Cardiometabolic‐Inflammatory Risk Factors and Cognitive Decline Among Older Indians—Report From a Nationally Representative, Longitudinal Study

ABSTRACT

Background

Rapid increase in cardiometabolic diseases in India may contribute to increased incidence of late‐life cognitive impairment. This study focuses on associations between baseline cardiometabolic risk factors and subsequent decline in cognitive function among older adults in India, leveraging data from two waves (Wave 1: 2017–2020, Wave 2: 2022–2024) of the Longitudinal Aging Study in India–Diagnostic Assessment of Dementia (LASI‐DAD).

Methods

Analysis included longitudinal data of 1554 study participants. A summary measure of different cognitive functional domains was used. Cognitive decline was defined as annual decline in cognitive score ≥ 0.05 times the standard deviation of the summary score. Cardiometabolic risk was characterized using cardiovascular, metabolic, and inflammatory biomarkers. Multivariate, multinomial logistic regression analysis was used to examine the associations between cardiometabolic risk and cognitive decline.

Results

At baseline, 71.7% of the sample had elevated homocysteine levels, 44.4% had elevated blood pressure, 23% had elevated glycosylated hemoglobin (HbA1c), and 6.7% had elevated uric acid levels. Between the two waves, 34.8% experienced significant cognitive decline, while 35.6% died. Multivariate multinomial logistic regression showed significant cognitive decline was associated with elevated blood pressure [odds ratio (OR): 1.7, 95% confidence interval (CI) 1.3–2.2], elevated HbA1c (OR: 1.1, 95% CI: 1.0–1.2), being overweight (OR: 1.4, 95% CI: 1.0–2.0), and elevated uric acid level (OR: 1.2, 95% CI: 1.0–1.3). Those with hypertension had 1.5 times higher odds of mortality (95% confidence interval: 1.2–2.0), while those with diabetes mellitus or elevated pro‐brain natriuretic peptide had 1.2 times (95% CI: 1.1–1.4), and 1.8 times (95% CI: 1.1–1.4) higher odds of mortality.

Conclusion

Cardiometabolic risk factors play a significant role in late‐life cognitive decline and death among older Indians. These longitudinal relationships from LASI‐DAD highlight potentially modifiable risk factors and inform potential prevention policies.

Summary

• key points

  • ○
    Aging is associated with both increased risk of cognitive decline and mortality.
  • ○
    The study showed that significant cognitive decline was associated with potentially modifiable cardiometabolic risk factors like elevated blood pressure (odds ratio: 1.7, 95% confidence interval: 1.3–2.2), elevated glycosylated hemoglobin, (odds ratio: 1.1, 95% confidence interval: 1.02–1.2), being overweight (odds ratio: 1.4, 95% confidence interval: 1.02–2.0), and elevated uric acid level (odds ratio: 1.2, 95% confidence interval: 1.0–1.3).
  • ○
    Hypertension, diabetes, being overweight, and elevated uric acid level increase the risk of both cognitive decline and mortality whereas elevated levels of Pro‐BNP, white blood cells (WBC), C‐reactive protein (CRP), and lower levels of HDL cholesterol and albumin are associated with increased mortality risk but not cognitive decline.
  • ○
    Risk factors for cognitive decline observed in this longitudinal data are largely modifiable by lifestyle interventions and medications. Our findings indicate that awareness, early diagnosis, and timely intervention of these risk factors in populous and aging nations like India are essential for preserving cognitive health as well as reducing mortality.
• why this paper matters

  • ○
    Longitudinal data provided by LASI‐DAD study shed light on how cardiometabolic risk factors may influence decline of cognitive function and development of dementia in older adults in India. This study shows that raising awareness and initiating preventive measures for these potentially modifiable risk factor like high blood pressure, diabetes, overweight, and high serum uric acid levels may help mitigate late life cognitive decline.

1. Background

The global demographic trend of population aging is predicted to be more pronounced in low‐and‐middle‐income countries (LMIC) like India [1]. With this demographic shift and an increase in life expectancy, the population at risk for developing cognitive impairment and dementia will increase exponentially [2]. According to World Health Organization (WHO) estimates, the number of individuals living with dementia globally will increase to a projected 82 million in 2030 and 152 million in 2050, and highly populated South Asian countries like India will be key contributors to this increased burden [3]. It has been reported that approximately 8.8 million older Indian adults (60 years and above) presently live with dementia [4].

