Impact of cardiometabolic conditions on the progression from mild cognitive impairment to dementia: A large cohort study

Abstract

INTRODUCTION

This study investigates the impact of cardiometabolic conditions, including type 2 diabetes, hyperlipidemia, hypertension, and obesity, on the progression of mild cognitive impairment (MCI) to dementia.

METHODS

The cohort included adults ≥ 50 years old with MCI and a cardiometabolic condition identified through electronic health records. Propensity score matching was applied to control for confounders, and Kaplan–Meier analysis was used to assess time‐to‐event outcomes.

RESULTS

During a 7‐year median follow‐up, type 2 diabetes was associated with the highest risk of all‐cause dementia (hazard ratio [HR] 1.18, 95% confidence interval [CI]: 1.06 to 1.31), followed by hypertension and hyperlipidemia. For vascular dementia, type 2 diabetes conferred the greatest risk (HR 1.33, 95% CI: 1.07 to 1.64). Hyperlipidemia was the sole cardiometabolic factor significantly associated with Alzheimer's disease (AD) risk (HR 1.21, 95% CI: 1.11 to 1.32).

CONCLUSIONS

Hyperlipidemia is primarily associated with AD dementia risk, while type 2 diabetes is the major contributor to vascular dementia and all‐cause risk in individuals with MCI.

Highlights

• Type 2 diabetes, hypertension, and hyperlipidemia are associated with a high risk of developing all‐cause dementia in participants already diagnosed with mild cognitive impairment (MCI).

• Type 2 diabetes was shown to pose a high risk for the progression from MCI to vascular dementia.

• Hyperlipidemia was associated with Alzheimer's disease progression in individuals with MCI.

1. BACKGROUND

As the global population ages, the number of individuals affected by dementia is steadily increasing. In 2015, ≈ 47 million individuals worldwide were affected by dementia, and this number is projected to triple by 2050. ¹ Alzheimer's disease (AD) is the most prevalent form, accounting for ≈ 80% of all dementia cases. ² , ³ Mild cognitive impairment (MCI) represents a pre‐dementia phase of cognitive decline but demonstrates variability in the underlying etiology and is estimated to affect > 15% of the global population. ³ Previous research has indicated that MCI typically progresses to dementia over a span of ≈ 9 years, ⁴ and each year, ≈ 10% to 15% of individuals with MCI progress to dementia. ⁵ Although MCI poses a significant health burden, patients with MCI may also revert to healthier cognitive states through the modification of risk factors. ⁶

Cardiometabolic health has garnered significant attention in recent years due to its increasing prevalence. ⁷ , ⁸ , ⁹ , ¹⁰ , ¹¹ , ¹² Numerous studies have identified an intricate relationship among cardiometabolic health, cognition, and neurocognitive decline. ¹³ , ¹⁴ The 2024 Lancet Commission on Dementia report proposed that 45% of dementia cases could be prevented, with 12% of dementia cases attributed to factors such as high low‐density lipoprotein cholesterol (LDL‐C), type 2 diabetes, hypertension, and obesity. ¹⁵ These conditions, individually and as part of overall cardiometabolic multimorbidity, have been shown to be associated with diffuse structural brain alterations and an increased risk of dementia as the number of conditions accumulates. ¹⁶ , ¹⁷

The majority of existing studies have focused on the cardiovascular risk factor mitigation of individuals without cognitive decline developing dementia, leaving an unmet clinical need to explore how cardiometabolic conditions, already established as modifiable risk factors for dementia, impact the progression of MCI to dementia. In this large cohort of propensity score‐matched individuals across numerous health‐care organization in the United States, we sought to estimate the risk of individual risk factors rather than overall cardiometabolic morbidity (in line with the 2024 report of the Lancet Standing Commission ¹⁸ ), that is, hyperlipidemia, type 2 diabetes, hypertension, and obesity in participants with MCI for developing all‐cause, vascular, and AD dementia.

