Abstract
Background:
Black adults have higher dementia risk than White adults. Whether tighter population-level blood pressure (BP) control reduces this disparity is unknown.
Objective:
Estimate the impact of optimal BP treatment intensity on racial disparities in dementia.
Methods:
A microsimulation study of US adults ≥18 across a life-time policy-planning horizon. BP treatment strategies were the Systolic Blood Pressure Intervention Trial (SPRINT) protocol, the Eighth Joint National Committee (JNC-8) recommendations, and usual care (non-intervention control). Outcomes were all-cause dementia, atherosclerotic cardiovascular disease (ASCVD), stroke, myocardial infarction, non-ASCVD death, global cognitive performance, and optimal brain health (being free of dementia, cognitive impairment, or stroke). Population-level and individual-level effects stratified by race were estimated.
Results:
Optimal population-level implementation of a SPRINT-based BP treatment strategy, compared to usual care, would increase average annual dementia incidence in White, but not Black, adults (1% vs. 0%), due to hypertensive individuals’ greater survival, and reduce annual ASCVD events more in Black than White adults (13% vs. 5%). Under a SPRINT-based strategy, individuals with hypertension gained more years lived without dementia, ASCVD, myocardial infarction, or stroke and more years lived in optimal brain health. A SPRINT-based strategy did not attenuate individual-level race disparities in outcomes, except stroke. Due to longer life expectancy, a SPRINT-based strategy did not substantially reduce lifetime dementia risk in either group. The JNC-8-based strategy had similar but smaller effects as the SPRINT-based strategy.
Conclusions:
Our results suggest that tighter population-level BP control would not reduce population-level disparities in dementia between US Black and White adults.
Keywords: Alzheimer’s disease, blood pressure treatment, dementia risk, microsimulation, disparities, equity
INTRODUCTION
Population-level interventions to reduce Black adults’ higher dementia risk are urgently needed. Black adults are approximately twice as likely as White adults to develop dementia.1, 2 Preventing dementia can lead to longer survival, less disability and institutionalization, lower health care costs, and better quality of life, outcomes that are worse in US Black adults than in US White adults. The US government needs evidence on viable population-level interventions to decrease disparities in dementia risk for Black adults.3
Intensive blood pressure (BP)-lowering is a potential strategy to reduce dementia and disparities between Black and White adults.4 High BP is associated with higher dementia risk and tends to develop at younger ages, be more severe and prevalent, and be less controlled in Black adults than White adults.4, 5 Black adults’ higher cumulative BP levels may contribute to racial differences in later-life cognitive decline.6 While the Systolic Blood Pressure Intervention Trial (SPRINT) showed that lowering BP to optimal levels (<120/80 mmHg) reduces atherosclerotic cardiovascular disease (ASCVD) events and mild cognitive impairment in adults with high ASCVD risk,7, 8 it is unknown whether population-level BP-lowering interventions reduce dementia risk disparities between US Black and White adults. As no strong evidence of a subgroup effect by race was detected in SPRINT, the SPRINT algorithm was applied similarly to both Black and White populations, resulting in the same relative effect on ASCVD events for Black and White populations. Estimating the individual and societal benefits of intensive BP treatment on dementia and ASCVD in US Black and White adults will inform clinical care and policy.
Leveraging our unique dementia and ASCVD microsimulation analysis platform,9 we compared the effects of three strategies for treating high BP on dementia and ASCVD events in US Black and White adults: a BP treatment strategy based on the SPRINT protocol, a BP treatment strategy based on the Eighth Joint National Committee (JNC-8) recommendations,10 and usual care (non-intervention). We examined how race differences in all-cause dementia and ASCVD events would differ with each BP treatment strategy for the average Black or White adult with hypertension and the US population.
METHODS
Overview
We conducted an individual-level microsimulation analysis using an open-source dementia and ASCVD microsimulation framework, the MIchigan ChROnic Disease SIMulation Model (MICROSIM), which has been previously described in detail elsewhere.9 MICROSIM is an agent-based Monte Carlo microsimulation model using regression-based models to model annual transitions in vascular risk factors (VRFs) and outcomes, including all-cause dementia and ASCVD events, in simulated US adults. The University of Michigan Institutional Review Board approved this study. Participating institutions’ review boards approved the cohort studies. All participants provided written informed consent.
Population
The MICROSIM population was constructed from the National Health and Nutrition Examination Survey (NHANES), a nationally representative probability sample of the US population with rich clinical information. The MICROSIM population includes adults aged ≥18 without dementia or ASCVD and has the distribution of demographic and ASCVD risk factors, including BP levels, and antihypertensive medication use observed in the US. Consequently, both mean levels and uncertainty in demographics and risk factors, as well as race differences, represent those in the US population as measured in NHANES.9
Data Sources
Model inputs were derived from the best available evidence, mainly from an individual participant data (IPD) meta-analysis of 5 pooled and harmonized US cardiovascular cohort studies.6 The cohort studies included in the IPD meta-analysis were the Atherosclerosis Risk in Communities Study (ARIC),11 Coronary Artery Risk Development in Young Adults Study (CARDIA),12 the Cardiovascular Health Study (CHS),13 Framingham Offspring Study (FOS),14 and Northern Manhattan Study (NOMAS).15 These longitudinal cardiovascular cohort studies were chosen because they measured CVD risk factors, ASCVD events, and cognition with or without dementia. They also had age, geography, or race with or without ethnic diversity.
Outcomes
The simulated outcomes of interest were all-cause dementia; fatal or non-fatal ASCVD, which includes stroke and myocardial infarction (MI); non-ASCVD death; stroke; and MI. MICROSIM is validated and reproduces population-level all-cause dementia, stroke, and MI incidence.9 MICROSIM updates the status of individuals annually into all-cause dementia, fatal or non-fatal ASCVD, non-ASCVD death, or remaining free of all events using Monte Carlo methods. MICROSIM has been validated to reproduce population-level all-cause dementia, stroke, and MI incidence.9 ASCVD risks were calculated using the revised ASCVD pooled cohort equation.16 ASCVD events were partitioned into stroke vs. MI using a separate regression model in the IPD meta-analyses of the pooled cardiovascular cohort studies. Estimates of dementia risk, non-ASCVD death, changes in global cognitive performance (GCP) scores, and changes in VRFs were estimated using predictive models derived from IPD meta-analyses of the pooled cardiovascular cohort studies.9 Specifically, dementia risk estimates were calculated using proportional hazard models based on demographics, VRF levels, initial GCP level, and GCP trajectories. Then, dementia incidence was recalibrated to population-level norms.17, 18 Non-CVD death was estimated using a logistic regression model developed using NHANES-linked mortality data adjusting for demographics and all VRFs.
GCP was estimated using a published linear mixed-effects model.6 Cognitive impairment was defined as a GCP score 1.5 standard deviations below the age-, sex-, and race-adjusted population mean GCP score.1, 19 This absolute cognitive score approach effectively defines cognitive impairment in the absence of a clinical diagnosis.19, 20 Optimal brain health was defined as being free of dementia, cognitive impairment, or stroke. Uncertainty was estimated by sampling from the population-level distribution of residuals.
