Association between cardiovascular and cognitive health among older Indigenous Africans: Data from an urban Nigerian settlement participating in the VALIANT study

Abstract

INTRODUCTION

The relationship between cardiovascular health (CVH) and cognitive health (CH) has been established in diverse populations but is understudied among indigenous Africans. We investigated the association between CVH and CH in 1031 older Nigerian Africans participating in the Vascular heAlth, fraiLty, and cognItion in Ageing Nigerians sTudy (VALIANT).

METHODS

CVH was assessed by the Life's Simple 7 (LS7) score. CH was measured using three validated metrics of general cognitive well‐being: Montreal Cognitive Assessment (MoCA), Community Screening Instrument for Dementia (CSID), and Identification and Intervention for Dementia in Elderly Africans (IDEA). Frailty was evaluated using Rockwood's Clinical Frailty Scale. Multivariable linear regression was performed.

RESULTS

Individual LS7 metrics, including poor diet, ideal body mass index (overall), and intermediate blood pressure (among female participants), showed independent relationships with poor CH. Although the LS7 composite score showed a significant univariate relationship with MoCA and IDEA scores, the independent determinants of general CH were older age, low educational attainment, and clinical frailty, but not LS7 composite score

DISCUSSION

In a sample of older Nigerian Africans, the LS7 composite score may not accurately characterize the relationship between CVH and CH. Afrocentric CVH composite scores incorporating measures that are more sensitive to outcomes in Africans are needed.

Highlights

• We determined the association between cardiovascular health (CVH) and cognitive health (CH) in 1031 urban‐dwelling Nigerian Africans.

• Although individual Life's Simple 7 (LS7) metrics such as poor diet, intermediate blood pressure, and ideal body mass index showed independent association with poor CH, the composite LS7 metric showed no association.

• In a sample of Nigerian Africans, the LS7 score as a composite vascular marker may not accurately characterize the relationship between CVH and CH.

• Afrocentric composite scores incorporating risk markers potentially unique to indigenous African populations are needed for robust characterization of CVH.

1. BACKGROUND

An estimated 47 million people worldwide were living with dementia in 2015, and this number is projected to triple by 2050 due to population growth, population aging, and clustering of vascular risk factors, particularly in low‐ and middle‐income countries (LMICs), including many African nations. ¹ , ² In the absence of a definitive cure or robust disease‐modifying therapy for most dementias, affordable and readily available, well‐tolerated, contextually‐tailored public health measures are urgently needed to mitigate this growing public health problem. ³ , ⁴ One such public health intervention is the modification of cardiovascular risk factors. Hence, further characterization of the association between cardiovascular health (CVH) and cognitive health (CH) remains imperative in a population with an increasing cardiovascular disease burden.

CVH, calculated using the Life's Simple 7 (LS7) score (as developed by the American Heart Association), underscores the importance of primary prevention and sets realistic targets for modifiable health factors and behaviors to reduce the global burden of cardiovascular disease. ⁵ The LS7 score is a composite measure with seven key vascular indicators: cardiometabolic—fasting plasma glucose (FPG), serum total cholesterol (TC), blood pressure (BP); behavioral—body mass index (BMI), physical activity, diet, and cigarette smoking. Each indicator is classified into ideal, intermediate, and poor levels to yield an aggregate CVH score. Higher scores are indicative of better CVH, associated with favorable outcomes for not only cardiovascular but also non‐cardiovascular disorders, including depression, cancer, and neurodegenerative dementia. ⁶ , ⁷ , ⁸ , ⁹

The association between CVH and disease outcomes has been established in many studies among Caucasian ancestries but is insufficiently explored among indigenous Africans—a population witnessing a “double burden” of cognitive and cardiovascular disorders. ¹ An understanding of this relationship offers a multi‐pronged window for primary prevention of neurocognitive disorders, especially in underserved urban populations. In addition, some studies have shown better CVH neuroprotection among Caucasian compared to African ancestry populations and among women compared to men ¹⁰ —suggesting racial, gender, and even geographic and familial influences. ¹¹ , ¹² The association has also varied by age group ¹³ and by specific components of the CVH metric considered ¹⁴ , ¹⁵ in a paradigm described as the multifactorial nature of vascular aging. ¹⁶ An improvement in population‐level CVH has the potential to reduce the burden of dementia. ¹⁷ , ¹⁸ , ¹⁹ , ²⁰

AIM: We examined the relationship between neurocognition and CVH metrics (LS7) among indigenous Africans in an urban settlement. We hypothesized that higher LS7 would be associated with healthier brain function (neurocognitive performance).

2. METHODS

2.1. Study population

Vascular heAlth, fraiLty and cognItion in Ageing Nigerians sTudy (VALIANT) is a longitudinal community‐based cohort study aimed at exploring the association between CVH, cognition, and frailty in a cohort of older Nigerians. Participants were recruited using a multistage sampling method from Yemetu, Ibadan North Local Government Area (LGA), Southwest Nigeria, over 12 months between November 2022 and October 2023. Using the African Rigorous Innovative Stroke Epidemiological Surveillance (ARISES) database ²¹ as a sampling frame, two wards (Wards 3 and 4) were purposively selected. ARISES is an ongoing observational cohort study in selected wards in Ibadan North and Ibarapa Central LGAs in Oyo State, Nigeria. VALIANT involves the Ibadan North LGA with a population of 352,270 ²² and divided into 10 administrative wards. Under the ARISES Project, ²¹ a de jure census of the Ibadan North LGA Wards 1,3, and 4 obtained a de jure population of 45,712 individuals in 13,531 households. Wards 3 and 4 were purposively selected for VALIANT because they host more indigenous low‐income urban dwellers than Ward 1, which is predominantly a government reservation area for offices and other establishments. Thereafter, 11 of the 16 clusters from the two wards were randomly selected, and all households within the selected clusters were visited to recruit eligible participants. Participants were included if they were 50 years of age or older, resided within the study area, and provided written informed consent. Overall, 1031 participants were enrolled and taken through a battery of cardiovascular, cognitive, and frailty assessment tools. Ethical approval was obtained from the University of Ibadan/University College Hospital (UI/UCH) Health Research Ethics Committee (HREC).

2.2. Sociodemographic data, frailty, and other baseline characteristics

Sociodemographic data were collected from all participants, including sex (men and women), age (in years), education level (up to primary [<8 years of studies], secondary [between 8 and 12 years of studies], and beyond secondary [>12 years of study]), marital status (married and others that includes single, widowed or separated), and indigenous ethnicity (yes or no).

