Linking perceived stress and cognitive aging: The influence of blood pressure in Black Americans

Abstract

Background

Although previous research suggests that perceptions of stress shape cognitive health outcomes, few studies have explored this association using cognitive domains versus a global measure of cognition, and fewer still have examined blood pressure (BP) as a moderator.

Objective

We explore whether perceived stress shapes cognitive health outcomes and the influence of BP.

Methods

Using the Baltimore Study of Black Aging—Patterns of Cognitive Aging, we applied linear regression models to assess whether perceived stress is associated with domain-specific cognitive changes over time and if BP acts as a moderator.

Results

Findings showed no direct effect of stress on cognitive changes over time. Systolic (F (3346) = 3.00, p = 0.031) and diastolic (F (3346) = 2.63, p = 0.050) BP modified the stress-inductive reasoning relationship but was not related to any other cognitive domains, i.e., working memory, processing speed, verbal memory, and vocabulary at follow-up.

Conclusions

The findings indicate that stress does not have a direct effect on cognitive change and a fragile interaction between stress and BP on inductive reasoning, such that those with high stress and high BP had high inductive reasoning scores, while those with low stress and high BP had low scores. These results are counterintuitive and require further examination, given that high stress may not be beneficial for people with high BP. Thus, additional research that documents whether stressors have an unforeseen protective effect on domain-specific cognitive decline in Black Americans is needed.

Keywords

African Americans Alzheimer's disease cognition hypertension subjective stress

Introduction

The risk of dementia and cognitive decline increases with advancing age,¹ making these neurodegenerative conditions of upmost importance for an aging U.S. population. Moreover, there are stark racial differences in the distribution of these conditions across the population. Specifically, Black Americans share a greater burden of dementia risk compared to other racial and ethnic groups. For example, compared to non-Hispanic Whites, Black Americans on average have lower baseline cognitive test scores, experience cognitive decline at an earlier age,^2,3 and are twice as likely to develop Alzheimer's or other related dementias (ADRD).¹ Yet, genetic risk factors^4,5 do not entirely explain these cognitive and ADRD disparities,⁴ which suggests that other factors such as perceptions of stress may better account for the higher risk of decline and dementia in U.S. Blacks. It is critical to explore these non-genetic risk factors as the aging U.S. population is expected to become more racially and ethnically diverse, particularly, older Black Americans will increase from 9% in 2014 to 12% in 2060.⁶

Stress and cognition

Repeated stress exposure leads to chronic activation of the hypothalamic-pituitary-adrenal (HPA) axis and high amounts of glucocorticoids (cortisol) are secreted.⁷ This prolonged glucocorticoid (cortisol) exposure and HPA-axis dysregulation triggers changes in the brain that may lead towards the development and progression of neurodegenerative pathologies such as Alzheimer's disease.⁸ Chronic stress and elevated glucocorticoids are also linked to atrophy of key brain structures essential for learning and memory (i.e., frontal lobes, hippocampus, and prefrontal cortex).⁸ Thus, cognitive processes that are heavily dependent on the frontal lobes (promotes reasoning, learning, creativity, decision-making, and executive control, i.e., the selection and coordination of goal-directed behaviors),⁹ hippocampus (declarative or episodic, spatial and contextual memory),¹⁰ and prefrontal cortex (supports higher order cognitive functions, i.e., executive functioning or working memory, thinking, reasoning, planning, and decision-making), may be particularly vulnerable to stress (via the damaging actions of the glucocorticoids that are secreted when exposed to chronic stress).¹⁰ Hence, greater exposure to stress is likely associated with cognitive decline and ADRD risk in Black Americans and may further explain racial disparities.^11,12

Research shows that compared to White Americans, Black Americans are disproportionately exposed to stress, including reporting higher levels of perceived stress,¹³ experiencing more early-life adversity,¹⁴ and greater accumulated stressors across the life course.^12,15 This higher level of stress exposure has likewise been shown to negatively impact cognitive functioning in older Black Americans.^11,13,16 Perceptions of stress (as measured by the Perceived Stress Scale; PSS), in particular, have been found to contribute to worse cognitive health and greater dementia risk in Black Americans.^11,13,16 The PSS is a widely used instrument for global stress assessment. It captures both the occurrence of stressful events and one's appraisal of stress in the past month.¹⁷ The appraisal of stress is measured by the frequency of one's feelings (e.g., upset, uncontrollability, nervousness) and thoughts when life is generally seen as stressful during the past 30 days. Higher perceived stress (4-item PSS) has been linked to declines in global cognition, episodic memory, and visuospatial abilities, controlling for demographics, vascular factors, and depressive symptoms in a sample of 467 Black Americans aged 61–98 years.¹¹ Higher levels of perceived stress (14-item PSS) have also been associated with more subjective memory complaints, independent of demographic factors in a sample of 579 Black adults aged 58–95 years.¹⁶ Another study reported that greater perceived stress (6-item PSS) was predictive of worse baseline global cognition and greater cognitive decline over approximately 7 years of follow-up, independent of sociodemographic, vascular, health behaviors, and psychosocial factors. The authors of this study used a primarily Black sample (N = 6207; 65.7%) and conducted additional sensitivity analyses for each cognitive domain of global, episodic memory (immediate and delayed recall), and executive functioning (processing speed). Their findings were similar in that higher levels of stress were related to worse global, episodic memory, and executive function at baseline and greater declines over time in both global and episodic memory.¹³ Collectively, these results suggest there is an association between perceptions of stress, as measured by various items on the PSS, and worse cognitive function in Black Americans. Yet, few studies have examined whether blood pressure (BP) modifies this association.

Blood pressure

BP is a known risk factor for cognitive decline and dementia in later life. It rises during stress responses and may influence brain perfusion and neural functioning, which in turn affects cognitive processes.¹⁸ A recent review of 149 studies across 26 countries found midlife (40–65 years) hypertension was significantly associated with worse cognition in later life, particularly for memory, executive functioning, and global cognition.¹⁹ Additionally, exposure to moderate to high levels of chronic stress over time contributes to a greater risk for hypertension in middle-aged Black Americans.²⁰ This research demonstrates that stress has a negative effect on health, particularly for Black Americans.

Yet it's unclear whether BP changes mediate or moderate the stress-cognition link.¹⁸ For instance, research shows that stress can provoke BP responses, and BP reactivity has been linked to cognitive outcomes independent of resting clinical BP.²¹ Specifically, systolic and diastolic BP reactivity has been associated with diminished performance on tests of immediate and delayed verbal memory and executive function (i.e., response inhibition).²¹ These findings suggests that BP reactivity might be an important pathway linking stress and cognition.

