You are here
Home > Blog > Endocrinology > Antidiabetic Medications And Dementia Risk

Antidiabetic Medications And Dementia Risk

Antidiabetic Medications And Dementia Risk


This umbrella review and meta-analysis aimed to assess the impact of diabetes on dementia risk and investigate the potential mitigating effects of various antidiabetic treatments. The study conducted a systematic review on diabetes and its treatment, along with a meta-analysis specifically focusing on treatment outcomes.

A comprehensive search of databases, including MEDLINE/PubMed, Embase, PsycINFO, CINAHL, and the Cochrane Library, was carried out up to July 2, 2023. The analysis involved 100 reviews and 27 cohort/case-control studies, encompassing a total of 3,046,661 participants.

Key findings revealed that certain antidiabetic medications, namely metformin, thiazolidinediones, pioglitazone, glucagon-like peptide-1 receptor agonists (GLP1RAs), and sodium-glucose cotransporter-2 inhibitors (SGLT2is), were associated with a significant reduction in the risk of dementia. Interestingly, the positive effect of metformin was particularly significant in Western populations, specifically the United States, while no significant impact was observed in Eastern populations.

Conversely, dipeptidyl peptidase-4 inhibitors, α-glucosidase inhibitors, and insulin did not exhibit a significant effect on dementia risk. Additionally, meglitinides and sulphonylureas were associated with an increased risk of dementia.

The study underscores the importance of more extensive longitudinal investigations to discern the relative benefits of these antidiabetic medications in diverse populations. The findings contribute valuable insights into potential therapeutic avenues for reducing dementia risk in individuals with diabetes, emphasizing the need for targeted research to tailor treatment strategies to specific demographic and geographical contexts.


Diabetes, currently the seventh leading global cause of disability, has witnessed a drastic surge, affecting over 450 million adults worldwide, with Type 2 diabetes constituting the majority of cases. This form is characterized by insulin resistance and impaired insulin secretion, leading to hyperglycemia and hyperinsulinemia. Meanwhile, Type 1 diabetes results from autoimmune destruction of pancreatic β-cells, necessitating insulin treatment for survival. The systemic impact of diabetes extends to macrovascular complications such as cardiovascular disease and cerebrovascular disease, as well as microvascular complications including retinopathy, nephropathy, and neuropathy.

Dementia, a prevalent and debilitating disease, sees around 4.6 million new cases annually globally, with estimates anticipating a rise to 75.6 million in 2030 and 135.5 million in 2050. Alzheimer’s disease (AD), the primary form of dementia, involves irreversible neuronal degeneration, manifesting as significant cognitive impairments, notably memory loss, physical disability, and behavioral alterations. The pathological hallmarks of AD include neuronal loss, the formation of amyloid β deposits, and neurofibrillary tangles of tau proteins. 

The interplay between diabetes and dementia is evident, with insulin resistance in the brain coined as ‘type 3 diabetes,’ a key factor in Alzheimer’s pathophysiology. Despite decades of recognition of diabetes as a risk factor for cognitive decline and dementia, the specific impact on dementia subtypes (especially AD and VaD) and the influence of treatment remain contentious. The study aims to address these gaps through a systematic umbrella review and an updated meta-analysis, scrutinizing the evidence linking diabetes and its treatments to the risk of dementia. This research is crucial for advancing our understanding of the intricate relationship between diabetes and dementia, contributing to more effective prevention and treatment strategies.


In this comprehensive systematic review and meta-analysis, researchers conducted an exhaustive search of multiple databases for systematic reviews and meta-analyses, as well as cohort and case-control studies related to diabetes and its treatment’s impact on the risk of dementia. The inclusion criteria involved rigorous selection of studies up to July 2, 2023, focusing on incident dementia as a clinically significant treatment outcome. The study design followed a pre-specified protocol registered in PROSPERO (CRD42021267541).

For the meta-analysis, data extraction involved aggregate information from published reports, prioritizing the largest sample size or the most recent publication in cases of duplicate data. The primary outcome examined was dementia, with secondary outcomes exploring changes in cognition. Various study characteristics were meticulously extracted, encompassing population demographics, treatment specifics, outcome assessments, and key results.

The AMSTAR tool gauged the quality of systematic reviews and meta-analyses, ensuring a rigorous evaluation of study credibility, while the Newcastle-Ottawa Scale assessed the quality of cohort and case-control studies. A meticulous approach to data analysis included meta-regression analyses and subgroup analyses to explore sources of heterogeneity and potential biases.

