Warning: fopen(/home/virtual/enm-kes/journal/upload/ip_log/ip_log_2026-06.txt): failed to open stream: Permission denied in /home/virtual/lib/view_data.php on line 100 Warning: fwrite() expects parameter 1 to be resource, boolean given in /home/virtual/lib/view_data.php on line 101 Early-Onset Dementia Risk Escalates with Diabetes Duration: Insights from a Nationwide Cohort Study
Skip Navigation
Skip to contents

Endocrinol Metab : Endocrinology and Metabolism

clarivate
OPEN ACCESS
SEARCH
Search
Advanced Search
mobile menu button

Articles

Page Path
HOME > Endocrinol Metab > Volume 41(2); 2026 > Article
Original Article
Diabetes, obesity and metabolism Early-Onset Dementia Risk Escalates with Diabetes Duration: Insights from a Nationwide Cohort Study
Keypoint
This nationwide cohort study investigated the association between diabetes and early-onset dementia, or EOD, demonstrating that diabetes was associated with an increased risk of EOD, particularly among patients with a diabetes duration of ≥5 years.
However, impaired fasting glucose was not significantly associated with EOD risk.
These findings highlight the importance of early intervention and cognitive screening in individuals with diabetes who are younger than 65 years.
Ji-Hong Park1*orcid, Sun-Joon Moon2*orcid, Da Yeon Lee2, Ji-Hee Ko2, Han Na Jang2, Hye-Mi Kwon2, Se-Eun Park2, Kyung-Do Han3, Eun-Jung Rhee2orcid, Won-Young Lee2orcid
Endocrinology and Metabolism 2026;41(2):235-244.
DOI: https://doi.org/10.3803/EnM.2024.2287
Published online: July 17, 2025

1Department of Internal Medicine, Kangbuk Samsung Hospital, Sungkyunkwan University School of Medicine, Seoul, Korea

2Division of Endocrinology and Metabolism, Department of Internal Medicine, Kangbuk Samsung Hospital, Sungkyunkwan University School of Medicine, Seoul, Korea

3Department of Statistics and Actuarial Science, Soongsil University, Seoul, Korea

Corresponding authors: Eun-Jung Rhee. Division of Endocrinology and Metabolism, Department of Internal Medicine, Kangbuk Samsung Hospital, Sungkyunkwan University School of Medicine, 29 Saemunan-ro, Jongno-gu, Seoul 03181, Korea Tel: +82-2-2001-2485, Fax: +82-2-2001-2049, E-mail: hongsiri@hanmail.net
Won-Young Lee. Division of Endocrinology and Metabolism, Department of Internal Medicine, Kangbuk Samsung Hospital, Sungkyunkwan University School of Medicine, 29 Saemunan-ro, Jongno-gu, Seoul 03181, Korea Tel: +82-2-2001-2579, Fax: +82-2-2001-2049, E-mail: drlwy@hanmail.net
These authors contributed equally to this work.
• Received: December 17, 2024   • Revised: April 24, 2025   • Accepted: May 19, 2025

