Abstract
Background
There is a lack of data about the treatment effect of glycemic control on incident dementia in patients with advanced age.
Methods
In a nationwide Korean cohort of 79,076 diabetic patients 75 years or older and a representative cohort of 74,672 diabetics aged 50 to 74 years, multivariable-adjusted incidence of overt dementia was estimated across yearly-averaged on-treatment fasting blood glucose (FBG) levels.
Results
During 9-year follow-up, overt dementia was noted in 24,710 (31.2%) patients 75 years or older and in 5237 (7.0%) patients aged 50 to 74 years. For dementia risk, J-shaped associations were observed across on-treatment FBG levels (80–99, 100–109, 110–125, 126–139, 140–159, 160–179, and 180–900 mg/dl) in patients 75 years or older (respective incidence: 49.3, 45.7, 45.9, 45.7, 48.5, 51.5, and 57.9 per 1000 person-years) and in those aged 50 to 74 years (respective incidence: 8.9, 8.3, 7.7, 7.6, 8.0, 8.6, and 10.6 per 1000 person-years) with a significant interaction of FBG level and age group (P = 0.001). For all-cause mortality, the J-shaped association curve was left-shifted in patients 75 years or older (respective incidence: 64.9, 59.1, 57.6, 60.4, 64.0, 70.9, and 90.4 per 1000 person-years) relative to that in patients aged 50 to 74 years (respective incidence: 15.7, 13.4, 12.3, 12.2, 13.4, 15.7, and 21.8 per 1000 person-years; P < 0.001 for interaction).
Conclusion
The achieved glycemic level with the lowest risk for dementia and mortality was lower in older patients, and absolute risk increase related to poorly controlled glucose was greater in the elderly compared with younger patients.
Supplementary Information
The online version contains supplementary material available at 10.1007/s13340-023-00684-4.
Keywords: Age groups, Dementia, Diabetes, Glycemic control, Mortality
Introduction
Dementia is the leading cause of disability among elderly people, and the global number of dementia has increased with population ageing [1]. Although there is currently no available treatment to cure dementia, the identification and modification of risk factors can potentially prevent dementia. Studies have shown that type 2 diabetes is associated with a higher risk for incident dementia [2, 3], and higher glycemic levels are associated with a higher risk of dementia or cognitive dysfunction in people with and without diabetes [4–8]. However, there is insufficient evidence that glycemic control can reduce the risk of cognitive impairment.
A 2017 Cochrane review of randomized controlled trials found no good evidence that any specific treatment for diabetes is superior to prevent cognitive impairment [9], while the review was limited by the small number of trials measuring cognition to assess treatment effects and the use of non-standardized tests to measure cognitive function. The Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial showed no difference between intensive glycemic control and standard control in changes in cognitive function test scores [10, 11]. Similarly, the Action in Diabetes and Vascular Disease: Preterax and Diamicron Modified Release Controlled Evaluation (ADVANCE) trial failed to show a significant difference between intensive and standard glycemic control in the rate of cognitive function decline or 5-year risk of dementia [12]. On the other hand, a large-scale cohort study in 40 to 74-year-old adults showed a J-shaped association between on-treatment fasting blood glucose (FBG) levels and the risk of dementia [13], while few studies examined appropriate glycemic targets for treatment of diabetes in patients with very old age. It remains unanswered whether glycemic control can reduce dementia risk in patients with advanced age.
To investigate what the treatment target of blood glucose is appropriate for the prevention of dementia in patients with diabetes aged above and below 75 years, 9-year risk of dementia was assessed according to on-treatment FBG levels in a nationwide cohort of diabetic patients aged 75 years or older and a representative sample of diabetics aged 50 to 74 years in Korea.
Methods
Participants
The present study was conducted in retrospective cohorts generated from the Korean National Health Information Database. The public database covering data for the whole population in Korea is managed by the National Health Insurance Service (NHIS) [14]. The data used were de-identified before access. Access to the data was approved by the Institutional Review Board of Kangwon National University Hospital (IRB File No. KNHU-A-2021–04-018). The need for informed consent was waived because all data were fully anonymized.