The Global Physical Activity Questionnaire (GPAQ) was used to assess physical activity. ²³ The GPAQ has been validated in various populations using accelerometers and/or pedometers with high reliability coefficients and criterion validity. The scoring of the GPAQ comprises the calculation of Metabolic Equivalent of Task (MET) min per week for each specific domain (work, travel, leisure). These values were then aggregated to obtain a total MET‐min score. This overall score is subsequently classified into high, moderate, or low physical activity levels in accordance with World Health Organization (WHO) guidelines. The Food Frequency Questionnaire (FFQ), used widely in studies of nutritional epidemiology, was used for dietary recall. ²⁴ To score the FFQ, numerical values were assigned to food consumption frequencies and a total score calculated. Frailty was measured using Rockwood's Clinical Frailty Scale (CFS). ²⁵ , ²⁶ , ²⁷ The CFS is a 9‐point scale that quantifies frailty in individuals, with higher scores indicating higher frailty (very fit 1, fit 2, managing well 3, living with very mild frailty 4, mild frailty 5, moderate frailty 6, severe frailty 7, very severe frailty 8, terminally ill 9). ²⁸ The CFS scores were assigned by trained neurologists after clinical encounters with each participant. The revised Lubben Social Network Scale (SNS), a 12‐item questionnaire designed to assess the size, closeness, and frequency of contacts within an individual's social network, was used to measure social isolation. Data on hand grip strength (HGS) were obtained using a digital hand dynamometer.

RESEARCH IN CONTEXT

1. Systematic review: The association between composite cardiovascular health (CVH) and cognitive health (CH) is insufficiently explored among indigenous Africans—a population witnessing a “double burden” of cognitive and cardiovascular disorders. Detailed characterization of this relationship offers a multi‐pronged window for risk profiling and primary prevention of neurocognitive disorders, in the absence of definitive cure.

2. Interpretation: CVH was assessed using the Life's Simple 7 (LS7) score—a composite metric of vascular function. Our findings revealed that individual LS7 metrics, including poor diet, intermediate blood pressure, and ideal body mass index, showed independent relationships with poor cognitive function, whereas the composite LS7 metric showed no association. Overall, older age, higher educational attainment, higher social network score, and clinical frailty were independent determinants of general CH in the population. Among Nigerian Africans, the LS7 score as a composite vascular marker may not accurately characterize the relationship between cardiovascular and CH.

3. Future directions: Healthy dietary practices and blood pressure control offer pragmatic strategies for population‐wide dementia risk reduction among indigenous Africans. Afrocentric composite scores incorporating risk markers specific to the African population are needed for robust characterization.

2.3. Assessment of cardiovascular health and cognitive health

Components of the CVH metric, the LS7 Score, include BP, FPG, TC, BMI, physical activity, diet, and cigarette smoking. Each CVH component was given a score of 0, 1, or 2 to represent poor, intermediate, or ideal health, respectively, according to pre‐defined categories. Based on the sum of all seven CVH components, an overall CVH score, ranging from 0 to 14, was categorized as inadequate (0–4), average (5–9), or optimal (10–14) CVH. Systolic BP and diastolic BP were measured using a mercury sphygmomanometer on the right arm with the subject in a sitting position after 10 min of rest. The average of two measurements 5 min apart was used in the statistical analyses. Venous blood samples were drawn for the measurement of glucose and lipid profiles after an overnight fast. Fasting lipid profile and fasting plasma glucose were assessed using standard laboratory techniques. BMI was calculated from the height and weight measurements of participants.

Cognitive assessment was undertaken using translated and validated neuro‐psychometric instruments: Community Screening Instrument for Dementia (CSID), ²⁹ , ³⁰ Montreal Cognitive Assessment (MoCA), ³¹ and the Identification and Intervention for Dementia in Elderly Africans (IDEA) cognitive screen. ³² The CSID, MoCA, and IDEA are tests of general cognitive functioning that have been well validated in the African setting.

2.4. Statistical analyses

Continuous variables were described as mean ± standard deviation (SD) and categorical variables as frequencies (percentages). Age (in years) was analyzed as a continuous and a categorical variable (<65 and ≥65). Using the independent t‐test/analysis of variance (ANOVA), the mean cognitive scores of the study population were compared using the baseline sociodemographic characteristics, CVH factors, CFS score, and HGS. A multivariable linear regression model was used to test the bivariate relationship of socio‐demographics, CVH risk markers, grip strength, and CFS score to cognitive performance. An adjusted multivariable linear regression analysis was done to identify independent factors related to cognitive performance. The multivariable model was then stratified by sex to explore sex differences in patterns of relationships. The strength and direction of the relationship were reported as a beta coefficient with a 95% confidence interval (CI). All statistical analyses were conducted using Stata SE version 16 with a significance level set at 0.05

3. RESULTS

3.1. Baseline characteristics, cardiovascular and cognitive health

The baseline characteristics of participants are shown in Table 1. The mean (SD) age (in years) of study participants was 64.5 (± 11.8), and 27.2% were male with 5.6 ± 4.8 mean (SD) years of education. Ninety‐eight percent of the study participants were of Yoruba ethnic extraction. The mean ± SD MoCA, IDEA, and CSID scores were 18.0 ± 6.5, 11.13 ± 3.1, and 50.16 ± 7.53, respectively. Older age, female sex, low level of education, marital status (widow/widower), living situation (alone), and abstinence from alcohol/smoking showed significant bivariate relationships with cognitive dysfunction across the three cognitive scales (MoCA, IDEA, and CSID), (Table 1). The respective mean (SD) MoCA, IDEA, and CSID scores were significantly lower among participants ≥65 years and females, compared to those <65 years and males (Table 1). Participants with worse CFS scores and with BMI <25 had worse cognitive performance using both MoCA (p < 0.001) and IDEA scales (p < 0.001) but not the CSID scale (Table 1). As shown in Table S1, participants >65 years had significantly lower LS7 scores, and were more likely to be frail, hypertensive, and socially isolated when compared to those <65 years (Table S1).

TABLE 1.

Mean cognitive scores of the study population according to socio‐demographic characteristics and cardiovascular and frailty markers.