The meta-analysis delved into ten distinct drug classes, including metformin, thiazolidinediones, dipeptidyl peptidase-4 inhibitors, α-glucosidase inhibitors, meglitinides, insulin, sulphonylureas, glucagon-like peptide-1 receptor agonists, and sodium-glucose cotransporter-2 inhibitors. Risk ratios and 95% confidence intervals were calculated, utilizing a random-effects model.


Sensitivity analyses were executed for specific drug classes, and potential small study effects and publication bias were assessed visually and quantitatively. Robust statistical methodologies, including meta-regression and subgroup analyses, were applied to unveil heterogeneity sources, ensuring a comprehensive and nuanced examination of the data.

This meticulous approach to the systematic review and meta-analysis adheres to pre-established protocols, ensuring a high standard of methodological rigor and transparency throughout the study.


The study conducted a thorough review, identifying 1264 unique reviews, from which 100 met eligibility criteria for the umbrella review. Additionally, 685 unique studies were identified for the meta-analysis, with eight studies meeting inclusion criteria and 19 studies from previous reviews included, amounting to a total sample of 3,046,661 participants.

The umbrella review encompassed diverse aspects, with 32 reviews exploring the link between diabetes and cognitive function in mixed diabetic populations, seven focusing on type 1 diabetes, 19 on type 2 diabetes, 14 examining brain structure in diabetes, two investigating genetic and biomarkers of dementia, and 26 concentrating on the cognitive effects of antidiabetic treatment.

The meta-analysis, including 27 studies, provided a nuanced understanding of various antidiabetic treatments’ impact on dementia risk. Notably, metformin, thiazolidinediones, pioglitazone, GLP1RAs, and SGLT2is exhibited a significant reduction in dementia risk. Metformin’s effect varied across countries, with a significant impact in the United States and Western populations.

Findings indicated that antidiabetic medications, in general, reduced the risk of cognitive impairment associated with diabetes. Specific medications, such as metformin and DPP-4is, showed consistent risk reduction in dementia across multiple analyses. Notably, GLP1RAs and SGLT2is demonstrated substantial protective effects.

Thorough analyses, including meta-regression, were conducted to explore sources of heterogeneity, with follow-up duration and sex emerging as significant factors in the metformin meta-analysis. The study employed rigorous quality assessment tools, ensuring robust evaluation of systematic reviews and meta-analyses.

Despite notable strengths, potential publication bias was indicated, prompting further scrutiny. Sensitivity analyses reinforced the robustness of the metformin effect. However, both thiazolidinedione and sulphonylurea meta-analyses lost significance in sensitivity analyses.

In conclusion, this study offers a comprehensive exploration of the intricate relationship between diabetes, its treatments, and the risk of dementia. The findings underscore the potential protective effects of specific antidiabetic medications, paving the way for nuanced discussions on the management of diabetes in relation to cognitive health.


The study provides a comprehensive qualitative synthesis of 100 systematic reviews and meta-analyses exploring the intricate relationship between diabetes and cognitive decline/dementia. Additionally, it includes an updated meta-analysis investigating the impact of nine antidiabetic drug classes on dementia risk.

The meta-analyses reveal a protective effect of several antidiabetic medications on dementia risk. Metformin, thiazolidinediones, pioglitazone, GLP1RAs, and SGLT2is demonstrated a significant reduction in dementia risk, while DPP-4is, α-glucosidase inhibitors, and insulin showed a neutral effect. 

Notably, metformin exhibited varied effects based on geographical regions, showing a protective effect in the United States and Western populations. This aligns with previous findings highlighting regional differences in the relative risk of dementia among individuals with diabetes.

The study sheds light on the structural brain impairments observed in individuals with diabetes and emphasizes the association between diabetes and an increased risk of both Alzheimer’s disease (AD) and vascular dementia (VaD). Thorough analyses of antidiabetic treatments reveal that thiazolidinediones, metformin, and DPP-4is have been linked to improved cognitive function, while insulin was associated with cognitive deficits.

Despite the strengths of the study, such as its comprehensive nature and large sample size, there are limitations, including potential heterogeneity in diabetes and dementia diagnoses among studies. The study calls for more rigorous investigations, emphasizing the need for randomized controlled trials (RCTs) with dementia as a prespecified outcome to strengthen the evidence.

In conclusion, the research underscores the significance of evaluating antidiabetic treatments not only for glycemic control but also for their potential impact on preserving cognitive function. The findings suggest that metformin, a common first-line medication for type 2 diabetes, exhibits benefit in reducing dementia risk, while highlighting the importance of exploring newer treatments for their potential cognitive benefits. The study advocates for continued research to better understand the relationship between antidiabetic treatments and dementia risk, emphasizing the need for prospective trials and mechanistic studies.

Oncology Related Tools


Latest Research


About Author

Similar Articles

Leave a Reply