Related Experiment Video
Updated: May 16, 2026
A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment
Published on: January 11, 2020
A sensorimotor function construct to assess mild cognitive impairment risk
Amal A Wanigatunga1,2, Hang Wang1,2, Di Huang1
1Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland, USA.
Introduction:
Identifying precursors to mild cognitive impairment (MCI)-the transitional stage between unimpaired cognition and dementia-is a public health priority. While sensory and motor impairments are each linked to MCI, their integrated contribution as a sensorimotor construct remains underexplored.
Methods:
We analyzed cross-sectional data from two US cohorts: Atherosclerosis Risk in Communities (ARIC; n = 880) and Baltimore Longitudinal Study of Aging (BLSA; n = 681), excluding individuals with stroke, Parkinson's disease, or dementia. A composite sensorimotor score-based on hearing, vision, olfaction, balance, gait speed, and grip strength-was created using factor analysis. Structural equation modeling assessed associations with MCI, adjusting for key covariates.
Results:
Higher sensorimotor function was associated with lower odds of MCI: ARIC odds ratio (OR) = 0.53 (95% confidence interval [CI]: 0.40-0.71); BLSA OR = 0.59 (95% CI: 0.43-0.81).
Discussion:
Sensorimotor function appears robustly related with MCI in a large sample of older adults. These findings highlight the potential value of incorporating sensorimotor assessments in early detection for cognitive decline. Longitudinal research is needed to understand temporality.