		MoCA		IDEA		CSID
	Freq (%)	Mean ± SD	p‐value	Mean ± SD	p‐value	Mean ± SD	p‐value
Age group (years)
<65	515 (50.4)	20.92 ± 5.38	<0.001**, b	12.18 ± 2.72	<0.001**, b	52.54 ± 6.19	< 0.001**, b
≥65	507 (49.6)	15.31 ± 6.37		10.07 ± 3.46		47.78 ± 8.86
Gender
Male	279 (27.2)	21.28 ± 5.73	<0.001**, b	11.81 ± 3.22	<0.001**, b	52.27 ± 8.17	< 0.001**, b
Female	746 (72.8)	16.93 ± 6.42		10.87 ± 3.28		49.36 ± 7.81
Marital status
Currently married	458 (44.4)	20.60 ± 5.65	<0.001**, a	11.96 ± 2.92	<0.001**, a	52.35 ± 6.68	< 0.001**, a
Widow/widower	437 (42.4)	15.39 ± 6.42		10.26 ± 3.47		47.89 ± 8.66
Others (single/saparated)	136 (13.2)	18.38 ± 6.24		10.95 ± 3.39		49.23 ± 9.94
Highest level of education
None	337 (32.8)	12.53 ± 5.03	<0.001**, a	9.39 ± 3.33	<0.001**, a	45.74 ± 8.60	< 0.001**, a
Primary	397 (38.7)	19.31 ± 5.42		11.61 ± 2.80		51.27 ± 6.51
Secondary	244 (23.8)	22.69 ± 4.38		12.47 ± 2.93		53.63 ± 6.86
Tertiary/postgraduate	48 (4.7)	24.54 ± 3.89		12.58 ± 3.37		54.31 ± 6.30
Living situation
Lives alone	206 (20.3)	16.28 ± 6.54	<0.001**, a	10.37 ± 3.52	<0.001**, a	48.08 ± 9.24	< 0.001**, a
Lives with spouse and children	288 (28.4)	21.65 ± 5.41		12.29 ± 2.74		53.17 ± 7.01
Lives in a nursing home	1 (0.1)	20 ± 0		14.00 ± 0.00		50.00 ± 0.0
Lives with spouse	125 (12.3)	19.00 ± 5.73		11.30 ± 3.39		50.24 ± 7.97
Lives with extended family	101 (9.9)	14.40 ± 6.21		9.59 ± 3.24		47.00 ± 8.52
Lives with children	295 (29.0)	16.83 ± 6.37		11.02 ± 3.21		49.74 ± 6.84
Diabetic
No	969 (94.0)	18.17 ± 6.55	0.217 b	11.13 ± 3.31	0.386 b	50.06 ± 8.38	0.882 b
Yes	62 (6.0)	17.08 ± 6.13		10.75 ± 3.52		49.90 ± 6.86
Alcohol use
No	692 (67.7)	17.01 ± 6.56	<0.001**, b	10.96 ± 3.22	0.002**, b	49.52 ± 7.58	< 0.001**, b
Yes	330 (32.3)	20.40 ± 5.87		11.63 ± 3.22		51.93 ± 7.19
Smoking
No	903 (88.9)	17.75 ± 6.49	<0.001**, b	11.07 ± 3.26	0.002**, b	50.14 ± 7.44	0.010**, b
Yes	113 (11.1)	20.95 ± 6.18		12.05 ± 2.80		52.04 ± 6.84
Dyslipidemia
No	670 (65.0)	18.18 ± 6.62	0.611 b	11.15 ± 3.28	0.574 b	50.20 ± 8.20	0.438 b
Yes	361 (35.0)	17.96 ± 6.35		11.03 ± 3.39		49.78 ± 8.47
Sleep disorder
No	944 (92.1)	18.22 ± 6.53	0.084 b	18.22 ± 6.53	0.084 b	50.57 ± 7.91	0.099 b
Yes	81 (7.9)	16.89 ± 6.54		16.89 ± 6.54		49.70 ± 8.54
Hypertensive
No	413 (40.1)	18.59 ± 6.48	0.055 b	11.33 ± 3.15	0.073 b	50.73 ± 8.25	0.439 b
Yes	618 (59.9)	17.78 ± 6.54		10.96 ± 3.42		67.75 ± 447.12
Clinical frailty scale
Very fit (1)	145 (14.4)	20.08 ± 5.71	<0.001**, a	11.99 ± 2.71	<0.001**, a	52.31 ± 4.94	0.943 a
Well (2)	343 (34.2)	20.21 ± 5.89		11.99 ± 2.59		84.62 ± 599.95
Managing well (3)	383 (38.2)	17.06 ± 6.41		11.04 ± 2.91		49.97 ± 7.80
Very mild frailty (4)	88 (8.8)	14.47 ± 6.47		9.64 ± 3.95		46.96 ± 9.09
Mildly frail (5)	27 (2.7)	12.74 ± 5.91		9.25 ± 3.91		44.92 ± 10.06
Moderately frail (6)	13 (1.3)	11.25 ± 4.65		6.76 ± 3.13		41.07 ± 10.95
Severely frail (7)	4 (0.4)	10.5 ± 5.44		10.00 ± 1.41		41.00 ± 12.54
Very severely frail (8)	1 (0.1)	4.00 ± 0		3.00 ± 0		29.00 ± 00
Smoking
Current smoker	26 (2.6)	22.15 ± 5.53	<0.001**, a	12.19 ± 2.81	0.009**, a	52.73 ± 6.87	0.031**, a
Former smoker	87 (8.6)	20.58 ± 6.36		12.01 ± 2.81		51.83 ± 6.85
Never	903 (88.9)	17.75 ± 6.49		11.07 ± 3.26		50.14 ± 7.44
Physical activity
Physically active <1 h per day during work or no leisure time	16 (12.5)	19.53 ± 6.47	0.955 a	10.00 ± 0.00	0.091 a	51.87 ± 5.78	0.777 a
Physically active between 1 and 6 h per day during work or 1 and 2 h per day at leisure	61 (47.7)	19.30 ± 6.33		10.91 ± 2.77		52.58 ± 5.65
Physically active ≥6 h per day during work or ≥2 h per day at leisure	51 (39.8)	19.04 ± 6.02		11.63 ± 2.64		52.98 ± 5.36
Healthy diet
<1 portion per day of fruits, green leafy vegetables, cooked vegetables, and <2 portions per week of fish	413 (40.1)	17.04 ± 6.23	<0.001**, a	10.68 ± 3.42	<0.001**, a	76.61 ± 546.92	0.493 a
≥1 portion per day of fruits, green leafy vegetables, cooked vegetables OR ≥2 portions per week of fish	252 (24.4)	17.93 ± 7.06		11.01 ± 3.42		49.57 ± 8.16
≥1 portion per day of each of fruits, green leafy vegetables, cooked vegetables AND ≥2 portions per week of fish	366 (35.5)	19.40 ± 6.25		11.66 ± 3.06		51.07 ± 7.96
BMI (kg/m2)
≥30	190 (19.5)	19.18 ± 5.87	<0.001**, a	11.81 ± 2.83	0.001**, a	52.02 ± 5.73	0.747 a
25–29.9	222 (22.8)	19.00 ± 6.16		11.47 ± 3.27		51.33 ± 8.01
<25	563 (57.7)	17.45 ± 6.78		10.88 ± 3.30		69.28 ± 468.44
Blood pressure
≥140/90 mmHg	524 (51.6)	17.66 ± 6.70	0.075 a	10.98 ± 3.34	0.244 a	47.88 ± 9.6	0.217 a
<120/80 mmHg treated or 120–139/80–89 mmHg	272 (26.8)	18.34 ± 6.45		11.30 ± 3.25		–
<120/80 mmHg	220 (21.6)	18.82 ± 6.22		11.35 ± 3.08		51.93 ± 6.29
Fasting plasma glucose
>126 mg/dL	60 (5.8)	20.23 ± 5.76	<0.001**, a	11.81 ± 3.05	0.194 a	70.98 ± 485.54	0.656 a
100–126 mg/dL or <100 mg/dL treated	348 (33.8)	16.85 ± 6.24		10.97 ± 3.04		50.43 ± 8.62
<100 mg/dL untreated	623 (60.4)	18.60 ± 6.65		11.12 ± 3.49		51.48 ± 6.66
Total Cholesterol
Optimal (<200)	677	17.79 ± 6.59	0.229 a	11.47 ± 3.10	0.139 a	49.97 ± 8.06	0.279 a
Borderline (200–239)	82	17.03 ± 6.69		10.75 ± 3.24		48.46 ± 9.54
High (>239)	22	19.72 ± 6.65		11.59 ± 3.09		49.91 ± 13.09
Hand grip strength	18.56 ± 8.28
Social Network	29.16 ± 9.89
LS7	6.60 ± 1.69