Present study

Although previous research^11,13,16 suggests that perceptions of stress shape cognitive health outcomes, few studies have explored this association using cognitive domains versus a global measure of cognition, and fewer still have examined BP as a moderator. The aim of this current study is to determine whether perceived stress is associated with cognitive changes over time and if BP moderates this relationship in a sample of older Black adults (51–96 years). We hypothesized that perceived stress would be inversely associated with domain-specific changes in cognitive functioning (Hypothesis 1). We further hypothesized that the strength of these associations would depend on BP (Hypothesis 2); however, because of the limited research in this area, no specific hypotheses were made about the direction of the moderation effect of BP.

Methods

Participants

The Baltimore Study of Black Aging—Patterns of Cognitive Aging (BSBA-PCA) aimed to identify whether various health and psychosocial factors contribute to individual differences in cognitive functioning in older Black Americans and explore changes in these factors over time. Community-dwelling participants were recruited from 29 predominantly Black/African American senior housing facilities in Baltimore, Maryland. A total of 602 individuals (ages 48–95; M = 69.1 years; SD = 9.8) were interviewed face-to-face at two time points (wave 1 and wave 2), approximately 3 years apart. The wave 1 interviews were conducted between 2005–2007, while the wave 2 interviews of 450 participants (ages 51–96; M = 71.4 years; SD = 9.2) were done between 2008–2010. On average, the face-to-face interviews took 2.5 h during which each subject completed a battery of cognitive tests, 3 BP measurements, and information on sociodemographic, stress and psychosocial factors, and physical and mental health. All subjects provided written informed consent to participate in the BSBA-PCA study. The current study included 450 BSBA-PCA participants who completed both baseline (wave 1) and follow-up (wave 2) interview and testing sessions (IRB approval from Duke University #1610).

Attrition

Between the two waves of data collection, 152 participants were lost due to attrition (i.e., death, refusal, relocation, sickness, and unable to locate). We conducted sensitivity analyses on key demographics and cognitive variables to compare those who remained in the study versus those lost to follow-up. We found no significant differences by age, sex, and education between participants who completed the second wave of data collection and those who dropped out.^22,23 However, there were differences on the cognitive measures in that drop-out participants had significantly lower working memory,t(596) = −1.99, p = 0.047; processing (perceptual) speed, t(598) = −2.64, p = 0.009; verbal memory, t(597) = −2.26, p = 0.024; vocabulary, t(588) = −3.70, p < 0.001; and inductive reasoning, t(595) = −2.37, p = 0.018 scores than those who remained in the study.

Measures

Cognitive domains

Cognitive status was assessed at baseline and follow-up and included 5 domains: working memory, processing speed, verbal memory, vocabulary, and inductive reasoning. Raw scores were used for all cognitive assessments. Below we describe each measure and the scoring norms.

Working memory (one's short-term memory and executive function) was measured using the Operation Span,²⁴ Alpha Span,²⁵ and Backward Digit Span Tasks.²⁶

The Operation Span Task²⁴ tests one's ability to solve arithmetic problems while remembering words (i.e., solving equation word-pairs). Participants listen as a series of equation-word pairs using addition or subtraction problems and a word (e.g., Is 7 + 6 = 13, “YES” “DRUM”) are read aloud. Participants are then instructed to choose “YES” if the equation is true and “NO” if it is false while also remembering the word (e.g., “DRUM”). After selecting the correct answer to several of the arithmetic problems and saying the word aloud, participants are then shown the word “RECALL” and instructed to recall all the target words (i.e., the word presented in each of the arithmetic problems) they remember. This exercise includes 12 trials, with the number of equation word-pairs varying between two to five from trial to trial. Computation span scores reflect the total number of correct responses.

The Alpha Span Task²⁵ is a 14-item test that measures working memory and executive function. Participants are read a list of two to eight words then asked to repeat the list in alphabetical order. Responses were coded as the total number of words correctly recalled, with no penalties for intrusion errors.

The Backwards Digit Span Task²⁶ also measures working memory and executive function and requires participants to repeat backwards a series of numbers that were read aloud. This test includes 14-individual digit strings with two to eight numbers per string. Scores reflect the total number of digits that the participant recalled in the correct (opposite) order. If a participant failed two consecutive attempts, the test was ended.

Perceptual speed (information processing speed) was assessed using the Identical Pictures Test,²⁷ Digit Symbol Substitution Test,²⁶ and Number Comparison Test.²⁷

The Identical Pictures Test²⁷ is a 48-item measure that asks participants to match a picture presented in the left margin to an identical picture shown in a row of five similar pictures. Participants are given 90 s to match as many of the pictures as possible. Scores reflect the total number of correct responses.

The Digit Symbol Substitution Test²⁶ provides participants with a coding scheme of symbols that correspond to the numbers of 1 through 9. Using this provided coding scheme, participants have 90 s to write as many matching symbols as possible below each number generated in random order. Scores reflect the total number of correct responses.

The Number Comparison Test²⁷ includes pairs of numbers ranging from 3 to 13 digits presented side-by-side. The participant is instructed to compare each pair of numbers and indicate the non-identical pairs (e.g., 54732___54742) by placing an “x” between the two strings of digits that are not the same. Participants are given 90 s to complete 48 number comparisons. Scores reflect the total number of items answered correctly.

Verbal memory (one's ability to learn and recall verbal information) was measured using three list learning tasks: Immediate Recall Test,²⁸ Rey Auditory Verbal Learning Test (RAVLT),²⁹ and Hopkins Verbal Learning Test (HVLT).³⁰

The Immediate Recall Test²⁸ requires participants to silently study a list of 20 words for 3.5 min. Participants are then given a 3.5-min free recall period where they write down all the words they can remember. Scores reflect the total number of words recalled correctly.

The RAVLT²⁹ requires participants to be read a list of 15 words once (with a pause of 2 s between each word) followed by a 1-mintue free recall period for a total of 5 trials. Scores reflect the total number of words correctly recalled, with no penalties for intrusion (a word that is written down that is NOT on the list) or perseveration (writing down a correct word more than once) errors.

The HVLT requires participants to be read a list of 12 words (with a pause of 2 s between each word) for a total of 3 trials. Subjects are then asked to recall as many of the words as possible. Scores reflect the total number of correctly recalled words, with no penalties for intrusion or perseveration errors.

Vocabulary (one's verbal ability and knowledge of word meanings) was assessed using parts II and IV of the Vocabulary Test²⁷ and Shipley Institute of Living Verbal Meaning Test.³¹

The Vocabulary Test²⁷ is a 36-item measure testing one's knowledge of word meanings. It is broken into two 18-item subtests that contain a prompt word and five-choice synonyms for that word. Participants are given four minutes per subtest to complete the 18-item synonyms, for a total of eight minutes to complete the full test. The number of correct responses from both subtests are summed to generate a total score.

The Shipley Institute of Living Verbal Meaning Test³¹ is a 40-item vocabulary test that includes four-choice synonyms. Participants are instructed to circle the one word out of four that is most similar to the prompting word. Scores reflect the sum of the total number of correct responses (scored as “1”).

