Maintenance of normal blood pressure from middle to older age: Results from the Atherosclerosis Risk in Communities Study

Abstract

Background:

It is unknown whether maintaining normal blood pressure (BP) from middle into older age is associated with improved health outcomes.

Methods:

We estimated the proportion of Atherosclerosis Risk in Communities study participants who maintained normal BP from 1987–1989 (visit 1) through 1996–1998 and 2011–2013 (over 4 and 5 visits, respectively). Normal BP was defined as systolic BP<120 mmHg and diastolic BP<80 mmHg, without antihypertensive medication. We estimated the risk of cardiovascular disease (CVD), dementia, and poor physical functioning after visit 5. In exploratory analyses, we examined participant characteristics associated with maintaining normal BP.

Results:

Among 2,699 participants with normal BP at baseline (mean age 51.3 years), 47.1% and 15.0% maintained normal BP through visits 4 and 5, respectively. The hazard ratios comparing participants who maintained normal BP through visit 4 but not visit 5 and through visit 5 versus those who did not maintain normal BP through visit 4 were 0.80 (95%CI 0.63, 1.03) and 0.60 (95%CI 0.42, 0.86), respectively, for CVD, and 0.85 (95%CI 0.71, 1.01) and 0.69 (95%CI 0.54, 0.90), respectively, for poor physical functioning. Maintaining normal BP through visit 5 was more common among participants with normal body mass index (BMI) versus obesity at visit 1, and those with normal BMI at visits 1 and 5 and with overweight at visit 1 and overweight or normal BMI at visit 5, compared to those with obesity at visits 1 and 5.

Conclusions:

Maintaining normal BP was associated with lower risk of CVD and poor physical functioning.

Graphical Abstract

INTRODUCTION

In contrast with several non-industrialized populations where systolic blood pressure (SBP) does not increase with age,^1–3 in most industrialized settings including the United States (US), SBP increases with age.⁴ However, some individuals may experience only a minimal increase in blood pressure (BP) as they age and maintain normal BP across their life course.^5,6 Determining whether maintaining normal BP from middle into older age is associated with lower CVD risk as compared to having normal BP only through midlife could inform public health programs focused on preventing the rise in BP.

Previous studies examining the maintenance of normal BP have not been able to follow participants into older age or have had relatively short follow up for CVD outcomes. In an analysis of the Coronary Artery Risk Development in Young Adults (CARDIA) study, which included adults aged 18–30 years with normal BP at baseline, 36% of participants maintained normal BP over 30 years.⁵ In the Jackson Heart Study, a cohort comprised of Black adults with mean age of 46.4 years at baseline, 35% of those with normal BP at baseline maintained normal BP over 8 years between study visits.⁶ Participants who maintained normal BP had a lower incidence of CVD over 5.9 years of subsequent follow up than those who did not have normal BP at baseline.⁶

We used data from the Atherosclerosis Risk in Communities (ARIC) Study to identify the proportion of participants who maintained normal BP through middle and into older age, over a period of more than 23 years from 1987–1989 to 2011–2013. We examined the association of maintaining normal BP with health outcomes important to older adults, including CVD, cognition, and physical functioning, with follow up from 2011–2013 through 2020. In exploratory analyses, we examined participant characteristics associated with maintaining normal BP.

METHODS

Study population

Instructions to access the dataset used for this study are available on the ARIC website (https://sites.cscc.unc.edu/aric). The ARIC Study enrolled 15,792 adults aged 45 to 64 years in 1987–1989 from Forsyth County, North Carolina, Jackson, Mississippi, Minneapolis, Minnesota, and Washington County, Maryland.⁷ To date, seven in-person visits have been conducted. The current analysis included Black and White participants with normal BP, as defined below, at visit 1 in 1987–1989 who attended visit 5 in 2011–2013 (Figure S1). We excluded participants missing information on the demographic characteristics, health behaviors, and psychosocial factors used in the current analysis. Institutional review boards at all sites approved the study procedures and all participants provided written informed consent at each visit.