For this study, 614.927 adults aged 75 years or older who participated in the nationwide health screening survey in 2009 or 2010 were enrolled, and their health screening and NHIS reimbursement records were collected from January 1, 2005 to December 31, 2019. From the 614,927 participants, 26,877 with missing or outlier data, 21,070 who died or had an estimated glomerular filtration rate (eGFR) < 15 ml/min/1.73 m2 or end-stage kidney disease, and 43,540 who developed dementia before baseline (January 1, 2011) were excluded. From the remaining 523,440 participants, 79,076 who had diabetes and on-treatment FBG records at baseline were included in the final analysis (Fig. 1A).
Fig. 1.
Flow chart of participants selection for diabetic patients 75 years or older (A) and those aged 50 to 74 years (B)
In addition, one-tenth of 50- to 74-year-old participants of the health screening survey were randomly selected to generate a sample of the middle-aged or young-old population. From the 759,053 participants, 27,014 with missing or outlier data, 11,766 with an eGFR < 15 ml/min/1.73 m2 or end-stage kidney disease, and 5744 who died or developed dementia before baseline were excluded. From the remaining 714,529 participants, 74,672 who had diabetes and on-treatment FBG records at baseline were included in the final analysis (Fig. 1B).
Fixed and time-varying covariates
Baseline data were obtained from the health screening and NHIS reimbursement records (Table S1) between 2005 and 2010. Using the baseline data, total cholesterol, HDL cholesterol, systolic blood pressure, body mass index, waist circumference, creatinine-based eGFR, dipstick albuminuria, income level, smoking status, frequency of physical exercise, alcohol consumption, medical history of cardiovascular disease, antihypertensive treatment status, statin use status, and onset year of diabetes were determined and categorized referring to the publications (Text S1).
In each year of follow-up, hypoglycemic treatment status was assessed and categorized into 3 groups (occasional, irregular, or regular medication). Regular medication was considered if the prescription was for > 2/3 of the period from the initiation of medication to each year, irregular medication if the prescription was for 1/3 to 2/3 of each period, occasional medication if the prescription was for < 1/3 of each period.
Blood glucose was measured after at least an 8-h fast, during biennial health screenings. On-treatment and untreated FBG values were separated as the risk thresholds were quite different depending on treatment status [13, 15, 16]. On-treatment FBG was considered in the case that received hypoglycemic agents for ≥ 90 days in the year of measurement. In each year of follow-up, the separated FBG values were averaged from 2005 and classified into 8 categories (30–79, 80–99, 100–109, 110–125, 126–139, 140–159, 160–179, or 180–900 mg/dl) primarily with 20 mg/dl intervals and additionally incorporating cutoffs for prediabetes and diabetes [17].
Outcome
The primary outcome was overt dementia, which was defined as a composite of moderate to severe dementia or death from dementia. The secondary outcome was all-cause mortality. Moderate to severe dementia was identified using NHIS reimbursement records as the case that received donepezil, galantamine, rivastigmine, or memantine with diagnosis of dementia (Table S1). For patients with the MMSE score ≤ 26, the Global Deterioration Scale ≥ 3, or the Clinical Dementia Rating ≥ 1, the cholinesterase inhibitors or memantine are reimbursed in Korea. All-cause deaths were confirmed using death certificates from Statistics Korea. Death from dementia was identified by the primary cause of death on the certificates. The outcome was identified from January 1, 2011 to December 31, 2019, and censoring occurred with death or end of study (December 31, 2019).
Statistical analysis
All statistical analyses were performed using SAS software version 9.4 (SAS Institute). This study estimated multivariable-adjusted hazard ratios in time-dependent Cox regression models. The models included yearly-updated, hypoglycemic treatment status and yearly-averaged, on-treatment FBG levels along with untreated levels, as time-varying covariates (Table S2). In addition, the models included a continuous variable of baseline age and categorized variables of baseline eGFR, albuminuria, total cholesterol, HDL cholesterol, systolic blood pressure, body mass index, waist circumference, income level, smoking status, drinking amount, exercise frequency, history of cardiovascular disease, antihypertensive treatment status, statin use status, onset year of diabetes, and sex as fixed covariates. Absolute risk was estimated as an incidence rate per person-year. The adjusted incidence rate and 95% confidence interval (CI) was calculated by multiplying the hazard ratio and its 95% CI by a constant to make the sum of the products of incidence rates and person-years in FBG categories equal the total number of observed events. In addition, spline regression analysis was conducted adjusting for the same covariates, to display associations of continuous FBG values and hazard ratios. The yearly-averaged, on-treatment FBG was modelled using a restricted cubic spline function with 5 knots at 5, 27.5, 50, 72.5, and 95 percentiles.