Copyright © 2026 Korean Endocrine Society

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

  • 3,012 Views
  • 81 Download
  • 1 Crossref
prev next
Full Article

Download PDF Download PDF

  • Background
    The prevalence of diabetes mellitus and early-onset dementia (EOD), defined as dementia diagnosed at an age <65 years, is increasing worldwide, with significant socioeconomic implications. We investigated the association between diabetes, prediabetes, and EOD, focusing on the influence of diabetes duration on EOD risk.
  • Methods
    Using the Korean National Health Insurance Service database, we analyzed data from 1,979,509 patients aged 40–60 years who underwent health checkups in 2009. Patients were categorized into five groups: normal, impaired fasting glucose (IFG), newly diagnosed diabetes, diabetes duration <5 years, and diabetes duration ≥5 years. Cox proportional hazard models were used to estimate the adjusted hazard ratios (aHRs) for EOD after adjusting for demographic and clinical covariates.
  • Results
    During the observation period (mean 7.75 years), 8,921 patients with EOD were identified. The diabetes group demonstrated a significantly higher incidence of EOD compared to the normal group (aHR, 1.334; 95% confidence interval [CI], 1.226 to 1.451). EOD risk increased with longer diabetes duration, with the highest risk observed in patients with diabetes ≥5 years (aHR, 1.543; 95% CI, 1.368 to 1.741). No significant difference was observed between the IFG and normal groups (aHR, 0.989; 95% CI, 0.938 to 1.043). Additionally, the hypertension group exhibited a significantly higher incidence of EOD compared to the non-hypertension group (aHR, 1.364; 95% CI, 1.291 to 1.442).
  • Conclusion
    Diabetes is independently associated with increased risk of EOD, and this risk increases with longer diabetes duration. This association remained significant regardless of the presence and duration of hypertension.
  • Keywords: Dementia; Diabetes mellitus; Prediabetic state
The prevalence of dementia and diabetes mellitus is increasing globally, accompanied by continuously rising socioeconomic costs [1,2]. Recent studies indicate that the prevalence of both dementia and diabetes begins to increase with age even among relatively young populations [1,3,4]. Early-onset dementia (EOD) is typically defined as dementia diagnosed or symptomatically manifested in individuals under 65 years of age, with its prevalence significantly increasing from the age of 45 years onwards [3,5,6]. Recent studies suggest that approximately 350,000 new cases of EOD are diagnosed annually worldwide [4,7]. EOD significantly affects patients’ social functioning, contributes to economic crises, exacerbates interpersonal problems, and diminishes quality of life [8]. Furthermore, distinguishing EOD from other psychiatric and neurocognitive disorders remains challenging, frequently delaying diagnosis and thus postponing timely referrals to appropriate treatment centers [9]. These factors collectively present substantial medical and social challenges.
Meanwhile, a global consensus has acknowledged diabetes as a well-established risk factor for dementia, prompting ongoing research into underlying mechanisms [10,11]. Follow-up studies have demonstrated that cognitive decline among individuals with type 2 diabetes mellitus occurs at about twice the rate observed in those without this disease [10]. Additionally, prospective population studies indicate that individuals with type 2 diabetes mellitus have a higher risk of developing mild cognitive impairment and progressing from mild cognitive impairment to dementia compared to those without this condition [12]. Several studies further demonstrate that elevated plasma glucose levels constitute a risk factor for dementia, even within the non-diabetes range [13]. Moreover, dementia risk is also increased in patients with newly diagnosed diabetes, underscoring the importance of vigilance given the rising prevalence of impaired fasting glucose (IFG) [14]. Notably, diabetes prevalence continues to rise, whereas the prevalence of other chronic conditions such as hypertension, cardiovascular disease, and stroke has remained stable or decreased among patients with dementia between 2008 and 2016 [15]. These findings reinforce the hypothesis that diabetes is a key risk factor for dementia.
Although numerous studies have explored the relationship between diabetes and dementia, investigations specifically focusing on EOD and diabetes—issues with growing medical and social significance—remain limited. To our knowledge, only two studies have directly examined this association. The first, involving participants from the United Kingdom Biobank, identified several risk factors for EOD, including diabetes [16]. The second, a recent longitudinal study using a large Korean cohort, examined risk factors for both EOD and late-onset dementia, identifying diabetes as significant [17]. However, that study did not specifically investigate diabetes, did not consider prediabetes, and did not explore the impact of diabetes duration on EOD onset.
Therefore, we aimed to determine whether prediabetes and diabetes are associated with increased EOD risk. Additionally, our study provides a detailed analysis of how diabetes duration affects the likelihood of developing EOD.
National Health Insurance health examination cohort
The National Health Insurance Service (NHIS) is a government-operated, mandatory social health insurance program that covers approximately 97% of the Korean population. It maintains comprehensive health data for approximately 50 million individuals [18]. The NHIS database includes detailed information on medical facility utilization, prescription records, and diagnostic codes based on the International Classification of Diseases, Tenth Revision (ICD-10) [19]. Additionally, the NHIS conducts health examinations at least biennially for all beneficiaries aged ≥40 years. These examinations include measurements of anthropometric parameters, a self-administered questionnaire about medical history and health-related behaviors, and various laboratory tests [20]. This database, comprising anonymized and de-identified data, is considered representative of the Korean population and is available for research purposes.
Study participants
In 2009, 4,234,415 individuals participated in the NHIS health examinations (Fig. 1). Participants younger than 40 or older than 60 years (n=2,133,554), those with a medical history of dementia (n=847), and those with any missing variables (n=117,176) were excluded. Additionally, 3,329 individuals were excluded due to a 1-year lag period. Ultimately, 1,979,509 participants were included in the analysis. Among these, 1,302,706 were classified as normal, 501,347 as IFG, and 175,456 as diabetes. The diabetes group was further subdivided into new-onset diabetes (n=77,843), diabetes duration <5 years (n=51,926), and diabetes duration ≥5 years (n=45,687). Participants were followed until December 31 of the year in which they turned 64 years old or until December 31, 2018, if they had not reached age 64.
Diabetes definition and medications
Diabetes was defined based on a prescription for antidiabetic medications combined with ICD-10 codes (E11–E14) or a fasting glucose level >126 mg/dL recorded in the health examination database [21]. Diabetes duration was calculated from the date of initial diabetes diagnosis to the baseline examination at study entry. Information on insulin use and the number of oral hypoglycemic agents prescribed was collected from medication claim records (ICD-10 codes E11–14) and antidiabetic medication prescriptions.
Outcomes
Dementia was determined based on prescriptions for anti-dementia medications (rivastigmine, galantamine, memantine, or donepezil) along with ICD-10 diagnostic codes (F00, F01, F02, F03, G30, or G31) from medical expense claims submitted to the NHIS. If all dementia codes were initially recorded only as secondary diagnoses, the determination of dementia subtype was deferred until a subsequent visit. If any dementia code was then listed as a primary diagnosis, the subtype was assigned based on that diagnosis. If all dementia codes remained classified as secondary diagnoses, the patient was classified as having ‘other dementia.’ Patients diagnosed with dementia before the age of 65, in accordance with these criteria, were classified as having EOD.
Definitions of covariates and comorbidities
Blood samples were collected after overnight fasting, and quality control procedures were performed in accordance with the guidelines of the Korean Association of Laboratory Quality Control. Smoking status, alcohol consumption, physical activity, baseline income level, and medical history were assessed using standardized self-report questionnaires. Participants were categorized as low income if their income fell within the lowest quartile of the population. Smoking status was classified as nonsmoker, ex-smoker, or current smoker. Heavy alcohol drinkers were defined as individuals who consumed ≥30 g of alcohol per day [22]. Regular physical activity was defined as high-intensity physical activity for a minimum of 20 minutes, at least 3 days per week, or moderate-intensity physical activity for a minimum of 30 minutes, at least 5 days per week [23].
Baseline comorbidities—including hypertension, dyslipidemia, depression, atrial fibrillation, and stroke—were identified through questionnaires regarding medical history, ICD-10 diagnostic codes, and prescription records. Hypertension was defined as a systolic/diastolic blood pressure ≥140/90 mm Hg or at least one annual claim for antihypertensive medications under relevant ICD-10 codes (I10–I13 and I15). Dyslipidemia was defined as a total cholesterol level ≥240 mg/dL (6.21 mmol/L) or at least one annual claim for antihyperlipidemic medications under ICD-10 code E78. Estimated glomerular filtration rate (eGFR) was calculated using the Modification of Diet in Renal Disease equation: eGFR=175×serum creatinine−1.154×age−0.203, multiplied by 0.742 for women [24]. Chronic kidney disease was defined as an eGFR <60 mL/min/1.73 m2 [25]. Atrial fibrillation and stroke histories were evaluated using standardized questionnaires. Depression, a well-established risk factor for dementia [26], was defined based on the ICD-10 codes F32 or F33 recorded at least once before the baseline health examination.
Statistical analysis
Data were expressed as mean±standard deviation, geometric mean (95% confidence interval [CI]), or percentage. Differences between the study groups were assessed using the Student t-test.
The chi-square test was used to evaluate differences in the distribution of categorical variables. The hazard ratio (HR) and 95% CI for EOD in each group were calculated using Cox proportional hazards analysis. Model 1 yielded the unadjusted HR; Model 2 was adjusted for age and sex; and Model 3 was further adjusted for body mass index (BMI), low baseline income, smoking status, alcohol consumption, regular physical activity, dyslipidemia, depression, atrial fibrillation, stroke, fasting glucose level, use of three or more oral hypoglycemic agents, and insulin use. Additionally, we conducted a stratified analysis to examine variations in risk according to sex, fasting glucose level, and medication status regarding antidiabetic drugs, which can indirectly reflect glycemic control status. All statistical analyses were performed using SAS version 9.4 (SAS Institute Inc., Cary, NC, USA), and P values less than 0.05 were considered to indicate statistical significance.
Ethics statement
This study was approved by the Institutional Review Board of Kangbuk Samsung Medical Center (KBSMC 2024-09-028), Seoul, Korea. The requirement for informed consent was waived as the analyses were conducted using de-identified data.
Data and resource availability
All raw data are accessible from designated terminals approved by the NHIS. Upon reasonable request, data can be made available through approval and oversight from the Korean NHIS.
Baseline characteristics
