs. multiple) and residing in Australia (vs. US).

CONCLUSIONS

High levels of “untreated” and “treated but uncontrolled” BP occur in healthy elderly people without CVD, suggesting there are opportunities for better BP control in the primary prevention of CVD in this population.

CLINICAL TRIALS REGISTRATION

NCT01038583.

Cardiovascular disease (CVD) is the leading global cause of mortality, responsible for approximately 17.6 million deaths worldwide in 2016.¹ Between 2006 and 2016, deaths from CVD increased by 14.5% globally, despite a decrease in the age-standardized CVD mortality rate. Majority of these deaths occur in those aged 70 years and older.¹ Between 1990 and 2016, there was a 54% increase in total number of deaths due to CVD in this age group.¹

Elevated blood pressure (BP) is one of the major risk factors for CVD. The prevalence of hypertension increases with age.^2,3 In high-income countries such as the United States (US) and Australia, the elderly population carries the greatest burden of hypertension. Among Australians aged “65–74 years” and “75 years and above,” 44% and 47%, respectively, who were treated with BP lowering medication, had uncontrolled hypertension in 2014–15.⁴ In the US, 32% of the hypertensive population aged 65 years and above, treated with BP lowering medication, had uncontrolled BP in 2013–14.⁵

The management of hypertension in the elderly has many challenges, including agreement on threshold and target BP levels, and the balancing of adverse effects and potential benefits of treatment.^6,7 While extensive studies have been undertaken to identify risk factors for high BP and CVD in predominantly middle-aged populations, gaps in our understanding of risk profiles and management of BP among older, generally “healthy” populations, still exist. A better understanding of the factors impacting uncontrolled hypertension in older adults is crucial to the development of interventions to manage high BP in this growing sector of the population.

The ASPirin in Reducing Events in the Elderly (ASPREE) study was a randomized placebo-controlled trial of low-dose aspirin in a cohort of 19,114 healthy Australian and US individuals mainly aged ≥70 years (US African-Americans and Hispanics were aged ≥65 years). Healthy was defined as the absence of disability and any previous history of CVD, dementia, or serious intercurrent illness expected to limit life expectancy to <5 years.⁸ As this study cohort was identified as representative of healthy elders in both countries,⁹ we sought to analyze the prevalence and control of hypertension among study participants. Additionally, we aimed to identify factors associated with: (i) not receiving antihypertensive treatment despite being hypertensive; and (ii) having uncontrolled hypertension despite being treated with antihypertensives.

METHODS

Study population

Details of the ASPREE study and enrolled cohort have been extensively described.^8,10 Briefly, ASPREE was a large-scale, placebo-controlled clinical trial in men and women aged 70 years or more (65+ for US minorities) designed to determine the balance of benefit and harm of daily low-dose aspirin as a primary preventive agent. The study participants were recruited between 2010 and 2014. Two baseline visits were conducted to determine the participants’ study eligibility. Recruitment in Australia (n = 16,703) was mostly done through general practice. In the US (n = 2,411) it was through community-based academic and clinical trial centers with an enriched sampling of minorities. ASPREE had multiple Institutional Review Board approvals in the US and Australia.

Study measurements

Participants were asked during two baseline visits (initial, and after 4-week placebo run-in) about their medical history/diagnoses and were examined for neurocognitive, physical function, lifestyle, anthropometric, and other clinical data. In addition, data on medication use were collected from self-reported medication review with the participant and through examination of their clinical records. Baseline BP for each participant was measured prior to randomization to aspirin or placebo treatment at the initial baseline visit at least three times by trained study staff, according to standard operating procedures.⁸ Individual BP was measured in the seated position after at least 5 min rest, with each reading 1 min apart, using an automatic sphygmomanometer with an occluding cuff of appropriate size for the participant’s measured upper arm. The average of all three BP measurements from the first baseline visit for each participant was used as the BP for baseline. Further, we classified the participants as having hypercholesterolemia (if receipt of cholesterol-lowering medication and/or as a serum total cholesterol level of at least 212 mg/dl (≥5.5 mmol/l) in Australia and at least 240 mg/dl (≥6.2 mmol/l) in the US), diabetes (if a fasting glucose level of at least 126 mg/dl (≥7 mmol/l) and/or receipt of treatment for diabetes), reduced renal function (if undergoing renal dialysis and/or estimated Glomerular Filtration Rate (eGFR) <60 ml/min/1.73 m²), and/or having depression (based on antidepressant drug use and/or participant history). We also classified participants grip strength (average of three readings from the dominant hand) into three categories: “normal” if male ≥31.83 kg and female ≥19.99 kg; “intermediate” if male: ≥25.99 and <31.83 kg and female: ≥15.92 kg and <19.99 kg; and “weak” if male <25.99 kg and female <15.92 kg.¹¹