Abbreviations: BMI, body mass index; CSID, Community Screening Instrument for Dementia; Freq, frequency; IDEA, Identification and Intervention for Dementia in Elderly Africans; LS7, Life's Simple 7; MoCA, Montreal Cognitive Assessment.

^** p‐value < 0.05 was considered statistically significant.

^a  = F‐test.

^b  = t‐test.

The mean LS7 score was 6.6 ± 1.6 (max = 14), and only 4% of the study population had optimal LS7 scores (Figure 1). The individual LS7 metrics are shown in Table 2. Remarkably, the predominant vascular risk marker was hypertension (systolic BP ≥140 mmHg or diastolic BP≥90 mmHg). A total of 639 participants (62.7% of the study population) had poor, 226 (22.18%) intermediate, and 154 (15.1%) had ideal BP status. In contrast, the majority of the participants (96.68%) had an ideal LS7 score for smoking compared to intermediate (0.81%) and poor (2.52%). Similarly, 89.1%, 94.3%, and 40.9% of the study population had ideal LS7 scores for TC, FPG, and BMI, respectively. As shown in Table S2, there was a lower proportion of participants >65 years with ideal LS7 scores across the individual metrics of systolic BP, total cholesterol, physical activity, and dietary habits.

FIGURE 1.

Cardiovascular health of study population using the LS7 score. LS7, Life's Simple 7.

TABLE 2.

Prevalence of individual LS7 metrics in the population.

LS7 metrics	Poor, n (%)	Intermediate, n (%)	Ideal, n (%)	Total, n
Physical activity	149 (14.45)	421 (40.8)	461 (44.7)	1031
BMI	190 (19.5)	222 (22.7)	399 (40.9)	785
Cigarette smoking	25 (2.52)	8 (0.81)	960 (96.68)	993
Dietary habits	75 (13)	332 (57.54)	170 (29.46)	577
Systolic BP	639 (62.7)	226 (22.18)	154 (15.1)	1019
Total cholesterol	0 (0)	89 (10.9)	726 (89.1)	815
FPG	0 (0)	56 (5.69)	929 (94.3)	985

Abbreviations: BMI, body mass index; BP, blood pressure; FPG, fasting plasma glucose; LS7, Life's Simple 7 (score).

3.2. Unadjusted determinants of cognitive performance

In the simple linear regression (Table 3), older age, female sex, being a widower/widow, and higher CFS were inversely (negatively) related to MoCA and IDEA scores, whereas more years of education, living with spouse/children, a higher SNS score, and alcohol use were positively associated with higher MoCA and IDEA scores, but not CSID scores. A higher LS7 composite score was significantly positively associated (β, 95% CI) with the MoCA (0.25, 0.015–0.492) and IDEA scores (0.12, 0.01–0.24), but not the CSID scores (–5.05, –17.36 to 7.26). As shown in Table 3, individual LS7 metrics such as unhealthy diet, BP ≥140/90 mmHg, BMI <25 kg/m², and physical inactivity showed significant bivariate association with poor CH.

TABLE 3.

Simple linear regression of the determinants of cognitive performance.