Inductive reasoning (how well a participant can find patterns in strings of numbers, letters, and words) was ascertained using three measures: Number Series Test,³² Letter Series Test,³² and Shipley Institute of Living Scale Abstract Thinking Test.³¹

Number Series Test³² required participants to identify the pattern in a series of numbers (e.g., 10, 11, 12, 13, 14, ___) and then produce the next number (from a list of five possible numbers) that will continue the series (i.e., 15). Participants are given 4.5 min to complete 15 number series, and the number of correct responses (coded as 1) are summed to create a total score.

Letter Series Test³² requires participants to identify the pattern in a series of letters (e.g., cdcdcd___) and then generate the next letter in the series (i.e., c). This timed test allows 4 min to complete 30 letter series. The total score is calculated by summing the number of correct responses.

The Shipley Inductive Reasoning Test³¹ assesses abstract thinking. This test requires participants to identify the pattern in a series of words, numbers, or letters (e.g., North South; up down; high ____), and then fill in the item that would most logically follow (i.e., low). Participants complete 20 words, numbers, or letters series. The number of correct responses (coded as 1) are summed to create a total score.

Perceived stress

Subjective perceptions of stress—in the past 30 days—was measured using a 14-item version of the Perceived Stress Scale (PSS).¹⁷ This classic measure, originally designed in 1983, remains one of the most widely used instruments for global stress assessment. The PSS scale items capture both the occurrence of stressful events and the appraisal of stress by an individual. Representative questions included: “In the last month, how often have you …: ‘been upset because of something that happened unexpectedly’; ‘felt nervous and stressed’; and ‘dealt successfully with irritating life hassles’ (reverse coded)?” Respondents indicate on a five-point Likert-type scale ranging from 1 (never) to 5 (very often) how often these feelings and thoughts have occurred in the last month. The response categories were recoded as 0 (never) to 4 (very often), with a sample range of 0–46. Positive valence items were reverse coded and summed resulting in a continuous measure, with higher scores suggesting greater perceived stress levels. The Cronbach's alpha = 0.85 for the wave 1 PSS used in the present study is similar to the reliability estimates for the original PSS (Cronbach's alpha range = 0.84 to 0.86 and test–retest reliability = 0.55 to 0.85, depending on time to retest).¹⁷

Blood pressure (moderator)

Three readings of orthostatic BP were collected using an oscillometer automated device (A&D model UA- 767).³³ Mean systolic (sample range = 92–230.3 mmHg) and diastolic (sample range = 35–144.7 mmHg) BP values were calculated for each participant and BP was treated as a continuous variable.

Covariates

A self-reported questionnaire was used to collect demographic (age, sex, and education) and health information (health status and depression). Age was measured in years and treated as continuous. Sex was dichotomized, with males as the reference category. Education was ascertained by asking participants to “Circle the number of the highest grade attained and provide the exact number of years of education.” Responses included 0 (none), 1–8 (grade school), 9–12 (high school), 12.5–16 (some college, vocational or trade school), and 17–20 (graduate school). Education was treated as continuous variable using the number of years (sample range 3–20; M = 11.5 years; SD = 2.9). Health conditions were measured using a self-reported count of whether participants had been told by a physician or nurse they had any of the following diseases (yes/no response): diabetes, arthritis, heart attack, angina, circulation problems, asthma, and kidney trouble. Due to the low response rate, a binary variable (0 = no and 1 = yes) representing heart troubles was created by combining participants who reported having a heart attack (n = 64) or angina (n = 47).³⁴ A dichotomous scale with two binary categories (i.e., two or more health conditions versus those with one or none) was created by summing all four health conditions. Other studies^34–36 have used this dichotomous classification to identify Black Americans with comorbidities, high vascular risk factors, and mobility limitations. Blood pressure medication use was measured using two self-reported questions that asked participants about a clinical diagnosis of hypertension (Have you ever been told by a doctor you have HBP? yes/no response) and a follow-up question about medication use for high BP (If yes, are you currently taking medicine for high blood pressure?). Responses included 0 (No), 1 (Did take but not anymore), 2 (Yes, when I feel that I need it), and 3 (Yes, all the time). We combined these two questions to create a composite variable for BP medication use with the following categories: 0 (No High BP/No BP medication use), 1 (Yes HBP/No BP medication use), 2 (Yes HBP/Inconsistent BP medication use), and 3 (Yes HBP/Yes BP medication use).

Depressive symptoms, in the past week, were assessed using the 20-item Center for Epidemiologic Studies Depression scale (CES-D)³⁷ The CES-D is designed to measure the extent to which a person experiences depressive symptoms. This brief well-validated self-report scale is also used as an indicator of negative psychological well-being in older Black Americans.^16,38 Sample items included: “I felt depressed”; “My sleep was restless”; and “I enjoyed life” (reverse coded). Responses were 0 (rarely or never), 1 (sometimes), 2 (frequently), or 3 (most of the time). A continuous measure was created by reverse coding and summing all positive valence items, with higher CES-D scores suggesting worse depressive symptoms. Cronbach's alpha of the CES-D is 0.88.

Statistical analysis

Descriptive statistics, including means and standard deviations for continuous variables and frequencies and percentages for categorical variables, were calculated for the BSBA-PCA sample. Ordinary least squares (OLS) linear regression³⁹ was used to test the independent relationships between perceived stress, cognitive domains, and whether these associations differ by BP, controlling for baseline (wave 1) age, sex, education, health conditions, BP medication use, and cognition. Empirical simulation studies have shown that a mediated effect can be statistically significant even in the absence of a significant total effect.⁴⁰ While these articles primarily address mediation analysis, its core principles are also applicable to moderation analysis, as both are extensions of basic regression models and are often used together in more complex analyses.⁴⁰ Moderation analysis is valuable even when no direct effect of X (predictor) on Y (outcome) is observed, as it can reveal conditional effects that are not apparent in the overall relationship.⁴⁰ Thus, if a direct effect is observed or not, we will test for moderation using an interaction term. This term will test whether BP modifies the impact of perceived stress on cognitive changes, followed by a contour plot analysis to determine if the difference between BP groups is significantly different for different values of perceived stress. We calculated the linear predictions—at specific values of stress (0, 10, 20, 30, and 40), systolic BP at 90 to 210, and diastolic BP at 60 to 130 (using increments of 10)—to examine the association between perceived stress and BP. The contour plot analyses of perceived stress predicting domain-specific changes in cognitive function at many different values of BP allows us to decompose significant interactions, i.e., break down the interaction term into its lower order components.⁴¹ Listwise deletion (complete-case analysis) was used to address missing cases due to the small number of missing values (3%) in those who completed both waves of the study.⁴² All p values < .05 were considered statistically significant. Analyses were conducted using Stata, version 18, software.⁴³

Results

Sample descriptive statistics

The demographic characteristics for wave 1 of the BSBA-PCA sample are shown in Table 1. The average age of participants was 69.1 years (range = 48–95; SD = 9.7). The majority of the sample was female (74.6%), and slightly more than half reported having a high school education (55.8%). More than half the sample reported having two or more health conditions (53.2%), while most participants reported having high BP and taking BP medication all the time (73.9%). The average depressive symptoms and perceived stress scores were 11.2 (SD = 9.3) and 17.8 (SD = 9.8), respectively. The mean systolic BP and diastolic BP readings were 146.2 mmHg (SD = 23.8) and 86.6 mmHg (SD = 12.9), respectively.

