New pace of aging measurement reveals trajectories of healthspan and lifespan in older people

Until now the metrics used in population health research on aging did not distinguish differences caused by early-life factors, such as prenatal care and nutrition, from those caused by ongoing changes in people's bodies due to aging. Findings from the study are published in Nature Aging.

"The Pace of Aging method is an important approach for understanding population aging," explained Arun Balachandran, PhD, a postdoctoral researcher at the Columbia Aging Center and lead author of the study. "Our existing toolkit doesn't include methods that can separate out the legacies of early life from the changes caused by aging," Daniel Belsky, PhD, associate professor of Epidemiology at Columbia Mailman School and member of the Robert N. Butler Columbia Aging Center elaborated.

"We originally developed the Pace of Aging method to evaluate the effectiveness of interventions targeting the biology of aging. The new approach introduced in this paper is designed to do the same for social policies and public health programs. Our method will enable researchers and public health professionals working with population data to better understand how policies, social structures, environments, and individual behaviors shape aging trajectories across populations worldwide."

The team analyzed data from two large-scale, nationally representative studies: the U.S. Health and Retirement Study (HRS)and theEnglish Longitudinal Study of Aging (ELSA). These long-term studies follow adults aged 50 and older -- along with their spouses -- and collect detailed information on health, cognition, socioeconomic status, and family dynamics. The studies have been ongoing for decades and periodically enroll new participants.

The new approach makes use of data from dried blood spots, physical exams, and performance tests given to participants in their homes at up to three timepoints over eight-year follow-up intervals. Pace of Aging was examined in 19,045 participants who contributed data over 2006-2016, with additional follow-up to determine disease, disability, and mortality through 2022. In the US study, Pace of Aging was measured from C-reactive protein (CRP), Cystatin-C, glycated hemoglobin (HbA1C), diastolic blood pressure, waist circumference, lung capacity (peak flow), balance, grip strength, and gait speed.

"Our findings establish that we can measure important variability in the pace of aging in older people with a relatively limited set of measurements," said Belsky. "Our findings open up possibilities to study pace of aging in cohorts around the world," expanded Balachandran. "These metrics consistently predict future health outcomes, including disease onset, disability, and death. And they reveal important differences in aging trajectories across population subgroups." For example, the study reported signs of accelerated aging in people with lower levels of education."

Originally developed using data from the Dunedin Study -- a longitudinal study of individuals born in 1972-73 -- the initial Pace of Aging tool focused on changes from young adulthood through midlife. The newly adapted method extends its utility to population-based studies of aging, offering planners and policymakers a valuable resource for monitoring and improving population health and longevity.

"Beyond medicine and gerontology, this work has important implications for sociology and economics," added Belsky. "It can help us understand how life transitions -- such as retirement, caregiving, and bereavement -- affect the aging process and support the development of more effective public health and social policies."

"The differences in aging speed we found weren't just statistically significant -- they were meaningful," Belsky said. "People aging faster were much more likely to get sick, become disabled, or die sooner, even if they were the same age on paper."

Other co-authors are Heming Pei, Yifan Shi, John Beard, Alan Cohen, Claire Eckstein Indik, Calen Ryan, Alex Furuya, Meerai Kothari, and Yuang Zhang, Butler Aging Center, Columbia Mailman School; Avshalom Caspi and Terrie Moffitt, University of London; Benjamin Domingue, Stanford University; Luigi Ferrucci, National Institute on Aging; and Vegard Skirbekk, Norwegian Institute for Public Health.

The study was supported by National Institutes of Health grant R01AG061378, T32AI114398 and U01AG009740; Russel Sage Foundation, BioSS Grant 1810-08987, and the Robert N Butler Columbia Aging Center.