A complex combination of genetic and environmental factors contributes to the development of dementia, with neurodegeneration and vascular processes being the most common direct causes [5]. India is undergoing an “epidemiological transition,” characterized by a swift increase in cardiovascular and metabolic diseases [6]. It is reported to have the second largest diabetic population in the world [7]. Hypertension, either self‐reported or elevated blood pressure on objective measurement, may be present in two‐thirds of older Indians [8]. Furthermore, diagnosis and management of these conditions are often sub‐optimal due to lack of health literacy and limited access to quality health care, particularly in rural India [9]. Therefore, cardiometabolic diseases may well play a more important role in the pathogenesis of dementia in India, compared to more developed countries.

Previous studies have focused on the cross‐sectional associations between cardiometabolic risks and cognitive impairment and dementia in India [10]. Nevertheless, robust longitudinal data—essential for observing the trajectory of how these cardiometabolic risk factors influence subsequent cognitive decline—is still lacking [11]. Recent longitudinal studies like the Tata Longitudinal Study on Aging (TLSA) and the Srinivaspura Aging, NeuroSenescence and COGnition (SANSCOG) study are limited by their geographic restrictions and lack of socio‐demographic diversity representative of the Indian subcontinent [12].

The Longitudinal Aging Study in India—Harmonized Diagnostic Assessment of Dementia (LASI‐DAD) is an ongoing, nationally representative study focusing on risk factors impacting late‐life cognition and dementia [13, 14]. It attempts to fill existing knowledge gaps by collecting rich epidemiological, clinical, lifestyle, and environmental data from community‐dwelling adults 60 years and above in India. LASI‐DAD has completed two waves of data collection. Previous cross‐sectional analysis using LASI‐DAD wave 1 data has shown that lower cognitive function was associated with older age, lower educational attainment, elevated levels of homocysteine and pro‐brain natriuretic peptide (proBNP), and lower levels of glycosylated hemoglobin (HbA1c) and albumin [15]. With the newly available data from the LASI‐DAD second wave, which was conducted at an interval of approximately 5 years after the first, we focus on how baseline cardiometabolic risk factors are associated with subsequent decline in cognitive function among older adults in India. We hypothesize that increased cardiometabolic diseases are associated with a more rapid subsequent decline in cognitive function.

2. Methods

2.1. Study Population

The study population is LASI‐DAD participants who have completed both waves. The details of study design and methods have been described elsewhere [13]. Briefly, LASI‐DAD is a nationally representative prospective cohort study of adults 60 years and older in India, covering 22 states and union territories. A multi‐stage stratified sampling strategy was used to select respondents from the broader Longitudinal Aging Study in India (LASI), which includes over 73,000 adults aged 45 years and older and their spouses. As the study focusses on late life cognitive impairment and its correlates, individuals at higher risk of cognitive impairment were oversampled to ensure sufficient numbers of respondents with dementia and mild cognitive impairment. To ensure a nationally representative sample, there were no exclusion criteria beyond age eligibility. The response rate was high, 83% and 84% respectively in the first and second waves. Among 2465 participants who had cognitive test scores and biomarker data from wave 1, 1005 had cognitive scores from both waves and 549 died during the follow‐up period. Therefore, a total of 1554 subjects have been included in analysis for this study (Figure 1).

FIGURE 1.

Depicts the flow of the LASI ‐DAD study (both waves 1 and 2) and the sample selection method for the present study. Abbreviations: LASI, Longitudinal Aging Study in India, LASI DAD, Longitudinal Aging Study in India—Harmonized Diagnostic Assessment of dementia.

2.2. Methods

2.2.1. Cognitive Function Measure and Death

Cognitive function was measured using a summary score based on the following tests: 10‐word learning, including immediate and delayed recall and recognition; [16] logical memory story recall, including immediate and delayed recall; [17] a validated Hindi version of the Mini Mental State Examination (MMSE) summary score; [18] verbal fluency score, which was the number of named animals within 60 s; [16] four items from the community screening instrument for dementia score (CSID), one each from object naming (i.e., elbow), concept nomination (“What do you do with a hammer”?), orientation‐(“where is the local market/store”?), and comprehension domains (asking to point first to the door and then the window), [19] and the Raven test (a count of the number of correct answers to a series of images that required the respondent to select the missing piece) [20]. Culturally appropriate modifications were made in the cognitive tests mentioned above as most of the tests have been validated on a predominantly western population. Further, translation and back‐translation of these questionnaires to different languages used in data collection have been done and validated during pilot studies before the collection of the wave 1 data. The same cognitive summary score was used in the previous cross‐sectional analysis of the associations between cognitive function and cardiometabolic risk factors [15]. The total cognitive scores ranged from 0 to 175.