The analyses were conducted separately in patients aged 75 years or older and those aged 50 to 74 years, and further analyses stratified by sex were conducted. Besides, the interaction effects of on-treatment FBG and age group (or sex) were analyzed. The proportional hazard assumption was tested by including a time-dependent interaction term of on-treatment FBG and a function of survival time. In addition, the analyses adding participants with no diabetes at baseline were performed to estimate risks of untreated FBG among people with and without diabetes and to estimate risks of on-treatment FBG among patients with and without baseline on-treatment FBG records including those with new-onset diabetes during the follow-up period. The analyses with covariates of baseline FBG levels rather than yearly-averaged levels were also performed to compare the risks of baseline levels with those of time-averaged levels. Data are presented as means and SDs, numbers and percentages, hazard ratios and 95% CIs, or incidence rates and 95% CIs. P values and 95% CIs were two-sided, and those were not adjusted for multiple testing. P < 0.05 was considered to be statistically significant.
Results
Baseline characteristics
Of 79,076 diabetic patients 75 years or older (mean age, 78.8 years; 40.2% men), 60,231 (76.2%) had a duration of diabetes longer than 4 years at baseline (Table 1). Of 74,672 patients aged 50 to 74 years (mean age, 62.3 years, 53.4% men), 54,729 (72.7%) had a diabetes duration of longer than 4 years. The proportions of patients with a medical history of cardiovascular disease and those who received hypoglycemic or antihypertensive treatment regularly were higher in patients 75 years or older compared with those aged 50 to 74 years. The older patients had lower FBG and eGFR levels but higher systolic blood pressure levels.
Table 1.
Baseline characteristics of the study participants
| Characteristic | Diabetics Aged ≥ 75 Years | Diabetics Aged 50 to 74 Years |
|---|---|---|
| No. of participants | 79,076 | 74,672 |
| Age, mean (SD), yr | 78.8 (3.1) | 62.3 (6,8) |
| Men, no. (%) | 31,827 (40.2%) | 39,865 (53.4%) |
| Onset year of diabetes, no. (%) | ||
| 2006 or less | 60,231 (76.2%) | 54,279 (72.7%) |
| 2007 or 2008 | 11,389 (14.4%) | 12,576 (16.8%) |
| 2009 or 2010 | 7,456 (9.4%) | 7,817 (10.5%) |
| Prior cardiovascular disease, no. (%) | 13,269 (16.8%) | 7,939 (10.6%) |
| Hypoglycemic treatment, no. (%) | ||
| Occasional medication | 5,373 (6.8%) | 6,741 (9.0%) |
| Irregular medication | 5,297 (6.7%) | 6,352 (8.5%) |
| Regular medication | 68,406 (86.5%) | 61,579 (82.5%) |
| Antihypertensive treatment, no. (%) | ||
| No or occasional medication | 16,056 (20.3%) | 27,857 (37.3%) |
| Irregular medication | 4,494 (5.7%) | 3,563 (4.8%) |
| Regular medication | 58,526 (74.0%) | 43,252 (57.9%) |
| Statin therapy, no. (%) | ||
| No or occasional medication | 57,354 (72.5%) | 50,127 (67.1%) |
| Irregular medication | 5,157 (6.5%) | 6,484 (8.7%) |
| Regular medication | 16,565 (20.9%) | 18,061 (24.2%) |
| FBG level, no. (%) | ||
| 30–79 mg/dl | 1,238 (1.6%) | 651 (0.9%) |
| 80–99 mg/dl | 10,728 (13.6%) | 6,430 (8.6%) |
