Summary: According to a PLOS ONE study, an individual’s variability in reaction time across repeated trials may independently predict mortality in older adults.
Source: PLOS
Common measures of cognitive change in later life may reveal more than declining thinking skills. A study published August 9, 2017 in PLOS ONE by Nicole A. Kochan and colleagues at the Centre for Healthy Brain Ageing (CHeBA), UNSW Sydney, found that intraindividual reaction time variability is associated with shorter survival in older adults.
Intraindividual reaction time variability (IIVRT) describes how much a person’s response times fluctuate across repeated trials of the same task. Higher IIVRT has previously been linked to mild cognitive decline, dementia and Parkinson’s disease. The research team asked whether this fluctuation in reaction time is also an independent predictor of mortality by tracking 861 community-dwelling adults aged 70 to 90 for eight years.
At baseline, participants completed two brief computerized cognitive tasks totaling 76 trials. From those tests, researchers derived two composite reaction-time measures: the average reaction time (mean RT) and IIVRT, calculated as the intraindividual standard deviation of reaction times after adjusting for age and time-on-task effects. Every two years, trained research psychologists conducted follow-up clinical assessments that included a broad neuropsychological test battery and a comprehensive medical review. An expert panel of clinicians made consensus dementia diagnoses at each follow-up interval, and mortality information was obtained from the relevant state registry.
During the eight-year follow-up, 191 of the 861 participants (22.2%) died. Using Cox proportional hazards models, the investigators examined whether baseline IIVRT or mean RT predicted survival time. After adjusting for key confounders — including age, sex, global cognitive performance, a cardiovascular risk index and apolipoprotein ɛ4 status — they found that greater IIVRT significantly predicted shorter survival time, whereas mean RT did not.
Importantly, the association between IIVRT and mortality remained essentially unchanged after excluding participants who developed dementia during follow-up. This indicates that the predictive value of IIVRT cannot be fully explained by lower overall cognition or by prodromal dementia. In other words, increased moment-to-moment variability in simple reaction time appears to be a behavioural marker linked to elevated mortality risk in later life, independent of several established biomedical and cognitive risk factors.
The authors emphasize that this was the first study to comprehensively account for overall cognitive level and incident dementia when testing whether intraindividual reaction time variability predicts mortality. Their findings suggest that monitoring IIVRT could provide additional, clinically relevant information about an older person’s health trajectory beyond traditional cognitive screening and cardiovascular risk assessment.
Funding: The study received support from the National Health and Medical Research Council of Australia Program Grant (grant number 350833), an NHMRC Early Career Fellowship (grant number 123148) and the Dementia Collaborative Research Centre. The funders did not influence study design, data collection and analysis, the decision to publish, or preparation of the manuscript.
Original research: Is intraindividual reaction time variability an independent cognitive predictor of mortality in old age? Findings from the Sydney Memory and Ageing Study. Authors: Nicole A. Kochan, David Bunce, Sarah Pont, John D. Crawford, Henry Brodaty, Perminder S. Sachdev. PLOS ONE. Published online August 9, 2017. DOI: 10.1371/journal.pone.0181719.
Key terms for search: intraindividual reaction time variability, IIVRT, reaction time variability, mortality prediction, older adults, Sydney Memory and Ageing Study, cognitive marker, dementia risk.
Abstract (condensed)
IIVRT, a proposed marker of neurobiological disruption that increases with age, has been linked to dementia and mortality. This study followed 861 adults aged 70–90 who completed two computerized reaction time tasks at baseline and received comprehensive medical and neuropsychological follow-up every two years. Using survival analysis over eight years (191 deaths), greater IIVRT—but not mean reaction time—predicted shorter survival after adjustment for age, sex, global cognition, cardiovascular risk and apolipoprotein ɛ4. Excluding incident dementia cases did not materially change the result. The authors conclude that increased IIVRT uniquely predicts shorter time to death and is not explained by lower global cognition or prodromal dementia.