Maintenance of normal BP

We used BP measurements from ARIC visits 1 through 5 (Figure 1). We did not investigate BP measurements at subsequent visits to allow follow up time for CVD, cognition, and physical functioning. BP measurement procedures are described in the Supplemental Methods.⁸ Normal BP was defined as SBP <120 mm Hg and diastolic BP (DBP) <80 mm Hg, without the self-reported use of antihypertensive medication. To be categorized as having maintained normal BP through a particular study visit, the participant was required to have normal BP at that visit and all available preceding visits. Participants with missing information on BP or antihypertensive medication use at any of the study visits were retained in the analysis, provided this information was available at visits 1 and 5. We categorized participants with normal BP at visit 1 as not maintaining normal BP through visit 4, maintaining normal BP through visit 4 but not visit 5, or maintaining normal BP through visit 5. These categories were chosen given the timing of the ARIC visits. The first 4 visits occurred at 3-year intervals (median of 8.9 years between visits 1 and 4), while visit 5 took place a median of 14.6 years after visit 4. This enabled us to examine maintenance of normal BP through middle age at visit 4 and into older age at visit 5. To retain as many participants in the analysis as possible, we allowed for missing data on BP measurements or antihypertensive medication use between visits 1 and 5, the handling of which is described in the Supplemental Methods.

Figure 1. Study schema showing the assessment of blood pressure from ARIC study visits 1 through 5 and follow up for incident outcomes.

Abbreviations: BP, Blood Pressure; CVD, Cardiovascular Disease.

Health outcomes

Incident CVD and dementia were assessed after visit 5. CVD outcomes included coronary heart disease (CHD), heart failure (HF), or stroke.^9–12 Participants were followed for CVD events and dementia from visit 5 through the date of onset, last contact with the study, or administratively censored on December 31, 2020, whichever occurred first. Information on CVD and dementia ascertainment is provided in the Supplemental Methods.^12–14 Physical functioning was assessed at visits 5, 6 and 7 using the Short Physical Performance Battery (SPPB), and poor physical functioning was defined as a SPPB score ≤ 6.¹⁵

Participant characteristics

Participant characteristics included in the analyses were chosen based on established relationships with BP. Demographic characteristics included age, sex, and race. We included race in this analysis as a marker for experiences of racialization and exposure to systemic racism, which influence opportunities for health over the life course.¹⁶ Education, height, weight, alcohol consumption, cigarette smoking, physical activity, and Dietary Approaches to Stop Hypertension (DASH) Diet score were assessed at visit 1.^17,18 Body mass index (BMI) was calculated from measured height and weight. Psychosocial factors were assessed at visit 2 and included perceived social support,^19,20 risk of isolation,^20,21 and trait anger.²² Information on ascertainment of these characteristics is provided in Table S1. We used BMI, alcohol consumption, cigarette smoking, and physical activity measured at visit 5 to determine changes in these variables from visit 1 to visit 5. Changes were categorized to reflect having the same, improved or worsened levels of these risk factors.

Statistical analysis

We estimated the proportion of participants who did not maintain normal BP through visit 4, who maintained normal BP through visit 4 but not visit 5, and who maintained normal BP through visit 5, and assessed the distribution of participant characteristics among these categories.

We used Cox proportional hazards regression to estimate hazard ratios (HR) for incident CVD, CHD, HF, stroke, and dementia after visit 5 for those who maintained normal BP through visit 4 but not visit 5 and who maintained normal BP through visit 5, each compared to those who did not maintain normal BP through visit 4. These analyses were restricted to participants who had not experienced events prior to or at visit 5 with follow up data available after visit 5. First, we examined unadjusted associations of maintaining normal BP through visit 4 but not visit 5 and maintaining normal BP through visit 5 with incident outcomes. Next, we examined associations adjusted for age at visit 5 and sex, race, and education assessed at visit 1. Then, we adjusted for sex, race, and education at visit 1, and age, BMI, alcohol consumption, smoking, and physical activity at visit 5. Finally, we included adjustment for changes in BMI, drinking status, smoking status, and physical activity from visit 1 to visit 5 instead of these values at visit 5. We estimated HRs for poor physical function using discrete time survival analysis with complementary log-log regression models. Participants were followed through the first study visit at which they had poor physical function or until the last visit they attended. The regression models used the same adjustments described for the other health outcomes.