Table 1 presents the baseline characteristics of the five groups: normal, IFG, new-onset diabetes, diabetes duration <5 years, and diabetes duration ≥5 years. Patient age increased with longer diabetes duration. The proportion of male participants was higher in the IFG and diabetes groups compared to the normal group. BMI and waist circumference were higher in the diabetes groups, with the highest values observed in the group with the disease for <5 years. Fasting glucose levels were significantly elevated in the newly diagnosed diabetes group and showed a trend toward further elevation as diabetes duration increased. Blood pressure and total cholesterol levels were highest in the newly diagnosed diabetes group. Significant differences were also observed among groups in terms of comorbidities, including dyslipidemia, depression, atrial fibrillation, and stroke.
Diabetes and risk of EOD
Over a mean follow-up period of 7.75±1.22 years, 8,921 individuals with EOD were identified among 1,979,509 participants (Table 2). The incidence of EOD was highest in the diabetes group (0.90%), followed by the IFG group (0.44%) and the normal group (0.39%).
In the unadjusted analysis, the diabetes group displayed a significantly higher risk of EOD compared to the normal group. This association remained significant after adjusting for age and sex in model 2 (adjusted hazard ratio [aHR], 1.765; 95% CI, 1.667 to 1.870) and persisted in the fully adjusted model, which further accounted for BMI, regular physical activity, low baseline income, smoking, and alcohol consumption (aHR, 1.334; 95% CI, 1.226 to 1.451). However, no significant difference was observed between the IFG and normal groups in the fully adjusted model.
When diabetes was further stratified by disease duration, a stepwise increase in EOD risk was observed. Compared to the normal group, individuals with newly diagnosed diabetes exhibited an elevated risk, which was even greater in those with a diabetes duration <5 years. The highest risk was observed in individuals with a diabetes duration ≥5 years, even after full adjustment for covariates (aHR, 1.543; 95% CI, 1.368 to 1.741). This pattern remained consistent across all models, reinforcing the association between prolonged diabetes duration and increased EOD risk.
Stratified analysis
The difference in aHR between the normal and diabetes groups was consistent between sexes (Supplemental Table S1). The aHR was slightly higher in men (1.358; 95% CI, 1.232 to 1.496) compared to women (1.295; 95% CI, 1.162 to 1.444). Additionally, a pattern of increasing EOD risk with longer diabetes duration was observed in both sexes.
To assess the impact of glycemic control status on EOD risk, a stratified analysis was conducted within the new onset, <5 years, and ≥5 years diabetes duration groups based on fasting glucose levels, the number of oral hypoglycemic agents used, and insulin use (Supplemental Table S2). When stratified by fasting glucose level (≥150 mg/dL vs. <150 mg/dL), a significantly elevated EOD risk was observed in the new-onset diabetes group with fasting glucose ≥150 mg/dL (aHR, 1.235; 95% CI, 1.037 to 1.471). However, no significant differences were found in the other diabetes duration groups. Similarly, when stratified by the number of oral hypoglycemic agents (≥3 vs. <3), no significant differences in EOD risk were observed across diabetes duration groups. In contrast, when stratified by insulin use, both the <5 and ≥5 years groups exhibited a significantly increased EOD risk among insulin users, with aHRs of 1.721 (95% CI, 1.338 to 2.214) and 1.611 (95% CI, 1.347 to 1.925), respectively.
Impact of hypertension and diabetes on EOD risk
Hypertension, like diabetes, is a well-known modifiable risk factor for dementia [17,27]. The relationship between hypertension and EOD risk was analyzed separately using Cox regression analyses, and the results are presented in Supplemental Table S3. Additionally, Table 3 provides a detailed comparison of the impact of hypertension and diabetes on the risk of EOD, emphasizing disease duration.
When hypertension was categorized into normal, pre-hypertension, and hypertension, individuals with hypertension had a higher risk of EOD. In the fully adjusted model, the hypertension group had a significantly increased risk of EOD (aHR, 1.410; 95% CI, 1.333 to 1.492), whereas no significant difference was observed in the pre-hypertension group.
When both conditions were considered together, a more pronounced pattern emerged. First, in the group with durations of ≥5 years for both diabetes and hypertension, the unadjusted HR for EOD was significantly elevated (7.052; 95% CI, 6.158 to 8.076). This elevated risk remained significant even after the adjustments in model 2 (aHR, 3.182; 95% CI, 2.774 to 3.649) and model 3 (aHR, 2.090; 95% CI, 1.777 to 2.458). Notably, in the group without hypertension but with a diabetes duration of ≥5 years, the aHR in model 3 was 1.705 (95% CI, 1.352 to 2.150). This significantly exceeded the aHR in model 3 for the group with hypertension ≥5 years but without diabetes (1.508; 95% CI, 1.359 to 1.674), approaching the values observed in the group where both diseases had durations ≥5 years.
Using data from the NHIS, our study included 1,979,509 individuals younger than 60 years, who were followed for 7.75±1.22 years. The findings revealed a significantly higher risk of EOD among those with diabetes (aHR, 1.33). Notably, this risk increased with longer diabetes duration, especially when exceeding 5 years (aHR, 1.54). These results were consistent in both male and female participants. The risk was further increased when fasting glucose was ≥150 mg/dL or when insulin was used, indirectly suggesting the negative impact of poor glycemic control on EOD. However, after adjusting for clinically significant variables, the IFG group exhibited no significant differences compared to the normal group. Consistent with previous studies confirming hypertension as a risk factor for dementia [17,27], our study also demonstrated that hypertension significantly increased the risk of EOD (aHR, 1.41). Both hypertension and diabetes showed a similar pattern, in which longer disease duration was associated with increased EOD risk. Within each hypertension category, the risk of EOD generally increased with longer diabetes duration. Interestingly, the highest risk of EOD was observed in individuals with newly diagnosed diabetes and hypertension. This finding may reflect the concept that patients diagnosed concurrently with both conditions might be in a metabolically unfavorable state or may have experienced rapid deterioration of their metabolic profile within a short timeframe [28].
Numerous experimental models and studies have shown that diabetes induces cognitive decline and represents a risk factor for dementia. According to a meta-analysis of 28 studies (89,708 participants with diabetes and 1,058,333 without diabetes), the adjusted relative risk for dementia of any type was significantly higher in patients with diabetes (1.73; 95% CI, 1.65 to 1.82) compared to those without diabetes [29]. However, most studies included in this meta-analysis focused on individuals aged 65 or older, limiting the generalizability of these findings to EOD. Although one study with a mean enrollment age of 44.5 years was included, its three-decade follow-up period also limits its applicability to EOD [30].
To our knowledge, only two longitudinal studies have investigated the potential association between diabetes and EOD. One study, using data from 356,052 participants retrieved from the United Kingdom Biobank, identified various risk factors for EOD [16]. Among these, diabetes significantly increased the risk, with an aHR of 1.65 (95% CI, 1.15 to 2.36). Additionally, a longitudinal study involving 546,709 participants selected from the Korean NHIS analyzed various risk factors for both EOD and late-onset dementia [17]. This study reported that patients with diabetes had a significantly elevated aHR of 1.680 (95% CI, 1.474 to 1.916) for EOD. However, these studies did not account for the duration of diabetes, and their cohorts were not diabetes-specific, thus precluding the analysis of prediabetes. In contrast, our study included a cohort encompassing individuals with IFG and stratified participants based on the duration of diabetes, utilizing a large nationwide dataset from national health examinations. We observed that the aHR for EOD significantly increased with longer diabetes duration. The aHR of 1.334 (95% CI, 1.226 to 1.451) in the diabetes group was comparable to the aHRs reported in the previous two longitudinal studies.
Aging is the most impactful risk factor and underlying mechanism for all types of dementia, especially Alzheimer disease and vascular dementia, which are highly prevalent in older populations [2,31]. However, the influence of aging is less pronounced in EOD, suggesting that diabetes may trigger alternative mechanisms that accelerate dementia onset. Various mechanisms have been identified in experimental models, including the accumulation of amyloid-β and amylin, neuroinflammation, and brain atrophy, all contributing to impairments in memory, learning, and motor functions [11,32-34]. Additionally, diabetes-related vascular endothelial dysfunction reduces cerebral blood flow, leading to hypoxic brain injury and blood-brain barrier disruption [11,35,36]. Insulin resistance, a hallmark of diabetes, has also been linked to cerebral insulin resistance, affecting amyloid-β production and tau hyperphosphorylation and potentially impairing learning ability via hippocampal insulin receptors [11,37,38]. The increased risk of EOD observed in the insulin-using subgroups in our study may further support this mechanism.
Recent studies have established a link between diabetes and increased dementia risk, prompting updates in clinical guidelines. The American Diabetes Association, Endocrine Society, and Japan Diabetes Society currently recommend regular cognitive function screening for patients with diabetes over 65 [39-41]. However, these guidelines do not address EOD or cognitive screening for those younger than 65 who have diabetes. Additionally, a Swedish Dementia Registry study found that patients with diabetes were less likely to receive dementia medications (odds ratios, 0.77 for cholinesterase inhibitors and 0.78 for memantine), suggesting clinical hesitancy in treatment [42]. Notably, diabetes has been found to be more strongly associated with EOD (aHR, 1.680) than with late-onset dementia (aHR, 1.208) [17]. Given the established link between diabetes and EOD, early cognitive assessment may benefit patients with diabetes under 65 who develop cognitive symptoms. Particularly for insulin-treated patients, early cognitive decline necessitates long-term treatment adjustments and proactive approaches for EOD management. Furthermore, recent studies suggest that metformin and sodium-glucose cotransporter 2 inhibitors may reduce the risk of dementia, underscoring the need for further research and updated guidelines regarding diabetes medication management based on cognitive function [43,44].
Our study has certain limitations. First, as an observational study, causal relationships could not be definitively established. Observed associations may have been influenced by residual confounding factors not considered in the analysis. Second, despite the large cohort, participants were limited to the Korean population, making the results potentially less applicable to populations with different genetic, environmental, or healthcare factors. However, this study also has several strengths. It is a large cohort study with an extensive sample size of 1,979,509 individuals and a relatively long average follow-up duration of 7.75 years, thus qualifying as robust. The substantial cohort size and the exclusion of only inappropriately aged individuals and those with prior dementia diagnoses effectively minimized selection bias. Additionally, the analysis accounted for various comorbidities, socioeconomic factors, and lifestyle variables from the NHIS database to ensure a comprehensive assessment. Finally, a notable strength was the detailed analysis of both diabetes duration and the presence of prediabetes.
In conclusion, this study demonstrated that diabetes, unlike IFG, is independently associated with increased risk of EOD. This risk rose significantly with longer diabetes duration. These findings highlight the potential benefits of early intervention and cognitive function monitoring in patients with diabetes, even before the age of 65. Further research is warranted to clarify the mechanisms underlying these findings.