Cognitive decline was defined as an annual decline in the cognitive summary score equal to or greater than 0.05 times the standard deviation, given prior evidence showing that the average annual decline in population‐based samples of older adults is often between 0.02 and 0.05 [21]. We calculated the exact time difference between two waves, using interview dates, to estimate annual change. Since some respondents died during the follow‐up period (N = 549), we also included death as a secondary outcome in analyses.

2.2.2. Cardiometabolic and Inflammatory Biomarkers

Cardiometabolic risk was measured using cardiovascular, metabolic, and inflammatory biomarkers based on physical measures and venous blood‐based assays. For this analysis, cardiovascular risk factors included high blood pressure on objective measurement (systolic blood pressure ≥ 140 mmHg or diastolic blood pressure ≥ 90 mmHg), tachycardia on measurement (pulse rate greater than 100 beats per minute), N terminal pro‐B‐type natriuretic peptide (NT pro‐BNP), and homocysteine. Metabolic risk was measured by calculated body mass index (BMI), glycosylated hemoglobin (HbA1c), high‐density lipoprotein (HDL) cholesterol, and lipoprotein (a). Inflammatory risk was indicted by white blood cell count (WBC), C‐reactive protein (CRP), albumin (a negative acute phase reactant), and uric acid.

2.2.3. Demographics

Sociodemographic characteristics included age (60–64, 65–69, 70–74, and ≥ 75 years), sex, and years of education (0, 1–11, 12, and ≥ 13 years).

2.3. Statistical Analysis

We applied commonly used clinical cut‐offs (Table 1) to define at‐risk categories for biomarkers. Baseline descriptive statistics were summarized. To examine how baseline cardiometabolic risk factors were related to subsequent cognitive decline, we performed multivariate multinomial logistic regression analysis, after controlling for age, gender, and education, where the outcomes of cognitive decline and death were compared to no decline in cognitive function. All statistical analyses were conducted using SAS, and statistical significance was defined as p < 0.05.

TABLE 1.

Characteristics of LASI‐DAD sample (N = 1554)–demographics and cardio‐metabolic inflammatory risk factor levels.

Variables	Percentage
Sociodemographic characteristics
Age, years
60–64	28.2%
65–69	32.3%
70–74	18.1%
≥ 75	21.4%
Sex (female)	47.4%
Education, years
0	50.2%
≤ 11	42.8%
≥ 12	6%
Cardiovascular risk factors
Systolic blood pressure ≥ 140 mmHg or diastolic blood pressure ≥ 90 mmHg (described as high blood pressure or hypertension in the study)	44.5%
Pulse > 100 beats/min (normal reference range 60–100/min)	5.9%
At risk pro‐ B‐type natriuretic peptide*	5.5%
Homocysteine > 15 μmol/L (reference range‐5–15 μmol/L)	71.8%
Metabolic risk factors
Body mass index, kg/m2
< 18.5 (underweight)	22.9%
18.5–24.9 (normal)	50.4%
25.0–29.9 (overweight)	19.0%
≥ 30 (obese)	7.7%
Glycosylated hemoglobin ≥ 6.5%**	23.0%
HDL‐cholesterol < 40 mg/dL (ideal range > 40 mg/dL)	38.1%
Lipoprotein (a) > 30 mg/dL (normal range < 30 mg/dL)	42.5%
Inflammatory risk factors
C‐reactive protein > 3 mg/L (reference range: 0–3 mg/dL‐ low‐ moderate risk of inflammation. > 3‐ high risk)	35.6%
White blood cell count > 11,000/mm3 (normal range‐ 4000–11,000/mm3)	5.6%
Albumin < 3.5 mg/dL (normal range 3.5–5.5 mg/dL)	3.3%
Uric acid > 7 mg/dL (normal range‐ 2.5–7 mg/dL)	6.8%
Cognitive outcome
Cognitively not declining	29.5%
Cognitively declining	34.9%
Died between wave 1 and wave 2	35.6%

Note: Reference serum levels of following risk factors ‐* At risk pro‐B‐type natriuretic peptide is ≥ 900 pg/mL for those aged 60–74 years and ≥ 1800 pg/mL for those aged 75 years and older. **Glycosylated hemoglobin HbA1c‐Normal: < 5.7%, Pre‐diabetes: 5.7%–6.4%, Diabetes: ≥ 6.5%.

3. Results

The characteristics of the study sample are summarized in Table 1. Almost half (47.4%) of the respondents were female, and 50.2% did not have any formal schooling. The most common at‐risk cardiovascular factor was elevated homocysteine level (71.8%), followed by elevated blood pressure on measurement (44.5%). Only 5.5% of respondents had elevated NT‐proBNP. Twenty‐three percent had HbA1c levels that meet the diagnostic criteria for diabetes mellitus. The prevalence for elevated uric acid levels was 6.8%. Between the two waves, 34.9% experienced significant cognitive decline and 35.6% had expired.