| 100–109 mg/dl | 10,886 (13.8%) | 7,990 (10.7%) |
| 110–125 mg/dl | 19,464 (24.6%) | 16,853 (22.6%) |
| 126–140 mg/dl | 12,807 (16.2%) | 12,988 (17.4%) |
| 140–159 mg/dl | 11,024 (13.9%) | 12,483 (16.7%) |
| 160–179 mg/dl | 5,676 (7.2%) | 7,162 (9.6%) |
| 180–900 mg/dl | 7,253 (9.2%) | 10,115 (13.5%) |
| Systolic blood pressure, mean (SD), mm Hg | 133.6 (13.6) | 130.0 (13.2) |
| Total cholesterol, mean (SD), mg/dl | 193.9 (31.9) | 195.0 (31.3) |
| HDL cholesterol, mean (SD), mg/dl | 49.6 (12.4) | 50.4 (12.0) |
| Body mass index, mean (SD), kg/m2 | 24.4 (3.0) | 25.1 (3.1) |
| Waist circumference, mean (SD), cm | 85.9 (8.2) | 85.8 (7.9) |
| eGFR, mean (SD), ml/min/1.73 m2 | 64.7 (16.4) | 79.4 (17.0) |
| Albuminuria, no. (%) | ||
| No albuminuria | 70,426 (89.1%) | 66,395 (88.9%) |
| Dipstick albumin 1 + or tracea | 5,046 (6.4%) | 4,660 (6.2%) |
| Dipstick albumin ≥ 2 + * | 3,604 (4.6%) | 3,617 (4.8%) |
| Income level, no. (%) | ||
| Highest | 27,347 (34.6%) | 15,411 (20.6%) |
| High | 19,238 (24.3%) | 18,904 (25.3%) |
| Middle | 14,132 (17.9%) | 18,076 (24.2%) |
| Low | 10,308 (13.0%) | 13,948 (18.7%) |
| Lowest | 8051 (10.2%) | 8333 (11.2%) |
| Smoking, no. (%) | ||
| Never smoked | 61,068 (77.2%) | 46,999 (62.9%) |
| Former smoker | 11,805 (14.9%) | 14,089 (18.9%) |
| Current smoker | 6203 (7.8%) | 13,584 (18.2%) |
| Physical exercise, no. (%) | ||
| < 1 day/week | 28,047 (35.5%) | 17,810 (23.9%) |
| 1 to 2 days/week | 13,075 (16.5%) | 13,400 (17.9%) |
| 3 to 4 days/week | 12,596 (15.9%) | 16,945 (22.7%) |
| ≥ 5 days/week | 25,358 (32.1%) | 26,517 (35.5%) |
| Drinking, no. (%) | ||
| Non drinking | 66,348 (83.9%) | 49,090 (65.7%) |
| 1–7 drinks/week | 6855 (8.7%) | 10,279 (13.8%) |
| 8–28 drinks/week | 4438 (5.6%) | 10,959 (14.7%) |
| ≥ 29 drinks/week | 1435 (1.8%) | 4344 (5.8%) |
aThe subgroups had dipstick albuminuria 1 + at least once or trace twice and dipstick albuminuria ≥ 2 + at least once during health screenings 2007 to 2010
eGFR estimated glomerular filtration rate, FBG fasting blood glucose, HDL high-density lipoprotein
Dementia risk
Over 9 years of follow-up, overt dementia was noted in 24,710 (31.2%) patients 75 years or older and in 5237 (7.0%) patients aged 50 to 74 years (Table S3). J-shaped associations (Fig. 2A, 3A, and 3C) were observed across on-treatment FBG levels (80–99, 100–109, 110–125, 126–139, 140–159, 160–179, and 180–900 mg/dl) for dementia risk in patients 75 years or older (incidence: 49.3, 45.7, 45.9, 45.7, 48.5, 51.5, and 57.9 per 1000 person-years, respectively) and in those aged 50 to 74 years (incidence: 8.9, 8.3, 7.7, 7.6, 8.0, 8.6, and 10.6 per 1000 person-years, respectively). The interaction effect (Table S4) of on-treatment FBG and age group in the association with dementia risk was statistically significant (P = 0.001). In analyses stratified by sex (Figure S1A and S2A), the relative risk increase in inadequately or excessively treated FBG levels did not clearly differ between men and women (Tables S3 and S4) among patients 75 years or older (P = 0.93 for interaction) or among those aged 50 to 74 years (P = 0.15 for interaction).
Fig. 2.
