Systematic analysis and prediction of the burden of Alzheimer’s disease and other dementias caused by hyperglycemia

Daqing He, Manting Liu, Yujin Tang, Xiaona Tian, Lisi Zhou, Yizhen Chen, Xiaoxia LiuFrontiers in Public Health 2025 · DOI

TL;DR

This study reveals that high fasting plasma glucose (HFPG) is a rapidly growing driver of Alzheimer’s disease and other dementias (ADOD), with related disability-adjusted life years (DALYs) skyrocketing by 271% since 1990. Projections suggest that by 2050, global cases of hyperglycemia-induced dementia will exceed one million, highlighting a critical need for metabolic health interventions to ensure healthy aging.

What Problem Does This Paper Address?

Dementia is currently the fifth leading cause of death globally, yet its relationship with metabolic health—specifically hyperglycemia—is often under-analyzed at a global scale. While we know that diabetes significantly increases dementia risk (sometimes referred to as "Type 3 diabetes"), we lack a granular, longitudinal understanding of how this specific burden varies across different regions and age groups over time.

The authors aim to fill this gap by analyzing 30 years of global data to determine how much of the dementia burden is attributable to high blood sugar, what demographic factors (age, period, or birth cohort) are driving these trends, and where we are headed by the year 2050.

Methods

The researchers utilized data from the Global Burden of Disease (GBD) 2021 study, covering 204 countries. Their analytical toolkit included:

  • Age-Period-Cohort (APC) Framework: This allows researchers to distinguish whether rising disease rates are due to people simply getting older (age), changes in medical technology or environment over time (period), or unique risks faced by specific generations (cohort).
  • BAPC (Bayesian Age-Period-Cohort) Model: Used for forward-looking projections to 2050.
  • Frontier Analysis: A method to identify "efficiency gaps"—the difference between a country's actual disease burden and the lowest possible burden achieved by other countries at a similar Socio-demographic Index (SDI).
  • Shiny Platform: The team developed an interactive web tool to visualize these complex datasets for policymakers and researchers.

Datasets

The primary dataset is the GBD 2021, which provides estimates for mortality, prevalence, and DALYs. The study also utilizes the Socio-demographic Index (SDI) to categorize regions based on income, education, and fertility rates.

Key Insights and Ideas

  • The "High SDI" Paradox: Counterintuitively, high-development regions are not escaping this trend. While medical care is better, these regions often face higher sugar intake and lower physical activity, which offsets healthcare improvements. In the last five years, high SDI regions showed the highest age-standardized mortality rates.
  • Aging vs. Epidemiology: In low-income regions, the rise in dementia is driven primarily by general population growth. However, in middle and high-income regions, aging and epidemiological changes (lifestyle-related risks) are the dominant drivers.
  • The Efficiency Gap: Frontier analysis showed that countries like Morocco and Qatar have significantly higher burdens than their peers at similar development levels, suggesting that socio-economic status alone doesn't dictate metabolic health outcomes.
  • Gender Disparity in Projections: By 2050, females are projected to bear a significantly higher burden, with nearly double the number of cases compared to males (approx. 657k vs 345k).

Key Results

  • Attributable Burden: In 2021, HFPG was responsible for approximately 15% of all ADOD deaths and DALYs.
  • Historical Surge: Between 1990 and 2021, the number of DALYs attributed to HFPG increased by 271.05%, and the mortality rate rose by 305.81%.
  • Age Shift: The primary burden has shifted toward older age groups, specifically the 85–89 age group, compared to the 80–84 group in 1990.
  • 2050 Projections: The global prevalence of HFPG-related dementia is projected to reach 1,003,018 cases by 2050.
  • Incidence: The incidence rate per 100,000 rose from 47.07 to 66.42 over the study period, indicating that current blood sugar control strategies are failing to keep pace with the dementia epidemic.

Limitations and Future Work

The authors acknowledge several limitations:

  • Data Quality: GBD data relies on the reporting accuracy of individual countries; underreporting in low-income regions may skew results.
  • Ecological Fallacy: As a population-level study, it cannot definitively prove that high blood sugar caused dementia in specific individuals, only that the two are strongly correlated at scale.
  • Risk Factor Complexity: While HFPG is a major factor, it often co-occurs with obesity and hypertension, which are difficult to disentangle completely.

Future Work: The authors advocate for region-specific policies—such as sugar taxes and exercise promotion in high-income areas, and building primary healthcare infrastructure in low-income areas—to mitigate the projected 2050 surge.

Summary updated

Abstract

The global burden of HFPG-related ADOD has been increasing, even in high SDI regions, over the past 5 years and is expected to continue rising until 2050. Implementing appropriate health policies to mitigate this trend could significantly reduce the substantial burden caused by HFPG-induced ADOD. Promoting the use of the Shiny prediction platform will contribute significantly to global healthy aging.
BibTeX
@inproceedings{he2025systematic,
  author = {Daqing He and Manting Liu and Yujin Tang and Xiaona Tian and Lisi Zhou and Yizhen Chen and Xiaoxia Liu},
  title = {{Systematic analysis and prediction of the burden of Alzheimer’s disease and other dementias caused by hyperglycemia}},
  booktitle = {Frontiers in Public Health},
  year = {2025},
  doi = {10.3389/fpubh.2024.1516267},
}
Figure