BP Treatment Strategies
We compared the optimal implementation of the SPRINT-based and JNC-8-based BP treatment strategies to usual care (non-intervention control). In the SPRINT-based BP treatment strategy,7 all adults with a high ASCVD risk (estimated 10-year ASCVD risk >15%), consistent with the trial’s participants, were treated to a systolic BP (SBP) target of <120 mmHg. For each individual, BP medications were added annually by assuming an average BP-lowering effect of 5.5/3.1 mmHg per medication21 up to a maximum of four BP medications. We capped treatment at four medications because the benefits of adding a fifth medication are uncertain,22 and guidelines recommend evaluating for secondary causes of hypertension at this threshold.23 To account for potentially higher average BP measurements taken in NHANES and the cohorts compared to those taken in SPRINT, which used a strict standardized protocol, we assumed that SBP measurements in NHANES and the cohorts were 6 mmHg higher than they would have been in SPRINT;24 thus, our effective SBP treatment target was 126 mmHg.
In the JNC-8-based BP strategy,10 simulated participants were assessed annually for eligibility for BP lowering using the JNC-8 eligibility criteria and treatment targets (BP 150/90 mmHg for most adults aged ≥60 vs. target BP 140/90 mmHg for adults aged <60, with diabetes or with a glomerular filtration rate <60 mL/min defined using a valid equation25). BP medications were added annually, assuming the same average BP-lowering effect per medication21 and a maximum of four BP medications as the SPRINT-based BP treatment strategy.
For usual care (non-intervention controls), we simulated the usual BP care that US patients would receive. We developed a regression model to predict subsequent BP treatment based on prior BP treatment and levels and all patient VRFs using an individual participant data meta-analysis of five pooled cohort studies—ARIC Study,11 CARDIA Study,12 CHS,13 FOS,14 and NOMAS15—herein called the CV pooled cohort.
The linkage between BP treatment strategies and dementia was implemented through a multi-step process in MICROSIM. First, BP levels were lowered according to the specific BP Treatment strategy. Second, BP levels were accounted for in global cognition using the best available observational data.6 Third, mean GCP and trends in GCP were accounted for in models predicting dementia risk.9
Analyses
We simulated three separate populations of 300,000 adults in the US Black and White populations (totaling six). Minimum population size was determined by exploring increasing population sizes until stable mean event rates were identified, and then the actual population size was approximately triple the level where stability was identified. We advanced each population into the future until all simulated people died, applying each BP treatment strategy (SPRINT-based, JNC-8-based, or usual care). All analyses were based on 100 bootstrapped analyses of these MICROSIM simulation runs.
First, we estimated the change in SBP levels in Black and White people and the differences in SBP levels between Black and White people over time with each BP treatment strategy separately in the overall population and the sub-population with hypertension. We calculated the number of BP medications, achieved BP levels, and the proportions of adults who receive BP treatment and have BP controlled based on different criteria (the assigned BP treatment strategy criteria, <130/80 mmHg, and <140/90 mmHg) at 1-year and 10-year follow-up intervals with each BP treatment strategy. We estimated the relative and absolute effects of a SPRINT-based or JNC-8-based BP treatment strategy, compared with usual care, on dementia, ASCVD events, and mortality by race and Black-White differences in outcomes for the overall US Black and White populations (with and without hypertension). We calculated Black-White disparities for each BP treatment strategy by subtracting the age- and sex-standardized event rate (per 100,000 population) in White adults from the rate in Black adults. We determined absolute differences in disparities by subtracting the Black-White disparity for each BP treatment strategy from the Black-White disparity in usual care and relative differences by taking one minus the Black-White disparity in each BP treatment strategy divided by the Black-White disparity in usual care expressed as a percent.
Next, we examined how the outcomes of a typical Black or White individual with hypertension and disparities in the average individual-level outcomes would be affected by implementing SPRINT-based or JNC-8-based BP treatment strategies. We summed the mean individual-level age- and sex-standardized years lived in total, without health conditions (dementia, ASCVD events, stroke, MI, all-cause mortality), and with optimal brain health in each population treated by a SPRINT-based or JNC-8-based BP strategy and subtracted the mean years lived in the usual care population (e.g., mean years lived without MI in the SPRINT-based BP treated population minus the mean years lived without MI in the usual care population). Credible intervals were estimated for each simulated quantity using the bounds within which 95% of each simulated quantity fell within bootstrapped samples within each treatment strategy/race group. Because MICROSIM is a probabilistic model that accounts for uncertainty at each level of the simulation (e.g., estimated VRFs accounted for error and outcomes based on VRFs were estimated with error), bootstrapped intervals from simulation outputs are conceptually similar to those derived from a multidimensional probabilistic sensitivity analysis of a decision analysis capturing the joint effect of uncertainty across all model parameters.26
RESULTS
Characteristics of the Baseline Simulated Population
Hypertension (BP >140/90 mmHg or anti-hypertensive medication use) was present in 35.9% of Black individuals and 31.3% of White individuals at baseline. Median (25th, 75th interquartile range) age was 40 (31, 52) years for Black individuals and 44 (33, 60) years for White individuals. Black individuals had higher SBP levels than White individuals, even among those with hypertension, despite Black individuals with hypertension having a higher number of BP medications than White individuals (medications per person, 1.1 vs. 1.0). In the top decile of SBP in the overall population with hypertension, the mean SBP was 160 mmHg for Black individuals vs. 158 mmHg for White individuals.
Differences in Population-Level SBP with BP Treatment Strategies
In the population with hypertension at baseline, Black people had a 4.7-mmHg-higher SBP level than White people with usual care. Each intensive BP treatment strategy reduced this SBP difference to 4.2 mmHg (10.6% reduction) (Figure 1). The Black-White difference in SBP under usual care and reduction with the intensive BP treatment strategies was smaller in those without hypertension at baseline (Figure 2).
Figure 1.
Temporal Trend in Systolic Blood Pressure in US Black and White Adults (Panel 1A) and the Difference in Systolic Blood Pressure Between US Black and White Adults (Panel 1B) by Blood Pressure Treatment Strategy in Those with Hypertension at Baseline.
Each panel shows BP time trends across simulation waves (x-axis). Panel 1A shows the absolute level of SBP for Black and White populations with each BP treatment strategy. Panel 1B shows the SBP differences between the Black and White adult populations. Abbreviations: BP, blood pressure; JNC-8, Eighth Joint National Committee; SBP, systolic blood pressure; SPRINT, Systolic Blood Pressure Intervention Trial.
Figure 2.
Temporal Trend in Systolic Blood Pressure in Black and White Adults (Panel 2A) and the Difference in Systolic Blood Pressure Between Black and White Adults (Panel 2B) over Time by Blood Pressure Treatment Strategy in the Overall US Population.