2. METHODS

2.1. Data sources

The study was conducted using data from the US Collaborative Network of the TriNetX Analytics platform. TriNetX is a platform that de‐identifies and aggregates electronic health record (EHR) data, including information from > 100 million patients across > 60 health‐care organizations in the United States. Participants included have commercial insurance, Medicare, or a Medicaid health‐care plan. As a federated network, research studies using the TriNetX research network do not require ethics approvals, as no patient identification is received. Informed consent was waived because the study used de‐identified secondary data.

2.2. Study design

We identified study cohorts of patients with a diagnosis of MCI and an age of ≥ 50 years. Participants with a diagnosis of individual risk factors, that is, hyperlipidemia, diabetes, hypertension, or obesity that was diagnosed at least 5 years before the MCI diagnosis, were compared to those without the diagnosis 5 years prior to the MCI diagnosis, yielding the following cohorts: MCI+type 2 diabetes versus MCI without type 2 diabetes (Cohort 1), MCI+hyperlipidemia versus MCI without hyperlipidemia (Cohort 2), MCI+obesity versus MCI without obesity (Cohort 3), and MCI+hypertension versus MCI without hypertension (Cohort 4; Figure 1). Diagnostic codes for inclusion criteria included International Classification of Diseases, 10th Revised (ICD‐10) codes for each cardiometabolic condition, along with laboratory and clinical measurements (hemoglobin A1c [HbA1c], LDL‐C, to tal serum cholesterol, serum triglycerides, systolic blood pressure, diastolic blood pressure, and body mass index) as a proxy for guideline‐based definitions of each condition. ¹⁹ , ²⁰ , ²¹ Blood‐based markers were collected in a fasting state. Further details on inclusion, exclusion criteria, and their diagnostic codes can be found in Tables S1–S3 in supporting information.

FIGURE 1.

Flowchart. Participants aged 50 years and above diagnosed with MCI were initially included. Four cohorts were then formed based on the four cardiometabolic conditions studied: type 2 diabetes, hyperlipidemia, obesity, and hypertension. MCI, mild cognitive impairment; PSM, propensity score matching. Created with BioRender.com.

Baseline data were derived from EHR records from participants with MCI from January 1, 2010 to August 31, 2017, as the cohort enrollment period, and participants were examined for dementia outcomes that occurred within 10 years of their cohort entry date. Individuals should have at least one health‐care visit within 1 year before their cohort entry. Patients were excluded if they had any diagnosis of dementia at any time before cohort entry (Tables S2 and S3).

2.3. Study outcomes

The primary outcome was all‐cause dementia, based on ICD‐10 codes F01, F02, F03, G30, G31.0, or G31.83, within 10 years of enrollment. In additional analyses, we investigated two dementia subtypes, that is, vascular dementia (ICD‐10: F01) and AD dementia (ICD‐10: G30) (Table S3).

2.4. Covariates

We measured potential confounders up to 20 years prior to the cohort entry date. Based on subject matter expertise and previous research on dementia, we identified variables that acted as confounders, proxies for confounders, or predictors of the outcome. The following variables were included in propensity score matching: age; sex; race; and, from diagnoses, three out of four cardiometabolic diseases included in the study (type 2 diabetes, essential primary hypertension, hyperlipidemia, overweight and obesity), ischemic heart diseases, cerebrovascular diseases, chronic kidney disease, acute kidney failure, venous embolism and thrombosis, nicotine dependency, alcohol dependence, history of healed traumatic brain injury, vision loss, hearing loss, sleep disorders, depressive episode, dietary counseling, activity, and neoplasms. Similarly, the following medications were included in the propensity score matching: antilipemic agents, beta blockers, calcium channel blockers, antiarrhythmics, diuretics, angiotensin‐converting enzyme (ACE) inhibitors, angiotensin II inhibitors, insulin, oral hypoglycemic agents, and other hypoglycemic agents. The cofounders included are based on the ICD‐10 diagnosis and generic drug names (Veterans Affairs drug classification system; Table S4 in supporting information).