Table 1.

Sample characteristics for adult respondents.

Characteristics	Mean (SD)
Demographic
Age (years)	71.43 (9.75)
Female, n (%)	449 (74.58)
Education, n (%)
Grade School	80 (13.29)
High School	336 (55.81)
Some College/Vocational or Trade School	161 (26.74)
Graduate School	24 (3.99)
Health Conditions, n (%)
Diabetes	205 (34.05)
Arthritis	393 (65.28)
Heart trouble	120 (19.93)
Circulation problems	162 (26.91)
Asthma	113 (18.77)
Kidney trouble	54 (8.97)
None or one health condition	282 (46.84)
Two or more health conditions	320 (53.16)
Blood Pressure Medication Use, n (%)
No High BP/No BP medication use	97 (16.11)
Yes HBP/No BP medication use	22 (3.65)
Yes HBP/Inconsistent BP medication use	31 (5.15)
Yes HBP/Yes BP medication use	445 (73.92)
Depressive symptoms^a	13.45 (4.49)
Perceived Stress	19.80 (7.50)
Blood Pressure ^b
Systolic	146.2 (23.81)
Diastolic	86.6 (12.86)
Cognitive Measures
Working memory	138.1 (33.14)
Processing speed	148.5 (28.27)
Verbal memory	147.7 (27.22)
Vocabulary	145.92 (33.36)
Inductive reasoning	132.9 (41.00)

Baltimore Study of Black Aging–Patterns of Cognitive Aging, Wave 1 (N = 602).

SD: standard deviation. ^aDepressive symptoms range = 0–51; Perceived stress range = 0–46.

Systolic range = 82–230.3 mmHg; Diastolic range = 51–136.7 mmHg.

OLS regression by cognitive domains

The main effect of perceived stress on each cognitive domain is displayed in Table 2. Perceived stress was not a significant predictor of cognitive change. This finding is contrary to our first hypothesis that perceived stress would be inversely associated with domain-specific changes in cognitive functioning. In terms of demographic characteristics, age was inversely related to all cognitive domains, except for vocabulary (b = -0.025, p = 0.778) and inductive reasoning (b = -0.179, p = 0.305). While female sex was positively related to working and verbal memory and vocabulary. Education was positively related to most cognitive domains, except for processing speed (b = 0.024, p = 0.945) and verbal memory (b = 0.141, p = 0.699).

Table 2.

Regression analysis for perceived stress predicting cognitive change among African Americans in the Baltimore Study of Black Aging–Patterns of Cognitive Aging.

	Working memory	Processing speed	Verbal memory	Vocabulary	Inductive reasoning
	Model 1 b(SE) n = 436	Model 2 b(SE) n = 439	Model 3b(SE) n = 437	Model 4b(SE)n = 429	Model 5b(SE)n = 436
Perceived Stress	0.045 (0.146)	0.115 (0.133)	0.085 (0.148)	0.092 (0.115)	0.051 (0.218)
Baseline Cognition	0.202*** (0.032)	0.644*** (0.040)	0.446*** (0.038)	0.307*** (0.027)	0.168*** (0.045)
Age	−0.218* (0.113)	−0.231* (0.109)	−0.506*** (0.115)	−0.025 (0.089)	−0.179 (0.174)
Female^a	3.946 (2.315)	1.155 (2.154)	6.092** (2.393)	5.230** (1.841)	0.037 (3.509)
Education	1.395*** (0.374)	0.024 (0.346)	0.141 (0.365)	1.015*** (0.310)	1.544** (0.599)
Two or More Health Conditions^b	−0.545 (2.000)	−2.210 (1.830)	−2.114 (2.049)	−1.412 (1.578)	1.152 (3.009)
High Blood Pressure Medication Use^c
Yes HBP/No BP medication use	6.624 (6.025)	4.522 (5.527)	7.838 (6.187)	−1.627 (4.753)	9.154 (9.070)
Yes HBP/ Inconsistent BP medication use	1.632 (4.852)	−1.984 (4.374)	5.026 (4.983)	−3.238 (3.831)	−11.923 (7.334)
Yes HBP/Yes BP medication use	4.869 (2.704)	−2.312 (2.481)	0.598 (2.768)	−3.025 (2.181)	0.951 (4.100)
Depressive Symptoms	−0.360* (0.152)	−0.102 (0.138)	−0.111 (0.155)	−0.053 (0.123)	−0.266 (0.230)

R²	115.264*** (10.636)	68.377*** (11.292)	109.215*** (11.498)	41.293*** (8.276)	118.586*** (15.883)

b: unstandardized coefficient; SE: standard error.

Reference category is male.

Reference category is one or no health conditions.

Reference category is no high blood pressure and does not use high BP medication.

*p < 0.05 **p < 0.01 ***p < 0.001.

Table 3.

Regression analysis for the moderating effect of blood pressure among African Americans in the Baltimore Study of Black Aging–Patterns of Cognitive Aging.^a

	Working memory		Processing speed		Verbal memory		Vocabulary		Inductive reasoning
	Model 1 b(SE)	Model 2 b(SE)	Model 3 b(SE)	Model 4 b(SE)	Model 5 b(SE)	Model 6 b(SE)	Model 7 b(SE)	Model 8 b(SE)	Model 9 b(SE)	Model 10 b(SE)
Perceived Stress	−0.720 (0.701)	0.191 (0.774)	−0.889 (0.637)	−0.261 (0.7115)	−0.123 (0.709)	0.083 (0.790)	−0.274 (0.555)	−0.378 (0.608)	−3.155*** (1.042)	−3.322*** (1.149)
Systolic Blood Pressure (SBP)	−0.157 (0.097)		−0.128 (0.087)		−0.015 (0.097)		−0.099 (0.076)		−0.539*** (0.144)
Perceived Stress×SBP Interaction	0.005 (0.005)		0.007 (0.004)		.001 (.005)		.003 (.004)		.022*** (.007)
Diastolic Blood Pressure (DBP)		−0.003 (0.180)		−0.075 (0.1643)		0.029 (0.183)		−0.131 (0.142)		−0.780*** (0.268)
Perceived Stress×DBP Interaction		−0.001 (0.008)		0.004 (0.008)		0.000 (0.009)		0.005 (0.007)		0.038*** (0.013)
R²	137.457*** (17.816)	115.770*** (20.183)	88.374 *** (17.651)	75.121*** (19.848)	111.765*** (18.592)	105.889*** (20.887)	54.411*** (14.066)	53.015*** (16.064)	195.673*** (26.124)	190.206*** (29.969)

b: unstandardized coefficient; SE: standard error.