Supplemental Table S1.

Comparison of Hazard Ratios for Early-Onset Dementia by Diabetes Status and Gender
enm-2024-2287-Supplemental-Table-S1.pdf

Supplemental Table S2.

Comparison of Hazard Ratios for Early- Onset Dementia by Fasting Glucose and Medication Status of Antidiabetic Drugs
enm-2024-2287-Supplemental-Table-S2.pdf

Supplemental Table S3.

Comparison of Hazard Ratios on Early-Onset Dementia according to Hypertension Status
enm-2024-2287-Supplemental-Table-S3.pdf

CONFLICTS OF INTEREST

Eun-Jung Rhee is a deputy editor of the journal. But she was not involved in the peer reviewer selection, evaluation, or decision process of this article. No other potential conflicts of interest relevant to this article were reported.

ACKNOWLEDGMENTS

The authors would like to thank the National Health Insurance Service and the Korean National Statistical Office for their invaluable cooperation and support.

AUTHOR CONTRIBUTIONS

Conception or design: J.H.P., S.J.M., K.D.H., E.J.R., W.Y.L. Acquisition, analysis, or interpretation of data: J.H.P., S.J.M., D.Y.L., J.H.K., H.N.J., H.M.K., S.E.P., K.D.H., E.J.R., W.Y.L. Drafting the work or revising: J.H.P., S.J.M. Final approval of the manuscript: J.H.P., S.J.M., D.Y.L., J.H.K., H.N.J., H.M.K., S.E.P., K.D.H., E.J.R., W.Y.L.

Fig. 1.
Flowchart of the participant selection process. IFG, impaired fasting glucose.
enm-2024-2287f1.jpg
enm-2024-2287f2.jpg
Table 1.
Baseline Characteristics of Study Participants
Characteristic Normal IFG New onset Diabetes <5 years Diabetes ≥5 years P value
Number 1,302,706 501,347 77,843 51,926 45,687
Age, yr 48.6±5.9 49.4±5.9 50.1±5.8 51.9±5.6 53.2±5.3 <0.0001
Male sex 595,422 (45.7) 301,042 (60.1) 55,831 (71.7) 33,701 (64.9) 28,977 (63.4) <0.0001
Height, cm 162.4±8.4 164.1±8.4 165.1±8.1 163.9±8.5 163.6±8.4 <0.0001
Weight, kg 62.4±10.3 66.0±10.7 68.6±11.0 68.8±11.2 66.7±10.8 <0.0001
BMI, kg/m2 23.6±2.9 24.5±3.0 25.1±3.2 25.5±3.3 24.8±3.2 <0.0001
Waist circumference, cm 79.2±8.4 82.3±8.3 85.0±8.2 86.0±8.5 84.9±8.2 <0.0001
SBP, mm Hg 120.5±14.4 125.1±14.9 129.4±16.0 126.6±14.8 126.7±15.1 <0.0001
DBP, mm Hg 75.7±10.1 78.6±10.3 81.1±10.7 79.0±9.8 78.2±9.7 <0.0001
Fasting glucose, mg/dL 88.2±7.5 107.9±6.6 159.6±46.7 137.2±46.8 156.2±57.5 <0.0001
Total cholesterol, mg/dL 197.2±35.4 204.5±37.2 210.7±41.7 192.0±40.2 190.0±40.8 <0.0001
HDL-C, mg/dL 56.8±29.0 55.4±26.4 53.4±26.0 51.8±28.2 51.9±28.2 <0.0001
LDL-C, mg/dL 116.7±37.6 119.7±38.4 118.3±44.0 106.4±42.5 105.7±42.0 <0.0001
Estimated GFR, mL/min/1.73 m2 87.2±41.6 85.1±34.1 86.6±36.2 87.1±40.1 85.4±36.0 <0.0001
Baseline income, low 212,373 (16.3) 75,126 (15.0) 13,172 (16.9) 9,035 (17.4) 7,456 (16.3) <0.0001
Regular physical activity 250,297 (19.2) 100,611 (20.1) 15,354 (19.7) 12,329 (23.7) 11,650 (25.5) <0.0001
Current smoking 278,289 (21.4) 131,830 (26.3) 27,728 (35.6) 15,102 (29.1) 12,497 (27.4) <0.0001
Heavy alcohol consumption 560,860 (43.1) 268,689 (53.6) 46,609 (59.9) 24,459 (47.1) 19,943 (43.7) <0.0001
Dyslipidemia 199,950 (15.4) 110,149 (22.0) 22,476 (28.9) 23,924 (46.1) 20,782 (45.5) <0.0001
Depression 38,593 (3.0) 14,116 (2.8) 1,783 (2.3) 2,387 (4.6) 2,266 (5.0) <0.0001
Atrial fibrillation 3,845 (0.3) 1,928 (0.4) 299 (0.4) 419 (0.8) 390 (0.9) <0.0001
Stroke 9,511 (0.7) 4,560 (0.9) 797 (1.0) 1,504 (2.9) 1,760 (3.9) <0.0001