RESEARCH IN CONTEXT

1. Systematic review: The authors conducted a literature search using the PubMed database and references from relevant articles. Although there is substantial evidence that type 2 diabetes, hyperlipidemia, obesity, and hypertension are individually associated with dementia, the impact of each of these cardiometabolic conditions on the progression to dementia from mild cognitive impairment (MCI) remains unclear.

2. Interpretation: In this 10‐year cohort study, with a median follow‐up of 7 years, we observed that type 2 diabetes, hypertension, and hyperlipidemia increase the risk of progression to all‐cause dementia in individuals with MCI. Type 2 diabetes was additionally associated with a higher risk of vascular dementia, while hyperlipidemia was associated with Alzheimer's disease dementia.

3. Future directions: Further research should investigate how cardiometabolic conditions influence cognitive decline in the early and pre‐symptomatic phases of dementia.

2.5. Statistical analysis

Baseline characteristics were assessed using chi‐squared tests for categorical variables and independent‐sample t tests for continuous variables. The TriNetX platform performed 1:1 propensity score matching for each cohort with logistic regression, using greedy nearest‐neighbor matching with a to lerance level of 0.01 and a maximum allowable difference of 0.1 between propensity scores. Kaplan–Meier analysis was used to estimate the likelihood of the outcome. If the last recorded event (outcome of interest or other medical encounter) in a patient's record was in the time window of the analysis period, the patient was censored on the day after that last recorded event. Hazard ratios (HRs), risk differences (RDs), and 95% confidence intervals (CIs) were used to describe the risk of the outcomes by comparing time‐to‐event rates. The proportional hazards assumption was tested using the generalized Schoenfeld approach on the TriNetX platform, with adjusted HRs recalculated for specific time intervals if the assumptions were violated. Data were lastly analyzed on July 17, 2025, within the TriNetX Analytics platform.

3. RESULTS

3.1. Characteristics of the study population

The four cohorts with patients with MCI consisted of: (1) 1867 with type 2 diabetes versus 1867 without type 2 diabetes, (2) 6643 with hyperlipidemia versus 6643 without hyperlipidemia, (3) 4118 with obesity versus 4118 without obesity, and (4) 3275 with hypertension versus 3275 without hypertension (Figure 1, Table 1).

TABLE 1.

Baseline characteristics of people with mild cognitive impairment, having either type 2 diabetes, hyperlipidemia, obesity, or hypertension, compared to controls after 1:1 propensity score matching.