All models control for baseline age, sex, education, health conditions, blood pressure medication use, and cognition.

*p < 0.05 **p < 0.01 ***p < 0.001.

Blood pressure as a moderator

The findings for the moderation model predicting each domain of cognitive functioning are displayed in Table 3. The overall interaction of systolic BP and perceived stress was significantly associated with inductive reasoning (F (3417) = 4.73, p = 0.003). Systolic BP modified the relationship between stress and inductive reasoning at follow-up but was not related to any other cognitive domains, adjusting for baseline age, sex, education, health conditions, BP medication use, depressive symptoms, and cognition. These findings support our second hypothesis that BP would interact with stress to predict cognitive changes. Figure 1 (visual representation of the interaction) plots the linear predictions of inductive reasoning by level of perceived stress and systolic BP. This figure indicates that at low stress levels and high average systolic BP, inductive reasoning decreases. At high stress levels and high average systolic BP, inductive reasoning increases; however, the number of participants who were in this category (high stress >31 and high SBP >176 mmHg) was very low (n = 4 people).

Figure 1.

The relationship between perceived stress and inductive reasoning as moderated by systolic blood pressure among black Americans (51–96 years). Baltimore Study of Black Aging-Patterns of Cognitive Aging.

The overall interaction of diastolic BP and perceived stress was significantly associated with inductive reasoning (F (3417) = 3.14, p = 0.025). Diastolic BP modified the relationship between stress and inductive reasoning at follow-up but was not related to any other cognitive domains, adjusting for baseline age, sex, education, health conditions, BP medication use, depressive symptoms, and cognition. As shown in Figure 2 at low stress levels and high average diastolic BP, inductive reasoning decreases. At high stress levels and high average diastolic BP, inductive reasoning increases; however, the number of participants who were in this category (high stress >31 and high DBP >103.83 mmHg) was low (n = 10 people). This finding lends further support to our second hypothesis but merits further investigation.

Figure 2.

The relationship between perceived stress and inductive reasoning as moderated by diastolic blood pressure among Black Americans (51–96 years). Baltimore Study of Black Aging-Patterns of Cognitive Aging.

Discussion

The current study examined if perceived stress is directly related to changes in domain-specific cognitive functioning and whether BP (systolic and diastolic) acts as a moderator in a sample of older Black Americans aged 51 to 96 years. Two findings are of particular interest. First, perceived stress did not predict changes in any of the 5 cognitive domains. Second, systolic and diastolic BP significantly modified the relationship between perceived stress and domain-specific cognitive changes. Specifically, high systolic and diastolic BP modified the association between stress and inductive reasoning but no other cognitive domains, adjusting for model covariates. Individuals with low stress and high BP had low inductive reasoning scores, while those with high stress and high BP had high inductive reasoning scores. Although, high stress and high BP are positively related to higher inductive reasoning this association is based on very few participants in this category. Thus, caution should be used when interpreting this interaction as our finding is counterintuitive and may be driven by the few individuals in the extremely high and low levels of perceived stress and BP. These findings highlight the nuanced effects of stress and systolic and diastolic BP on changes in inductive reasoning. Future studies with larger samples of high-stress, hypertensive individuals are needed to confirm if this modifying effect of BP on the stress-cognition association remains stable in a more robust population.

Perceived stress did not predict domain-specific cognitive changes over a three-year period. However, this null main effect should be interpreted in the context of the significant moderation results. Our analysis revealed a crossover interaction, where the relationship between stress and inductive reasoning trended in opposite directions depending on one's BP level. Specifically, low stress was associated with lower inductive reasoning scores at high BP levels, but higher inductive reasoning scores at high stress and high BP levels. These opposing associations likely cancel each other out when aggregated, resulting in a statistically non-significant main effect. Thus, the lack of a main effect is not an absence of signal, but rather an artifact of the opposing conditional effects masked by aggregation.

We also consider this first finding in the context of existing longitudinal studies of older Black Americans.^11,12 Similar to Turner and colleagues¹¹ and Zuelsdorff et al.,¹² the present results show no direct effect of stress on older Black Americans’ working memory and processing speed over time. Conversely, Turner and colleagues¹¹ found that perceived stress contributed to greater decline in memory performance, whereas we found no relationship between perceived stress and memory performance.¹¹ Likewise, Zuelsdorff et al.¹² found in a small sample of Black Americans (N = 50) that greater lifetime stress was associated with faster decline in verbal learning & memory (episodic memory).¹² It may be that differences in memory assessments contribute to the differences in findings. The present study measured verbal memory (immediate recall only) exclusively using list learning tasks (including both reading and listening), while Turner and colleagues¹¹ assessed episodic memory (both immediate and delayed recall) with list learning tasks (including both reading and listening) as well as memory within the context of two stories, i.e., East Boston story and Story A from the Logical Memory subtest of the Wechsler Memory Test-R. Zuelsdorff et al.¹² also assessed verbal memory (delayed recall only) using a list learning task (reading only) to represent episodic memory.

Vocabulary and inductive reasoning were also not addressed by Turner and colleagues¹¹ or Zuelsdorff et al.,¹² preventing comparisons across studies for these domains. The lack of a relationship between perceived stress and vocabulary was not surprising, given that this is a measure of crystallized intelligence that peaks later in life and is generally resistant to change.⁴⁴ Further, perceived stress did not predict changes in inductive reasoning (a measure of fluid intelligence); however, it was the only cognitive domain revealing an interaction between perceived stress and systolic and diastolic BP. Although we cannot definitively conclude why inductive reasoning is the only cognitive domain for which this interaction occurred, we consider three potential explanations for this second finding.