Values are expressed as mean±standard deviation or number (%).

IFG, impaired fasting glucose; BMI, body mass index; SBP, systolic blood pressure; DBP, diastolic blood pressure; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; GFR, glomerular filtration rate.

Table 2.
Comparison of Hazard Ratios for Early-Onset Dementia according to Diabetes Status
Group Total no. No. of events (%) HR (95% CI)
Model 1a Model 2b Model 3c
Normal 1,302,706 5,144 (0.39) 1 (Reference) 1 (Reference) 1 (Reference)
IFG 501,347 2,194 (0.44) 1.132 (1.077–1.190) 1.018 (0.968–1.071) 0.989 (0.938–1.043)
Diabetes 175,456 1,583 (0.90) 2.491 (2.355–2.636) 1.765 (1.667–1.870) 1.334 (1.226–1.451)
 New-onset diabetes 77,843 513 (0.66) 1.744 (1.593–1.909) 1.461 (1.333–1.601) 1.254 (1.122–1.401)
 Diabetes duration <5 years 51,926 465 (0.90) 2.493 (2.268–2.742) 1.669 (1.516–1.836) 1.312 (1.176–1.464)
 Diabetes duration ≥5 years 45,687 605 (1.32) 3.908 (3.592–4.252) 2.272 (2.086–2.474) 1.543 (1.368–1.741)

HR, hazard ratio; CI, confidence interval; IFG, impaired fasting glucose.

a Model 1: unadjusted;

b Model 2: adjusted for age and sex;

c Model 3: adjusted for factors in model 2 along with body mass index, low baseline income, smoking status, alcohol consumption, regular physical activity, dyslipidemia, depression, atrial fibrillation, stroke, fasting glucose, use of three or more oral hypoglycemic agents, and insulin use.

Table 3.
Comparison of Hazard Ratios for Early-Onset Dementia according to the Duration of Diabetes and hypertension
Hypertension Diabetes No. of events HR (95% CI)
Model 1a Model 2b Model 3c
Normal (n=681,966) Normal 1,591 1 (Reference) 1 (Reference) 1 (Reference)
IFG 485 1.203 (1.087–1.332) 1.082 (0.977–1.198) 1.046 (0.944–1.160)
New onset 83 2.008 (1.610–2.504) 1.646 (1.320–2.053) 1.398 (1.110–1.761)
<5 years 69 2.784 (2.187–3.542) 1.821 (1.431–2.318) 1.441 (1.125–1.845)
≥5 years 88 4.273 (3.447–5.296) 2.433 (1.961–3.017) 1.705 (1.352–2.150)
Pre-hypertension (n=800,593) Normal 1,831 1.154 (1.079–1.235) 1.04 (0.972–1.112) 1.063 (0.994–1.138)
IFG 799 1.249 (1.147–1.359) 1.069 (0.981–1.165) 1.073 (0.983–1.172)
New onset 161 1.722 (1.464–2.025) 1.398 (1.188–1.645) 1.236 (1.037–1.473)
<5 years 102 2.462 (2.016–3.008) 1.611 (1.318–1.970) 1.340 (1.088–1.651)
≥5 years 136 4.132 (3.468–4.922) 2.332 (1.956–2.781) 1.730 (1.422–2.105)
New-onset hypertension (n=208,890) Normal 665 1.928 (1.761–2.110) 1.514 (1.382–1.658) 1.573 (1.435–1.724)
IFG 379 1.909 (1.707–2.135) 1.481 (1.323–1.658) 1.517 (1.352–1.702)
New onset 138 3.135 (2.634–3.730) 2.407 (2.021–2.868) 2.163 (1.792–2.611)
<5 years 22 2.055 (1.349–3.129) 1.253 (0.822–1.909) 1.062 (0.693–1.625)
≥5 years 38 4.402 (3.191–6.072) 2.355 (1.706–3.251) 1.802 (1.291–2.515)
Hypertension duration <5 years (n=162,463) Normal 565 2.436 (2.213–2.681) 1.526 (1.385–1.681) 1.448 (1.313–1.597)
IFG 271 2.008 (1.765–2.284) 1.241 (1.090–1.413) 1.167 (1.022–1.331)
New onset 68 2.809 (2.204–3.581) 1.733 (1.359–2.211) 1.486 (1.157–1.909)
<5 years 151 3.62 (3.063–4.277) 2.101 (1.776–2.485) 1.648 (1.379–1.969)
≥5 years 102 4.818 (3.944–5.886) 2.505 (2.048–3.063) 1.634 (1.308–2.040)
Hypertension duration ≥5 years (n=125,597) Normal 492 3.216 (2.906–3.558) 1.63 (1.472–1.806) 1.508 (1.359–1.674)
IFG 260 2.739 (2.402–3.122) 1.378 (1.208–1.573) 1.281 (1.119–1.467)
New onset 63 3.107 (2.415–3.996) 1.592 (1.236–2.049) 1.366 (1.053–1.772)
<5 years 121 5.286 (4.394–6.359) 2.53 (2.102–3.047) 1.999 (1.648–2.425)
≥5 years 241 7.052 (6.158–8.076) 3.182 (2.774–3.649) 2.090 (1.777–2.458)

HR, hazard ratio; CI, confidence interval; IFG, impaired fasting glucose.

a Model 1: unadjusted;

b Model 2: adjusted for age and sex;

c Model 3: adjusted for factors in model 2 along with body mass index, low baseline income, smoking status, alcohol consumption, regular physical activity, dyslipidemia, depression, atrial fibrillation, stroke, fasting glucose, use of three or more oral hypoglycemic agents, and insulin use.