Hypertension and BP control definition

Hypertension was defined as a mean systolic BP (SBP) ≥140 mm Hg and/or a mean diastolic BP (DBP) ≥90 mm Hg, and/or use of any medication to lower BP. Antihypertensive medications were classified according to the World Health Organization Anatomical Therapeutic Chemical Classification System (i.e., Antihypertensives (C02), Diuretics (C03), Peripheral vasodilators (C04), Beta blocking agents (C07), Calcium channel blockers (C08), and Agents acting on the renin–angiotensin system (C09)). We did not use the recent 2017 American Heart Association guideline suggested BP cut-off point (≥130/80 mm Hg), nor the 2014 Eighth Joint National Committee report BP cut-off (≥150/90 mm Hg), to define hypertension as these guidelines did not exist at the time of participant recruitment into ASPREE, and most national level data available for comparison used a 140/90 mm Hg cut-point.^12,13 An individual was identified as “BP controlled” if on antihypertensive medication and SBP <140 mm Hg and DBP <90 mm Hg.

Statistical analyses

Our analyses were restricted to individuals in the cohort who met criteria for hypertension (Figure 1 flow diagram). We used descriptive analyses (frequency/sample proportions or mean with estimated standard deviations) to summarize the distribution of sociodemographic and clinical characteristics of the hypertensive participants by use of antihypertensive status as well as by BP control status. Student’s t-test, analysis of variance, or Chi-square tests were used for unadjusted comparisons of the distributions of characteristics or risk factors within different participant subgroups. Logistic regression models were applied to investigate factors associated with (i) hypertensive, but untreated, and (ii) treated, but uncontrolled hypertension. All analyses were adjusted for the clustering effects of patients within general practices/trial centers. For this, we first explored the univariate association of the variables in Table 1 using regression models with the outcome of interest (treated vs. untreated, controlled vs. uncontrolled). Thereafter, variables with a P-value of <0.10 for their univariate association were included in a multiple logistic regression model to identify the factors that were strongly associated with untreated hypertension and uncontrolled hypertension. Further, all these logistic regression analyses were repeated only among participants aged 70 years and older considering the participants aged <70 years were predominantly representing minorities from US. All statistical analyses were performed using Stata version 15.1 for Windows (StataCorp LP, College Station, TX).

Figure 1.

Flow diagram of ASPREE participants’ hypertension treatment and control. Abbreviations: ASPREE, ASPirin in Reducing Events in the Elderly; DBP, diastolic blood pressure; SBP, systolic blood pressure.

Table 1.

Baseline characteristics of the hypertensive participants by treatment and BP control