	Type 2 diabetes vs. control cohorts			Hyperlipidemia vs. control cohorts			Obesity vs. control cohorts			Hypertension vs. control cohorts
	Disease	Controls	SMD	Disease	Controls	SMD	Disease	Controls	MD	Disease	Controls	SMD
Total, n	1867	1867		6643	6643		4118	4118		3275	3275
Age, mean ± SD	69.4 ± 8.0	69.3 ± 8.0	0.018	69.2 ± 8.1	69.2 ± 8.2	0.004	68.5 ± 8.3	68.5 ± 8.4	0.005	68.1 ± 8.6	68.0 ± 8.4	0.003
Sex (%)
Female	953 (51.0)	948 (50.8)	0.005	3593 (54.1)	3534 (53.2)	0.018	2249 (54.6)	2285 (55.5)	0.018	1792 (54.7)	1794 (54.8)	0.001
Male	914 (49.0)	919 (49.2)	0.005	3050 (45.9)	3109 (46.8)	0.018	1869 (45.4)	1833 (44.5)	0.018	1483 (45.3)	1481 (45.2)	0.001
Ethnicity (%)
Hispanic/Latinx	100 (5.4)	87 (4.7)	0.032	288 (4.3)	283 (4.3)	0.004	177 (4.3)	178 (4.3)	0.001	133 (4.1)	155 (4.7)	0.033
Not Hispanic/Latinx	1391 (74.5)	1416 (75.8)	0.031	4889 (73.6)	4901 (73.8)	0.004	3327 (80.8)	3328 (80.8)	0.001	2546 (77.7)	2528 (77.2)	0.013
Unknown	376 (20.1)	364 (19.5)	0.016	1466 (22.1)	1459 (22.0)	0.003	614 (14.9)	612 (14.9)	0.001	596 (18.2)	592 (18.1)	0.003
Race (%)
Asian	65 (3.5)	67 (3.6)	0.006	166 (2.5)	159 (2.4)	0.007	50 (1.2)	51 (1.2)	0.002	82 (2.5)	88 (2.7)	0.012
Black	333 (17.8)	322 (17.2)	0.015	859 (12.9)	883 (13.3)	0.011	490 (11.9)	496 (12.0)	0.004	287 (8.8)	280 (8.5)	0.008
White	1309 (70.1)	1316 (70.5)	0.008	5117 (77.0)	5122 (77.1)	0.002	3210 (78.0)	3198 (77.7)	0.007	2580 (78.8)	2566 (78.4)	0.010
Other	29 (1.6)	29 (1.6)	<0.001	96 (1.4)	101 (1.5)	0.006	79 (1.9)	85 (2.1)	0.010	43 (1.3)	52 (1.6)	0.023
Cardiovascular and other risks/conditions (%)
Hyperlipidemia	1195 (64.0)	1151 (61.6)	0.049	N/A	N/A	N/A	2765 (67.1)	2768 (67.2)	0.002	859 (26.2)	885 (27.0)	0.018
Type 2 diabetes	N/A	N/A	N/A	2264 (34.1)	2228 (33.5)	0.011	1709 (41.5)	1662 (40.4)	0.023	451 (13.8)	437 (13.3)	0.012
Overweight and obesity	520 (27.9)	518 (27.7)	0.002	1651 (24.9)	1633 (24.6)	0.006	N/A	N/A	N/A	227 (6.9)	213 (6.5)	0.017
Essential hypertension	1511 (80.9)	1450 (77.7)	0.081	4799 (72.2)	4769 (71.8)	0.010	3261 (79.2)	3257 (79.1)	0.002	N/A	N/A	N/A
Ischemic heart diseases	685 (36.7)	678 (36.3)	0.008	1978 (29.8)	1979 (29.8)	<0.001	1410 (34.2)	1385 (33.6)	0.013	339 (10.4)	324 (9.9)	0.015
Cerebrovascular diseases	598 (32.0)	588 (31.5)	0.012	1987 (29.9)	1999 (30.1)	0.004	1341 (32.6)	1351 (32.8)	0.005	469 (14.3)	458 (14.0)	0.010
Other venous embolism and thrombosis	150 (8.0)	137 (7.3)	0.026	445 (6.7)	434 (6.5)	0.007	351 (8.5)	342 (8.3)	0.008	81 (2.5)	78 (2.4)	0.006
Personal history of other (healed) physical injury and trauma	42 (2.2)	38 (2.0)	0.015	129 (1.9)	125 (1.9)	0.004	92 (2.2)	79 (1.9)	0.022	44 (1.3)	40 (1.2)	0.011
Chronic kidney disease	425 (22.8)	384 (20.6)	0.053	1099 (16.5)	1,102 (16.6)	0.001	827 (20.1)	817 (19.8)	0.006	174 (5.3)	165 (5.0)	0.012