  • 1. GBD 2021 Diabetes Collaborators. Global, regional, and national burden of diabetes from 1990 to 2021, with projections of prevalence to 2050: a systematic analysis for the Global Burden of Disease Study 2021. Lancet 2023;402:203–34.ArticlePubMedPMC
  • 2. Rizzi L, Rosset I, Roriz-Cruz M. Global epidemiology of dementia: Alzheimer’s and vascular types. Biomed Res Int 2014;2014:908915.ArticlePubMedPMCPDF
  • 3. Harvey RJ, Skelton-Robinson M, Rossor MN. The prevalence and causes of dementia in people under the age of 65 years. J Neurol Neurosurg Psychiatry 2003;74:1206–9.ArticlePubMedPMC
  • 4. Hendriks S, Peetoom K, Bakker C, van der Flier WM, Papma JM, Koopmans R, et al. Global prevalence of young-onset dementia: a systematic review and meta-analysis. JAMA Neurol 2021;78:1080–90.ArticlePubMedPMC
  • 5. Kuruppu DK, Matthews BR. Young-onset dementia. Semin Neurol 2013;33:365–85.ArticlePubMedPMC
  • 6. Vieira RT, Caixeta L, Machado S, Silva AC, Nardi AE, Arias-Carrion O, et al. Epidemiology of early-onset dementia: a review of the literature. Clin Pract Epidemiol Ment Health 2013;9:88–95.ArticlePubMedPMC
  • 7. Alzheimer’s Association. Global dementia cases forecasted to triple by 2050. Alzheimer’s Association International Conference 2021 2021 Jul 26-30; Denver, CO. Available from: https://aaic.alz.org/releases_2021/global-prevalence.asp.Article
  • 8. Lim L, Zhang A, Lim L, Choong TM, Silva E, Ng A, et al. High caregiver burden in young onset dementia: what factors need attention? J Alzheimers Dis 2018;61:537–43.ArticlePubMed
  • 9. Nwadiugwu M. Early-onset dementia: key issues using a relationship-centred care approach. Postgrad Med J 2021;97:598–604.ArticlePubMedPMCPDF
  • 10. Biessels GJ, Staekenborg S, Brunner E, Brayne C, Scheltens P. Risk of dementia in diabetes mellitus: a systematic review. Lancet Neurol 2006;5:64–74.ArticlePubMedPMC
  • 11. Biessels GJ, Despa F. Cognitive decline and dementia in diabetes mellitus: mechanisms and clinical implications. Nat Rev Endocrinol 2018;14:591–604.ArticlePubMedPMCPDF
  • 12. Cheng G, Huang C, Deng H, Wang H. Diabetes as a risk factor for dementia and mild cognitive impairment: a meta-analysis of longitudinal studies. Intern Med J 2012;42:484–91.ArticlePubMedPMC
  • 13. Crane PK, Walker R, Hubbard RA, Li G, Nathan DM, Zheng H, et al. Glucose levels and risk of dementia. N Engl J Med 2013;369:540–8.ArticlePubMedPMC
  • 14. Haroon NN, Austin PC, Shah BR, Wu J, Gill SS, Booth GL. Risk of dementia in seniors with newly diagnosed diabetes: a population-based study. Diabetes Care 2015;38:1868–75.ArticlePubMedPDF
  • 15. Choi YJ, Kim S, Hwang YJ, Kim C. Prevalence of dementia in Korea based on hospital utilization data from 2008 to 2016. Yonsei Med J 2021;62:948–53.ArticlePubMedPMCPDF
  • 16. Hendriks S, Ranson JM, Peetoom K, Lourida I, Tai XY, de Vugt M, et al. Risk factors for young-onset dementia in the UK Biobank. JAMA Neurol 2024;81:134–42.ArticlePubMedPMC
  • 17. Chun MY, Chae W, Seo SW, Jang H, Yun J, Na DL, et al. Effects of risk factors on the development and mortality of early- and late-onset dementia: an 11-year longitudinal nationwide population-based cohort study in South Korea. Alzheimers Res Ther 2024;16:92.ArticlePubMedPMCPDF
  • 18. Lee J, Lee JS, Park SH, Shin SA, Kim K. Cohort profile: the National Health Insurance Service-National Sample Cohort (NHIS-NSC), South Korea. Int J Epidemiol 2017;46:e15.ArticlePubMedPMC
  • 19. Shin DW, Cho B, Guallar E. Korean National Health Insurance Database. JAMA Intern Med 2016;176:138.Article
  • 20. Seong SC, Kim YY, Park SK, Khang YH, Kim HC, Park JH, et al. Cohort profile: the National Health Insurance Service-National Health Screening Cohort (NHIS-HEALS) in Korea. BMJ Open 2017;7:e016640.ArticlePubMedPMC
  • 21. Lee YH, Han K, Ko SH, Ko KS, Lee KU; Taskforce Team of Diabetes Fact Sheet of the Korean Diabetes Association. Data analytic process of a nationwide population-based study using National Health Information Database established by National Health Insurance Service. Diabetes Metab J 2016;40:79–82.ArticlePubMedPMCPDF
  • 22. Agarwal DP. Cardioprotective effects of light-moderate consumption of alcohol: a review of putative mechanisms. Alcohol Alcohol 2002;37:409–15.ArticlePubMed
  • 23. Lobelo F, Rohm Young D, Sallis R, Garber MD, Billinger SA, Duperly J, et al. Routine assessment and promotion of physical activity in healthcare settings: a scientific statement from the American Heart Association. Circulation 2018;137:e495–522.ArticlePubMedPMC
  • 24. Levey AS, Coresh J, Greene T, Stevens LA, Zhang YL, Hendriksen S, et al. Using standardized serum creatinine values in the modification of diet in renal disease study equation for estimating glomerular filtration rate. Ann Intern Med 2006;145:247–54.ArticlePubMedPMC
  • 25. Levey AS, Coresh J, Balk E, Kausz AT, Levin A, Steffes MW, et al. National Kidney Foundation practice guidelines for chronic kidney disease: evaluation, classification, and stratification. Ann Intern Med 2003;139:137–47.ArticlePubMedPDF
  • 26. Fernandez Fernandez R, Martin JI, Anton MA. Depression as a risk factor for dementia: a meta-analysis. J Neuropsychiatry Clin Neurosci 2024;36:101–9.ArticlePubMed
  • 27. Daugherty AM. Hypertension-related risk for dementia: a summary review with future directions. Semin Cell Dev Biol 2021;116:82–9.ArticlePubMed
  • 28. Charoensri S, Kritmetapak K, Tangpattanasiri T, Pongchaiyakul C. The impact of new-onset diabetes mellitus and hypertension on all-cause mortality in an apparently healthy population: a ten-year follow-up study. J Diabetes Res 2021;2021:3964013.ArticlePubMedPMCPDF
  • 29. Gudala K, Bansal D, Schifano F, Bhansali A. Diabetes mellitus and risk of dementia: a meta-analysis of prospective observational studies. J Diabetes Investig 2013;4:640–50.ArticlePubMedPMCPDF
  • 30. Schnaider Beeri M, Goldbourt U, Silverman JM, Noy S, Schmeidler J, Ravona-Springer R, et al. Diabetes mellitus in midlife and the risk of dementia three decades later. Neurology 2004;63:1902–7.ArticlePubMed
  • 31. Hou Y, Dan X, Babbar M, Wei Y, Hasselbalch SG, Croteau DL, et al. Ageing as a risk factor for neurodegenerative disease. Nat Rev Neurol 2019;15:565–81.ArticlePubMedPMCPDF
  • 32. Devi L, Alldred MJ, Ginsberg SD, Ohno M. Mechanisms underlying insulin deficiency-induced acceleration of β-amyloidosis in a mouse model of Alzheimer’s disease. PLoS One 2012;7:e32792.ArticlePubMedPMC
  • 33. Wang JQ, Yin J, Song YF, Zhang L, Ren YX, Wang DG, et al. Brain aging and AD-like pathology in streptozotocin-induced diabetic rats. J Diabetes Res 2014;2014:796840.ArticlePubMedPMCPDF
  • 34. Ilaiwy A, Liu M, Parry TL, Bain JR, Newgard CB, Schisler JC, et al. Human amylin proteotoxicity impairs protein biosynthesis, and alters major cellular signaling pathways in the heart, brain and liver of humanized diabetic rat model in vivo. Metabolomics 2016;12:95.ArticlePubMedPMCPDF
  • 35. Ly H, Verma N, Wu F, Liu M, Saatman KE, Nelson PT, et al. Brain microvascular injury and white matter disease provoked by diabetes-associated hyperamylinemia. Ann Neurol 2017;82:208–22.ArticlePubMedPMCPDF
  • 36. Iadecola C. The pathobiology of vascular dementia. Neuron 2013;80:844–66.ArticlePubMedPMC
  • 37. Li ZG, Zhang W, Sima AA. Alzheimer-like changes in rat models of spontaneous diabetes. Diabetes 2007;56:1817–24.ArticlePubMedPDF
  • 38. Marks DR, Tucker K, Cavallin MA, Mast TG, Fadool DA. Awake intranasal insulin delivery modifies protein complexes and alters memory, anxiety, and olfactory behaviors. J Neurosci 2009;29:6734–51.ArticlePubMedPMC
  • 39. American Diabetes Association Professional Practice Committee. 13. Older adults: standards of care in diabetes-2024. Diabetes Care 2024;47(Suppl 1):S244–57.ArticlePubMedPMC
  • 40. LeRoith D, Biessels GJ, Braithwaite SS, Casanueva FF, Draznin B, Halter JB, et al. Treatment of diabetes in older adults: an Endocrine Society clinical practice guideline. J Clin Endocrinol Metab 2019;104:1520–74.ArticlePubMedPMCPDF
  • 41. Araki E, Goto A, Kondo T, Noda M, Noto H, Origasa H, et al. Japanese Clinical Practice Guideline for Diabetes 2019. J Diabetes Investig 2020;11:1020–76.ArticlePubMedPMCPDF
  • 42. Secnik J, Cermakova P, Fereshtehnejad SM, Dannberg P, Johnell K, Fastbom J, et al. Diabetes in a large dementia cohort: clinical characteristics and treatment from the Swedish Dementia Registry. Diabetes Care 2017;40:1159–66.ArticlePubMedPMCPDF
  • 43. Orkaby AR, Cho K, Cormack J, Gagnon DR, Driver JA. Metformin vs sulfonylurea use and risk of dementia in US veterans aged ≥65 years with diabetes. Neurology 2017;89:1877–85.ArticlePubMedPMC
  • 44. Wu CY, Iskander C, Wang C, Xiong LY, Shah BR, Edwards JD, et al. Association of sodium-glucose cotransporter 2 inhibitors with time to dementia: a population-based cohort study. Diabetes Care 2023;46:297–304.ArticlePubMedPDF

