Physical activity complexity, cognition, and risk of cognitive impairment and dementia in the Baltimore Longitudinal Study of Aging

Yurun Cai; Junhong Zhou; Paul W Scott; Qu Tian; Amal A Wanigatunga; Lewis Lipsitz; Eleanor M Simonsick; Susan M Resnick; Luigi Ferrucci; Dianxu Ren; Jennifer H Lingler; Jennifer A Schrack

doi:10.1002/trc2.70077

Physical activity complexity, cognition, and risk of cognitive impairment and dementia in the Baltimore Longitudinal Study of Aging

Alzheimers Dement (N Y). 2025 Apr 9;11(2):e70077. doi: 10.1002/trc2.70077. eCollection 2025 Apr-Jun.

Authors

Yurun Cai¹, Junhong Zhou², Paul W Scott¹, Qu Tian³, Amal A Wanigatunga^{4

5}, Lewis Lipsitz², Eleanor M Simonsick³, Susan M Resnick³, Luigi Ferrucci³, Dianxu Ren^{1

6}, Jennifer H Lingler^{1

6}, Jennifer A Schrack^{4

5}

Affiliations

¹ Department of Health and Community Systems University of Pittsburgh School of Nursing Pittsburgh Pennsylvania USA.
² Hinda and Arthur Marcus Institute for Aging Research Harvard Medical School Boston Massachusetts USA.
³ Intramural Research Program National Institute on Aging Baltimore Maryland USA.
⁴ Department of Epidemiology Johns Hopkins Bloomberg School of Public Health Baltimore Maryland USA.
⁵ Center on Aging and Health Johns Hopkins University Baltimore Maryland USA.
⁶ University of Pittsburgh Alzheimer's Disease Research Center Pittsburgh Pennsylvania USA.

Abstract

Introduction: Studies on physical activity (PA) and dementia mainly focus on activity quantity or intensity. Yet PA requires neuro-coordination of movement, and it is unclear whether complexity of daily activity varies by cognitive status. Thus, we examined the association between PA complexity, using multiscale entropy, and cognitive function, mild cognitive impairment (MCI), and dementia in older adults in the Baltimore Longitudinal Study of Aging (BLSA).

Methods: A total of 637 older adults (age 73.9 ± 11.3 years) in the BLSA completed a 7-day wrist-worn accelerometer assessment and neuropsychological tests from 2015 to 2020. Using logistic regression and structural equation modeling, we examined cross-sectional associations of PA complexity with MCI/dementia and cognition. Cross-lagged panel models (CLPMs) were used to assess bidirectional associations at baseline and 2-year follow-up. Multivariable models were adjusted for age, sex, race, education years, body mass index, and comorbidities.

Results: Participants in the lowest tertile of PA complexity had over double the odds of MCI/dementia (odds ratio = 2.63, 95% confidence interval [CI]: 1.02 to 6.79, p = 0.045) compared to those in the highest tertile in the fully adjusted model. Structural equation modeling showed that PA complexity was associated with global cognitive function (standardized B [SB] = 0.102, 95% CI: 0.033 to 0.171, p = 0.004), executive function (SB = 0.119, 95% CI: 0.049 to 0.189, p = 0.001), and visuospatial ability (SB = 0.096, 95% CI: 0.026 to 0.167, p = 0.008). CLPMs showed bidirectional associations between lower PA complexity and poorer executive function.

Discussion: Lower complexity of accelerometry-detected movement is associated with poorer cognition and higher risk of MCI/dementia. Future studies should explore whether low PA complexity is an early indicator of dementia.

Highlights: Prior studies mainly focused on quantity or intensity of physical activity.Poorer cognitive function was associated with lower complexity of daily activity.Lower complexity of physical activity may be an early indicator of dementia.

Keywords: cognitive function; cognitive impairment; dementia; physical activity; structural equation modeling.