The JNC-8- and SPRINT-based BP treatment strategies, compared with usual care, would reduce the differences in SBP levels between Black and White people in the overall population over time. The difference in the mean SBP levels between Black and White people was 3.8 mmHg with usual care, 3.7 mmHg with the JNC-8-based BP treatment strategy, and 3.5 mmHg for the SPRINT-based BP treatment strategy. Abbreviations: BP, blood pressure; JNC-8, Eighth Joint National Committee; SBP, systolic blood pressure; SPRINT, Systolic Blood Pressure Intervention Trial.
Table 1 shows participants’ BP treatment characteristics by BP treatment strategy at 1-year and 10-year follow-up. The difference in the number of BP medications between Black and White people with hypertension increased with both intensive BP strategies. In the first 5 years after initiation, the mean number of BP medications increased for Black and White people with hypertension by 0.27 and 0.15 under a SPRINT-based BP treatment strategy and 0.14 and 0.05 under a JNC-8-based BP treatment strategy.
Table 1.
Participant Blood Pressure Treatment Characteristics by Blood Pressure Treatment Strategy at 1-Year and 10-Year Follow-Up.
| After 1 year of follow-up (%) except where noted |
Population Treated with a JNC-8-based BP Treatment Strategy |
Population Treated with a SPRINT-based BP Treatment Strategy |
Population Treated with Usual Care |
|||
|---|---|---|---|---|---|---|
| Black Adults |
White Adults |
Black Adults |
White Adults |
Black Adults |
White Adults |
|
| Number of BP meds, mean (SD) | 1.8 (1.6) |
1.3 (1.2) |
2.4 (1.7) |
1.7 (1.5) |
1.2 (1.1) |
1.1 (1.0) |
| Adults who receive BP treatment | 77% | 72% | 85% | 79% | 67% | 67% |
| 1 medication | 28% | 36% | 24% | 32% | 34% | 37% |
| 2 medications | 20% | 21% | 18% | 21% | 21% | 20% |
| 3 medications | 9% | 7% | 10% | 9% | 8% | 7% |
| 4 medications | 11% | 6% | 20% | 12% | 3% | 2% |
| Adults with BP controlled by the BP treatment strategy criteria | 55% | 60% | 20% | 24% | NA | NA |
| Adults with BP controlled <130/80 | 22% | 25% | 22% | 25% | 22% | 25% |
| Adults with BP controlled <140/90 | 40% | 44% | 46% | 49% | 40% | 44% |
| SBP achieved, mean (SD) | 141.0 (20.2) |
139.2 (20.4) |
139.3 (19.9) |
137.9 (20.0) |
142.9 (21.9) |
139.9 (21.3) |
| DBP achieved, mean (SD) | 81.9 (11.3) |
77.6 (11.1) |
80.9 (10.4) |
76.8 (10.1) |
83.0 (12.7) |
78.0 (11.9) |
| After 10 years of follow-up | ||||||
| Number of BP meds, mean (SD) | 1.9 (1.6) |
1.5 (1.3) |
2.6 (1.7) |
2.0 (1.5) |
1.2 (1.1) |
1.2 (1.1) |
| Adults who receive BP treatment | 80% | 77% | 87% | 83% | 70% | 72% |
| 1 medication | 28% | 35% | 22% | 31% | 35% | 38% |
| 2 medications | 20% | 23% | 17% | 22% | 22% | 23% |
| 3 medications | 10% | 9% | 11% | 11% | 9% | 8% |
| 4 medications | 12% | 7% | 20% | 13% | 3% | 2% |
| Adults with BP controlled by the BP treatment strategy | 57% | 70% | 23% | 28% | NA | NA |
| Adults with BP controlled <130/80 | 22% | 30% | 22% | 30% | 22% | 30% |
| Adults with BP controlled <140/90 | 44% | 51% | 49% | 54% | 44% | 51% |
| SBP achieved, mean (SD) | 139.9 (18.6) |
136.5 (18.9) |
138.6 (18.4) |
135.4 (18.3) |
141.3 (20.2) |
136.9 (19.5) |
| DBP achieved, mean (SD) | 83.3 (10.9) |
78.1 (11.1) |
82.7 (10.4) |
77.6 (10.6) |
84.0 (11.8) |
78.3 (11.7) |
Population-Level Changes in Events with Different BP Treatment Strategies
The SPRINT-based and JNC-8-based BP treatment strategies, compared to usual care, would result in similar and small relative increases in average annual dementia incidence of 1% for the White population and 0% for the Black population mainly due to the small increase in survival among hypertensive individuals (Table 2). Both intensive BP treatment strategies reduced ASCVD events more than usual care in the Black and White populations. A SPRINT-based BP treatment strategy reduced ASCVD events in both groups more than a JNC-8-based BP treatment strategy. A SPRINT-based BP treatment strategy compared to usual care was associated with ~40,000 fewer MIs and ~42,000 fewer strokes in White people and ~10,000 fewer MIs and ~18,000 fewer strokes in Black people, representing on average a 5% and 7% reduction in the population-level event MI and stroke rates for the White population vs. 12% and 15% reductions for the Black population (Table 2) in the US population per year, respectively. Compared to usual care, the SPRINT-based and JNC-8-based BP treatment strategies have small relative risk reductions in mortality.
Table 2.
Population-Level Change in the Number and Percentage of Events Prevented Per Year, on Average across a Lifetime Perspective, with Each Blood Pressure Treatment Strategy Standardized to the Overall US Populations of Black and White Adults.
| Percentage Reductions in Events Prevented (Relative Effects of BP Treatment Strategies) | ||||
|---|---|---|---|---|
| Population Treated with a JNC-8- based BP Treatment Strategy versus Usual Care |
Population Treated with a SPRINT- based BP Treatment Strategy versus Usual Care |
|||
| Black | White | Black | White | |
| Dementia | 0% [−0%, 1%] |
1% [0%, 1%] |
0% [0%, 1%] |
1% [0%, 1%] |
| Any ASCVD event | −5% [−6%, −4%] |
−1% [−2%, −1%] |
−13% [−14%, −12%] |
−5% [−6%, −4%] |
| Myocardial infarction | −4% [−6%, −3%] |
−1% [−2%, 0%] |
−12% [−14%, −11%] |
−5% [−6%, −3%] |
| Stroke | −6% [−7%, −4%] |
−2% [−3%, −1%] |
−15% [−16%, −13%] |
−7% [−9%, −6%] |
| Reduction in number of events prevented (Absolute Effect of BP Treatment Policies) | ||||
| Dementia | 1,728 [−96, 3,881] |
15,271 [2,952, 27,402] |
2,075 [234, 3,634] |
11,968 [22, 23,224] |
| Any ASCVD Event | −9,441 [−11,193, −7,613] |
−18,992 [−31,904, −7,086] |
−23,829 [−25,370, −22,012] |
−64,526 [−76,862, −51,624] |
| Myocardial infarction | −3,523 [−4,767, −2,177] |
−8,775 [−19,432, 353] |
−9,666 [−10,856, −8,370] |
−39,628 [−48,965, −29,328] |
| Stroke | −7,252 [−8,982, −5,592] |
−13,272 [−20,251, −4,223] |
−17,533 [−18,929, −15,825] |
−41,803 [−51,447, −32,247] |
Abbreviations: ASCVD, atherosclerotic cardiovascular disease. BP, blood pressure. JNC-8, Eighth Joint National Committee. SPRINT, Systolic Blood Pressure Intervention Trial.