Figure & Data

References

    Citations

    Citations to this article as recorded by  
    • Research Progress on Time in Range as a Predictor of Chronic Complications in Type 2 Diabetes Mellitus
      晓雨 李
      Advances in Clinical Medicine.2026; 16(05): 2100.     CrossRef

    • PubReader PubReader
    • ePub LinkePub Link
    • Cite
      Cite
      export Copy Download
      Close
      Download Citation
      Download a citation file in RIS format that can be imported by all major citation management software, including EndNote, ProCite, RefWorks, and Reference Manager.
      Format:
      • RIS — For EndNote, ProCite, RefWorks, and most other reference management software
      • BibTeX — For JabRef, BibDesk, and other BibTeX-specific software
      Include:
      • Citation for the content below
      Early-Onset Dementia Risk Escalates with Diabetes Duration: Insights from a Nationwide Cohort Study
      Endocrinol Metab. 2026;41(2):235-244.   Published online July 17, 2025
      DOI: https://doi.org/10.3803/EnM.2024.2287
      Close
    • XML DownloadXML Download
    Figure
    • 0
    • 1
    Related articles
    Early-Onset Dementia Risk Escalates with Diabetes Duration: Insights from a Nationwide Cohort Study
    Image Image
    Fig. 1. Flowchart of the participant selection process. IFG, impaired fasting glucose.
    Graphical abstract

    Fig. 1.

