Transition Across Cognitive States and Death Among Older Adults With Type 2 Diabetes: Determining the Predictors of Cognitive Decline With a Multistate Markov Model
Author(s)
Wen Ye, PhD1, Yanbing Zhong, BS2, Jing Li, MS2, Deborah Levine, MD, MPH2.
1Research Scientist, University of Michigan, Ann Arbor, MI, USA, 2University of Michigan, Ann Arbor, MI, USA.
1Research Scientist, University of Michigan, Ann Arbor, MI, USA, 2University of Michigan, Ann Arbor, MI, USA.
OBJECTIVES: Type 2 diabetes (T2D) is the most prevalent co-existing comorbidity (47%) in people with dementia. Even at midlife, individuals with T2D have twice the risk of developing dementia than those without T2D. Here we explore a set of potential variables associated with progression of cognitive decline in T2D patients.
METHODS: We modeled cognitive decline in T2D patients using a multi-state Markov model (MSMM) to investigate the significance of potential risk factors, including sex, race, age and duration of T2D, physical examination variables, laboratory variables, smoking status, and medical history. The MSMM include four transient states: normal cognition (NC), mild cognitive impairment (MCI), dementia, and death. We compared the Akaike information criterion (AIC) of all possible models and selected the one with the smallest AIC.
RESULTS: A total of 2137 Atherosclerosis Risk in Communities (ARIC) study participants with T2D (mean [SD] age: 76 [6]) were followed for a median 5.7 years. At baseline, 1451 (67.9%), 504 (23.6%), 185 (8.5 %) had NC, MCI, and dementia. Among modifiable risk factors, history of heart failure (HF) (hazard ratio [HR] 2.6 [95% CI 1.7-3.8]), history of stroke (HR 1.3; 95% CI 0.9-1.8), and cigarette smoking (HR 1.7; 95% CI 1.2-2.4) were associated with a higher risk of transitioning from NC to MCI. History of HF (HR 1.6; 95% CI 1.1-2.5) was associated with a higher risk of transitioning from MCI to dementia. Lower urinary albumin:creatinine ratio level (HR 0.69; 95% CI 0.55-0.87 per 10 fold decrease) and HF history (HR 2.0; 95% CI 1.2-3.3) were associated with higher probability of the reverse transitioning from MCI to NC.
CONCLUSIONS: In this population-based study, pre-existing vascular disease played a critical role for progression to MCI and dementia in older adults with T2D.
METHODS: We modeled cognitive decline in T2D patients using a multi-state Markov model (MSMM) to investigate the significance of potential risk factors, including sex, race, age and duration of T2D, physical examination variables, laboratory variables, smoking status, and medical history. The MSMM include four transient states: normal cognition (NC), mild cognitive impairment (MCI), dementia, and death. We compared the Akaike information criterion (AIC) of all possible models and selected the one with the smallest AIC.
RESULTS: A total of 2137 Atherosclerosis Risk in Communities (ARIC) study participants with T2D (mean [SD] age: 76 [6]) were followed for a median 5.7 years. At baseline, 1451 (67.9%), 504 (23.6%), 185 (8.5 %) had NC, MCI, and dementia. Among modifiable risk factors, history of heart failure (HF) (hazard ratio [HR] 2.6 [95% CI 1.7-3.8]), history of stroke (HR 1.3; 95% CI 0.9-1.8), and cigarette smoking (HR 1.7; 95% CI 1.2-2.4) were associated with a higher risk of transitioning from NC to MCI. History of HF (HR 1.6; 95% CI 1.1-2.5) was associated with a higher risk of transitioning from MCI to dementia. Lower urinary albumin:creatinine ratio level (HR 0.69; 95% CI 0.55-0.87 per 10 fold decrease) and HF history (HR 2.0; 95% CI 1.2-3.3) were associated with higher probability of the reverse transitioning from MCI to NC.
CONCLUSIONS: In this population-based study, pre-existing vascular disease played a critical role for progression to MCI and dementia in older adults with T2D.
Conference/Value in Health Info
2025-11, ISPOR Europe 2025, Glasgow, Scotland
Value in Health, Volume 28, Issue S2
Code
EPH257
Topic
Epidemiology & Public Health
Disease
Diabetes/Endocrine/Metabolic Disorders (including obesity), Neurological Disorders