	MoCA			IDEA			CSID
	$\beta$	95% CI	p‐value	$\beta$	95% CI	p‐value	$\beta$	95% CI	p‐value
Age, years	−0.27	−0.31, −0.24	< 0.001 *	−0.11	−0.12, −0.09	< 0.001 *	0.65	−1.28, 2.60	0.507
Gender
Male	Ref.			0.00			0.00
Female	−4.35	−5.23, −3.47	< 0.001 *	−0.93	−1.38, −0.48	< 0.001 *	12.09	−35.73, 59.92	0.620
Marital Status
Currently married	Ref.			0.00			0.00
Widow/widower	−5.21	−6.02, −4.40	< 0.001 *	−1.70	−2.12, −1.27	< 0.001 *	−28.63	−74.07, 16.79	0.216
Others	−2.22	−3.42, −1.03	< 0.001 *	−1.01	−1.63, −0.39	0.001 *	−27.46	−93.81, 38.88	0.417
Years of education	0.88	0.81, 0.94	< 0.001 *	0.28	0.24, 0.32	< 0.001 *	0.67	0.58−0.77	<0.001 *
Living situation
Lives alone	Ref.			0.00			0.00
Lives with spouse and children	5.36	4.25, 6.48	< 0.001 *	1.92	1.35, 2.49	< 0.001 *	43.75	−18.75, 106.25	0.170
Lives in a nursing home	3.71	−8.20, 15.62	0.541	3.62	−2.63, 9.88	0.256	1.91	−684.71, 688.53	0.996
Lives with spouse	2.71	1.33, 4.09	< 0.001 *	0.93	0.22, 1.63	0.010 *	2.16	−75.49, 79.82	0.956
Lives with extended family	−1.88	−3.35, −0.41	0.012 *	−0.77	−1.53, −0.02	0.044	−1.07	−84.28, 82.12	0.980
Lives with children	0.54	−0.55, 1.65	0.330	0.64	0.08, 1.21	0.025 *	1.93	−60.25, 64.12	0.951
Diabetic
No	Ref.			0.00			0.00
Yes	−1.08	−2.82, 0.64	0.810	−0.37	−1.23, 0.47	0.387	−11.41	−100.44, 77.62	0.801
Alcohol use
No	Ref.			0.00			0.00
Yes	3.38	2.53, 4.23	< 0.001 *	0.66	0.24, 1.09	0.002 *	−13.72	−59.38, 31.93	0.555
Smoking
No	Ref.			0.00			0.00
Yes	3.19	1.91, 4.47	< 0.001 *	0.98	0.35, 1.60	0.002 *	−10.52	−78.83, 57.78	0.762
Dyslipidemia
No	Ref.			0.00			0.00
Yes	−0.22	−1.07, 0.63	0.612	−0.12	−0.54, 0.30	0.574	−17.13	−61.49, 27.22	0.449
Sleep disorder
No	Ref.			0.00			0.00
Yes	−1.32	−2.82, 0.17	0.084	−1.66	−2.40, −0.93	< 0.001 *		−	−
Hypertensive
No	Ref.			0.00			0.00
Yes	−0.81	−1.64, 0.01	0.056	−0.37	−0.79, 0.03	0.074	17.01	−26.17, 60.19	0.440
Smoking
Current smoker	4.39	1.87, 6.91	0.001 *	0.00			0.00
Former smoker	2.82	1.37, 4.27	< 0.001 *	−0.18	−1.59, 1.23	0.801
Never	Ref.			−1.11	−2.37, 0.13	0.081
Physical activity
Physically active <1 h per day during work or no leisure time	Ref.			0.00			0.00
Physically active between 1 and 6 h per day during work or 1 and 2 h per day at leisure	−0.22	−3.79, 3.33	0.899	0.91	−0.60, 2.43	0.235	0.70	−2.38, 3.80	0.651
Physically active ≥6 h per day during work or ≥2 h per day at leisure	−0.49	−4.12, 3.13	0.788	1.63	0.09, 3.17	0.038 *	1.10	−2.03, 4.24	0.488
Healthy diet
<1 portion per day of fruits, green leafy vegetables, cooked vegetables, and <2 portions per week of fish	Ref.			0.00			0.00
≥1 portion per day of fruits, green leafy vegetables, cooked vegetables OR ≥2 portions per week of fish	0.88	−0.14, 1.91	0.092	0.33	−0.18, 0.85	0.204	−27.04	−81.35, 27.27	0.329
≥1 portion per day of each of fruits, green leafy vegetables, cooked vegetables AND ≥2 portions per week of fish	2.36	1.43, 3.29	< 0.001 *	0.98	0.52, 1.45	< 0.001 *	−25.53	−74.32, 23.24	0.305
BMI
<25	Ref.			0.00			0.00
25–29.9	1.34	0.33, 2.34	0.009 *	0.75	0.24, 1.26	0.004 *	1.58	0.41, 2.74	0.008 *
≥30	2.00	0.96, 3.04	< 0.001 *	1.10	0.58, 1.63	< 0.001 *	2.65	1.43, 3.86	< 0.001 *
Blood pressure
<120/80 mmHg	Ref.			0.00			0.00
<120/80 mmHg treated or 120–139/80–89 mmHg	−0.49	−1.87, 0.88	0.482	−0.70	−1.40, −0.01	0.047 *	−1.04	−2.65, 0.57	0.206
≥140/90 mmHg	−0.86	−2.05, 0.31	0.151	−0.88	−1.47, −0.28	0.004 *	−1.47	−2.86, −0.09	0.037 *
Fasting glucose
>126 mg/dL	Ref.			0.00			0.00
100–126 mg/dL or <100 mg/dL treated	−3.37	−5.17, −1.58	< 0.001 *	−0.83	−1.75, 0.07	0.071	29.79	−65.17, 124.76	0.538
<100 mg/dL untreated	−1.63	−3.36, 0.09	0.065	−0.69	−1.57, 0.18	0.121	−1.46	−93.30, 90.36	0.975
Total cholesterol
Optimal (<200)	Ref.			Ref.			Ref.
Borderline (200–239)	−0.75	−2.27, 0.75	0.326	−0.71	−1.43, −0.01	0.050	−1.51	−3.36, 0.34	0.111
High (>239)	1.93	−0.87, 4.74	0.178	0.11	−1.21, 1.44	0.864	−0.05	−3.55, 3.44	0.974
CFS	−2.07	−2.43, −1.70	< 0.001 *	−0.83	−1.00, −0.66	< 0.001 *	−7.62	−28.01, 12.76	0.463
SNS	0.10	0.05, 0.14	< 0.001	0.06	0.04, 0.08	< 0.001 *	0.68	−1.71, 3.09	0.574
Life's Simple Seven (LS7)	0.25	0.015, 0.492	0.038 *	0.12	0.01, 0.24	0.034 *	−5.05	−17.36, 7.26	0.421

Abbreviations: BMI, body mass index; BP, blood pressure; CFS, Clinical Frailty Scale score; CSID, Community Screening Instrument for Dementia; IDEA, Identification and Intervention for Dementia in Elderly Africans; LS7, Life's Simple 7; MoCA, Montreal Cognitive Assessment; RBG, random blood glucose; SNS, Social Network Scale (score).

^* p−value < 0.05 was considered statistically significant.

3.3. Independent determinants of cognitive performance using composite LS7 score

In the multiple linear regression analysis (Table 4A), the independent determinants of general CH (using total MoCA score) with corresponding beta coefficients (95% CI) were age (−0.11, −0.17 to −0.04), CFS score (−0.59, −1.24 to 0.04), years of education (0.66, 0.52–0.79)—overall; and CFS score (−1.93, −3.06 to −0.80)—for males; age (−0.13, −0.21 to −0.05) and years of education (0.77, 0.59–0.94)—for females. Results also showed that the independent determinants of general CH (using total IDEA score) with corresponding beta coefficients (95% CI) were age (−0.05, −0.09 to −0.04); years of education (0.16, 0.09–0.23) and SNS score (0.06, 0.03–0.08)—overall; CFS (–0.77, −1.35 to −0.20)—males; age (−0.07, −0.11 to −0.03), years of education (0.18, 0.10–0.27), and SNS score (0.07, 0.04–0.10)—females (Table 4B). The composite LS7 score had a non‐significant independent relationship with MoCA and IDEA scores in the overall study population and when stratified by sex (Tables 4A, B). There was no significant sex differential in the relationship between the composite LS7 score and cognitive performance.

TABLE 4A.

Multiple linear regression of the LS7 composite score and cognitive performance (MoCA)—overall and stratified per gender.

	Overall	Overall	Male		Female
	$\beta$	95% CI	$\beta$	95% CI	$\beta$	95% CI
Age, years	−0.11***	−0.17, −0.04	−0.01	−0.13, 0.10	−0.13***	−0.21, −0.05
Gender, female	−1.36	−2.78, 0.59	NA	NA
Marital status (ref: married)
Widow/widower	−0.42	−2.47, 1.62	−0.19	−5.81, 5.43	−0.35	−2.62, 1.92
Others	−0.26	−2.58, 2.06	0.66	−3.78, 5.11	−1.26	−4.15, 1.62
Years of education	0.66***	0.52, 0.79	0.47	0.24, 0.69	0.77***	0.59, 0.94
Living situation (ref: alone)
With spouse and children	1.25	−0.98, 3.50	1.53	−2.89, 5.95	1.65	−1.03, 4.35
With spouse	1.85	−0.60, 4.31	1.75	−3.03, 6.54	1.86	−1.09, 4.81
With extended family	−1.02	−3.13, 1.08	−2.39	−7.54, 2.74	−0.58	−2.96, 1.78
With children	0.79	−0.83, 2.42	−0.66	−5.98, 4.65	1.29	−0.52, 3.10
Alcohol use (ref: no)	−0.41	−1.69, 0.87	−1.17	−3.28, 0.92	0.33	−1.32, 1.98
Clinical frailty score	−0.59	−1.24, 0.04	−1.93	−3.06, −0.80	0.02	−0.77, 0.82
Social Network Scale score	0.03	−0.01, 0.08	0.01	−0.09, 0.09	0.04	−0.01, 0.10
LS7 score (ref: poor)	−0.19	−0.53, 0.13	0.01	−0.59, 0.60	−0.29	0.71, 0.11

^*** p−value < 0.05 was considered statistically significant.

