Longitudinal trends in functional biological age : Impact of lifestyle factors

Composites of measures of biological aging (e.g., Anstey and colleagues, Wahlin and colleagues) can be more meaningful than simple chronological age and provide insights into the aging process and its covariates. The Swedish Adoption/Twin Study of Aging provides longitudinal data on measures of vision, hearing, gait, grip strength, and lung function from 642 individuals ranging in age from 47 to 87 at wave 1. Individuals were included who participated in at least one of 5 measurement waves covering 16 years of follow-up; 69% participated in at least 3 waves. The 5 measures are combined and transformed to T-scores to create FBioAge. A two-slope age-based latent growth curve model (LGCM) was applied to the data (note that results for a time-based LGCM were similar). Phenotypic analyses indicated an inflection point in rates of change at age 75: the rate of increase in FBioAge was twice as fast after age 75, compared with prior to age 75. Analysis of the impact of several covariates on the LGCM parameters indicated that most impacted the intercept, only. Thus, on average higher (i.e., older) FBioAge was indicated for women, individuals with less education, smokers, drinkers, individuals who reported more illnesses, and individuals who reported poorer subjective health. Two variables impact the rate of change in FBioAge. Faster rate of change was predictive of mortality and childhood SES impacted the rate of change prior to age 75, only. In future research we will examine how FBioAge relates longitudinally to aging-sensitive functions, such as cognitive abilities.