    Graphical abstract

    Early-Onset Dementia Risk Escalates with Diabetes Duration: Insights from a Nationwide Cohort Study
    Characteristic Normal IFG New onset Diabetes <5 years Diabetes ≥5 years P value
    Number 1,302,706 501,347 77,843 51,926 45,687
    Age, yr 48.6±5.9 49.4±5.9 50.1±5.8 51.9±5.6 53.2±5.3 <0.0001
    Male sex 595,422 (45.7) 301,042 (60.1) 55,831 (71.7) 33,701 (64.9) 28,977 (63.4) <0.0001
    Height, cm 162.4±8.4 164.1±8.4 165.1±8.1 163.9±8.5 163.6±8.4 <0.0001
    Weight, kg 62.4±10.3 66.0±10.7 68.6±11.0 68.8±11.2 66.7±10.8 <0.0001
    BMI, kg/m2 23.6±2.9 24.5±3.0 25.1±3.2 25.5±3.3 24.8±3.2 <0.0001
    Waist circumference, cm 79.2±8.4 82.3±8.3 85.0±8.2 86.0±8.5 84.9±8.2 <0.0001
    SBP, mm Hg 120.5±14.4 125.1±14.9 129.4±16.0 126.6±14.8 126.7±15.1 <0.0001
    DBP, mm Hg 75.7±10.1 78.6±10.3 81.1±10.7 79.0±9.8 78.2±9.7 <0.0001
    Fasting glucose, mg/dL 88.2±7.5 107.9±6.6 159.6±46.7 137.2±46.8 156.2±57.5 <0.0001
    Total cholesterol, mg/dL 197.2±35.4 204.5±37.2 210.7±41.7 192.0±40.2 190.0±40.8 <0.0001
    HDL-C, mg/dL 56.8±29.0 55.4±26.4 53.4±26.0 51.8±28.2 51.9±28.2 <0.0001
    LDL-C, mg/dL 116.7±37.6 119.7±38.4 118.3±44.0 106.4±42.5 105.7±42.0 <0.0001
    Estimated GFR, mL/min/1.73 m2 87.2±41.6 85.1±34.1 86.6±36.2 87.1±40.1 85.4±36.0 <0.0001
    Baseline income, low 212,373 (16.3) 75,126 (15.0) 13,172 (16.9) 9,035 (17.4) 7,456 (16.3) <0.0001
    Regular physical activity 250,297 (19.2) 100,611 (20.1) 15,354 (19.7) 12,329 (23.7) 11,650 (25.5) <0.0001
    Current smoking 278,289 (21.4) 131,830 (26.3) 27,728 (35.6) 15,102 (29.1) 12,497 (27.4) <0.0001
    Heavy alcohol consumption 560,860 (43.1) 268,689 (53.6) 46,609 (59.9) 24,459 (47.1) 19,943 (43.7) <0.0001
    Dyslipidemia 199,950 (15.4) 110,149 (22.0) 22,476 (28.9) 23,924 (46.1) 20,782 (45.5) <0.0001
    Depression 38,593 (3.0) 14,116 (2.8) 1,783 (2.3) 2,387 (4.6) 2,266 (5.0) <0.0001
    Atrial fibrillation 3,845 (0.3) 1,928 (0.4) 299 (0.4) 419 (0.8) 390 (0.9) <0.0001
    Stroke 9,511 (0.7) 4,560 (0.9) 797 (1.0) 1,504 (2.9) 1,760 (3.9) <0.0001
    Group Total no. No. of events (%) HR (95% CI)
    Model 1a Model 2b Model 3c
    Normal 1,302,706 5,144 (0.39) 1 (Reference) 1 (Reference) 1 (Reference)
    IFG 501,347 2,194 (0.44) 1.132 (1.077–1.190) 1.018 (0.968–1.071) 0.989 (0.938–1.043)
    Diabetes 175,456 1,583 (0.90) 2.491 (2.355–2.636) 1.765 (1.667–1.870) 1.334 (1.226–1.451)
     New-onset diabetes 77,843 513 (0.66) 1.744 (1.593–1.909) 1.461 (1.333–1.601) 1.254 (1.122–1.401)
     Diabetes duration <5 years 51,926 465 (0.90) 2.493 (2.268–2.742) 1.669 (1.516–1.836) 1.312 (1.176–1.464)
     Diabetes duration ≥5 years 45,687 605 (1.32) 3.908 (3.592–4.252) 2.272 (2.086–2.474) 1.543 (1.368–1.741)
    Hypertension Diabetes No. of events HR (95% CI)
    Model 1a Model 2b Model 3c
    Normal (n=681,966) Normal 1,591 1 (Reference) 1 (Reference) 1 (Reference)
    IFG 485 1.203 (1.087–1.332) 1.082 (0.977–1.198) 1.046 (0.944–1.160)
    New onset 83 2.008 (1.610–2.504) 1.646 (1.320–2.053) 1.398 (1.110–1.761)
    <5 years 69 2.784 (2.187–3.542) 1.821 (1.431–2.318) 1.441 (1.125–1.845)
    ≥5 years 88 4.273 (3.447–5.296) 2.433 (1.961–3.017) 1.705 (1.352–2.150)
    Pre-hypertension (n=800,593) Normal 1,831 1.154 (1.079–1.235) 1.04 (0.972–1.112) 1.063 (0.994–1.138)
    IFG 799 1.249 (1.147–1.359) 1.069 (0.981–1.165) 1.073 (0.983–1.172)
    New onset 161 1.722 (1.464–2.025) 1.398 (1.188–1.645) 1.236 (1.037–1.473)
    <5 years 102 2.462 (2.016–3.008) 1.611 (1.318–1.970) 1.340 (1.088–1.651)
    ≥5 years 136 4.132 (3.468–4.922) 2.332 (1.956–2.781) 1.730 (1.422–2.105)
    New-onset hypertension (n=208,890) Normal 665 1.928 (1.761–2.110) 1.514 (1.382–1.658) 1.573 (1.435–1.724)
    IFG 379 1.909 (1.707–2.135) 1.481 (1.323–1.658) 1.517 (1.352–1.702)
    New onset 138 3.135 (2.634–3.730) 2.407 (2.021–2.868) 2.163 (1.792–2.611)
    <5 years 22 2.055 (1.349–3.129) 1.253 (0.822–1.909) 1.062 (0.693–1.625)
    ≥5 years 38 4.402 (3.191–6.072) 2.355 (1.706–3.251) 1.802 (1.291–2.515)
    Hypertension duration <5 years (n=162,463) Normal 565 2.436 (2.213–2.681) 1.526 (1.385–1.681) 1.448 (1.313–1.597)
    IFG 271 2.008 (1.765–2.284) 1.241 (1.090–1.413) 1.167 (1.022–1.331)
    New onset 68 2.809 (2.204–3.581) 1.733 (1.359–2.211) 1.486 (1.157–1.909)
    <5 years 151 3.62 (3.063–4.277) 2.101 (1.776–2.485) 1.648 (1.379–1.969)
    ≥5 years 102 4.818 (3.944–5.886) 2.505 (2.048–3.063) 1.634 (1.308–2.040)
    Hypertension duration ≥5 years (n=125,597) Normal 492 3.216 (2.906–3.558) 1.63 (1.472–1.806) 1.508 (1.359–1.674)
    IFG 260 2.739 (2.402–3.122) 1.378 (1.208–1.573) 1.281 (1.119–1.467)
    New onset 63 3.107 (2.415–3.996) 1.592 (1.236–2.049) 1.366 (1.053–1.772)
    <5 years 121 5.286 (4.394–6.359) 2.53 (2.102–3.047) 1.999 (1.648–2.425)
    ≥5 years 241 7.052 (6.158–8.076) 3.182 (2.774–3.649) 2.090 (1.777–2.458)
    Table 1. Baseline Characteristics of Study Participants

    Values are expressed as mean±standard deviation or number (%).

    IFG, impaired fasting glucose; BMI, body mass index; SBP, systolic blood pressure; DBP, diastolic blood pressure; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; GFR, glomerular filtration rate.

    Table 2. Comparison of Hazard Ratios for Early-Onset Dementia according to Diabetes Status

    HR, hazard ratio; CI, confidence interval; IFG, impaired fasting glucose.

    Model 1: unadjusted;

    Model 2: adjusted for age and sex;

    Model 3: adjusted for factors in model 2 along with body mass index, low baseline income, smoking status, alcohol consumption, regular physical activity, dyslipidemia, depression, atrial fibrillation, stroke, fasting glucose, use of three or more oral hypoglycemic agents, and insulin use.

    Table 3. Comparison of Hazard Ratios for Early-Onset Dementia according to the Duration of Diabetes and hypertension

    HR, hazard ratio; CI, confidence interval; IFG, impaired fasting glucose.

    Model 1: unadjusted;

    Model 2: adjusted for age and sex;

    Model 3: adjusted for factors in model 2 along with body mass index, low baseline income, smoking status, alcohol consumption, regular physical activity, dyslipidemia, depression, atrial fibrillation, stroke, fasting glucose, use of three or more oral hypoglycemic agents, and insulin use.

    Table 1.

    Table 2.

    Table 3.

    Endocrinol Metab : Endocrinology and Metabolism
    © Korean Endocrine Society.
    TOP