TABLE 4B.

Multiple linear regression of the LS7 composite score and cognitive performance (IDEA)—overall and stratified per gender.

	Overall	Overall	Male		Female
	$\beta$	95% CI	$\beta$	95% CI	$\beta$	95% CI
Age, years	−0.05**	−0.09, −0.02	−0.01	−0.07, 0.05	−0.07**	−0.11, −0.03
Gender, female	−0.39	−1.09, 0.30	NA		NA	NA
Marital status (ref: married)
Widow/widower	−0.83	−1.89, 0.22	−1.20	−4.56, 2.16	−0.73	−1.91, 0.44
Others	−0.90	−2.10, 0.28	−1.27	−3.69, 1.13	−1.07	−2.56, 0.41
Years of education	0.16**	0.09, 0.23	0.08	−0.03, 0.19	0.18**	0.10, 0.27
Living situation (ref: alone)
With spouse and children	−0.51	−1.66, 0.63	−0.86	−3.27, 1.55	−0.33	−1.70, 1.04
With spouse	−0.07	−1.32, 1.17	−0.99	−3.59, 1.59	0.13	−1.04, 1.63
With extended family	−1.34**	−2.37, −0.31	−1.31	−3.84, 1.20	−1.22**	−2.39, −0.05
With children	0.56	−0.25, 1.38	0.35	−2.57, 3.28	0.72	−0.19, 1.63
Alcohol use (ref: no)	−0.82**	−1.45, −0.19	−1.26**	−2.29, −0.23	−0.43	−1.28, 0.38
Clinical Frailty Scale score	−0.28	−0.60, 0.04	−0.77**	−1.35, −0.20	−0.09	−0.50, 0.30
Social Network Scale score	0.06**	0.03, 0.08	0.04	−0.01, 0.09	0.07**	0.04, 0.10
LS7 score (ref: poor)	0.04	−0.12, 0.20	0.02	−0.26, 0.31	0.03	−0.17, 0.23

Abbreviations: IDEA, Identification and Intervention for Dementia in Elderly Africans; LS7, Life's Simple 7; MoCA, Montreal Cognitive Assessment.

^** p‐value < 0.05 was considered statistically significant.

3.4. Independent determinants of cognitive performance using individual LS7 metrics

As shown in Table 5A and B, age, years of education, CFS score, and SNS score showed a significant association with cognitive impairment using either the MoCA or IDEA scales. Of the individual LS7 metrics, only ideal diet (1.06, 0.16–0.98) and ideal BMI (−0.73, −1.42 to −0.05) maintained an independent association with CH using the IDEA score (Table 5B), but not with the MoCA score (Table 5A). Among female participants, ideal diet (1.04, 0.01–2.08) and intermediate BP (– 2.05, –3.78 to –0.33) showed an independent relationship with IDEA and MoCA score, respectively (Table 5A, B). There was otherwise no significant sex differential in the relationship between individual LS7 metrics and CH.

TABLE 5A.

Multiple linear regression of the individual LS7 metrics and cognitive performance (MoCA)—overall and by gender.

	Overall	All	Male		Female
Factors	$\beta$	95% CI	$\beta$	95% CI	$\beta$	95% CI
Age, years	−0.12**	−0.19, −0.05	−0.01	−0.15, 0.12	−0.14**	−0.22, −0.05
Gender, female	−1.45	−3.00, 0.08	NA	NA	NA	NA
Marital status (ref: married)
Widow/widower	−0.24	−2.34, 1.86	−0.72	−7.14, 5.69	0.01	−2.33, 2.35
Others	−0.03	−2.42, 2.35	0.71	−4.25, 5.68	−0.98	−3.93, 1.97
Years of education	0.64**	0.51, 0.78	0.49**	0.24, 0.73	0.74**	0.56, 0.92
Living situation (ref: alone)
With spouse and children	1.35	−0.93, 3.64	1.60	−3.30, 6.51	2.01	−0.75, 4.78
With spouse	1.86	−0.65, 4.37	1.45	−3.84, 6.74	2.17	−0.87, 5.22
With extended family	−0.93	−3.12, 1.26	−2.16	−7.94, 3.60	−0.37	−2.83, 2.08
With children	0.77	−0.89, 2.44	−0.32	−6.18, 5.53	1.38	−0.46, 3.23
Alcohol use (ref: no)	−0.32	−1.64, 0.99	−1.11	−3.42, 1.20	0.43	−1.24, 2.12
CFS	−0.72**	−1.41, −0.03	−1.93**	−3.24, −0.62	−0.20	−1.06, 0.64
SNS score	0.03	−0.01, 0.09	0.01	−0.08, 0.12	0.05	−0.01, 0.12
RBG (ref: poor)
Intermediate	—		—		—
Ideal	−0.32	−2.60, 1.95	−0.86	−5.15, 3.42	0.20	−2.55, 2.96
BP (ref: poor)
Intermediate	−1.30	−2.65, 0.04	0.24	−2.25, 2.74	−2.05**	−3.78, −0.33
Ideal	−0.37	−2.06, 1.30	1.28	−2.08, 4.66	−1.04	−3.02, 0.94
Total cholesterol (ref: poor)
Intermediate	—				—
Ideal	−0.50	−2.29, 1.28	−1.06	−8.39, 6.27	−0.25	−2.15, 1.63
Smoking (ref: poor)
Intermediate	−1.19	−6.95, 4.56	−1.85	−7.90, 4.19	—
Ideal	−0.23	−3.62, 3.16	−0.13	−3.78, 3.51	—
Physical activity (ref: poor)
Intermediate	−0.34	−2.18, 1.50	−0.14	−4.33, 4.04	−0.41	−2.53, 1.71
Ideal	−1.11	−2.99, 0.76	−0.98	−5.48, 3.51	−1.18	−3.30, 0.94
Diet (ref: poor)
Intermediate	−0.16	−1.73, 1.41	0.24	−4.33, 4.04	−0.39	−2.29, 1.50
Ideal	−0.22	−1.94, 1.49	−0.41	−5.48, 3.51	−0.37	−2.45, 1.70
BMI (ref: poor/high BMI)
Intermediate	−0.26	−1.74, 1.21	1.69	−2.78, 3.28	−0.88	−2.56, 0.79
Ideal	0.13	−1.28, 1.55	1.18	−3.67, 2.84	0.13	−1.46, 1.74

Abbreviations: BMI, body mass index; BP, blood pressure; CFS, Clinical Frailty Scale score; LS7, Life's Simple 7; MoCA, Montreal Cognitive Assessment; RBG, random blood glucose; SNS, Social Network Scale score.