First, as a complex form of cognition, inductive reasoning tasks require the integration and coordination of multiple cognitive skills, including basic sensation and perception, attention, working memory, and reasoning.⁴⁵ Other domains included in this study use some of these same cognitive skills, e.g., working memory and processing speed. Studies have shown that inductive reasoning or fluid reasoning (Gf) is inter-related with working memory and processing speed and share common cognitive processes.^46,47 However, visual working memory was found to be the strongest predictor of fluid reasoning, while processing speed was able to predict Gf only when visual working memory was controlled.⁴⁶ The authors argue that these results can be explained by two reasons: (1) the shared relationship between working memory and processing speed, and (2) the influence of time constraints in the subtests used to measure fluid intelligence (Gf).⁴⁶ Working memory and processing speed are closely related, meaning that working memory may overshadow processing speed as a predictor of Gf. On the other hand, processing speed reflects the ability to process information quickly.⁴⁶ Therefore, the time constraints in Gf-related tasks, particularly in the Figure Weight subtest (a supplementary subtest emphasizing the importance of the Gf within the perceptual reasoning index), may impact the relationship between processing speed and Gf.⁴⁶ Hence, a cognitive deficit in inductive reasoning identified through neuropsychological assessment may result from dysfunctions in various underlying cognitive skills. A deficit in inductive reasoning could then be interpreted in one of two ways⁴⁶: (1) a genuinely low capacity for Gf or inductive reasoning, or (2) a deficiency in the underlying cognitive skills that support Gf, such as working memory. Inductive reasoning or Gf is a complex, higher-order cognitive function that relies on several fundamental abilities, including working memory and processing speed, to function effectively.^46,47 In other words, an impairment in a lower-level cognitive skill, such as working memory, can contribute to an overall decline in inductive reasoning or Gf.⁴⁶ Thus, interpreting inductive reasoning in the context of low-order cognitive factors may explain why this single domain, rather than the combination of these cognitive skills, is impacted by the interaction of stress and BP. The combined effects of high stress and high BP facilitated improved inductive reasoning over time. Yet, at lower levels of stress and high BP inductive reasoning worsened over time. Thus, depending on one's level of stress, there may be a more pronounced effect of BP on inductive reasoning.

Second, we consider our findings within the current measure of stress, which was self-reported, general rather than specific in nature, and revealed low to moderate levels of perceived stress in the past month in the study sample. It may be that “higher” levels of perceived stress in the present sample were sufficiently moderate (mean wave 1 PSS score = 17.8, SD = 9.8, range = 0–46), such that, when combined with elevated BP, performance on complex inductive reasoning tasks was facilitated, though this facilitating effect was not present for those with low stress and high BP. It should also be noted that this association may have primarily been driven by the few people in the extremely high and low categories of perceived stress and BP. Moreover, it is possible that individuals who report high stress and high BP are high achievers who have high functioning jobs with a lot of stress but are still able to perform complex cognitive tasks as examined in the Whitehall II Study.⁴⁸

Based on the existing stress and cognition literature⁴⁹ we expected that elevated BP combined with high to very high levels of stress would be less advantageous to Black Americans cognitive health. Yet this was not the case in the present study as we found an unexpected, nuanced effect of stress and BP on cognitive functioning based on a small number of participants with high stress and high BP. Black Americans compared to Whites report a higher frequency of stress across their life course (e.g., childhood parental abuse or failing out of school)^12,15 and more specific life stressors (including financial insecurity, legal issues, being fired from a job),^12,50,51 which are associated with greater domain-specific cognitive decline (verbal learning & memory).¹² Moreover, moderate to high levels of chronic stress over time contributes to greater risk for hypertension in middle-aged Black Americans,²⁰ which is also a known risk factor for cognitive decline and dementia in later life. Studies have also consistently identified hypertension as a risk factor for worse global and domain-specific cognition (e.g., memory and executive functioning).¹⁹ Therefore, it is critical that we understand the individual and combined impact of stress and hypertension on cognitive functioning and decline in older Black Americans. In a similar vein, attention to the effects of specific sources of stress on cognitive abilities and whether BP moderates those effects is also needed.

Third, Black older adults who survive to volunteer for research on late-life health may be particularly resilient to social adversity, physical adversity, or both. Better-than-average cognitive abilities among those facing multiple challenges may reflect the sources of resilience, whether innate, interpersonal, or material, that support them to thrive. At least one study on childhood adversity and cognition in a large, population-based cohort found that among Black older adults (N = 3772; 61.7%; M_age = 74 years) specifically, some childhood stressors (i.e., food deprivation and being thinner than average in early life) were associated with slowed rates of decline over time.¹⁴ Hence, it is possible that early life adversity can lead to hardiness and resiliency in later life; yet less is known about these mechanisms.⁵² Additionally, some older Black Americans who experienced the most extreme adversity may not have survived to old age. Accordingly, individuals who participated in the BSBA-PCA may be more resilient than others. Brown and colleagues⁵³ have reported a complex and mixed relationship between self-reported stress, mental health, coping, and resilience in Black populations.⁵³ Although, these post hoc explanations (e.g., survivor bias, stress-related hardiness, resilience) are speculative and not directly testable within the current data. Future studies should explore the hypothesis of whether some stressors have an unexpected protective effect on cognitive decline in Black Americans.

The study's strengths and limitations should be considered when interpreting the findings. First, the BSBA-PCA was a community-dwelling convenience sample, and the findings may not apply to other settings (e.g., rural areas, other regions of the country, etc.) or be generalizable to the broader population of middle to older age Black Americans. For example, the BSBA-PCA sample would not be generalizable to middle age Black Americans under 48 years old given it included adults 48–95 years. According to PEW research center, the U.S. Black population is young with a median age of 38.2 years in 2023.⁵⁴ Roughly 30% of the entire Black population was below the age of 20, while 12% were 65 or older.⁵⁴ Thus, the current sample may not be generalizable to younger age groups. Second, there was some attrition in the BSBA-PCA sample and differences on the cognitive measures. Future studies should consider additional retention efforts for those with lower cognitive scores to minimize their losses to follow-up.

Third, the present study is limited in that it used a single subjective measure of global perceived stress. The current stress measure was based on the participant's subjective appraisal of stressful experiences in the past 30 days, and different findings might result, if additional objective measures were included. Subjective appraisal of stress versus a measure of stress based upon the objective environment has been shown to assess different components of stress.⁵⁵ Moreover, studies show that subjectively identified stress exposure may not operate the same as objective interview-based measures, such as discrete stressful life events or life stress checklists (e.g., the two measures may have different underlying biological mechanisms and be associated with different cognitive outcomes or other diseases).⁵⁵ Another limitation of the PSS is that it is influenced by a range of factors such as personality or social support aside from environmental stressors. Thus, scholars argue that “it is not possible to determine whether the variation in one's response to stress is due to external environmental conditions per se versus other individual differences in variables that contribute to the stress process.”⁵⁵ Fourth, while our current findings are conservative and are independent of disease-specific vascular pathways (i.e., no clinical diagnosis of hypertension was present in the BSBA-PCA cohort), it is important to note that we did not categorize individuals by hypertensive status. Rather, BP was treated as continuous. Future studies should use both continuous as well as dichotomous (hypertensive versus normotensive) measures of BP to better understand its influence on the stress-cognition association. Additionally, we used an oscillometer (automated) BP device. A device developed during the 1970s as an alternative way of measuring BP versus the gold-standard cuff method using a mercury sphygmomanometer.⁵⁶ A limitation of oscillometer automated devices is their potential to over- or under-estimate BP, especially in older adults, due to age-related arterial changes, such as increased arterial stiffness.^56,57 Therefore, an automated BP monitor with proven accuracy in a general population of adults may not be as accurate in a specialized population, such as older adults.⁵⁸ Future work should consider separate validation of these measures for better accuracy of BP measurement.⁵⁸

Fifth, we did not control for the use of medications associated with psychological or emotional conditions (e.g., anticholinergics), nor did we account for additional health-related variables such as body mass index (BMI) or lifestyle factors (e.g., physical activity, diet, smoking, etc.), which may influence the observed associations among perceived stress, BP, and cognition. Anticholinergic medications are known to affect cognition. Nishtala et al.,⁵⁹ for instance, found a significant association between high anticholinergic burden and poor cognitive performance in a community-dwelling sample of older adults in New Zealand (N = 14,198; 62.4% female; Mage = 82.5 years).⁵⁹ Moreover, studies controlling for health characteristics and lifestyle factors (literacy, BMI, smoking status) report an independent association between lifetime stressors and memory decline in Black Americans (N = 50; 4%; 40–65 years).¹² Thus, future work should consider medication usage for both BP and psychological conditions, along with other health factors and behaviors that influence cognition. Lastly, the current study measured a 2-point change in cognition. Having more than two time points may provide a stronger association between perceived stress and cognitive changes. Despite these limitations BSBA's large sample size and diverse cognitive domains support our conclusions.