In exploratory analyses, we used Poisson regression models with robust variance estimates to examine associations of participant factors at baseline with maintaining normal BP through visit 4 and through visit 5, each versus not maintaining normal BP through visit 4. Also, we estimated the relative risk for maintaining normal BP through visit 5 versus maintaining normal BP through visit 4 but not visit 5. In post hoc analyses, we examined associations of obesity defined based on waist circumference and waist-to-hip ratio, separately, with maintaining normal BP.^23,24 In sensitivity analyses, we used multilevel joint modeling multiple imputation to impute missing values for participant characteristics measured at baseline or visit 5 (Table S2) and repeated all analyses described above.^25–27 Additional information on the exploratory and sensitivity analyses is available in the Supplemental Methods.

Analyses were conducted using Stata version 18 (StataCorp LLC, College Station, TX), SAS version 9.4 (SAS Institute Inc, Cary, NC), and R (R Foundation for Statistical Computing, Vienna, Austria).

RESULTS

Proportion maintaining normal BP

Among 2,699 participants with normal BP at baseline who attended visit 5 (mean age at baseline, 51.3 years), 52.4% did not maintain normal BP through visit 4 which occurred a median of 8.9 years after the baseline visit (mean age at visit 4, 60.2 years). Among the remaining participants, 32.6% maintained normal BP through visit 4 but not visit 5 and 15.0% maintained normal BP through visit 5 over a median total follow up of 23.5 years (mean age at visit 5, 74.9 years). Characteristics of participants who did not maintain normal BP through visit 4, maintained normal BP through visit 4 but not visit 5, or maintained normal BP through visit 5 are provided in Table 1. Among those who maintained normal BP through visit 5, the mean SBP/DBP at visits 1 and 5 were 100.5/63.9 mm Hg and 110.2/59.2 mm Hg, respectively.

Table 1.

Visit 1 and visit 5 characteristics of participants who maintained and did not maintain normal blood pressure from ARIC visit 1 through ARIC visits 4 and 5.

Characteristic, %	Did not maintain normal BP through visit 4	Maintained normal BP through visit 4 but not visit 5*	Maintained normal BP through visit 5
N (%)	1415 (52.4%)	880 (32.6%)	404 (15.0%)
Age at visit 1, mean (SD)	51.8 (5.0)	50.8 (4.6)	50.5 (4.7)
Age at visit 5, mean (SD)	75.4 (5.1)	74.5 (4.7)	74.1 (4.8)
Female	60.7%	61.6%	61.1%
Black	13.6%	9.7%	6.2%
Education
<HS graduate	11.2%	8.6%	6.9%
HS/GED/vocational school	43.0%	40.8%	39.6%
At least some college	45.8%	50.6%	53.5%
BMI category, visit 1
Obesity	18.3%	9.1%	9.9%
Overweight	40.4%	35.6%	35.6%
Normal	41.3%	55.3%	54.5%
BMI category, visit 5
Obesity	34.1%	24.4%	18.8%
Overweight	40.3%	41.6%	38.4%
Normal	25.7%	34.0%	42.8%
Drinking status, visit 1
Current drinker	65.2%	67.8%	67.6%
Former drinker	14.7%	14.1%	12.6%
Never drinker	20.1%	18.1%	19.8%
Drinking status, visit 5
Current drinker	56.0%	57.6%	56.4%
Former drinker	26.2%	24.9%	24.5%
Never drinker	17.7%	17.5%	19.1%
Smoking status, visit 1
Current	18.6%	21.5%	17.6%
Former	34.2%	31.4%	28.5%
Never	47.2%	47.2%	54.0%
Smoking status, visit 5
Current	5.8%	8.2%	8.4%
Former	54.1%	52.5%	46.0%
Never	40.1%	39.3%	45.5%
Physical activity at visit 1, MVPA/week
0 minutes	29.0%	26.1%	24.8%
1–149 minutes	27.2%	28.9%	30.0%
≥150 minutes	43.7%	45.0%	45.3%
Physical activity at visit 5, MVPA/week
0 minutes	30.0%	25.5%	23.0%
1–149 minutes	18.9%	17.5%	15.8%
≥150 minutes	51.1%	57.0%	61.1%
DASH diet score at visit 1, mean (SD)	24.8 (4.8)	24.8 (4.8)	25.2 (4.7)
Perceived social support at visit 2
Quartile 1 (Lowest)	30.9%	33.0%	34.7%
Quartile 2	27.0%	26.0%	25.2%
Quartile 3	24.8%	22.5%	22.8%
Quartile 4 (Highest)	17.3%	18.5%	17.3%
Lubben social network scale at visit 2
Isolated or high risk for isolation	5.9%	6.4%	6.2%
Moderate risk	11.8%	11.9%	11.4%
Low risk	82.3%	81.7%	82.4%
Spielberger Trait Anger at visit 2
High	7.8%	6.0%	5.4%
Moderate	56.5%	56.6%	55.2%
Low	35.7%	37.4%	39.4%
SBP at visit 1, mean (SD)	109.0 (6.9)	102.9 (8.0)	100.5 (7.8)
DBP at visit 1, mean (SD)	68.5 (6.3)	65.0 (6.8)	63.9 (7.0)
SBP at visit 5, mean (SD)	131.3 (18.0)	129.4 (13.6)	110.2 (7.4)
DBP at visit 5, mean (SD)	66.6 (10.6)	66.8 (9.8)	59.2 (7.3)
Antihypertensive medication use at visit 5	66.6%	43.3%	0.0%

Abbreviations: ARIC, Atherosclerosis Risk in Communities; BMI, body mass index; BP, blood pressure; DASH, Dietary Approaches to Stop Hypertension; DBP, Diastolic Blood Pressure; HS, High School; MVPA, Moderate-Vigorous Physical Activity; SBP, Systolic Blood Pressure; SD, Standard Deviation.

^*Thirteen participants with missing BP at visit 4 who met criteria for normal BP at previous study visits, but had BP above normal at visit 5, are included in this category. Additionally, there was one participant missing BP at visit 4 who met criteria for normal BP at previous study visits who had normal BP at visit 5.

Association of maintaining normal BP with CVD, dementia, and physical function

In unadjusted analyses, maintaining normal BP through visit 4 but not visit 5 and through visit 5, versus not maintaining normal BP through visit 4, were associated with a lower risk of CVD, CHD, HF, dementia, and poor physical function (Table S3). After adjustment for characteristics at visit 5, the HRs comparing participants who maintained normal BP through visit 4 but not visit 5 and through visit 5 versus those who did not maintain normal BP through visit 4 were 0.80 (95% CI 0.63–1.03) and 0.60 (95% CI 0.42–0.86), respectively, for incident CVD; 0.69 (95% CI 0.47–1.02) and 0.36 (95% CI 0.18–0.73), respectively, for incident CHD, and; 0.68 (95% CI 0.50–0.94) and 0.64 (95% CI 0.41–1.00), respectively, for incident HF (Figure 2). Maintaining normal BP through visit 4 but not visit 5 and maintaining normal BP through visit 5 were not associated with the incidence of stroke. After adjustment, there was no evidence of an association of maintaining normal BP through visit 4 but not visit 5 or maintaining normal BP through visit 5 with incident dementia. After adjustment and compared with not maintaining normal BP through visit 4, the HRs for poor physical function associated with maintaining normal BP through visit 4 but not visit 5 and with maintaining normal BP through visit 5 were 0.85 (95% CI 0.71–1.01) and 0.69 (95% CI 0.54–0.90), respectively. The HRs adjusted for changes in BMI, drinking status, smoking status, and physical activity from visit 1 to visit 5, rather than at visit 5, are provided in Figure S2.

Figure 2. Adjusted hazard ratios for incident cardiovascular disease, dementia, and poor physical functioning after visit 5 associated with maintenance of normal blood pressure at visits 4 and 5 among participants who had normal blood pressure at visit 1 in midlife.

Abbreviations: BP, Blood Pressure; CI, Confidence Interval; HR, Hazard Ratio.