Multivariate multinomial logistic regression analysis shows that cognitive decline was significantly associated with elevated blood pressure on measurement [odds ratio (OR): 1.7, 95% confidence interval: 1.3–2.2], diabetes mellitus (OR: 1.1, 95% confidence interval: 1.02–1.2), overweight (OR: 1.4, 95% confidence interval: 1.02–2.0), and elevated uric acid level (OR: 1.2, 95% confidence interval: 1.1–1.4) (Table 2).

TABLE 2.

Association of cardiometabolic biomarkers with cognitive decline and mortality, after adjustment of age, sex, and education.

Biomarkers	Cognitive decline				Mortality
	OR	95% CI		p	OR	95% CI		p
Hypertension	1.70	1.31	2.20	< 0.00	1.58	1.21	2.07	0.00
Tachycardia	0.99	0.98	1.00	0.86	1.01	1.00	1.03	0.00
Elevated homocysteine	0.99	0.98	1.00	0.53	1.009	0.99	1.00	0.85
Elevated NT‐ pro BNP	0.92	0.80	1.06	0.29	1.88	1.63	2.16	< 0.00
Underweight	0.73	0.52	1.03	0.07	1.26	0.91	1.75	0.15
Overweight	1.43	1.02	2.02	0.03	1.54	1.06	2.23	0.02
Obesity	1.49	0.92	2.41	0.09	1.45	0.84	2.52	0.17
Diabetes	1.19	1.01	1.23	0.01	1.29	1.17	1.42	< 0.00
Elevated HDL cholesterol	0.99	0.98	1.00	0.26	0.98	0.97	0.99	0.02
Elevated lipoprotein (a)	1.00	0.99	1.00	0.49	1.00	0.99	1.00	0.24
Elevated CRP	1.03	0.93	1.14	0.52	1.33	1.19	1.47	< 0.00
Elevated WBC	1.03	0.67	1.58	0.87	2.01	1.28	3.14	0.00
Elevated albumin	0.92	0.59	1.44	0.73	0.13	0.09	0.22	< 0.00
Elevated uric acid	1.21	1.09	1.34	0.00	1.32	1.18	1.47	< 0.00

Abbreviations: CI‐ confidence interval; CRP, C reactive protein; NT‐proBNP, N‐terminal pro B type natriuretic peptide; OR, Odds ratio; WBC, white blood cell count.

Compared to LASI‐DAD respondents who passed away between two waves, the survivors were significantly younger, more likely to be females, and had higher education levels. Increased mortality was also associated with hypertension on measurement, diabetes, being overweight, and elevated uric acid level. Moreover, mortality was positively associated with elevated proBNP levels (OR: 1.9, 95% confidence interval: 1.6–2.2), CRP (OR: 1.3, 95% confidence interval: 1.2–1.5), and WBC levels (OR: 2.0, 95% confidence interval: 1.2–3.1), but inversely associated with HDL cholesterol levels (OR: 0.98, 95% confidence interval: 0.97–0.99) and albumin levels (OR: 0.1, 95% confidence interval: 0.09–0.2).

4. Discussion

The study results highlight that cardiometabolic biomarkers, such as elevated blood pressure, increased blood levels of glucose and uric acid, and being overweight, were independently associated with cognitive decline. These risk factors, as well as elevated pro‐BNP, CRP and WBC counts—together with lower HDL cholesterol and albumin levels—were also related to mortality risk. The changing epidemiological and demographic milieu in India and genetic factors contribute to increase in cardiometabolic risk factors like diabetes, hypertension, obesity, and metabolic syndrome (MetS) and may have adverse effect on the onset and progression of cognitive impairment and dementia [22]. A systematic review estimated burden of cardiometabolic syndrome in India as 30% due to swift urbanization, unhealthy dietary transitions, and sedentary lifestyles translating into obesity [23]. Recent studies from India have highlighted increased prevalence and suboptimal management of these risk factors mentioned above, more so in rural India [7]. The present study also observed association between high serum uric acid levels and cognitive decline in a representative sample of older Indians. It has been suggested that high serum uric acid levels can cross the blood brain barrier, causing inflammation in specific brain areas, leading to neuro‐degeneration and cognitive impairment [24]. Hyperuricemia has been correlated with high risk of MetS, which in turn increases risk of neurodegenerative diseases through induction of brain insulin resistance, neuroinflammation, mitochondrial dysfunction, and oxidative stress [25]. A recent Indian study by Singh et al. observed the rising prevalence of increased serum uric acid and its association with known MetS factors. This shift is attributable to major dietary and lifestyle changes due to rapid urbanization [26].