Adjusted incidence rates of overt dementia (A) and all-cause death (B) across on-treatment fasting blood glucose (FBG) levels in patients 75 years or older and those aged 50 to 74 years. The incidence rates and 95% CIs (error bars) in FBG categories (30–79, 80–99, 100–109, 110–125, 126–139, 140–159, 160–179, or 180–900 mg/dl) were calculated by multiplying multivariable-adjusted hazard ratios and their 95% CIs by a constant to make the sum of the products of incidence rates and person-years in FBG categories equal the total number of observed events. The FBG 30–79 mg/dl was excluded from the plots. The FBG 110–125 mg/dl was set as the reference
Fig. 3.
Associations of on-treatment fasting blood glucose (FBG) levels with risks of overt dementia (A and C) and all-cause mortality (B and D) in patients aged 75 years or older and patients aged 50 to 74 years. The hazard ratios (solid lines) and 95% CIs (dotted lines) were estimated using a restricted cubic spline function with adjustment for confounders. The FBG 120 mg/dL was set as the reference (dashed lines)
All-cause mortality
During the study follow-up, all-cause death was noted in 37,062 (46.9%) patients 75 years or older and in 9205 (12.3%) patients aged 50 to 74 years (Table S5). The J-curve (Fig. 2B, 3B, and 3D) for the association of on-treatment FBG (80–99, 100–109, 110–125, 126–139, 140–159, 160–179, and 180–900 mg/dl) and all-cause mortality in patients 75 years or older (incidence: 64.9, 59.1, 57.6, 60.4, 64.0, 70.9, and 90.4 per 1000 person-years, respectively) shifted to the left relative to that in patients aged 50 to 74 years (incidence: 15.7, 13.4, 12.3, 12.2, 13.4, 15.7, and 21.8 per 1000 person-years, respectively). The interaction effect (Table S4) of FBG level and age group on all-cause mortality was statistically significant (P < 0.001). In sex-stratified analyses (Figure S1B and S2B), the J-curve was left-shifted in women versus men among patients 75 years or older (P = 0.002 for interaction), while the shift was not substantial among patients aged 50 to 74 years (P = 0.060 for interaction).
Additional analyses
In Cox models that included an interaction term of on-treatment FBG and a function of time, the interaction was not statistically significant indicating that the proportional hazard assumption was not violated severely (Table S6). In analyses including participants with and without diabetes (Table S7), the associations of untreated FBG with dementia and mortality risks were roughly linear (Figure S3), whereas the associations of on-treatment FBG with the risks were J-shaped similarly to those in the primary analysis. When baseline FBG rather than yearly-averaged levels were introduced in Cox models, the associations of baseline on-treatment FBG with the risks were similar to those in the primary analysis, although the association with all-cause mortality was somewhat attenuated (Figure S4, and Table S8).
Discussion
This cohort study in Korea for diabetic patients aged above and below 75 years assessed risks of dementia and mortality across yearly-averaged on-treatment glycemic levels. For the risk of overt dementia, J-shaped associations were observed across on-treatment FBG levels. Compared with a level of 110–125 mg/dl, the FBG levels under 100 mg/dl and those over 140 mg/dl were associated with a higher risk for dementia among patients 75 years or older, and the levels under 100 mg/dl and those over 160 mg/dl were among patients aged 50 to 74 years. For all-cause mortality, on-treatment FBG levels under 110 mg/dl and those over 126 mg/dl were associated with a higher risk among patients 75 years or older, and the levels under 110 mg/dl and those over 160 mg/dl were among 50- to 74-year-old patients. As for absolute risks of dementia and mortality, the incidence rate difference between the reference and inadequately lowered FBG was consistently greater among patients 75 years or older than among patients aged 50 to 74 years. This study provides epidemiologic evidence that appropriate glycemic control can reduce dementia risk in the elderly as well as in middle-aged patients, the risk threshold of achieved glycemic level is lower in older patients, and the absolute risk reduction associated with appropriate glycemic control is greater in the elderly than in younger patients with diabetes.
In this 9-year follow-up study, on-treatment FBG levels between 100 mg/dl and 160 mg/dl had the lowest risk for overt dementia among patients aged 50 to 74 years, while the levels between 100 mg/dl and 140 mg/dl had the lowest risk among patients 75 years or older. In the ADVANCE trial for patients with a mean age of 66 years, intensive versus standard glycemic control (mean achieved FBG, 116–129 mg/dl versus 141–147 mg/dl) did not reduce the rate of incident dementia (risk ratio [RR], 1.27; 95% CI, 0.87 to 1.85) or cognitive function decline (RR, 0.98; 95% CI, 0.89 to 1.07) [12]. Similarly, in the ACCORD trial for patients aged 45 to 79 years, there was little or no difference between intensive and standard control groups (mean achieved FBG, 117 mg/dl and 153 mg/dl, respectively) in changes in cognitive function test scores [10, 11]. Although the trials failed to show a effect of intensive glycemic control on the incidence of dementia or cognitive dysfunction, the results do not oppose the present finding observed in 50- to 74-year-old patients, and the present study provides evidence that appropriate glycemic control can reduce dementia risk in patients with advanced age as well as in middle-aged and young-old patients.
Notably, this study found the J-shaped associations of on-treatment FBG with dementia and mortality risks in comparison to the roughly linear associations of untreated FBG with the risks. Lowering of blood glucose in patients with diabetes would have the benefits of risk reduction of adverse outcomes. On the other hand, intensive lowering of blood glucose with hypoglycemic agents may increase the risk of treatment-related adverse events such as hypoglycemia leading to poor outcomes [18, 19]. As the blood glucose lowered progressively, the adverse effects of hypoglycemic treatment may increase progressively and outweigh the benefits of treatment when below a certain blood glucose threshold. In this way, excessive hypoglycemic treatment that lowers blood glucose below a certain threshold might result in an increased risk of adverse clinical outcomes.
In this study, interestingly, on-treatment FBG levels with the lowest risks for dementia and mortality were lower among patients 75 years or older compared with those among patients aged 50 to 74 years. Despite a lack of data on the appropriate glycemic target in patients with advanced age, the treatment target of blood glucose is commonly recommended to be relaxed in patients with limited life expectancy due to advanced age [20, 21]. In the ADVANCE trial that enrolled patients at high cardiovascular risk, the relative effects of intensive glycemic control on vascular events and mortality were consistent, irrespective of age subgroups (≥ 65 and < 65 years) [22]. Moreover, the ACCORD trial, another trial for high risk patients, showed that intensive versus standard glycemic control was associated with a higher mortality risk among patients younger than 65 years, whereas no such association was found among those 65 years or older [23]. The present population-based study supports the results of the trials and suggests that more or at least equally strict control of blood glucose is needed to reduce risks of dementia and mortality in older patients with diabetes.
This study has several limitations. First, FBG was only used to assess glycemic status due to the lack of data on hemoglobin A1c. Although hemoglobin A1c is widely used as a treatment target for blood glucose, it cannot represent glucose variability or predict a risk of hypoglycemia, There is a practical need to consider FBG in addition to hemoglobin A1c, and the information on appropriate FBG levels will be relevant in treatment of diabetes. Second, although this analysis would have benefited from verification of dementia diagnoses using data on the prescription of cholinesterase inhibitors or memantine, there might remain a portion of patients with undetected dementia potentially leading to misclassification bias. In addition, this study did not provide information about adverse or drug-specific effects of hypoglycemic treatment, and the analysis did not distinguish type 1 and type 2 diabetes. Moreover, caution is required when applying the findings to the population with other ethnicities or nationalities as the study included residents in Korea who were aged 50 years or older. Finally, given the observational nature of this study, there remains uncertainty about the causal relationship between glycemic control and dementia, and it may be difficult to set precise glycemic targets for the age groups. Large-scale randomized trials including older and younger patients will be required to determine glycemic targets for those conclusively.
In this cohort study in Korea, the thresholds of on-treatment FBG, above which dementia risk increased significantly, among patients 75 years or older and those aged 50 to 74 years were 140 mg/dl and 160 mg/dl, respectively. For all-cause mortality, the respective thresholds were 126 mg/dl and 160 mg/dl. Further, the absolute risk increase associated with inadequately lowered FBG was consistently greater among patients 75 years or older than among patients aged 50 to 74 years. These findings do not support the common recommendation that higher glycemic targets are needed for older patients. More studies including interventional trials are warranted to confirm the causal relationship between glycemic control and risk of dementia and to determine appropriate glycemic targets in elderly patients.