	Treatment status N = 14,213		BP control status (among treated) N=10,062
	Untreated	Treated	Uncontrolled	Controlled
N	4,151	10,062	5,330	4,732
Age (years, mean ± SD)	74.6 ± 4.5	74.9 ± 4.7*	75.2 ± 4.8	74.5 ± 4.6¥
Age category		*		¥
65–69	79 (1.9%)	350 (3.5%)	142 (2.7%)	208 (4.4%)
70–74	2,344 (56.5%)	5,262 (52.3%)	2,693 (50.5	2,569 (54.3%)
75–79	1,082 (26.1%)	2,741 (27.2%)	1,506 (28.3	1,235 (26.1%)
80–85	491 (11.8%)	1,255 (12.5%)	710 (13.3%)	545 (11.5%)
85 and above	155 (3.7%)	454 (4.5%)	279 (5.2%)	175 (3.7%)
Gender
Men	2,185 (52.6%)	4,088 (40.6%)*	2,360 (44.3%)	1,728 (36.5%)¥
Women	1,966 (47.4%)	5,974 (59.4%)*	2,970 (55.7%)	3,004 (63.5%)¥
Racial category		*
White	3,901 (94.0%)	9,025 (89.7%)	4,836 (90.7%)	4,189 (88.5%)
Black	105 (2.5%)	609 (6.0%)	268 (5.0%)	341 (7.2%)
Hispanic	82 (2.0%)	275 (2.7%)	139 (2.6%)	136 (2.9%)
Asian	39 (0.9%)	89 (0.9%)	52 (1.0%)	37 (0.8%)
Other	24 (0.6%)	64 (0.6%)	35 (0.7%)	29 (0.6%)
Years of education		*
<12	1,799 (43.4%)	4,823 (47.9%)	2,598 (48.7%)	2,225 (47.0%)
12–15	1,228 (29.6%)	2,926 (29.1%)	1,524 (28.6%)	1,402 (29.6%)
16+	1,123 (27.1%)	2,313 (23.0%)	1,208 (22.7%)	1,105 (23.4%)
Born overseas	1,052 (25.3%)	2,164 (21.5%)*	1,226 (23.0%)	938 (19.8%)¥
English first language	3,962 (95.4%)	9,607 (95.5%)	5,063 (95.0%)	4,544 (96.0%)¥
Current smoker	173 (4.2%)	359 (3.6%)	179 (3.4%)	180 (3.8%)
Current drinker	3,322 (80.0%)	7,482 (74.4%)*	4,000 (75.0%)	3,482 (73.6%)
Obese (BMI ≥30 kg/m2)	950 (23.0%)	3,796 (37.9%)*	1,959 (36.9%)	1,837 (39.0%)¥
Physical state
Grip strength category€		*
Normal	2,611 (63.9%)	5,771 (58.1%)	3,078 (58.4%)	2,693 (57.8%)
Intermediate	5,771 (22.9%)	2,389 (24.1%)	1,277 (24.2%)	1,112 (23.9%)
Weak	542 (13.2%)	1,766 (17.8%)	913 (17.3%)	853 (18.3%)
Any walk outside home in past 2 weeks	3,998 (96.4%)	9,499 (94.4%)*	5,046 (94.7%)	4,453 (94.1%)
Any difficulty in walking (distance of 1 mile/1.6 km)	616 (15.1%)	2,457 (25.1%)*	1,282 (24.7%)	1,175 (25.5%)
Comorbidity/risk factors
Family history CVD	2,413 (58.1%)	6,386 (63.5%)*	3,371 (63.2%)	3,015 (63.7%)
Hypercholesterolemia	2,586 (62.3%)	6,949 (69.1%)*	3,713 (69.7%)	3,236 (68.4%)
Diabetes	208 (5.0%)	1,246 (12.4%)*	637 (12.0%)	609 (12.9%)
Reduced renal function (eGFR <60 ml/min/1.73 m2)	556 (13.8%)	2,337 (23.7%)*	1,222 (23.4%)	1,115 (24.0%)
Depression	467 (11.3%)	1,499 (14.5%)*	724 (13.6%)	775 (16.4%)¥
Regular aspirin use	332 (8.0%)	1,290 (12.8%)*	659 (12.4%)	631 (13.3%)
Number of antihypertensive drug		*
None	4,151 (100%)	—	—	—
One	—	5,019 (49.9%)	2,727 (51.1%)	2,292 (48.4%)¥
Two	—	3,519 (35.0%)	1,800 (33.8%)	1,719 (36.3%)¥
Three or more	—	1,524 (15.1%)	803 (15.1%)	721 (15.2%)¥
Australian participants	3,785 (91.2%)	8,748 (86.9%)	4,765 (89.4%)	3,983 (84.2%)¥

Abbreviations: BMI, body mass index; BP, blood pressure; CVD, cardiovascular disease; SD, standard deviation.

Note: Percentages reported for each characteristic are based on the denominator for all hypertensives within that column; i.e., treated vs. untreated; uncontrolled vs. controlled.

*Significant difference (P < 0.05) observed between those treated and those not treated.

^¥Significant difference (P < 0.05) observed between those with uncontrolled and controlled BP among treated hypertensive participants.

^€Grip strength (average 3 readings, dominant hand) “Normal” if male ≥31.83 kg and female ≥19.99 kg; “intermediate” if male: ≥25.99 and <31.83 kg and female: ≥15.92 kg and <19.99 kg; and “weak” if male <25.99 kg and female <15.92 kg.