Conclusion

Although this study found no direct effect of stress on cognitive changes, it suggests a potential joint effect of stress and BP for specific cognitive domains. Particularly, stress is more strongly associated with higher inductive reasoning among those with high stress and BP. These counterintuitive findings of individuals with high systolic and diastolic BP having significantly higher inductive reasoning scores at increasing levels of stress merit further investigation. Examination of additional stressors that may have an unexpected protective effect on the cognitive health of Black Americans warrants further study.

Supplemental Material

sj-docx-1-alz-10.1177_13872877261440135 - Supplemental material for Linking perceived stress and cognitive aging: The influence of blood pressure in Black Americans

Supplemental material, sj-docx-1-alz-10.1177_13872877261440135 for Linking perceived stress and cognitive aging: The influence of blood pressure in Black Americans by DeAnnah R. Byrd, Molly Maxfield, Megan Zuelsdorff, David W. Coon, Roland J. Thorpe and Keith E. Whitfield in Journal of Alzheimer's Disease

Footnotes

Acknowledgements

The authors have no acknowledgments to report.

ORCID iDs

DeAnnah R. Byrd

Megan Zuelsdorff

Ethical considerations

This study received ethical approval from the Duke University IRB (approval #1610) on January 28, 2012.

Consent to participate

All subjects provided written informed consent to participate in the BSBA-PCA study.

Consent for publication

Not applicable

Author contribution(s)

DeAnnah R. Byrd: Conceptualization; Formal analysis; Investigation; Methodology; Project administration; Visualization; Writing – original draft; Writing – review & editing.

Molly Maxfield: Conceptualization; Writing – original draft; Writing – review & editing.

Megan Zuelsdorff: Conceptualization; Writing – original draft.

David W. Coon: Conceptualization; Writing – review & editing.

Roland J. Thorpe: Conceptualization; Writing – review & editing.

Keith E. Whitfield: Data curation; Funding acquisition; Writing – review & editing.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Institute on Aging [1 K01 AG068376-01A1—D.R. Byrd, R01 AG24108—K. E. Whitfield, R01AG054363—R. J. Thorpe, Jr. and K. E. Whitfield, P30AG059298 —R. J. Thorpe, Jr.]; the National Institute on Minority Health and Health Disparities [U54MD000214—R. J. Thorpe, Jr.]; and the Alzheimer's Association [AARFD-21–852652—D.R. Byrd]. The content is solely the responsibility of the authors and does not necessarily represent the official views of the Alzheimer's Association.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data availability statement

The dataset generated during and/or analyzed during the current study are not publicly available due to privacy or ethical restrictions but are available from the senior author (K. E. Whitfield) on request.

Supplemental material

Supplemental material for this article is available online.

References

Alzheimer’s Association. 2025 Alzheimer’s disease facts and figures. Alzheimers Dement 2025; 21: e70235.

Byrd

Gee

Tarraf

. Black-white mental status trajectories: what ages do differences emerge? SSM Popul Health 2018; 6: 169–177.

Weuve

Barnes

Mendes de Leon

, et al. Cognitive aging in Black and White Americans: cognition, cognitive decline, and incidence of Alzheimer Disease dementia. Epidemiology 2018; 29: 151–159.

Beydoun

Weiss

Beydoun

, et al. Pathways explaining racial/ethnic disparities in incident all-cause and Alzheimer's Disease dementia among older US men and women. Alzheimers Dement (N Y) 2022; 8: e12275.

Eubank

Oberlin

Alto

, et al. Effects of lifestyle factors on cognition in minority population of older adults: a review. Front Nutr 2022; 9: 841070.

Federal Interagency Forum on Aging-Related Statistics. Older Americans 2016: Key indicators of well-being. Washington, DC: U.S. Government Printing Office, 2016.

Kino

. Stress, glucocorticoid hormones, and hippocampal neural progenitor cells: implications to mood disorders. Front Physiol 2015; 6: 230.

Vyas

Rodrigues

Silva

, et al. Chronic stress and glucocorticoids: from neuronal plasticity to neurodegeneration. Neural Plast 2016; 2016: 6391686.

Collins

Koechlin

. Reasoning, learning, and creativity: frontal lobe function and human decision-making. PLoS Biol 2012; 10: e1001293.

10.

McEwen

. Plasticity of the hippocampus: adaptation to chronic stress and allostatic load. Ann N Y Acad Sci 2001; 933: 265–277.

11.

Turner

James

Capuano

, et al. Perceived stress and cognitive decline in different cognitive domains in a cohort of older African Americans. Am J Geriatr Psychiatry 2017; 25: 25–34.

12.

Zuelsdorff

Okonkwo

Norton

, et al. Stressful life events and racial disparities in cognition among middle-aged and older adults. J Alzheimers Dis 2020; 73: 671–682.

13.

Aggarwal

Wilson

Beck

, et al. Perceived stress and change in cognitive function among adults 65 years and older. Psychosom Med 2014; 76: 80–85.

14.

Barnes

Wilson

Everson-Rose

, et al. Effects of early-life adversity on cognitive decline in older African Americans and whites. Neurology 2012; 79: 2321–2327.

15.

Chen

Weuve

Misra

, et al. Racial disparities in cognitive function among middle-aged and older adults: the roles of cumulative stress exposures across the life course. J Gerontol A Biol Sci Med Sci 2022; 77: 357–364.

16.

Sims

Whitfield

Ayotte

, et al. Subjective memory in older African Americans. Exp Aging Res 2011; 37: 220–240.

17.

Cohen

Kamarck

Mermelstein

. A global measure of perceived stress. J Health Soc Behav 1983; 24: 385–396.

18.

Forte

De Pascalis

Favieri

, et al. Effects of blood pressure on cognitive performance: a systematic review. J Clin Med 2020; 9: 34.

19.