Absolute event rates represent the approximate per-year change in the number of events observed in the United States in a typical year with the implementation of each BP treatment strategy compared to usual care. Differences were estimated by tabulating the difference in the number of each event (dementia, ASCVD events, MI, stroke) between simulated subpopulations (e.g., the number of events in White people with a JNC-8-based BP strategy minus the number of events in White people with a usual care strategy) and then normalizing the event number to the size of the subpopulation of adults in that subpopulation in the United States. Relative effects were determined by dividing the absolute reduction in the number of events by the total number of events in the usual care group. In the model structure, more intense BP treatment results in a slightly lower per year risk of dementia via improved global cognition. More intense BP treatment also results, through ASCVD event prevention, in slightly longer life expectancy. The incidence of dementia increases markedly in older adults. The paradoxical increase in dementia, then, reflects that the latter effect (survival to older ages resulting in increased dementia incidence) outweighs the former effect (reduced dementia risk via improved BP control).
The SPRINT-based and JNC-8-based BP treatment strategies would slightly reduce the disparity in dementia cases between the Black and White populations because the White population would have more dementia cases than the Black population (Table 3). Conversely, both intensive BP treatment strategies reduced the Black-White disparities in ASCVD events and stroke, as the Black population had a larger relative reduction of ASCVD events than the White population with more intensive BP treatment strategies than usual care. The SPRINT-based BP treatment strategy reduced the Black-White ASCVD disparities more than the JNC-8-based BP treatment strategy (e.g., a relative reduction of 24% vs. 12% in stroke disparities). Compared to usual care, the SPRINT-based and JNC-8-based BP treatment strategies produced small and similar decreases in the age- and sex-standardized mortality rate in the Black and White populations but did not change the mortality disparity between the two groups.
Table 3.
Differences in Age and Sex Standardized Event Rates between Black and White People with Each Blood Pressure Treatment Strategy versus Usual Care in the Overall US Population.
| JNC-8-based BP Treatment Strategy versus Usual Care | |||
|---|---|---|---|
| Black-White Difference | Change in Black- White Difference with JNC-8-based BP Treatment Strategy vs. Usual Care |
||
| Usual Care | JNC-8-based BP Treatment Strategy |
Absolute Change | |
| Dementia | 212.5 [214.8, 209.8] |
212.9 [215.3, 209.8] |
0.5 [−4.0, 4.1] |
| ASCVD event | 32.0 [37.1, 27.0] |
15.1 [20.3, 10.5] |
−16.9 [−23.2, −9.8] |
| Myocardial infarction | −63.4 [−59.8, −67.1] |
−69.1 [−65.5, −72.6] |
5.6 [0.2, 10.9,] |
| Stroke | 95.4 [97.9, 92.4] |
84.2 [86.4, 81.6] |
−11.3 [−15.6, −7.0] |
| All-cause mortality | 29.6 [36.2, 23.8] |
16.9 [22.7, 9.9] |
−12.7 [−19.1, −3.3] |
| SPRINT-based BP Treatment Strategy versus Usual Care | |||
| Black-White Difference | Change in Black- White Difference with SPRINT-based BP Treatment Strategy vs. Usual Care |
||
| Usual Care | SPRINT-based BP Treatment Strategy |
Absolute Change | |
| Dementia | 212.5 [214.8, 209.8] |
214.9 [217.4, 211.8] |
2.4 [−1.0, 5.8] |
| ASCVD event | 32.0 [37.1, 27.0] |
0.0 [3.9, −3.9] |
−32.0 [−38.5, −25.8] |
| Myocardial infarction | −63.4 [−59.8, −67.1] |
−72.3 [−69.3, −75.4] |
8.9 [4.0, 13.8] |
| Stroke | 95.4 [97.9, 92.4] |
72.3 [74.8, 70.0] |
−23.1 [−26.8, −19.4] |
| All-cause mortality | 29.6 [36.2, 23.8] |
17.8 [23.7, 11.7] |
−11.7 [−20.9, −3.1] |
Abbreviations: ASCVD, atherosclerotic cardiovascular disease. BP, blood pressure. JNC-8, Eighth Joint National Committee. SPRINT, Systolic Blood Pressure Intervention Trial.
Optimal brain health is defined as without stroke, dementia, or cognitive impairment.
This table summarizes age-sex standardized differences in the rates of dementia, ASCVD events, myocardial infarction, stroke, and dementia. The two left columns represent the differences in age and sex standardized event rates between Black and White people (age and sex standardized rate in Black adults minus the age and sex standardized rate in White adults) with usual care, a JNC-8-based BP treatment strategy (top set of rows) and a SPRINT-based BP treatment strategy (bottom set of rows). Positive values reflect a higher rate of events in Black people compared to White people, and negative values represent a higher rate of events in White people compared to Black people. The 3rd column (Absolute Change) represents the absolute change in the Black-White difference in event rates with the implementation of each BP treatment strategy compared to usual care (e.g., the magnitude of the Black-White difference with a JNC-8-based BP treatment strategy minus the magnitude of the Black-White difference with usual care). Thus, negative values represent a reduction in the absolute Black-White difference in event rates, and positive values represent an increase in the absolute Black-White difference in event rates.
Changes in Outcomes for Black and White Hypertensive Individuals with Different BP Treatment Strategies
In those with hypertension, Black people, compared to White people, have fewer years lived overall (−3.7 years [−3.8. −3.5]), without dementia (−5.1 years [−5.2, −4.9]), without ASCVD (−3.8 years [−4.0, −3.6]), without MI (−3.1 years [−3.3, −2.9]), without stroke (−4.4 years [−4.6, −4.2]) and with optimal breath health (−3.1 years [−3.2, −2.9]) under usual care (Table 4). The SPRINT-based and JNC-8-based BP treatment strategies would modestly reduce the magnitude of these disparities. For example, the absolute reduction in the Black-White disparity in overall survival is 0.11 years vs. 0.04 years with a SPRINT-based vs. JNC-8-based BP treatment strategy. Reductions in the Black-White disparity in survival without ASCVD events (absolute reduction, 0.36 years [0.29-0.39]; 9.3% relative reduction) and without dementia (0.37 years [0.40, 0.34]; 8.3% relative reduction) were the largest under the SPRINT-based BP treatment strategy.
Table 4.
Differences in Age- and Sex-Standardized Years of Life Lived in Total, Without Health Conditions (Dementia, ASCVD event, Myocardial Infarction, Stroke) and With Optimal Brain Health among US Individuals with Hypertension with Each BP Treatment Strategy Compared to Usual Care by Race.
| JNC-8-based BP Treatment Strategy versus Usual Care | |||
|---|---|---|---|
| Additional Years Lived |
Black-White Difference with Usual Care |
Black-White Difference with JNC- 8-based BP Treatment Strategy |
Absolute Reduction of Black-White Difference with JNC-8- based BP Treatment Strategy vs. Usual Care |
| Total | −3.7 [−3.8, −3.5] |
−3.6 [−3.8, −3.5] |
0.04 [−0.19, 0.28] |
| Without dementia | −5.1 [−5.2, −4.9] |
−5.0 [−5.1, −4.8] |
0.16 [−0.09, 0.37] |
| Without ASCVD event | −3.8 [−4.0, −3.6] |
−3.7 [−3.9, −3.5] |
0.11 [−0.20, 0.36] |
| Without myocardial infarction | −3.1 [−3.3, −2.9] |
−3.1 [−3.3, −2.9] |
0.09 [−0.09, 0.34] |
| Without stroke | −4.4 [−4.6, −4.2] |
−4.3 [−4.4, −4.1] |
0.01 [−0.33, 0.25] |
| With optimal brain health * | −3.1 [−3.2, −2.9] |
−3.0 [−3.1, −2.8] |
0.08 [−0.13, 0.32] |
| SPRINT-based BP Treatment Strategy versus Usual Care | |||
| Additional Years Lived |
Black-White Difference with Usual Care |
Black-White Difference with SPRINT-based BP Treatment Strategy |
Absolute Reduction of Black-White Differenc with SPRINT- based BP Treatment Strategy vs. Usual Care |
| Total | −3.7 [−3.8, −3.5] |
−3.6 [−3.7, −3.4] |
0.11 [−0.13, 0.31] |
| Without dementia | −5.1 [−5.2, −4.9] |
−5.0 [−5.2, −4.9] |
0.37 [0.14, 0.59] |
| Without ASCVD event | −3.8 [−4.0, −3.6] |
−3.5 [−3.6, −3.3] |
0.36 [0.10, 0.59] |
| Without myocardial infarction | −3.1 [−3.3, −2.9] |
−3.0 [−3.2, −2.8] |
0.05 [−0.19, 0.22] |
| Without stroke | −4.4 [−4.6, −4.2] |
−4.1 [−4.2, −3.9] |
0.12 [−0.14, 0.35] |
| With optimal brain health * | −3.1 [−3.2, −2.9] |
−3.1 [−3.3, −2.9] |
−0.03 [−0.28, 0.17] |
Abbreviations: ASCVD, atherosclerotic cardiovascular disease. BP, blood pressure. JNC-8, Eighth Joint National Committee. SPRINT, Systolic Blood Pressure Intervention Trial.
Optimal brain health is defined as without dementia, cognitive impairment, or stroke.
This table summarizes the Black-White disparities in age-sex standardized life years lived in each state with usual care (leftmost column) and each BP treatment strategy (second column from the left, JNC-8 in top rows, SPRINT in bottom rows). The rightmost columns summarize differences in disparities comparing each BP treatment strategy to usual care. The 3rd column (absolute difference) represents the absolute change in years in the Black-White disparity with implementing each BP treatment strategy. A positive numeric result reflects a reduction in the magnitude of the disparity. These values were calculated by subtracting the Black-White disparity in life years in each BP treatment strategy from the Black-White disparity in life years with usual care.
Black individuals with hypertension gained slightly more age- and sex-standardized dementia-free years than White individuals from SPRINT-based and JNC-8-based BP treatment strategies compared to usual care (Table 5). Among hypertensive individuals, the SPRINT-based and JNC-8-based BP treatment strategies, compared to usual care, produced greater gains in survival, stroke-free years, MI-free years, and years in optimal brain health for Black individuals than White individuals (Table 5). For all outcomes except dementia, gains were larger with the SPRINT-based BP treatment strategy than with the JNC-8-based strategy. Black individuals’ simulated gains from more intensive BP treatment strategies were greater for stroke-free years than MI-free years. For example, Black individuals, compared with White individuals, gained more years of stroke-free life (0.83 years vs. 0.46 years) than years of MI-free life (0.59 vs. 0.46 years) with a SPRINT-based BP treatment strategy compared to usual care.
Table 5.
Age and Sex Standardized Absolute Changes in Years of Life Lived in Total, Without Health Conditions (Dementia, ASCVD event, Myocardial Infarction, Stroke) and With Optimal Brain Health among US Individuals with Hypertension with Each BP Treatment Strategy Compared to Usual Care by Race.
| Hypertensive Individuals Treated with a JNC-8- based BP Treatment Strategy versus Usual Care |
Hypertensive Individuals Treated with a SPRINT- based BP Treatment Strategy versus Usual Care |
|||
|---|---|---|---|---|
| Additional Years Lived | White | Black | White | Black |
| Total | 0.2 [0.045, 0.39] |
0.25 [0.098, 0.38] |
0.23 [0.035, 0.44] |
0.33 [0.21, 0.47] |
| Without dementia | 0.1 [−0.09, 0.28] |
0.2 [0.073, 0.3] |
0.11 [−0.047, 0.28] |
0.16 [0.058, 0.27] |
| Without ASCVD event | 0.42 [0.22, 0.61] |
0.53 [0.37, 0.68] |
0.64 [0.43, 0.84] |
1.0 [0.86, 1.2] |
| Without myocardial infarction | 0.33 [0.14, 0.53] |
0.34 [0.16, 0.48] |
0.46 [0.28, 0.71] |
0.59 [0.43, 0.72] |
| Without stroke | 0.32 [0.12, 0.52] |
0.48 [0.34, 0.63] |
0.46 [0.27, 0.66] |
0.83 [0.7, 0.97] |
| In optimal brain health * | 0.25 [0.061, 0.43] |
0.33 [0.18, 0.45] |
0.44 [0.29, 0.64] |
0.41 [0.3, 0.54] |
Abbreviations: ASCVD, atherosclerotic cardiovascular disease. BP, blood pressure. JNC-8, Eighth Joint National Committee. SPRINT, Systolic Blood Pressure Intervention Trial.
Optimal brain health is defined as without dementia, cognitive impairment, or stroke.
Each cell value represents the mean of the additional age and sex standardized years lived with each treatment strategy (JNC-8 in the leftmost columns, SPRINT in the rightmost columns) compared to usual care for White and Black adults with hypertension. Cell values were calculated by determining the distribution of life years for each outcome within each population (treatment strategy and race combination), subtracting the life years in each state in usual care and calculating the mean of the difference.
DISCUSSION
In this nationally representative microsimulation, the population-level disparity in dementia between US Black and White people did not narrow with the SPRINT-based and JNC-8-based BP treatment strategies. However, the SPRINT- and JNC-8-based BP treatment strategies were more effective than usual care in reducing population-level disparities in ASCVD events, particularly stroke, between US Black and White people with larger effects for the SPRINT-based treatment strategy. Intensive BP control also produced gains in survival, stroke-free years, MI-free years, and years in optimal brain health (by reducing stroke) for Black individuals than White individuals, but smaller gains in dementia-free years. These relatively modest differences in outcome disparity reduction are partially due to Black individuals still having higher SBP levels than White individuals with SPRINT-based and JNC-8-based BP treatment strategies.
To our knowledge, no modeling study has compared the impact of intensive vs. usual care BP treatment strategies on dementia in Black and White adults. Our results are consistent with a UK simulation modeling study showing that lower hypertension prevalence would lead to a modest reduction in deaths, a small increase in dementia cases, no significant effect on disability burden, and a modest gain in healthy life expectancy at age 65.27 Our results also agree with a cross-sectional analysis showing that the SPRINT-based BP treatment strategy reduced racial disparities in BP control in the trial.28 A prior study 29 found that US Black people had a similar relative risk reduction and greater absolute risk reduction in mortality than White people with a SPRINT BP treatment strategy than usual care. The study,29 compared with our study, estimated more deaths averted for Black people (8,800 vs. ~2,800) and White people (88,900 vs. ~8,600). Compared with our study, a prior microsimulation study estimated similar averted deaths and more averted ASCVD events with a JNC-8-based BP treatment strategy than usual care.30 Numerous differences in assumptions and methods between the studies may contribute to these differences, including different treatment approaches if low DBP, accounting for baseline medications, capping BP treatment at four medications, different assumptions linking BP lowering to mortality (in our study, BP linked to mortality through ASCVD event prevention), and different strategies for competing mortality. Our effects assume that BP lowering has a causal effect on dementia, reflecting the size of the effect of BP lowering on cognition in the best available observational data. However, our estimates will be inaccurate if the true effect size is different (i.e., the trial-based effect is smaller or greater than the observational effect).
Our dementia results do not detract from the substantial benefits of more intensive BP control for individuals treated differently under the two BP treatment strategies. Intensive BP control would help these individuals live longer and have fewer years living with stroke, MI, and, to a lesser extent, dementia, with greater benefits for Black individuals than White individuals. By increasing disease-free survival, intensive BP control would likely result in a compression of morbidity.31 However, the overall population-level effects of the SPRINT-based and JNC-8-based BP treatment strategies, even with optimal implementation, are estimated to be small; however, because a small proportion of individuals met the criteria for more intensive BP treatment with the SPRINT-based or the JNC-8-based than with usual care. Black people are less likely to meet the SPRINT eligibility criteria than White people because of age <50 with hypertension, diabetes, stroke, heart failure, and severe chronic kidney disease.32 Also, the effect of BP lowering on dementia is small.27, 33 Black individuals, compared with White individuals, gained more years of stroke-free life than years of MI-free life with more intensive BP treatment compared to usual care, reflecting that BP is a stronger risk factor for stroke than MI.
Our study’s strengths include estimating the impact of the population-level implementation of BP treatment strategies on cognitive outcomes, including dementia and GCP. We created and used an individual-level simulation model that includes both dementia and ASCVD events, applies to the US population, and includes all adults, not only those with diabetes.34 All individuals’ attributes in the simulation were initially established using real-world data and then updated over time using a series of regression models. Treatment algorithms were implemented and then validated against external standards.9 When trial evidence was unavailable, IPD meta-analyses of well-characterized US cohort studies were used to estimate effect sizes.
The results of our simulation model-based analysis are limited by the assumptions and input data. Although trial evidence is limited in some areas (e.g., the effect of BP treatment on global cognition), we tried to mitigate this limitation by estimating effects using IPD meta-analyses of US cardiovascular cohort studies.6 We assumed optimal and equitable implementation of the SPRINT- and JNC-8-based BP treatment strategies. The model structure assumes linear relationships between BP and mortality because the potential J-shaped curve relating diastolic BP to mortality might not be causal and exist in the SPRINT trial. The microsimulation model was based on the best estimates of the age-adjusted dementia incidence in Black and White adults. Although suboptimal precision of dementia incidence could bias results toward the null, incident dementia was estimated using IPD data meta-analyses of US cohort studies. Dementia is complex and incompletely understood. This simulation models the pathway between BP control, cognition, and dementia. Other potential pathways (e.g., the direct effect of stroke on dementia, genetic susceptibility, and head injury) are not included. These cohorts measured incident dementia by physician adjudication using standard diagnostic criteria, study-specific protocols, and all available data, including in-person neuropsychological and neurologic assessments, telephone interviews (participant or informant), brain imaging, and medical record review. While the model might overestimate dementia incidence, this overestimation should not affect the relative race-dementia and BP-dementia relationships.
Although the SPRINT trial measured expert-adjudicated MCI, our simulation study estimated cognitive impairment because some cohort studies in the pooled cardiovascular cohort studies did not measure MCI. The simulation model does not separately estimate the direct effect of stroke on dementia risk or the effect of BP lowering on cerebral small vessel disease. Loss to follow-up after stroke might underestimate dementia risk. We used the observational pooled cardiovascular cohort studies to estimate the effect of BP lowering on global cognition and dementia because direct trial evidence is limited and heterogeneous.35 The “true” effect sizes of BP-lowering on global cognition and dementia are unknown. If the actual effect of BP lowering on cognitive impairment or dementia equals the effect size found in SPRINT MIND (relative risk reduction of 15%) or a legacy effect of BP lowering on MCI and dementia exists, our approach might have underestimated the effect of BP lowering on dementia. The mean SBP achieved with the SPRINT-based intensive BP protocol was higher in the simulation than the SPRINT trial because of differences in BP measurement and patient populations between the simulation and the SPRINT trial (at one year, SBP was 139.3 mm Hg in Black adults and 137.9 mm Hg in White adults in the simulation vs. 122.6 mm Hg in Black participants and 121.8 mm Hg in White participants in the SPRINT trial).36 As described in the Methods, the simulation assumed that SBP measurements in NHANES and the pooled cardiovascular cohort studies were systematically 6 mm Hg higher than they would have been in the SPRINT trial;24 so, the simulation’s effective SBP treatment target was 126 mm Hg. Also, Black individuals had higher mean SBP levels at baseline in the simulated population than in the SPRINT trial (~142.5 mm Hg vs. 139.5 mm Hg).36 The simulation included individuals with higher baseline SBPs than the SPRINT trial (>180 mm Hg vs. <180 mm Hg). The simulation had a maximum of four BP medications, but the SPRINT trial did not. Finally, and perhaps most importantly, a relatively small proportion of individuals, especially Black individuals, in the simulated population met the eligibility criteria for SPRINT and BP treatment intensification (i.e., they had hypertension, high baseline ASCVD risk, and a current BP above treatment thresholds and were not already taking a maximum number of BP medications), consistent with prior research.32, 37 A prior study showed that, in the overall US population, only 4.8% of Black adults and 9.0% of White adults would meet the SPRINT eligibility criteria.32 Relatively few US adults, especially Black adults, would be treated more intensively with optimal implementation of a SPRINT-based BP treatment strategy.
From 1999 to 2018, disparities in hypertension between US Black and White adults increased significantly.38 Our results suggest that tighter population-level BP control based on the SPRINT and JNC-8-based BP treatment strategies are unlikely to reduce population-level disparities in dementia between US Black and White people because the likely effect of BP-lowering on dementia risk is small,33 lowering BP helps people live longer, and dementia risk soars at older ages, where the majority of improved life expectancy accrues and BP differences between Black and white adults lessen. Alternative population-level strategies to prevent dementia and reduce dementia disparities between Black and White people are needed.