RESULTS

Overall, 74% (14,213/19,114) of the ASPREE study participants (Australia, 75% and US, 70%) met the criteria for hypertension (BP ≥140/90 mm Hg and/or on BP lowering medication) based on first baseline visit BP. The characteristics of the 19,114 ASPREE participants at baseline have been published previously.¹⁰ In brief, participants identified as hypertensive were older and more commonly men, had higher body mass index, higher comorbidities/risk factors such as a family history of CVD (either stroke or heart attack), hypercholesterolemia, diabetes, and reduced renal function, compared to those who were normotensive. In contrast, normotensive participants were more likely to be White/Caucasian, current drinkers or physically active compared to those hypertensives (see Supplementary Table S1 online).

Of those meeting criteria for hypertension, 71% (Australia 70% and US 78%) were receiving an antihypertensive at baseline, while 29% (4,151/14,213) of participants were not receiving antihypertensives but had a BP ≥140/90 mm Hg. Among individuals treated with at least one antihypertensive at baseline, 47% (4,732/10,062) (Australia 46% and US 57%) had controlled BP (SBP <140 and DBP <90 mm Hg).

Characteristics of the hypertensive population by treatment and BP control status

The baseline characteristics of all hypertensive participants and their distribution across several categories are summarized in Table 1. As expected, the mean BP (SBP/DBP) in those treated was lower than in untreated hypertensives (141 ± 17/77 ± 10 mm Hg vs. 152 ± 10/83 ± 9 mm Hg, respectively). Similar rates of treated vs. untreated were observed across quintiles of age groupings, with the exception of those aged 65–69 years (exclusively minorities from the US) where rates of treated were higher than untreated. Overall, treated participants were slightly older, had higher presence of other comorbidities/risk factors such as a family history of CVD (either stroke or heart attack), obesity, hypercholesterolemia, diabetes, reduced renal function, depression, and/or were taking aspirin prior to entry into the trial compared to those who were not treated. Those participants who were treated also had weaker grip strength and more commonly reported difficulty in walking a distance of one mile (i.e., 1.6 km). In contrast, higher rates of untreated (vs. treated) were observed for men, those born overseas, and/or current drinkers at baseline. Figure 2a shows the percentages of untreated hypertensives by gender for age and ethnicity groups. In general, the prevalence of untreated hypertensives was higher in men compared to women across all age groupings, except for the age group 65–69 years and 85+ years. A higher prevalence of untreated hypertension was observed among White and Black men participants than women.

Figure 2.

Distribution of “untreated” and “treated but uncontrolled” BP among ASPREE hypertensive participants by age and ethnicity. (Note: Percentage are based on the overall N within that specific category/grouping.) Abbreviations: ASPREE, ASPirin in Reducing Events in the Elderly; BP, blood pressure.

Among the participants treated for hypertension, the mean BP (SBP/DBP) in those uncontrolled was 153 ± 11/82 ± 10 mm Hg compared to 126 ± 9/72 ± 8 mm Hg for those who were controlled. The treated, but uncontrolled, hypertensives were slightly older, and more commonly men, born overseas, and less obese. Single antihypertensive use was more common in those hypertensives who were treated, but uncontrolled, compared to those who were controlled (Table 1). Unlike the treated vs. untreated comparisons, there was no significant difference in hypertension control by the presence of comorbidity and/or CVD risk factors. Results of further exploration of BP control status by gender for participant’s age and ethnicity are summarized in Figure 2b. The percentage of participants with uncontrolled BP increased with age in both men and women. Higher prevalence of uncontrolled BP was observed among men compared to women for the 70–74 year age group. The percentage of participants with treated, but uncontrolled BP between men and women were similar across ethnic groups, except for the White participants.

Factors associated with untreated hypertension

Univariate logistic regression analysis findings on the association of different baseline characteristics with untreated hypertension among the hypertensive participants are shown in Table 2. All the variables in univariate logistic regression analysis having a P-value of <0.10, were included in the multivariate analysis to identify the factors having association with untreated hypertension. Findings from the multivariate analysis adjusted for other variables suggest participants’ age was not associated with hypertension treatment (Table 3). Men were more likely to have untreated hypertension than women (odds ratio 1.40, 95% confidence interval 1.29–1.51, P < 0.001). Additionally, those with higher educational status or born overseas were also more likely to be untreated for hypertension. In contrast, Black participants, those being obese, the presence of comorbidities (e.g., diabetes, hypercholesterolemia, and depression), family history of CVD, or poor physical health (i.e., weak grip strength and difficulty in walking) were more likely to receive treatment for hypertension. Further, analysis restricted among participants aged ≥70 years resulted in similar findings (see Supplementary Table S2 online).