^** p‐value < 0.05 was considered statistically significant.

TABLE 5B.

Multiple linear regression of the individual LS7 metrics and cognitive performance (IDEA)—overall and by gender.

	Overall	All	Male		Female
Factors	$\beta$	95% CI	$\beta$	95% CI	$\beta$	95% CI
Age, years	−0.05**	−0.08, −0.17	0.01	−0.6, 0.06	−0.06**	−0.10, −0.01
Gender, female	−0.66	−1.40, 0.08	NA	NA	NA	NA
Marital status (ref: married)
Widow/widower	−0.56	−1.63, 0.50	−1.33	−4.93, 2.26	−0.41	−1.62, 0.78
Others (separated/single)	−0.55	−1.75, 0.64	−0.87	−3.47, 1.73	−0.82	−2.32, 0.68
Years of education	0.15**	0.08, 0.22	0.08	−0.03, 0.20	0.18**	0.09, 0.27
Living situation (ref: alone)
With spouse and children	−0.50	−1.65, 0.64	−1.03	−3.66, 1.59	−0.23	−1.63, 1.15
With spouse	0.04	−1.21, 1.30	−1.05	−3.86, 1.76	0.36	−1.17, 1.90
With extended family	−1.36**	−2.40, −0.31	−1.05	−3.75, 1.63	−1.22**	−2.41, −0.03
With children	0.47	−0.33, 1.29	0.33	−2.69, 3.37	0.64	−0.27, 1.56
Alcohol use (ref: no)	−0.93**	−1.57, −0.30	−1.37**	−2.47, −0.27	−0.56	−1.39, 0.26
CFS	−0.30	−0.64, 0.03	−0.77**	−1.41, −0.13	−0.17	−0.59, 0.25
SNS score	0 .06**	0.03, 0.08	0.04	−0.01, 0.09	0.07** *	0.03, 0.10
RBG (ref: poor)
Intermediate	—
Ideal	0.39	−0.66, 1.46	−0.32	−2.30, 1.66	0.55	−0.75, 1.86
BP (ref: poor)
Intermediate	−0.27	−0.92, 0.37	−0.13	−1.29, 1.02	−0.52	−1.36, 0.31
Ideal	0.68	−0.12, 1.48	1.12	−0.44, 2.70	0.46	−0.50, 1.43
Total cholesterol (ref: poor)
Intermediate	—		—
Ideal	0.41	−0.43, 1.26	−0.20	−3.65, 3.24	0.42	−0.49, 1.35
Smoking (ref: poor)
Intermediate	−1.21	−4.01, 1.57	−1.58	−4.42, 1.24	—
Ideal	−0.67	−2.37, 1.03	−0.96	−2.72, 0.79	—
Physical activity (ref: poor)
Intermediate	0.23	−0.64, 1.11	−0.53	−2.53, 1.45	0.42	−0.60, 1.45
Ideal	−0.05	−0.94, 0.82	−0.78	−2.91, 1.34	0.06	−0.95, 1.08
Diet (ref: poor)
Intermediate	0.64	−0.12, 1.40	−0.17	−1.58, 1.22	0.90	−0.05, 1.85
Ideal	1.06**	0.22, 1.89	0.85	−0.65, 2.35	1.04**	0.01, 2.08
BMI (ref: poor/high BMI)
Intermediate	−0.44	−1.16, 0.26	−0.98	−2.75, 0.78	−0.46	−1.29, 0.36
Ideal	−0.73**	−1.42, −0.05	−1.31	−3.06, 0.43	−0.59	−1.39, 0.19

Abbreviations: BMI, body mass index; BP, blood pressure; CFS, Clinical Frailty Scale score; IDEA, Identification and Intervention for Dementia in Elderly Africans; LS7, Life's Simple 7; RBG, random blood glucose; SNS, Social Network Scale score.

^** p−value < 0.05 was considered statistically significant.

4. DISCUSSION

Among ≈1000 older adults, with predominantly suboptimal CVH, older age, lower years of education, clinical frailty, and social network, but not LS7 score, showed an independent association with CH. Although there was an independent association between some individual metrics within the LS7 composite measure (poor diet and ideal BMI [overall]; intermediate BP [female participants]) and CH at baseline, we did not find an independent association between the composite LS7 score and CH. It is possible that cardiovascular risk markers measured at baseline influence the cognitive trajectory, rather than the baseline CH status, as shown previously in a prospective study in the same population. ³³ Furthermore, results from the unadjusted and adjusted analyses suggest that the influence of individual components of LS7 is negating or cancelling out one another when the variables/items are used for the composite index. This implies that the generation of the LS7 composite measure may need further refinement, especially for application among indigenous Africans. Our findings are consistent with those from the Multi‐Ethnic Study of Atherosclerosis (MESA) and other studies, where the association of vascular risk factors and cognitive performance was shown to differ based on the risk score used, ³⁴ and further modified by racial and ethnic differences. ³⁴ , ³⁵ , ³⁶ In addition, the lack of association between LS7 score and CH observed in our study may be attributable to the pattern of cardiovascular risk in this urban‐dwelling older population, specific population demographic characteristics, the phenomenon of reverse causality, and the heterogeneity of vascular risk burden among indigenous Africans.

The heterogeneity of vascular risk factors based on population setting meant that the LS7 score as a composite marker may not be ideal for certain populations/age groups, ¹³ , ³⁷ , ³⁸ especially populations with low levels of education, poor nutrition, and lower prevalence of obesity. ¹⁹ , ³³ Although some of the individual vascular metrics (such as systolic BP, physical inactivity, low BMI, and unhealthy diet in the bivariate analysis; poor diet, intermediate BP, ideal BMI in the multivariate analysis) showed an association with CH, the LS7 score, originally modeled from datasets from a Caucasian population, ⁵ may not be ideal for an indigenous African populations. ³⁹ , ⁴⁰ , ⁴¹ , ⁴² For instance, in this study, an ideal BMI (<25 kg/m2) was associated with worse CH. Older persons (at risk of dementia) have less muscle mass and are significantly more frail (Table S1). This observed relationship may be a consequence of poor nutrition in advanced dementia. These findings are similar to those from a larger cohort of 1.3 million individuals, where higher BMI was associated with reduced dementia risk in a phenomenon known as reverse causation. ⁴³ Among female participants, the observed relationship between intermediate BP and poor CH may similarly be attributable to reverse causality. In our study, and as reported in another previous study in the same population, ⁴⁴ there was a lower prevalence of obesity (19.5%), compared to that of a similar U.S. population (39.6%). ⁴⁵ Precisely 16% of our study population was underweight, and this is higher than that reported in the same population (5.3%) but among much younger individuals. ⁴⁴ Weight loss, not obesity, has been associated with dementia in the African setting. ³⁹ , ⁴⁶ Per initial design, persons who are underweight are not captured on the LS7 scale, ⁸ and this may be contributing to the observed non‐significant association.