Acute kidney failure	298 (16.0)	293 (15.7)	0.007	770 (11.6)	751 (11.3)	0.009	603 (14.6)	617 (15.0)	0.010	86 (2.6)	89 (2.7)	0.006
Depressive episode	686 (36.7)	672 (36.0)	0.016	2583 (38.9)	2555 (38.5)	0.009	1846 (44.8)	1900 (46.1)	0.026	726 (22.2)	753 (23.0)	0.020
Nicotine dependence	282 (15.1)	282 (15.1)	<0.001	976 (14.7)	929 (14.0)	0.020	654 (15.9)	630 (15.3)	0.016	218 (6.7)	212 (6.5)	0.007
Alcohol related disorders	135 (7.2)	138 (7.4)	0.006	490 (7.4)	492 (7.4)	0.001	326 (7.9)	317 (7.7)	0.008	135 (4.1)	126 (3.8)	0.014
Neoplasms	969 (51.9)	938 (50.2)	0.033	3761 (56.6)	3800 (57.2)	0.012	2528 (61.4)	2580 (62.7)	0.026	1114 (34.0)	1179 (36)	0.042
Unspecified vision loss	54 (2.9)	53 (2.8)	0.003	191 (2.9)	182 (2.7)	0.008	136 (3.3)	125 (3.0)	0.015	30 (0.9)	31 (0.9)	0.003
Unspecified hearing loss	206 (11.0)	205 (11.0)	0.002	759 (11.4)	790 (11.9)	0.015	595 (14.4)	592 (14.4)	0.002	185 (5.6)	189 (5.8)	0.005
Sleep disorders	720 (38.6)	694 (37.2)	0.029	2599 (39.1)	2620 (39.4)	0.006	1984 (48.2)	1952 (47.4)	0.016	692 (21.1)	692 (21.1)	<0.001
Dietary counseling and surveillance	46 (2.5)	53 (2.8)	0.023	138 (2.1)	147 (2.2)	0.009	122 (3.0)	117 (2.8)	0.007	21 (0.6)	20 (0.6)	0.004
Activity codes	23 (1.2)	17 (0.9)	0.031	73 (1.1)	80 (1.2)	0.010	69 (1.7)	63 (1.5)	0.012	26 (0.8)	21 (0.6)	0.018
Medications (%)
Insulin	621 (33.3)	650 (34.8)	0.033	1530 (23.0)	1541 (23.2)	0.004	1136 (27.6)	1054 (25.6)	0.045	219 (6.7)	214 (6.5)	0.006
Oral hypoglycemic agents	393 (21.0)	421 (22.5)	0.036	1509 (22.7)	1452 (21.9)	0.021	1150 (27.9)	1121 (27.2)	0.016	267 (8.2)	275 (8.4)	0.009
Oral hypoglycemic agents (other)	18 (1.0)	18 (1.0)	<0.001	71 (1.1)	68 (1.0)	0.004	50 (1.2)	53 (1.3)	0.007	12 (0.4)	13 (0.4)	0.005
Antilipemic agents	1215 (65.1)	1169 (62.6)	0.051	4487 (67.5)	4483 (67.5)	0.001	2789 (67.7)	2776 (67.4)	0.007	1078 (32.9)	1112 (34.0)	0.022
Beta blockers	1020 (54.6)	1001 (53.6)	0.020	3486 (52.5)	3490 (52.5)	0.001	2353 (57.1)	2331 (56.6)	0.011	766 (23.4)	762 (23.3)	0.003
Calcium channel blockers	698 (37.4)	667 (35.7)	0.034	2281 (34.3)	2277 (34.3)	0.001	1609 (39.1)	1575 (38.2)	0.017	351 (10.7)	362 (11.1)	0.011
Antiarrhythmics	859 (46.0)	851 (45.6)	0.009	3158 (47.5)	3197 (48.1)	0.012	2275 (55.2)	2322 (56.4)	0.023	763 (23.3)	717 (21.9)	0.034
Diuretics	938 (50.2)	901 (48.3)	0.040	3218 (48.4)	3249 (48.9)	0.009	2365 (57.4)	2350 (57.1)	0.007	654 (20.0)	605 (18.5)	0.038
Ace inhibitors	811 (43.4)	762 (40.8)	0.053	2653 (39.9)	2635 (39.7)	0.006	1884 (45.8)	1857 (45.1)	0.013	413 (12.6)	409 (12.5)	0.004
Angiotensin II inhibitors	459 (24.6)	413 (22.1)	0.058	1402 (21.1)	1397 (21.0)	0.002	1007 (24.5)	975 (23.7)	0.018	267 (8.2)	265 (8.1)	0.002

