Cardiometabolic factors and risk of dementia in people with type 2 diabetes in England: a large retrospective cohort study
File(s)
Published version
Author(s)
Type
Journal Article
Abstract
Introduction: People with type 2 diabetes (T2D) have a higher dementia risk. Cardiometabolic factors may influence dementia development, but long-term links among people with T2D remain unclear. Prior studies show mixed findings and have not considered how the time before dementia diagnosis at which cardiometabolic factors are measured affects their association with dementia risk.
Methods: We examined the associations between eight routinely measured cardiometabolic factors (including blood pressure, lipid and glycaemic markers) and the risk of dementia in 249,958 adults with T2D from the Clinical Practice Research Datalink (1999-2018). Multivariable Cox proportional hazards regression was applied with age as the timescale and stratified by follow-up time at 5, 5-10 and ≥10 years before dementia diagnosis. Cardiometabolic factors were analysed continuously and categorically using clinically relevant cut-offs.
Results: 22,492 incident dementia cases were observed (1,257,998 person-years of follow up). Higher systolic blood pressure was associated with a lower dementia risk, particularly when measured closer to the time of dementia diagnosis. Higher diastolic blood pressure
(DBP) (90 vs. 80 mmHg) was associated with a 21% greater risk of dementia (95% CI: 15-27%). Dementia risk was 32% (27-36%) lower for those with body mass index (BMI) ≥30 (vs. <25 kg/m2) when measured within five years of dementia diagnosis, but 15% (7-23%)
and 43% (31-56%) higher if measured 5-10 and ≥10 years before diagnosis, respectively. Higher levels of glycaemic measures were consistently associated with a greater dementia risk. Findings for total cholesterol and low-density lipoprotein were largely null.
Conclusions: Higher DBP, BMI, and glycaemic measures measured over 10 years before dementia diagnosis were linked to a greater dementia risk among adults with T2D.
Associations varied depending on when cardiometabolic factors were measured relative to dementia diagnosis, demonstrating the importance of distinguishing potential long-term risk factors from short-term markers of dementia development.
Methods: We examined the associations between eight routinely measured cardiometabolic factors (including blood pressure, lipid and glycaemic markers) and the risk of dementia in 249,958 adults with T2D from the Clinical Practice Research Datalink (1999-2018). Multivariable Cox proportional hazards regression was applied with age as the timescale and stratified by follow-up time at 5, 5-10 and ≥10 years before dementia diagnosis. Cardiometabolic factors were analysed continuously and categorically using clinically relevant cut-offs.
Results: 22,492 incident dementia cases were observed (1,257,998 person-years of follow up). Higher systolic blood pressure was associated with a lower dementia risk, particularly when measured closer to the time of dementia diagnosis. Higher diastolic blood pressure
(DBP) (90 vs. 80 mmHg) was associated with a 21% greater risk of dementia (95% CI: 15-27%). Dementia risk was 32% (27-36%) lower for those with body mass index (BMI) ≥30 (vs. <25 kg/m2) when measured within five years of dementia diagnosis, but 15% (7-23%)
and 43% (31-56%) higher if measured 5-10 and ≥10 years before diagnosis, respectively. Higher levels of glycaemic measures were consistently associated with a greater dementia risk. Findings for total cholesterol and low-density lipoprotein were largely null.
Conclusions: Higher DBP, BMI, and glycaemic measures measured over 10 years before dementia diagnosis were linked to a greater dementia risk among adults with T2D.
Associations varied depending on when cardiometabolic factors were measured relative to dementia diagnosis, demonstrating the importance of distinguishing potential long-term risk factors from short-term markers of dementia development.
Date Issued
2026-07-01
Date Acceptance
2026-08-02
Citation
BMJ Open Diabetes Research & Care, 2026, 14 (4)
ISSN
2052-4897
Publisher
BMJ Publishing Group
Journal / Book Title
BMJ Open Diabetes Research & Care
Volume
14
Issue
4
Copyright Statement
© Author(s) (or their employer(s)) 2026. Re-use permitted under CC BY. Published by BMJ Group.
License URL
Identifier
10.1136/bmjdrc-2025-005743
Publication Status
Published
Article Number
e005743
Date Publish Online
2026-08-17