Table 2.

Association of different baseline characteristics with lack of treatment and uncontrolled BP status, findings from univariate logistic regression analyses

	Untreated status (vs. treatment) N = 14,213		Uncontrolled BP status (vs. control) N = 10,062
	OR (95% CI)	P-value	OR (95% CI)	P-value
Age category (years)
65–69	0.51 (0.39–0.65)	<0.001	0.65(0.52–0.81)	<0.001
70–74	1.00		1.00
75–79	0.89 (0.81–0.97)	0.01	1.16 (1.06–1.28)	0.001
80–85	0.88 (0.78–0.99)	0.03	1.24 (1.10–1.41)	0.001
85 and above	0.77 (0.63–0.93)	0.01	1.52 (1.25–1.85)	<0.001
Men (vs. women)	1.62 (1.51–1.75)	<0.001	1.38 (1.27–1.50)	<0.001
Racial category
White/Caucasian	1.00		1.00
Black/African American	0.40 (0.32–0.49)	<0.001	0.68 (0.58–0.80)	<0.001
Hispanic	0.69 (0.54–0.89)	0.004	0.89 (0.70–1.13)	0.32
Asian	1.01 (0.69–1.48)	0.94	1.22 (0.80–1.86)	0.36
Other	0.87 (0.54–1.39)	0.55	1.05 (0.64–1.71)	0.86
Years of education
<12	1.00		1.00
12–15	1.13 (1.03–1.23)	0.01	0.93 (0.85–1.02)	0.13
16+	1.30 (1.19–1.42)	<0.001	0.94 (0.85–1.03)	0.19
Born overseas	1.24 (1.14–1.35)	<0.001	1.21 (1.10–1.33)	<0.001
English first language	0.99 (0.83–1.18)	0.94	0.78 (0.65–0.95)	0.01
Current smoker	1.18 (0.98–1.41)	0.087	0.88 (0.71–1.08)	0.23
Current drinker	1.38 (1.26–1.51)	<0.001	1.08 (0.99–1.18)	0.09
Obese (vs. not obese)	0.49 (0.45–0.53)	<0.001	0.92 (0.85–0.99)	0.04
Physical state
Grip strength category
Normal	1.00		1.00
Intermediate	0.86 (0.79–0.95)	0.001	1.00 (0.91–1.11)	0.92
Weak	0.68 (0.61–0.75)	<0.001	0.94 (0.84–1.04)	0.23
Any walk outside home in past 2 weeks	1.58 (1.32–1.91)	<0.001	1.11 (0.93–1.31)	0.25
Any difficulty in walking (a distance of 1 mile /1.6 km)	0.53 (0.48–0.58)	<0.001	0.96 (0.87–1.05)	0.37
Comorbidity/risk factors
Family history CVD	0.80 (0.74–0.86)	<0.001	0.98 (0.90–1.06)	0.63
Hypercholesterolemia	0.74 (0.69–0.80)	<0.001	1.06 (0.98–1.16)	0.17
Diabetes	0.37 (0.32–0.43)	<0.001	0.92 (0.82–1.03)	0.16
Reduced renal function (eGFR <60 vs. ≥60 ml/min/1.73 m2)	0.51 (0.47–0.57)	<0.001	0.97 (0.88–1.06)	0.50
Depression	0.72 (0.65–0.81)	<0.001	0.80 (0.72–0.90)	<0.001
Regular aspirin use	0.59 (0.52–0.67)	<0.001	0.92 (0.82–1.03)	0.15
Number of antihypertensive drug use
None	NA		NA
One	NA		1.00
Two	NA		0.88 (0.81–0.96)	0.004
Three or more	NA		0.94 (0.83–1.05)	0.26
Australian (vs. US)	1.55 (1.37–1.75)	<0.001	1.59 (1.41–1.78)	<0.001

Abbreviations: BP, blood pressure; CI, confidence interval; CVD, cardiovascular disease; NA, not applicable; OR, odds ratio; US, United States.

Table 3.

Association of different factors with receiving BP lowering treatment among ASPREE hypertensive participants (findings from multivariable logistic regression analysis)

Abbreviations: ASPREE, ASPirin in Reducing Events in the Elderly; BP, blood pressure; CI, confidence interval; CVD, cardiovascular disease; US, United States.