Any cardiovascular disease (CVD) defined as coronary heart disease (CHD), heart failure (HF), or stroke. Analyses of incident CVD, CHD, HF, stroke, and dementia exclude participants with the relevant condition at or before visit 5. Poor physical function defined as Short Physical Performance Battery (SPPB) score ≤6.

Models adjusted for sex, race, and education at visit 1, and age, body mass index, drinking status, smoking status, and physical activity at visit 5.

Factors associated with maintaining normal BP

Factors at baseline

In unadjusted, demographics-adjusted, and fully adjusted models, older age and Black race were associated with a lower likelihood of maintaining normal BP through visit 4 (Table S4, Table 2). Having overweight or normal BMI versus obesity and low versus high trait anger were associated with a higher likelihood of maintaining normal BP through visit 4. Older age and Black race were associated with a lower likelihood, while having a normal BMI was associated with a higher likelihood, of maintaining versus not maintaining normal BP through visit 5 in unadjusted and adjusted models. Among those who maintained normal BP through visit 4, Black participants were less likely to maintain normal BP through visit 5. No other factors were associated with maintaining normal BP through visits 4 and 5.

Table 2.

Baseline characteristics associated with maintaining normal blood pressure through visit 4 and visit 5.

Baseline characteristics	Maintained normal BP through visit 4 versus did not maintain normal BP through visit 4 Relative risk (95% CI)*	Maintained normal BP through visit 5 versus did not maintain normal BP through visit 5 Relative risk (95% CI)*	Maintained normal BP through visit 5 versus maintained normal BP through visit 4 but not visit 5 Relative risk (95% CI)*
Age, per 5 years	0.87 (0.83, 0.91)	0.81 (0.73, 0.90)	0.93 (0.85, 1.02)
Female sex (vs male)	0.95 (0.87, 1.04)	0.88 (0.71, 1.07)	0.92 (0.76, 1.11)
Black race (vs White)	0.77 (0.66, 0.90)	0.52 (0.35, 0.78)	0.67 (0.47, 0.97)
Education
<HS graduate	REF	REF	REF
HS/GED/vocational school	1.05 (0.89, 1.23)	1.11 (0.76, 1.63)	1.04 (0.73, 1.47)
At least some college	1.12 (0.95, 1.32)	1.23 (0.84, 1.81)	1.08 (0.76, 1.53)
BMI category
Obesity	REF	REF	REF
Overweight	1.38 (1.17, 1.62)	1.27 (0.91, 1.78)	0.94 (0.70, 1.26)
Normal	1.67 (1.42, 1.95)	1.48 (1.08, 2.04)	0.90 (0.68, 1.20)
Alcohol consumption
Current	REF	REF	REF
Former	1.01 (0.89, 1.14)	0.97 (0.74, 1.29)	0.98 (0.76, 1.27)
Never	1.04 (0.93, 1.16)	1.16 (0.91, 1.48)	1.09 (0.88, 1.35)
Smoking status
Current	REF	REF	REF
Former	0.92 (0.82, 1.03)	0.96 (0.73, 1.27)	1.04 (0.81, 1.35)
Never	0.99 (0.89, 1.10)	1.20 (0.93, 1.54)	1.22 (0.96, 1.54)
Physical activity, MVPA/week
0 mins	REF	REF	REF
1–149 mins	1.05 (0.95, 1.17)	1.06 (0.83, 1.36)	1.01 (0.81, 1.27)
≥150 mins	1.03 (0.93, 1.14)	0.99 (0.79, 1.26)	0.96 (0.77, 1.18)
DASH diet score, per point	1.00 (0.99, 1.01)	1.02 (0.995, 1.04)	1.02 (0.997, 1.03)
Perceived social support†
Quartile 1 (Lowest)	REF	REF	REF
Quartile 2	0.94 (0.85, 1.04)	0.90 (0.71, 1.13)	0.95 (0.77, 1.17)
Quartile 3	0.93 (0.83, 1.03)	0.92 (0.72, 1.18)	0.99 (0.80, 1.24)
Quartile 4 (Highest)	1.00 (0.88, 1.13)	0.94 (0.71, 1.25)	0.94 (0.73, 1.22)
Lubben social network scale†
Isolated or high risk for isolation	REF	REF	REF
Moderate risk	0.97 (0.80, 1.17)	0.93 (0.59, 1.47)	0.95 (0.63, 1.45)
Low risk	0.96 (0.81, 1.14)	0.93 (0.62, 1.40)	0.98 (0.68, 1.42)
Spielberger Trait Anger†
High	REF	REF	REF
Moderate	1.19 (0.99, 1.42)	1.21 (0.81, 1.83)	1.02 (0.71, 1.49)
Low	1.22 (1.02, 1.47)	1.30 (0.85, 1.98)	1.07 (0.73, 1.57)