In this study, we similarly observed that an unhealthy diet was independently associated with poor CH. This is in tandem with other population‐based and epidemiologic studies in the region, which have highlighted the benefit of a healthy diet as protective against vascular diseases, such as hypertension, stroke occurrence, and stroke severity. ⁴⁷ , ⁴⁸ , ⁴⁹ , ⁵⁰ An ideal and balanced diet is known to reduce oxidative stress, supply micronutrients, and improve body weight, which in turn improves brain volume, brain mass, CH, and cognitive resilience. ⁵⁰ , ⁵¹ The protective effects of green leafy vegetable consumption on vascular risk reduction have been attributed to high folic acid content and effects on hyper‐homocysteinemia. ⁵² Improved dietary patterns with folic acid supplementation may potentially be a low‐cost strategy for population‐wide dementia risk reduction, as has been suggested with stroke. ⁵¹ Indeed, the independent association between frailty and CH noted in our study supports the emerging importance of frailty as a construct in geriatric medicine and may similarly be linked to the finding of unhealthy diet and low BMI, which were independently associated with poor CH in our study population. Frailty, a surrogate marker of poor cardiometabolic and CH, is linked to poor physical fitness and nutrition, and may reflect cognitive vulnerability. In contrast to our findings, authors in the MESA study, however, noted that adherence to a healthy (Dietary Approaches to Stop Hypertension [DASH]) diet was not associated with cognitive decline, ⁵³ although these results are not entirely consistent across different studies. ¹⁶

Approximately 85% of our study population had intermediate LS7 scores. Despite this, the mean cognitive score in our population was lower compared to studies in Caucasian populations with poor LS7 scores and higher vascular burden. ²⁰ , ⁵⁴ , ⁵⁵ This suggests that some other factors (not hitherto captured on the LS7) may be contributory to poor cognitive performance in African populations, and these should be targeted for specific intervention. These factors include various environmental, socio‐economic factors, as well as life events that may be peculiar to the African population. ⁴⁶ The role of life events was documented in a cross‐sectional study among 977 elderly Africans, where the death of a parent during childhood and recent relocation were shown to be associated with cognitive impairment. ⁴⁶ In addition, the association between CVH and CH may vary based on the specific CVH risk marker considered and specific population, genetic, lifestyle, diet, or environmental characteristics ¹² , ³³ , ⁴² , ⁵⁶ —for instance, air pollution, visual and hearing impairment—which were not captured in this population. As expected, however, older age, level of education, and clinical frailty showed independent association with CH. ¹ , ² This underlies the established contribution of older age, low levels of education, and clinical frailty to cognitive impairment in the African context ⁵⁷ and has implications for strengthening public health intervention and policy formulation to reduce the burden of neurocognitive disorders. ³⁹ , ⁴⁶ , ⁵⁸

It is also important to highlight some specific characteristics of our study population that make our study findings unique. Overall, our study population was much older, with an observed prevalence of hypertension of 60%. This is higher than the national average of 30% ⁵⁹ as well as that of a similar study in the same community, ⁴⁴ which involved a much younger population with a mean age of 38.8 ± 15.6 years. In addition, our study population comprised predominantly poorly educated older women who were widows, whereas the men were fewer, more educated, and with spouses (Table S1). This has implications for the effect of gender and educational attainment on CH. In the bivariate analysis (Table 3), systolic BP ≥140/90 mmHg and physical inactivity were associated with poor cognitive health, similar to reports in literature, ⁹ , ³³ whereas other cardiometabolic markers such as diabetes mellitus, dyslipidemia, and hypercholesterolemia were not. Of note, in this study, the prevalence of diabetes mellitus and hypercholesterolemia (Table 2) was significantly lower than that from Western cohorts, regional studies, ⁶⁰ and the national average. ⁶¹ , ⁶² This may be reflective of the population setting, demographic characteristics, and dietary pattern in this low‐income urban Nigerian community. ⁴⁰ , ⁴⁶ Indeed, in a recent nationwide study, the overall prevalence of elevated low‐density lipoprotein cholesterol (LDL‐C), hypertriglyceridemia, and hypercholesterolemia was similarly low—13.6%, 21.4%, and 7.5%, respectively. ⁶² Behavioral components of the LS7 metrics, like smoking, Western diet, and reduced physical activity, were similarly less prevalent in our study population. As shown, the prevalence, pattern of, and effect size of vascular risk factors for cognitive impairment and dementia may differ in sub‐Saharan Africa, where a myriad of poorly quantified life events, environmental (air pollution), educational, and socio‐economic factors hold sway. ⁴⁶

Our study has several notable strengths. Being a population‐based design, our findings may be generalizable to the prototypal low‐income urban African community. However, given the diverse socio‐ethno‐cultural peculiarities of the African continent, follow‐up studies will be needed to tease out context‐specific differences within other African sub‐populations. In addition, the use of multiple cognitive assessment measures and rigorous analytic methods ensures that our findings are consistent across several cognitive batteries. However, we acknowledge a few limitations. Our study population was predominantly female, who had longer life expectancy than their male counterparts, were more willing to take part in the study, and were readily found in the home setting (less nomadic). We, however, note that the female gender has a higher risk for neurocognitive disorders. We also noted incomplete individual metrics in a few participants, thus rendering the LS7 composite scoring incomplete. To mitigate this, we not only analyzed LS7 as a composite marker but also analyzed per‐individual metrics.

This cohort of older Nigerians, with predominantly average CVH profiles, older age, lower educational attainment, and clinical frailty, but not the LS7 composite score, showed an independent association with poor cognitive health. The individual LS7 metrics of poor diet, intermediate BP, and ideal BMI showed independent relationships with poor cognitive function. Improved dietary approaches to improve physical health (and reduce frailty) offer a low‐cost strategy for population‐wide dementia risk reduction. The LS7 score as a composite marker of CVH may not be ideal for certain populations/age groups. Afrocentric CVH composite scores incorporating measures that are more sensitive to cognitive outcomes in Africans are desirable. To adequately characterize the association between vascular and cognitive health, population‐specific, age‐appropriate composite markers are required to enable targeted, contextually tailored primary prevention measures. This association may vary based on the population setting; hence a follow‐up study in different indigenous African population settings is needed. It may also be useful to investigate associations between the LS7 score and other endophenotypes of CVH, such as carotid intima media thickness (CIMT) in the same population.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflicts of interest. Any author disclosures are available in the supporting information.

CONSENT STATEMENT

All human subjects provided written informed consent

Supporting information

Supporting Information

Supporting Information

Akinyemi RO, Olalusi OV, Ogunde GO, et al. Association between cardiovascular and cognitive health among older Indigenous Africans: Data from an urban Nigerian settlement participating in the VALIANT study. Alzheimer's Dement. 2025;21:e70669. 10.1002/alz.70669