Note: Data are numbers (%) unless stated otherwise.

Abbreviations: SD, standard deviation; SMD, standard mean difference.

3.2. Cardiometabolic conditions and risk for all‐cause dementia

A to tal of 11,500 dementia cases were identified during a median 7‐year follow‐up among individuals with MCI: 749 in participants with type 2 diabetes (incidence rate [IR] 65, 95% CI: 60–70; mean follow‐up time: 2251 days), 1479 in participants with hyperlipidemia (IR 32, 95% CI: 31–34 per 1000 person‐years; mean follow‐up time: 2515 days), 1520 in participants with obesity (IR 59 95% CI: 56–62; mean follow‐up time: 2280 days), and 1256 in participants with hypertension (IR 59, 95% CI: 55–62; mean follow‐up time: 2387 days). Full characteristics are summarized in Table 1, and all covariates had a standardized mean difference < 0.1 after weighting.

Type 2 diabetes showed the highest HR for all‐cause dementia (HR 1.18, 95% CI: 1.06 to 1.31; RD 7.98, 95% CI: 2.30 to 13.67), followed by hypertension (HR 1.14, 95% CI: 1.05 to 1.23; RD 3.37, 95% CI: −0.72 to 7.47) and then by hyperlipidemia (HR 1.07, 95% CI: 1.02 to 1.14; RD −0.31, 95% CI: −3.32 to 2.71). These associations for type 2 diabetes remained significant for male participants with type 2 diabetes and participants aged 50 to 74 years. This pattern for hypertension persisted only in the sensitivity analysis for males. Concerning hyperlipidemia, this pattern persisted only for participants aged 50 to 74 years and did not show sex‐related differences. The association of obesity and all‐cause dementia was not significant across all sensitivity analyses (Figures 2 and S1), with the exception of males with obesity (HR 1.13, 95% CI: 1.01 to 1.26; RD 3.20, 95% CI: −2.37 to 8.78).

FIGURE 2.

Kaplan–Meier curves of medical encounters for all‐cause dementia diagnosis for type 2 diabetes, hyperlipidemia, obesity, and hypertension groups within 10 years of baseline. Pairwise comparisons were conducted using controls without each condition, respectively, as the reference. CI, confidence interval; HR, hazard ratio; RD, risk difference.

3.3. Cardiometabolic conditions and risk for vascular dementia

A to tal of 2836 cases of vascular dementia were reported. Type 2 diabetes was associated with the greatest risk of developing vascular dementia (HR 1.33, 95% CI: 1.07 to 1.64; RD 3.86, 95% CI: 0.97 to 6.75; Figures 3 and S2 in supporting information). In the sensitivity analysis, the associations with type 2 diabetes remained significant in males with type 2 diabetes. Furthermore, females with hypertension showed a slight reduced risk for vascular dementia (HR 0.75, 95% CI: 0.58 to 0.97; RD −2.78, 95% CI: −5.15 to −0.41).

FIGURE 3.

Kaplan–Meier curves of medical encounters for vascular dementia diagnosis for type 2 diabetes, hyperlipidemia, obesity, and hypertension groups within 10 years of baseline. Pairwise comparisons were conducted using controls without each condition, respectively, as the reference. CI, confidence interval; HR, hazard ratio; RD, risk difference.

3.4. Cardiometabolic conditions and risk for AD dementia

During the follow‐up period, 5007 cases of AD dementia were reported. Hyperlipidemia was the only cardiometabolic condition (included in this study) that was positively associated with the risk of progressing to AD dementia (HR 1.21, 95% CI: 1.11 to 1.32; RD 2.77, 95% CI: 0.81 to 4.73; Figures 4 and S3 in supporting information). This association persisted across all sensitivity analyses for hyperlipidemia, with the exception of males and participants aged ≥ 75 years.

FIGURE 4.

Kaplan–Meier curves of medical encounters for Alzheimer's disease dementia diagnosis for type 2 diabetes, hyperlipidemia, obesity, and hypertension groups within 10 years of baseline. Pairwise comparisons were conducted using controls without each condition, respectively, as the reference. CI, confidence interval; HR, hazard ratio; RD, risk difference.