CONCLUSION
Our results suggest that tighter population-level BP control reduces population-level disparities in stroke risk but not dementia risk between US Black and White adults.
ACKNOWLEDGEMENTS
Funding/Support
This research project is supported by grant R01NS102715 from the National Institute of Neurological Disorders and Stroke (NINDS), National Institutes of Health (NIH), Department of Health and Human Service. The NINDS was not involved in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NINDS or NIH. Additional funding was provided by the National Institute of Aging (NIA) R01AG051827 (Levine), NIA Claude Pepper Center P30AG024824 (Gałecki), NIA K01AG050699 (Gross), and NIA Michigan Alzheimer's Disease Research Center P30AG053760 (Giordani).
Cohort Funding/Support
The Atherosclerosis Risk in Communities (ARIC) Study has been funded in whole or in part with federal funds under NHLBI 75N92022D00001, 75N92022D00002, 75N92022D00003, 75N92022D00004, and 75N92022D00005. The ARIC Neurocognitive Study is supported by grant numbers U01HL096812, U01HL096814, U01HL096899, U01HL096902, and U01HL096917 from the NHLBI, NINDS, NIA, and National Institute on Deafness and Other Communication Disorders (NIDCD).
The Coronary Artery Risk Development in Young Adults (CARDIA) Study is conducted and supported by the NHLBI in collaboration with the University of Alabama at Birmingham (HHSN268201800005I and HHSN268201800007I), Northwestern University (HHSN268201800003I), University of Minnesota (HHSN268201800006I), and Kaiser Foundation Research Institute (HHSN268201800004I). A neurocognitive ancillary study is funded by NIA R01AG063887.
The Cardiovascular Health Study (CHS) was supported under NHLBI contracts HHSN268201200036C, HHSN268200800007C, HHSN268201800001C, N01HC55222, N01HC85079, N01HC85080, N01HC85081, N01HC85082, N01HC85083, N01HC85086, and 75N92021D00006, and grants U01HL080295, U01HL130114, and R01HL094555, with additional contribution from the NINDS. Additional support was provided by NIA R01AG023629, R01AG15928, and R01AG20098.
The Framingham Heart Study (FHS) is a project of the NHLBI and Boston University School of Medicine. This project has been funded in whole or in part with federal funds under NHLBI contracts N01HC25195, HHSN268201500001I, and 75N92019D00031; NINDS R01NS017950; and NIA R01AG016495, RF1AG059421, R01AG033193, R01AG008122, and R01AG054076.
The Northern Manhattan Stroke (NOMAS) Study has been funded at least in part with federal funds under NINDS R01NS29993.
The Multi-Ethnic Study of Atherosclerosis (MESA) was supported under NHLBI contracts 75N92020D00001, HHSN268201500003I, N01-HC-95159, 75N92020D00005, N01-HC-95160, 75N92020D00002, N01-HC-95161, 75N92020D00003, N01-HC-95162, 75N92020D00006, N01-HC-95163, 75N92020D00004, N01-HC-95164, 75N92020D00007, N01-HC-95165, N01-HC-95166, N01-HC-95167, N01-HC-95168, and N01-HC-95169, and National Center for Advancing Translational Sciences (NCATS) grants UL1-TR-000040, UL1-TR-001079, and UL1-TR-001420. Cognitive testing at Exam 6 in MESA has been funded by NHLBI grants R01HL127659 and R01AG054069.
Footnotes
CONFLICT OF INTEREST
Deborah Levine is an Associate Editor of the Journal of Alzheimer’s Disease, but was not involved in the peer-review process of this article nor had access to any information regarding its peer-review. The authors have no other relevant conflicts of interest.
DATA AVAILABILITY
NHANES data are publicly available. The model’s code and input parameters are publicly available at https://github.com/jburke5/microsim.
REFERENCES
- 1.Manly JJ, Jones RN, Langa KM, Ryan LH, Levine DA, McCammon R, et al. Estimating the prevalence of dementia and mild cognitive impairment in the US: The 2016 Health and Retirement Study Harmonized Cognitive Assessment Protocol Project. JAMA Neurol. 2022; 79(12):1242–1249.. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.The Alzheimer's Association. 2023 Alzheimer's disease facts and figures. Alzheimers Dement. 2023;19(4):1598–695. [DOI] [PubMed] [Google Scholar]
- 3.United States Department of Health and Human Services. National plan to address Alzheimer's disease: 2022 update. In: Department of Health and Human Services; editor. Washington, DC: US Department of Health and Human Services; 2022. [Google Scholar]
- 4.Lee M, Whitsel E, Avery C, Hughes TM, Griswold ME, Sedaghat S, et al. Variation in population attributable fraction of dementia associated with potentially modifiable risk factors by race and ethnicity in the US. JAMA Network Open. 2022;5(7):e2219672. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Flack JM, Sica DA, Bakris G, Brown AL, Ferdinand KC, Grimm RH Jr., et al. Management of high blood pressure in Blacks: An update of the International Society on Hypertension in Blacks consensus statement. Hypertension. 2010;56(5):780–800. [DOI] [PubMed] [Google Scholar]
- 6.Levine DA, Gross AL, Briceno EM, Tilton N, Kabeto MU, Hingtgen SM, et al. Association between blood pressure and later-life cognition among Black and white individuals. JAMA Neurology. 2020;77(7):810–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Wright JT Jr., Williamson JD, Whelton PK, Snyder JK, Sink KM, Rocco MV, et al. A randomized trial of intensive versus standard blood-pressure control. N Engl J Med. 2015;373(22):2103–16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.SPRINT MIND Investigators Research Group, Williamson JD, Pajewski NM, Auchus AP, Bryan RN, Chelune G, et al. Effect of intensive vs standard blood pressure control on probable dementia: A randomized clinical trial. JAMA. 2019;321(6):553–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Burke JF, Copeland LL, Sussman JB, et al. Development and validation of the Michigan Chronic Disease Simulation Model (MICROSIM). PLoS One. 2024;19(5):e0300005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.James PA, Oparil S, Carter BL, Cushman WC, Dennison-Himmelfarb C, Handler J, et al. 2014 evidence-based guideline for the management of high blood pressure in adults: Report from the panel members appointed to the Eighth Joint National Committee (JNC 8). JAMA. 2014;311(5):507–20. [DOI] [PubMed] [Google Scholar]
- 11.The Atherosclerosis Risk in Communities (ARIC) study: Design and objectives. The ARIC investigators. American journal of epidemiology. 1989;129(4):687–702. [PubMed] [Google Scholar]
- 12.Friedman GD, Cutter GR, Donahue RP, Hughes GH, Hulley SB, Jacobs DR Jr., et al. CARDIA: Study design, recruitment, and some characteristics of the examined subjects. Journal of Clinical Epidemiology. 1988;41(11):1105–16. [DOI] [PubMed] [Google Scholar]
- 13.Fried LP, Borhani NO, Enright P, Furberg CD, Gardin JM, Kronmal RA, et al. The Cardiovascular Health Study (CHS): Design and rationale. Annals of Epidemiology. 1991;1(3):263–76. [DOI] [PubMed] [Google Scholar]