Abbreviations: BMI, Body Mass Index; BP, blood pressure; CI: Confidence Interval; DASH, Dietary Approaches to Stop Hypertension; HS, High School; MVPA, Moderate-Vigorous Physical Activity. Bold indicates statistically significant at P < 0.05.

^*Adjusted for: Age, sex, race, education, BMI, alcohol consumption, smoking status, physical activity, and DASH diet score measured at visit 1, and psychosocial factors measured at visit 2.

^†Measured at visit 2.

When defined based on waist circumference or waist-to-hip ratio, not having versus having obesity was associated with a higher likelihood of maintaining normal BP through visit 4 (Table S5).

Changes in factors from visit 1 to visit 5

In unadjusted, demographics-adjusted, and fully adjusted models, participants with normal BMI at visits 1 and 5, and those with overweight at visit 1 who had overweight or normal BMI at visit 5, were more likely to maintain normal BP through visit 5 compared to those who had obesity at visits 1 and 5 (Table S6). The changes in other factors were not associated with maintaining normal BP through visit 5 from baseline or from visit 4 in fully adjusted models.

Associations of changes in obesity status defined based on waist circumference and waist-to-hip ratio with maintaining normal BP through visit 5 from baseline or from visit 4 are presented in Table S7.

Sensitivity analyses – Multiple imputation

Multiple imputation of missing covariate information did not substantively change the results. After performing multiple imputation, 51.4% of participants did not maintain normal BP through visit 4, 34.0% of participants maintained normal BP through visit 4 but not visit 5, and 14.6% of participants maintained normal BP through visit 5. Maintaining normal BP through visit 4 but not visit 5 and through visit 5 was associated with lower risk of incident CVD, CHD, HF, and poor physical function (Table S8). After adjustment, there was no association of maintaining normal BP with incident stroke or dementia. The associations of baseline participant characteristics, and changes in participant characteristics between visits 1 and 5, with maintaining normal BP after multiple imputation are presented in Tables S9 and S10.

DISCUSSION

The current study has several important findings, which have clinical and public health implications. Overall, 15.0% of participants 45–64 years old at baseline maintained normal BP over 23.5 years of follow-up, at which time their mean age was nearly 75 years. The risks of incident CVD, CHD, HF and poor physical function were lower in participants who maintained versus did not maintain normal BP. In exploratory analyses, having normal BMI at baseline, as well as having normal BMI at both visit 1 and visit 5, were associated with a higher likelihood of maintaining normal BP. Additionally, participants who were older at baseline and Black participants were less likely to maintain normal BP.

A recent modeling study estimated 40% of all CVD events are attributable to SBP and DBP above the normal range.²⁸ Therefore, maintaining normal BP across the life course is an important public health goal for preventing CVD. The current study extends previous work conducted among younger and middle-aged adults^5,6 and provides evidence for maintaining normal BP into older age as an achievable goal. In the subgroup who maintained normal BP from midlife through older adulthood, SBP was on average 100.5 mm Hg at baseline and increased by a mean of 9.7 mm Hg over 23.5 years, or approximately 4 mm Hg per decade. The current results support prior studies showing that SBP in early life is a strong predictor of future SBP and the development of hypertension.²⁹ Having lower SBP in early and midlife may be important for mitigating the increase in SBP that typically occurs with age.^29,30