4. DISCUSSION

4.1. Principal finding

This large‐scale, real‐world population study demonstrates the associations of four prevalent cardiometabolic conditions (type 2 diabetes, hyperlipidemia, obesity, and hypertension) with the progression from MCI to all‐cause dementia, vascular dementia, and AD dementia. In this cohort study, we demonstrated that, among these four cardiometabolic conditions, type 2 diabetes, hypertension, and hyperlipidemia increase the risk of all‐cause dementia. Type 2 diabetes was additionally associated with a higher risk of vascular dementia, while hyperlipidemia was associated with AD dementia. In contrast, obesity was not associated with dementia progression overall, except amongmales, for which obesity conferred an increased risk for all‐cause dementia.

4.2. Relationship to prior studies

It is well established that each of the four studied cardiometabolic conditions serves as a risk factor for dementia progression from a cognitively normal state. In addition to cardiometabolic conditions, ²² white matter atrophy, ventricular enlargement, and low cerebrospinal fluid (CSF) amyloid beta (Aβ)42:Aβ40 ratios ²³ have been implicated in the pathophysiology of progression to MCI. Although the mechanisms underlying progression from MCI to dementia are not fully elucidated, ²⁴ neurodegeneration of the entorhinal–hippocampal pathway ²⁵ and perforant fibers ²⁶ has been identified as early markers of AD disease.

Our findings align with and extend this literature by demonstrating that hyperlipidemia poses a high risk for all‐cause dementia (HR 1.07, 95% CI: 1.02 to 1.14) and AD dementia (HR 1.21, 95% CI: 1.11 to 1.32). These findings align with the recent report from the Lancet Standing Commission for Dementia Prevention, Intervention, and Care, identifying high LDL‐C as the most potent modifiable cardiometabolic risk factor for dementia in midlife. ¹⁵ Each standard deviation increase in LDL‐C has been associated with higher risks for all‐cause, vascular, and AD dementia. ²⁷ The mechanisms linking lipids to dementia remain elusive. It is speculated that brain cholesterol may increase Aβ production, but when the blood–brain barrier is intact, as in cognitively unimpaired individuals, blood lipid levels might not correlate with brain cholesterol levels. ²⁸ Furthermore, the apolipoprotein E (APOE) gene is implicated in cholesterol metabolism, and in particular, the APOE ε4 variant is strongly associated with high to tal and LDL‐C levels, subsequently increasing the risk for AD. ²⁹ On the contrary, the APOE ε2 variant is associated with low LDL‐C and lower AD risk. ³⁰

Regarding type 2 diabetes, large meta‐analyses have highlighted its association with a higher risk of conversion from MCI to all‐cause dementia and AD dementia. ³¹ , ³² In our cohort, participants diagnosed with type 2 diabetes and MCI exhibited a higher risk for all‐cause dementia and vascular dementia. Type 2 diabetes, responsible for 2% out of 45% of preventable dementia cases, ¹⁵ is a strong risk factor for vascular dementia. ³³ Large epidemiological studies support a significant association between diabetes duration and dementia risk, ³⁴ likely mediated by factors such as peripheral and central insulin resistance and inflammation, which contribute to vascular damage and cerebrovascular insults. ³³

Prior research has established that, in cognitively healthy individuals, obesity is associated with a significantly higher risk of dementia compared to individuals with normal weight over long‐term follow‐up. ¹⁴ , ³⁵ , ³⁶ , ³⁷ However, these associations were not observed in our cohort, which consisted entirely of participants diagnosed with MCI at baseline, as we showed that only males with obesity present with a higher risk of all‐cause dementia. Furthermore, hypertension in a population without cognitive impairment is linked to a higher risk for all types of dementia. ³⁸ A meta‐analysis, in line with our findings, identified hypertension as a risk factor for the progression from MCI to all‐cause dementia (relative risk: 1.41 [95% CI, 1.00 to 1.99], I ² = 33%). ³² , ³⁹

