Physical frailty is well known to be strongly associated with malnutrition, but the combined impact of physical frailty and cognitive impairment among non-demented older persons (cognitive frailty) on malnutrition prevalence is not well documented.

Cross-sectional cohort study.

Community-dwelling older Singaporeans aged ≥55y (n=5414) without dementia in the Singapore Longitudinal Ageing Study (SLAS-1 and SLAS-2).

The Mini Nutritional Assessment–short form (MNA-SF) and Nutrition Screening Initiative (NSI) Determine Checklist were used to determine their nutritional status. Participants were categorized as cognitive normal (CN) or cognitive impaired (CI) by Mini Mental State Examination (MMSE<=23), as pre-frail (PF) (score=1-2) or frail (F) (score=3-5) using Fried’s criteria, and as cognitive pre-frail (PF+CI) or cognitive frail (F+CI).

The prevalence of cognitive frailty was 1.6%, and cognitive pre-frailty was 5.5% (total, 7.1%). The prevalence of MNA malnutrition was 2.4%, and NSI high nutritional risk was 6.3%. The prevalence of MNA malnutrition was lowest among Robust-CN and highest among Frail-CI (0.5% in Robust- CN, 0.6% in Robust-CI, 2.8% in Pre-frail-CN, 7.3% in Prefrail- CI, 15.4% in Frail-CN, and 23.1% in Frail-CI). Similarly, the prevalence of NSI high nutritional risk was lowest in Robust-CN (3.7%) and highest in Frail-CI (13.6%). Adjusted for sociodemographic and health status, pre-frailty/frailty-CI versus Robust-CN was associated with the highest odds ratio of association with MNA malnutrition (OR=8.16, p<0.001), although not the highest with NSI high nutritional risk (OR=1.48, p=0.017).

An extraordinary high prevalence of malnutrition was observed among older adults with cognitive frailty who should be specially targeted for active intervention.