The current study provides evidence of the importance of maintaining normal BP into older age for the prevention of incident CVD, CHD, HF, and poor physical function. However, there was no association with the risk of stroke or dementia. A prior analysis of ARIC study data found no difference in stroke risk in those with SBP <120 mm Hg at visit 1 whose SBP increased to ≥120 mm Hg at visit 4 versus those with SBP <120 mm Hg from visits 1 to 4.³¹ Another analysis of BP patterns in ARIC found no difference in dementia risk for participants who did not have hypertension in midlife but had hypertension in late-life compared to those without hypertension in mid- and late-life.¹³ The current study results are consistent with these prior analyses. As we restricted the study population to participants with normal BP at baseline, we were unable to examine the relationship between midlife BP and the risk for stroke and dementia associated, which may be stronger than the relationship in late-life.^32–34 The current study differs from previous analyses because it focused on the maintenance of normal BP into older age, not only the absence of hypertension.

Participants who maintained normal BMI from midlife to older adulthood and those with overweight in midlife who maintained or lost weight were more likely to maintain normal BP through older adulthood compared to those with obesity.^5,6,35 In the US, more than 40% of adults have obesity,³⁶ which suggests that maintenance of normal weight or weight loss may be important public health interventions for preventing elevated BP and hypertension. Randomized controlled trials have shown weight loss reduces BP.³⁷ However, there are few data demonstrating weight loss interventions in randomized trials can be translated into community-based settings and reach populations before they develop hypertension. The American Heart Association-funded RESTORE (addREssing Social determinants TO pRevent hypErtension) Network, which is testing the implementation of community-based approaches to prevent hypertension, including dietary and physical activity interventions, in populations with elevated BP.³⁸

The current study has several potential and known limitations. We did not have BP data available before the ARIC baseline visit when participants were 45 to 64 years of age. We may have included participants who had elevated BP or hypertension before but not at baseline or excluded participants who maintained normal BP into their sixties but had elevated BP or hypertension at baseline. Due to missing data, some misclassification of the maintenance of normal BP may have been possible. We did not consider BP measurements at visits 6 or 7 and some participants who maintained normal BP through visit 5 may not have continued to maintain normal BP. This study may be subject to bias due to selective attrition and death. For example, if participants who did not maintain normal BP were more likely to have dementia and die prior to a cognitive evaluation at visit 6 or 7. However, in that situation, the effect estimates would likely be attenuated. We estimated associations of a large number of participant characteristics with maintenance of normal BP. Therefore, the analyses should be considered exploratory, and the results interpreted with caution.³⁹ Additionally, changes in participant characteristics between visits 1 and 5 may not capture potentially clinically relevant fluctuations in these characteristics that may have occurred between timepoints. We acknowledge the need for better, direct measures of racialized experiences and exposure to systemic racism, which may affect BP levels. Many of the factors we examined were assessed by self-report. The psychosocial variables were assessed at visit 2 and not at visit 1. We were unable to assess potentially important variables such as usual sodium intake, which is not captured well by food frequency questionnaires,⁴⁰ or sleep characteristics which were not assessed in the full ARIC cohort.

Perspectives

The current study suggests that maintaining normal BP into older age is achievable and is associated with a lower risk for incident CVD and poor physical function.

Novelty and Relevance.

What Is New?

• We assessed whether maintaining normal blood pressure (BP) from middle into older age is associated with health outcomes important to older adults, including CVD, dementia, and poor functioning.

What Is Relevant?

• Among participants 45–64 years of age at baseline with normal BP, 15.0% maintained normal BP over a median of 23.5 years.

• Maintaining normal BP was associated with lower risk of CVD and poor physical function.

• Having a normal body mass index (BMI) at baseline and maintaining a normal BMI were associated with maintaining normal BP.

Clinical/Pathophysiological Implications?

• Maintaining normal BP into older age is achievable and associated with lower risk of incident CVD and poor physical functioning.

• Maintaining normal BMI may be important for maintaining normal BP.

ABBREVIATIONS

ARIC

Atherosclerosis Risk in Communities

BMI

Body mass index

BP

Blood pressure

CHD

Coronary heart disease

CVD

Cardiovascular disease

DASH

Dietary Approaches to Stop Hypertension

DBP

Diastolic blood pressure

HF

Heart failure

HR

Hazard ratio

SBP

Systolic blood pressure

SPPB

Short Physical Performance Battery

US

United States