4.3. Strengths and limitations

This study has several strengths. First, it is the largest cohort study investigating the impact of four prevalent cardiometabolic conditions on the progression from MCI to dementia. Second, we reduced reverse causality bias by using a median follow‐up of 7 years, exceeding the median progression time from MCI to dementia. ⁴⁰ Third, we focused on the impact of cardiometabolic diseases with a minimum duration of 5 years to avoid confounding effects of recent diagnoses, which may affect dementia progression. For example, late‐onset type 2 diabetes does not significantly increase the risk of subsequent dementia. ³⁴

Our study presents several limitations. The most significant limitation is bias and inconsistency in ICD diagnostic codes, which could be affected by regional physician practice standards and thoroughness in documentation. to limit phenotypic biases associated with ICD diagnosis codes, we supplemented our data with baseline laboratory and clinical measurements (HbA1c, LDL‐C, to tal cholesterol, triglycerides, systolic blood pressure, diastolic blood pressure, body mass index) as a proxy for guideline‐based definitions of each condition. Although body mass index may have different thresholds for each race, race‐specific cut‐off values were not applied. Specifically, for type 2 diabetes, we used a single HbA1c ≥ 6.5% or ICD‐10 E11 codes, which may lead to some misclassification relative to American Diabetes Association diagnostic recommendations, which generally recommend confirmatory testing. ⁴¹ Our dataset was derived from health‐care organizations based in the United States, potentially limiting the generalizability of our findings to other populations. Ethnic and regional differences may influence dementia progression and should be addressed in future studies. Additionally, unmeasured confounding variables not included in the propensity score matching, as well as diagnostic inconsistencies across health‐care organizations, may pose limitations. Lastly, the models used did not account for time‐dependent variations in the laboratory measurements, which may reduce the accuracy of the predictions. Future research should validate our findings over longer follow‐up periods (a decade or more) and investigate the mechanisms ⁴² underlying the influence of each cardiometabolic condition on the brain in MCI patients.

5. CONCLUSION

This large‐scale cohort study provides critical insights into the influence of cardiometabolic conditions such as hypertension, hyperlipidemia, and type 2 diabetes on the progression from MCI to dementia, with type 2 diabetes and hyperlipidemia associated with vascular and AD dementia, respectively. Future research is needed to further explore underlying mechanisms and develop targeted interventions to slow cognitive decline in at‐risk populations.

AUTHOR CONTRIBUTIONS

Filippos Anagnostakis had access to and verified the underlying study data and was responsible for ensuring the integrity and accuracy of the data analysis and is the guarantor. Filippos Anagnostakis drafted the manuscript. All authors contributed to the study design, analysis, and interpretation of data, as well as the critical revision of the manuscript. Christos Davatzikos supervised the study. The corresponding author confirms that all listed authors meet the criteria for authorship and that no other authors meeting the criteria have been excluded.

CONFLICT OF INTEREST STATEMENT

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

CONSENT STATEMENT

This study used population‐level aggregate and de‐identified data collected by the TriNetX Platform, which are available from TriNetX (https://trinetx.com/); however, third‐party restrictions apply to the availability of these data. The data were used under license for this study, with restrictions that do not allow for data to be redistributed or made publicly available. to gain access to the data, a request can be made to TriNetX (join@trinetx.com), but costs might be incurred, and a data‐sharing agreement would be necessary. As the data are routinely collected and fully anonymized, patient consent is not required. Data specific to this study, including diagnosis codes and group characteristics in aggregated format, are included in the paper as tables, figures, and supplementary online content.

Supporting information

Supporting Information

Supporting Information

Supporting Information

Supporting Information

Supporting Information

Supporting Information

Anagnostakis F, Kokkorakis M, Asvestis C, et al. Impact of cardiometabolic conditions on the progression from mild cognitive impairment to dementia: A large cohort study. Alzheimer's Dement. 2025;21:e70692. 10.1002/alz.70692