- 14.Feinleib M, Kannel WB, Garrison RJ, McNamara PM, Castelli WP. The Framingham Offspring Study (FOS): Design and preliminary data. Preventive Medicine. 1975;4(4):518–25. [DOI] [PubMed] [Google Scholar]
- 15.Sacco RL, Boden-Albala B, Gan R, Chen X, Kargman DE, Shea S, et al. Stroke incidence among white, black, and hispanic residents of an urban community: The Northern Manhattan Stroke Study (NOMAS). American Journal of Epidemiology. 1998;147(3):259–68. [DOI] [PubMed] [Google Scholar]
- 16.Yadlowsky S, Hayward RA, Sussman JB, McClelland RL, Min YI, Basu S. Clinical implications of revised pooled cohort equations for estimating atherosclerotic cardiovascular disease risk. Ann Intern Med. 2018;169(1):20–9. [DOI] [PubMed] [Google Scholar]
- 17.Brookmeyer R, Gray S, Kawas C. Projections of Alzheimer's disease in the United States and the public health impact of delaying disease onset. Am J Public Health. 1998;88(9):1337–42. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Brookmeyer R, Johnson E, Ziegler-Graham K, Arrighi HM. Forecasting the global burden of Alzheimer's disease. Alzheimer's and Dementia. 2007;3(3):186–91. [DOI] [PubMed] [Google Scholar]
- 19.Gillett SR, Thacker EL, Letter AJ, McClure LA, Wadley VG, Unverzagt FW, et al. Correlates of incident cognitive impairment in the Reasons for Geographic and Racial Differences in Stroke (REGARDS) study. Clin Neuropsychol. 2015;29(4):466–86. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Ivnik RJ, Smith GE, Petersen RC, Boeve BF, Kokmen E, Tangalos EG. Diagnostic accuracy of four approaches to interpreting neuropsychological test data. Neuropsychology. 2000;14(2):163–77. [DOI] [PubMed] [Google Scholar]
- 21.Blood pressure-lowering treatment based on cardiovascular risk: A meta-analysis of individual patient data. Lancet. 2014;384(9943):591–8. [DOI] [PubMed] [Google Scholar]
- 22.Basu S, Sussman JB, Hayward RA. Detecting heterogeneous treatment effects to guide personalized blood pressure treatment: A modeling study of randomized clinical trials. Ann Intern Med. 2017;166(5):354–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Carey RM, Calhoun DA, Bakris GL, Brook RD, Daugherty SL, Dennison-Himmelfarb CR, et al. Resistant hypertension: Detection, evaluation, and management: A scientific statement from the American Heart Association. Hypertension. 2018;72(5):e53–e90. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Drawz PE, Agarwal A, Dwyer JP, Horwitz E, Lash J, Lenoir K, et al. Concordance between blood pressure in the systolic blood pressure intervention trial and in routine clinical practice. JAMA Internal Medicine. 2020;180(12):1655–63. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Levey AS, Stevens LA, Schmid CH, Zhang YL, Castro AF 3rd, Feldman HI, et al. A new equation to estimate glomerular filtration rate. Annals of Internal Medicine. 2009;150(9):604–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Groot Koerkamp B, Weinstein MC, Stijnen T, Heijenbrok-Kal MH, Hunink MG. Uncertainty and patient heterogeneity in medical decision models. Medical Decision Making. 2010;30(2):194–205. [DOI] [PubMed] [Google Scholar]
- 27.Chen Y, Araghi M, Bandosz P, Shipley MJ, Ahmadi-Abhari S, Lobanov-Rostovsky S, et al. Impact of hypertension prevalence trend on mortality and burdens of dementia and disability in England and Wales to 2060: A simulation modelling study. The Lancet Healthy Longevity. 2023;4(9):e470–e7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Zheutlin AR, Mondesir FL, Derington CG, King JB, Zhang C, Cohen JB, et al. Analysis of therapeutic inertia and race and ethnicity in the Systolic Blood Pressure Intervention Trial: A secondary analysis of a randomized clinical trial. JAMA Network Open. 2022;5(1):e2143001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Bress AP, Kramer H, Khatib R, Beddhu S, Cheung AK, Hess R, et al. Potential deaths averted and serious adverse events incurred from adoption of the SPRINT (Systolic Blood Pressure Intervention Trial) intensive blood pressure regimen in the United States: Projections from NHANES (National Health and Nutrition Examination Survey). Circulation. 2017;135(17):1617–28. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Bress AP, Bellows BK, King JB, Hess R, Beddhu S, Zhang Z, et al. Cost-effectiveness of intensive versus standard blood-pressure control. New England Journal of Medicine. 2017;377(8):745–55. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Ryan J, Ernst ME. The quid pro quo of blood pressure control and dementia. The Lancet Healthy Longevity. 2023;4(9):e444–e5. [DOI] [PubMed] [Google Scholar]
- 32.Bress AP, Tanner RM, Hess R, Colantonio LD, Shimbo D, Muntner P. Generalizability of SPRINT results to the US adult population. Journal of the American College of Cardiology. 2016;67(5):463–72. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Peters R, Xu Y, Fitzgerald O, Aung HL, Beckett N, Bulpitt C, et al. Blood pressure lowering and prevention of dementia: An individual patient data meta-analysis. European Heart Journal. 2022;43(48):4980–90. [DOI] [PubMed] [Google Scholar]
- 34.Breeze P, Thomas C, Thokala P, Lafortune L, Brayne C, Brennan A. The impact of including costs and outcomes of dementia in a health economic model to evaluate lifestyle interventions to prevent diabetes and cardiovascular disease. Med Decis Making. 2020;40(7):912–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Hughes D, Judge C, Murphy R, Loughlin E, Costello M, Whiteley W, et al. Association of blood pressure lowering with incident dementia or cognitive impairment: A systematic review and meta-analysis. JAMA. 2020;323(19):1934–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Still CH, Rodriguez CJ, Wright JT Jr., Craven TE, Bress AP, Chertow GM, et al. Clinical outcomes by race and ethnicity in the Systolic Blood Pressure Intervention Trial (SPRINT): A randomized clinical trial. Am J Hypertens. 2017;31(1):97–107. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Burke JF, Sussman JB, Yaffe K, Hayward RA, Giordani BJ, Galecki AT, et al. Effect of population-level blood pressure treatment strategies on cardiovascular and cognitive outcomes. Circ Cardiovasc Qual Outcomes. 2024;17(6):e010288. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Odlum M, Moise N, Kronish IM, Broadwell P, Alcántara C, Davis NJ, et al. Trends in poor health indicators among Black and Hispanic middle-aged and older adults in the United States, 1999-2018. JAMA Network Open. 2020;3(11):e2025134. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
NHANES data are publicly available. The model’s code and input parameters are publicly available at https://github.com/jburke5/microsim.