Factors associated with uncontrolled hypertension

Ten thousand and sixty-two hypertensive participants who were on antihypertensives at baseline were included in the multivariate analysis to investigate the factors associated with uncontrolled hypertension. All the variables in univariate logistic regression analysis (Table 2) having P-value of <0.10 for uncontrolled BP were included in the multivariate analysis. Among ASPREE hypertensive participants who were receiving treatment, uncontrolled hypertension was associated with older age and male gender (Table 4). Black participants were more likely to have uncontrolled baseline BP ≥140/90 mm Hg compared to Whites (odds ratio = 1.39, 95% confidence interval 1.03–1.87, P = 0.03). Better hypertension control was observed among those with English as a first language and/or receiving more than one antihypertensive. Further, US participants were more likely to have controlled BP compared to Australian participants. No significant association was observed between hypertension control and obesity, physical activity status, and presence of comorbidity (except depression, which was associated with better control). Restricting analyses to those participants aged ≥70 years resulted in similar findings (see Supplementary Table S3 online).

Table 4.

Association of different factors with BP control (<140 and <90 mm Hg) among treated ASPREE hypertensive participants (findings from multivariable logistic regression analysis)

Abbreviations: ASPREE, ASPirin in Reducing Events in the Elderly; BP, blood pressure; CI, confidence interval; US, United States.

DISCUSSION

Both Australia and the US have made substantive improvements in hypertension treatment indicators such as BP lowering medication use and BP control over the past two decades.^4,14,15 However, despite these improvements nearly one-third of the elderly hypertensive participants in our study were not treated with antihypertensives. Additionally, more than half of those hypertensive individuals who were treated, were uncontrolled. We found that hypertension control among the treated hypertensives differed by factors such as age, gender, ethnicity, number of medications used to treat hypertension, presence of depression, and participants’ country of residence (US/Australia). These findings highlight potential areas where health professionals and policy makers can focus efforts to improve BP control in otherwise healthy elderly hypertensive people.

We observed several characteristics associated with receiving treatment for hypertension, including gender, education level, ethnicity, place of birth (overseas or not), physical health, and presence of comorbidity. Interestingly, age itself was not associated with treatment, as we observed similar rates of treated/untreated across age strata, including in those aged 85 and older. Additionally, the use of antihypertensives was more common among hypertensive individuals with more comorbidities (i.e., diabetes, depression) and other CVD risk factors such as obesity, or a family history of CVD. Collectively, these findings suggest that physicians are weighing risk/benefit of antihypertensive treatment and individualizing the approach in the elderly. An alternate theory to explain the higher rates of no treatment in individuals with fewer comorbidity is that physicians may be wary of the consequences of initiating medication and quality of life by “medicalizing” an otherwise healthy person.¹⁶

In our study, gender differences were present both in treatment and control of hypertension, with a higher proportion of women in both countries receiving BP lowering medication. This finding is similar to that reported previously from the US and Australia among hypertensive populations in general, irrespective of age.^4,17 Among the treated hypertensive participants, men were less likely to have controlled BP compared to women. The underlying reasons could be that women are more likely to visit their healthcare provider and are more adherent to their prescription/medication compared to men.^18,19 Evidence from non-institutionalized population surveys from US, Canada, and UK participants suggest that the higher use of BP lowering medication among women may be also due to their higher awareness about the consequences of hypertension.²⁰

While participants’ age was not associated with receiving treatment for hypertension in our study, older age was correlated with uncontrolled hypertension among the treated hypertensive participants. Aging has been previously identified as one of the key reasons for poor BP control worldwide.³ The reasons for this are likely multifactorial, including specific age-related changes in blood vessel structure (i.e., increasing arterial stiffness and reduced compliance) which often necessitate use of multiple agents to control BP, controversy surrounding the optimal goal BP in older patients, and overall therapeutic inertia.²¹

In our study, Blacks were more likely to receive treatment for hypertension compared to the White participants. Evidence shows that Black are more likely to be diagnosed with hypertension and die from CVD compared to their White counterparts.²² This could be one of the underlying reasons for our observation of ethnic variation for BP lowering medication use in our study. Despite higher uptake of antihypertensive medication, “Black” participants were more likely to have uncontrolled hypertension than White participants. Similar to our findings, previous studies have shown Blacks and Hispanics are less likely to have controlled hypertension compared to Whites in the US.^14,23 The difference may be due to factors such as treatment adherence, underlying physiological differences, or participants’ lifestyle.^24,25 Our finding suggests that further research is necessary to explore the relationship between ethnicity and BP lowering prescription patterns and BP control.