Supplementary Information
Below is the link to the electronic supplementary material.
Author contributions
Conceptualization: JHH. Data curation: JHH. Formal analysis: JHH. Methodology: JHH. Validation: JHH. Visualization: JHH. Writing: JHH.
Data availability
The datasets generated during and/or analyzed during the current study are not publicly available due to the sensitive nature of the data, but requests to access the National Health Information Database may be sent to the data provision review committee of the National Health Insurance Sharing Service (http://nhiss.nhis.or.kr).
Declarations
Conflict of interest
The author declares no conflict of interest to disclosure.
Ethical approval
All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and/or with the Helsinki Declaration of 1964 and later versions. The data used were de-identified before access. Access to the data was approved by the Institutional Review Board of Kangwon National University Hospital (IRB File No. KNHU-A-2021–04-018) on May 12, 2021. The need for informed consent was waived because all data were fully anonymized.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Nichols E, Szoeke CEI, Vollset SE, Abbasi N, Abd-Allah F, Abdela J, et al. Global, regional, and national burden of Alzheimer’s disease and other dementias, 1990–2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet Neurol. 2019;18:88–106. doi: 10.1016/S1474-4422(18)30403-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.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;4:640–650. doi: 10.1111/jdi.12087. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Chatterjee S, Peters SAE, Woodward M, Mejia Arango S, Batty GD, Beckett N, et al. Type 2 diabetes as a risk factor for dementia in women compared with men: a pooled analysis of 2.3 million people comprising more than 100,000 cases of dementia. Diabetes Care. 2016;39: 300–307. 10.2337/dc15-1588 [DOI] [PMC free article] [PubMed]
- 4.Cukierman-Yaffe T, Gerstein HC, Williamson JD, Lazar RM, Lovato L, Miller ME, et al. Relationship between baseline glycemic control and cognitive function in individuals with type 2 diabetes and other cardiovascular risk factors: the action to control cardiovascular risk in diabetes-memory in diabetes (ACCORD-MIND) trial. Diabetes Care. 2009;32:221–226. doi: 10.2337/dc08-1153. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Zheng B, Su B, Price G, Tzoulaki I, Ahmadi-Abhari S, Middleton L. Glycemic control, diabetic complications, and risk of dementia in patients with diabetes: results from a large UK Cohort Study. Diabetes Care. 2021;44:1556–1563. doi: 10.2337/dc20-2850. [DOI] [PubMed] [Google Scholar]
- 6.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;369:540–548. doi: 10.1056/NEJMoa1215740. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Rawlings AM, Sharrett AR, Schneider ALC, Coresh J, Albert M, Couper D, et al. Diabetes in midlife and cognitive change over 20 years: a cohort study. Ann Intern Med. 2014;161:785–793. doi: 10.7326/M14-0737. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Kim WJ, Lee SJ, Lee E, Lee EY, Han K. Risk of incident dementia according to glycemic status and comorbidities of hyperglycemia: a nationwide population-based cohort study. Diabetes Care. 2022;45:134–141. doi: 10.2337/dc21-0957. [DOI] [PubMed] [Google Scholar]
- 9.Areosa Sastre A, Vernooij RW, González‐Colaço Harmand M, Martínez G. Effect of the treatment of Type 2 diabetes mellitus on the development of cognitive impairment and dementia. Cochrane Database Syst Rev. 2017;2017: 003804. 10.1002/14651858.CD003804.pub2 [DOI] [PMC free article] [PubMed]