Our findings of a positive association between higher levels of pro BNP and CRP and mortality are consistent with the literature. Recent results published from a longitudinal, nationally representative Finnish study with a 10‐year follow up has shown that higher levels of Pro‐BNP increases the risk of cardiovascular and all‐cause mortality in older adults [27]. Inflammaging, a state of chronic, low‐grade inflammation associated with aging, usually causes a rise in inflammatory biomarkers like CRP [28]. Alterations in serum albumin levels are commonly observed with aging due to altered metabolism, chronic inflammation, nutritional status, and other health conditions. The findings in this study emphasize the importance of routine serum albumin testing as a potential prognostic marker in the aging population, as observed in previous work [29]. High prevalence of atherogenic dyslipidaemia—consisting of high triglyceride and LDL levels combined with low HDL cholesterol attributable to sedentary lifestyle, carbohydrate rich diets and deficiency of polyunsaturated fatty acids (PUFA) in Indian diets has also been implicated in developing the cardiometabolic risks in this population [30].

The risk factors mentioned above—hypertension, diabetes mellitus, obesity, and elevated uric acid levels—are largely modifiable by lifestyle interventions and medications. Our findings indicate that early diagnosis and appropriate treatment of cardiometabolic diseases in populous and aging nations like India are essential for preserving cognitive health as well as reducing morbidity and mortality. As India continues to undergo rapid economic development, this should be a public health priority at a policy and planning level of the government.

Significant mortality was seen between the two waves of the LASI‐DAD study. This could be related to the corona virus disease (COVID) pandemic. Literature on all‐cause mortality due to the COVID pandemic in India suggests that 26%–29% extra deaths occurred during 2020–2021 [31]. This equated to the deaths of more than 3.2 million Indians within a time span of approximately 1 year, peaking between April and June of 2021. These deaths presumably would have been the highest among older people, who were the target population of LASI‐DAD. Experts also believe that reported death rates might be much lower than actual numbers.

This study has several strengths. It is one of the first longitudinal analyses that has examined the associations between cardiometabolic risks and subsequent cognitive decline in India. LASI‐DAD is representative of the diverse Indian population from both rural and urban regions and heterogenous sociodemographic population from different states. Therefore, the results are generalisable for older Indian population. LASI‐DAD has implemented standardized venous blood collection, processing, shipment, and assay protocols to ensure high quality biomarker data. There are several limitations of the current study. First, the sample size for longitudinal analysis of cognitive decline is relatively small and the follow‐up is only approximately 5 years. Part of the reason for small sample size is the significant mortality between the two study waves, magnified by the COVID epidemic in India. Second, it is possible that some participants might have had decline in cognitive functional performance over time though the decline in cognitive summary score might be less than 0.05 standard deviation per year, leading to misclassification. However, we think this type of misclassification should be relatively uncommon. Third, the prevalence for some cardiometabolic risk factors, such as elevated proBNP and obesity is low, which may have further limited our ability to detect the associations between these and subsequent cognitive decline.

Despite these limitations, our study has shown that cardiometabolic risk factors play a significant role in the pathogenesis of dementia among older Indians. Going forward, we look ahead to the third wave of data collection of the LASI‐DAD study, which can shed more light on the longitudinal exposures to cardiometabolic risk factors and their associations with late life cognitive status and dementia in older adults in India.

Author Contributions

P.H., J.L., and J.B. conceived the idea for this manuscript. J.L., A.B.D., J.B., P.H., P.K. were critical in the conduct of the clinical study. J.B. and P.H. drafted the manuscript, performed revisions, and contributed their expertise in the area. J.K.K, E.N., P.H. contributed to the analysis of the results and important inputs in the revision of the manuscript. J.B., J.K.K., E.N., P.K., A.B.D., S.D., D.F., K.M.L., E.C., J.L., and P.H. were responsible for the critical revision of the manuscript for important intellectual content and their content expertise. All authors have read and approved the submission of this manuscript.

Funding

This project is funded by the National Institute on Aging, the National Institute of Health (R01 AG051125).

Disclosure

Funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Ethics Statement

The study was conducted according to the guidelines of the Declaration of Helsinki. Ethics approval was obtained from the Indian Council of Medical Research (2202–16,741/F1) and the collaborating institution, University of Southern California (UP‐15‐00684). Informed consent to take part in the study was obtained from all individual participants included in the study.

Conflicts of Interest

The authors declare no conflicts of interest.