We also observed differences in hypertension control by country. Hypertension control was better among the US participants compared to the Australian participants (57% vs. 46%). This observed difference could be due to differences in participant recruitment procedure as the Australian participants were general practice based whereas the US participants were from academic trial centers. There might be also differences in the physician approaches for identification and treating individuals with hypertension (i.e., country specific protocols/guidelines for managing hypertension). However, the overall hypertension control among the treated hypertensive participants in our study was lower than estimates of earlier national hypertension control rates in Australia and the US at comparable age ranges.^4,20

We observed that use of multiple antihypertensives was related to better hypertension control. Reliance on monotherapy has been reported as the single most important factor contributing to low rates of hypertension control.²⁶ Physician’s attitudes and assessing benefits for the management of hypertensive patients, especially for the elderly and those with comorbidities, might influence their prescribing decisions.²⁷ Furthermore, lack of consistency in guidelines regarding the management of hypertension in the elderly makes it difficult for physicians to incorporate the evidence-based recommendations in their daily clinical practice.^28,29 In addition to these, cost of medication or therapeutic inertia and patient preferences might influence prescriber’s decisions.³⁰

Our study has a number of limitations owing to its cross-sectional nature. Firstly, we used the conventional office BP measurement from a single visit (average of three readings taken 1 min apart) and/or self-reported use of BP lowering medications to classify participants’ hypertension status. While in most cases, treatment with BP lowering medication would correctly identify someone as having hypertension, in certain situations, use of the antihypertensive for another reason (e.g., alpha blocker for benign prostatic hyperplasia, or loop diuretic for edema), might have overestimated the true prevalence of hypertension. Defining hypertension status based on clinic BP measures, especially those who were not on antihypertensives, compared to defining hypertension status based on using ambulatory BP measure, may be subject to white-coat hypertension and lead to an overestimated prevalence. However, baseline BP measures in the ASPREE study were performed by trained research assistants according to standardized procedures which should reduce the “white-coat” effect. Secondly, due to introduction of Eighth Joint National Committee guideline in the US with a recommended treatment at BP cut-off of ≥150/90 during the study trial period, which is higher than the cut-off point used in this paper to define and treat hypertension, some participants might have been listed as “untreated” but actually did not require treatment as per the Eighth Joint National Committee guideline at the time of the trial. This might have led to overestimation of the “untreated” category. However, considering this guideline was unique to the US and there were far fewer ASPREE participants enrolled from the US and especially after publication of the Eighth Joint National Committee guideline (363 US ASPREE participants were enrolled), the potential for overestimating the “untreated” group is small. In this study, we did not have information regarding why people in “untreated” group did not receive treatment e.g. if they had prior episodes of hypotension, pharmacotherapy intolerance, and/or due to personal preferences. Lastly, we defined hypertension control status among the treated hypertensive participants based on their BP level at first baseline visit. We did not have information about drug adherence and duration of treatment with BP lowering, which might have affected hypertension control (BP <140/90 mm Hg).

This study extends our understanding of hypertension treatment and control in an ethnically and geographically diverse group of healthy older adults free of CVD. There are high levels of “untreated” and “treated, but uncontrolled” BP in these individuals, suggesting that opportunities for better BP control exist in this population.

FUNDING

The work was supported by the National Institute on Aging and the National Cancer Institute at the National Institutes of Health (grant number U01AG029824); the National Health and Medical Research Council (grant numbers 334047 and 1127060); Monash University (Australia); and the Victorian Cancer Agency (Australia).

DISCLOSURE

Bayer AG supplied study drug (aspirin) and matching placebo also produce BP lowering medication (no link with ASPREE’s participants) and had no other role in the trial. EKC has received High Blood Pressure Research Council Australia early career research transition grant to support current work. MRN received travel and consultancy support from Bayer to attend a meeting in Berlin. AMT has received unrelated research support and honoraria/travel expenses from Bayer. CM Reid is supported on a NHMRC Principal Research Fellowship (1136372). All other authors have no conflict of interest to declare in relation to this study.