Dietary and Serum Biomarkers, and Glycaemic status: Relationships with Type 2 Diabetes Mellitus and Incident Dementia

Background: Dementia is a well-recognized complication of T2D, with both conditions sharing pathophysiological pathways. However, the role of T2D in mediating the development of dementia remains underexplored. Aims: The three aims of the study were (1) to investigate associations between serum biomarkers and glycaemic status in older Australians using traditional and machine learning methods (2) to examine how daily dietary intake, serum biomarkers, and glycaemic status predict dementia progression over 14 years and (3) to assess whether diabetes status moderates or mediates the relationship between biomarkers, dietary quality, dietary diversity, and dementia progression. Methods: Data were drawn from the Sydney Memory and Ageing Study, involving 1037 community-dwelling participants aged 70–90 years. Seven waves of dementia status data, serum biomarkers, dietary intake, diet quality, and diet diversity scores were analysed. Descriptive statistics, regression models, survival analysis, and machine learning methods were applied. Multinomial logistic regression assessed biomarker-glycaemic status associations, while elastic net regression selected key predictors; and cox proportional hazards models examined dementia risk over time. Mediation and moderation analyses explored the role of T2D in biomarker and dietary relationships with dementia. Results: Malondialdehyde showed the strongest association with T2D, surpassing insulin, triglycerides, and waist-to-hip ratio. T2D was associated with lower vitamin E and higher macrophage inhibiting cytokine- 1 . Age, apolipoprotein ε4 carrier status, macrophage inhibiting cytokine- 1, and homocysteine showed positive associations with dementia risk, while urate showed a negative association. CoxBoost model identified fructosamine, MIC-1, urate, plasminogen activator inhibitor-1, and vitamin E as key dementia predictors. T2D moderated associations between dietary factors and dementia, though no mediation effects were found. Conclusion: Oxidative stress and inflammatory markers like malondialdehyde and MIC-1 are linked to T2D and dementia risk. T2D moderates the relationship between dietary factors and dementia, suggesting potential dietary interventions, without any mediation effect. Further research is needed to explore these findings in diverse populations.

Grantor dc:publisher

UNSW, Sydney

Year dc:date

2026

Author dc:creator

• Krishnamurthy, Madhu

• Nutrition
• Diabetes
• Dementia
• Biomarkers
• anzsrc-for: 32 BIOMEDICAL AND CLINICAL SCIENCES

Statement dc:rights

• open access
• CC BY 4.0
• free_to_read

Licence

• https://purl.org/coar/access_right/c_abf2
• https://creativecommons.org/licenses/by/4.0/

Language dc:language

en

Identifier

https://doi.org/10.26190/unsworks/32199

OAI identifier oai:identifier

oai:unsworks.library.unsw.edu.au:1959.4/107507