Self-Reported Cognitive Frailty Predicts Adverse Health Outcomes for Community-Dwelling Older Adults Based on an Analysis of Sex and Age

J Nutr Health Aging. 2019;23(7):654-664. doi: 10.1007/s12603-019-1217-7.


Objectives: The present study examined whether the combination of self-reported mobility decline (SR-MD) and cognitive decline (SR-CD) was associated with mortality and new long-term care insurance (LTCI) service certifications based on sex and age.

Design: A prospective cohort study.

Setting and participants: We analyzed cohort data from a sample of older adult residents in Kami Town, Japan. The response rate was 94.3%, and we followed 5,094 older adults for 3 years. Full analyses were conducted on 5,076 participants.

Measures: A total of four groups were determined through self-reported responses on the Kihon Checklist for SR-MD (a score of 3 or more on 5 items) and SR-CD (a score of 1 or more on 3 items): non-SR-cognitive frailty, non-SR-MD and SR-CD, SR-MD and non-SR-CD, and SR-cognitive frailty.

Results: Main outcomes included mortality (n = 262) or new certifications for LTCI services (n = 708) during the 3-year period. Excluding overlapping, this included 845 older adults (16.6%). Among men, prevalence of non-SR-cognitive frailty, non-SR-MD and SR-CD, SR-MD and non-SR-CD, and SR-cognitive frailty (SR-MD and SR-CD) was 48.2%, 26.4%, 11.5%, and 13.8%, respectively. Respective rates for women were 45.7%, 15.5%, 23.1%, and 15.7%. Multivariate analyses revealed that for men, SR-MD and non-SR-CD significantly affected adverse health outcomes, leading to earlier negative outcomes relative to the non-SR-MD and SR-CD group. For women, non-SR-MD and SR-CD and SR-MD and non-SR-CD had similar slopes.

Conclusions: The impact of SR-MD or SR-CD on adverse health outcomes differed as a function of age and sex. Thus, we need to consider preventive approaches according to these specific target group features.

Keywords: Frailty; cognitive impairment; long-term care insurance; mobility disorders; mortality; older adults.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Aged
  • Aged, 80 and over
  • Checklist
  • Cognition / physiology*
  • Cohort Studies
  • Female
  • Frail Elderly / statistics & numerical data*
  • Frailty / diagnosis
  • Frailty / mortality*
  • Humans
  • Independent Living
  • Insurance, Long-Term Care
  • Japan / epidemiology
  • Male
  • Mobility Limitation*
  • Prevalence
  • Prospective Studies
  • Self Report / statistics & numerical data*