- 10.Launer LJ, Miller ME, Williamson JD, Lazar RM, Gerstein HC, Murray AM, et al. Effects of intensive glucose lowering on brain structure and function in people with type 2 diabetes (ACCORD MIND): a randomised open-label substudy. Lancet Neurol. 2011;10:969–977. doi: 10.1016/S1474-4422(11)70188-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Murray AM, Hsu F-C, Williamson JD, Bryan RN, Gerstein HC, Sullivan MD, et al. ACCORDION MIND: results of the observational extension of the ACCORD MIND randomised trial. Diabetologia. 2017;60:69–80. doi: 10.1007/s00125-016-4118-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.ADVANCE Collaborative Group, Patel A, MacMahon S, Chalmers J, Neal B, Billot L, et al. Intensive blood glucose control and vascular outcomes in patients with type 2 diabetes. N Engl J Med. 2008;358: 2560–2572. doi:10.1056/NEJMoa0802987 [DOI] [PubMed]
- 13.Jung HH, Lee S. Optimal fasting glucose levels with regard to cardiovascular and mortality outcomes in people treated with or without antidiabetic medication. Diabetes Obes Metab. 2020;22:243–253. doi: 10.1111/dom.13892. [DOI] [PubMed] [Google Scholar]
- 14.Cheol Seong S, Kim Y-Y, Khang Y-H, Heon Park J, Kang H-J, Lee H, et al. Data Resource Profile: The National Health Information Database of the National Health Insurance Service in South Korea. Int J Epidemiol. 2017;46:799–800. doi: 10.1093/ije/dyw253. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Currie CJ, Peters JR, Tynan A, Evans M, Heine RJ, Bracco OL, et al. Survival as a function of HbA(1c) in people with type 2 diabetes: a retrospective cohort study. Lancet. 2010;375:481–489. doi: 10.1016/S0140-6736(09)61969-3. [DOI] [PubMed] [Google Scholar]
- 16.Emerging Risk Factors Collaboration, Sarwar N, Gao P, Seshasai SRK, Gobin R, Kaptoge S, et al. Diabetes mellitus, fasting blood glucose concentration, and risk of vascular disease: a collaborative meta-analysis of 102 prospective studies. Lancet. 2010;375: 2215–2222. doi:10.1016/S0140-6736(10)60484-9 [DOI] [PMC free article] [PubMed]
- 17.Organization WH, Others. International Diabetes Federation (2006) Definition and diagnosis of diabetes mellitus and intermediate hyperglycemia: report of a WHO/IDF consultation. IDF Consult. 2008.
- 18.Zoungas S, Patel A, Chalmers J, de Galan BE, Li Q, Billot L, et al. Severe Hypoglycemia and Risks of Vascular Events and Death. N Engl J Med. 2010;363:1410–1418. doi: 10.1056/NEJMoa1003795. [DOI] [PubMed] [Google Scholar]
- 19.Whitmer RA, Karter AJ, Yaffe K, Quesenberry CP, Selby JV. Hypoglycemic episodes and risk of dementia in older patients with type 2 diabetes mellitus. JAMA. 2009;301:1565–1572. doi: 10.1001/jama.2009.460. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Qaseem A, Wilt TJ, Kansagara D, Horwitch C, Barry MJ, Forciea MA, et al. Hemoglobin A1c Targets for Glycemic Control With Pharmacologic Therapy for Nonpregnant Adults With Type 2 Diabetes Mellitus: A Guidance Statement Update From the American College of Physicians. Ann Intern Med. 2018;168:569–576. doi: 10.7326/M17-0939. [DOI] [PubMed] [Google Scholar]
- 21.Association AD. 6. Glycemic Targets: Standards of Medical Care in Diabetes—2021. Diabetes Care. 2021;44: S73–S84. doi:10.2337/dc21-S006 [DOI] [PubMed]
- 22.Ohkuma T, Chalmers J, Cooper M, Hamet P, Harrap S, Marre M, et al. The comparative effects of intensive glucose lowering in diabetes patients aged below or above 65 years: Results from the ADVANCE trial. Diabetes Obes Metab. 2021;23:1292–1300. doi: 10.1111/dom.14339. [DOI] [PubMed] [Google Scholar]
- 23.Miller ME, Williamson JD, Gerstein HC, Byington RP, Cushman WC, Ginsberg HN, et al. Effects of randomization to intensive glucose control on adverse events, cardiovascular disease, and mortality in older versus younger adults in the ACCORD Trial. Diabetes Care. 2014;37:634–643. doi: 10.2337/dc13-1545. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The datasets generated during and/or analyzed during the current study are not publicly available due to the sensitive nature of the data, but requests to access the National Health Information Database may be sent to the data provision review committee of the National Health Insurance Sharing Service (http://nhiss.nhis.or.kr).