Joyce

McHugh

Mockler

, et al. Midlife hypertension is a risk factor for some, but not all, domains of cognitive decline in later life: a systematic review and meta-analysis. J Hypertens 2024; 42: 205–223.

20.

Spruill

Butler

Thomas

, et al. Association between high perceived stress over time and incident hypertension in Black adults: findings from the Jackson Heart Study. J Am Heart Assoc 2019; 8: e012139.

21.

Waldstein

Katzel

. Stress-induced blood pressure reactivity and cognitive function. Neurology 2005; 64: 1746–1749.

22.

Aiken-Morgan

Gamaldo

Sims

, et al. Education desegregation and cognitive change in African American older adults. J Gerontol B Psychol Sci Soc Sci 2014; 70: 348–356.

23.

Gamaldo

Allaire

, et al.

Sleep complaints in older blacks: do demographic and health indices explain poor sleep quality and duration?

J Clin Sleep Med 2014; 10: 725–731.

24.

Turner

Engle

. Is working memory capacity task dependent? J Mem Lang 1989; 28: 127–154.

25.

Craik

. Changes in memory with normal aging: a functional view. Adv Neurol 1990; 51: 201–205.

26.

Wechsler

. WAIS-R manual: Wechsler adult intelligence scale-revised. New York, NY: Psychological Corporation, 1981.

27.

Ekstrom

Harman

. Manual for kit of factor-referenced cognitive tests, 1976. Princeton, NJ: Educational Testing Service, 1976.

28.

Zelinski

Gilewski

Schaie

. Individual differences in cross-sectional and 3-year longitudinal memory performance across the adult life span. Psychol Aging 1993; 8: 176–186.

29.

Rey

. L’examen psychologique dans les cas d’encephopathie traumatique (The psychological examination of cases of traumatic encephalopathy). Arch Psychol (Geneve) 1941; 28: 286–340.

30.

Brandt

. The Hopkins Verbal Learning Test: development of a new memory test with six equivalent forms. Clin Neuropsychol 1991; 5: 125–142.

31.

Shipley

. Shipley institute of living scale. Los Angeles, CA: Western Psychological Services, 1986.

32.

Schaie

Thurstone

. Schaie-Thurstone adult mental abilities test. Palo Alto, CA: Consulting Psychologists Press, 1985.

33.

Beevers

Lip

GYH

O'Brien

. ABC Of hypertension - Blood pressure measurement - Part I - Sphygmomanometry: factors common to all techniques. Br Med J 2001; 322: 981–985.

34.

Thorpe

Jr Clay

Szanton

, et al. Correlates of mobility limitation in African Americans. J Gerontol A Biol Sci Med Sci 2011; 66: 1258–1263.

35.

Carmasin

Mast

Allaire

, et al. Vascular risk factors, depression, and cognitive change among African American older adults. Int J Geriatr Psychiatry 2014; 29: 291–298.

36.

Byrd

Thorpe

Jr Whitfield

. Greater disease burden, greater risk? Exploring cognitive change and health status among older blacks. J Aging Health 2020; 32: 807–816.

37.

Radloff

. The CES-D scale: a self-report depression scale for research in the general population. Appl Psychol Meas 1977; 1: 385–401.

38.

Sims

Thorpe

Jr Gamaldo

, et al. Cognition and health in African American men. J Aging Health 2015; 27: 195–219.

39.

Vickers

Altman

. Analysing controlled trials with baseline and follow up measurements. Br Med J 2001; 323: 1123–1124.

40.

O'Rourke

MacKinnon

. Reasons for testing mediation in the absence of an intervention effect: a research imperative in prevention and intervention research. J Stud Alcohol Drugs 2018; 79: 171–181.

41.

McCabe

Kim

King

. Improving present practices in the visual display of interactions. Adv Methods Pract Psychol Sci 2018; 1: 147–165.

42.

Schafer

. Multiple imputation: a primer. Stat Methods Med Res 1999; 8: 3–15.

43.

StataCorp

LLC

. Stata: Release 18.5 [Computer software]. College Station, TX: StataCorp LLC, 2024.

44.

Hartshorne

Germine

. When does cognitive functioning peak? The asynchronous rise and fall of different cognitive abilities across the life span. Psychol Sci 2015; 26: 433–443.

45.

Harvey

. Domains of cognition and their assessment. Dialogues Clin Neurosci 2019; 21: 227–237.

46.

Park

. Relationship of working memory, processing speed, and fluid reasoning in psychiatric patients. Psychiatry Investig 2018; 15: 1154–1161.

47.

Lohman

. Fluid intelligence, inductive reasoning, and working memory: Where the theory of Multiple Intelligences falls short. Seattle, WA: American Educational Research Association, 2001, pp. 1–8.

48.

Elovainio

Ferrie

Singh-Manoux

, et al. Cumulative exposure to high-strain and active jobs as predictors of cognitive function: the Whitehall II Study. Occup Environ Med 2009; 66: 32–37.

49.

Sandi

. Stress and cognition. Wiley Interdiscip Rev Cogn Sci 2013; 4: 245–261.

50.

Zuelsdorff

Gleason

Kind

, et al. Lifetime stressful experiences, racial disparities, and cognitive performance: findings from the Wisconsin Registry for Alzheimer’s Prevention (WRAP) Study. Alzheimers Dement 2017; 13: P212.

51.

Allen

Mezuk

Byrd

, et al.

Mechanisms of cardiometabolic health outcomes and disparities: what characteristics of chronic stressors are linked to HPA-axis dysregulation?

J Aging Health 2022; 34: 448–459.

52.

Thomas Tobin

Gutiérrez

Bell

, et al. Early life racial discrimination, racial centrality, and allostatic load among African American older adults. Gerontologist 2022; 62: 721–731.

53.

Brown

García

Reeves

, et al. Stress is a latent construct: exploring the differential experience of stress and discrimination on depressive symptoms among Black older adults. J Aging Health 2022; 34: 334–346.

54.

Martinez

Passel

. Facts about the U.S. Black population. Washington, DC: Pew Research Center, 2025.

55.

Monroe

. Modern approaches to conceptualizing and measuring human life stress. Annu Rev Clin Psychol 2008; 4: 33–52.

56.

Sharman

Tan

Stergiou

, et al. Automated ‘oscillometric’ blood pressure measuring devices: how they work and what they measure. J Hum Hypertens 2023; 37: 93–100.

57.

Finnegan

Spence

Wong

, et al. Blood pressure measurement in the elderly: correlation of arterial stiffness with difference between intra-arterial and cuff pressures. J Hypertens 1985; 3: 231–235.

58.

Stergiou

Dolan

Kollias

, et al. Blood pressure measurement in special populations and circumstances. J Clin Hypertens (Greenwich) 2018; 20: 1122–1127.

59.

Nishtala

Allore

Han

, et al. Impact of anticholinergic burden on cognitive performance: a cohort study of community-dwelling older adults. J Am Med Dir Assoc 2020; 21: 1357–1358.e1353.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB