Joint association of the frailty index and phenotypic age with all-cause and cause-specific mortality: A prospective cohort study.

Abstract

Little is known about whether frailty index (FI) and phenotypic age (PhenoAge), two independent indicators of biological age, may jointly or interactively predict death risk. This prospective study was conducted using National Health and Nutrition Examination Survey (NHANES) data of four survey cycles (2003 to 2010) and mortality records until December 31, 2019. FI was calculated for each participant based on his/her responses to 34 questions, and PhenoAge was derived from algorithms using data of nine biomarkers and chronological age. We used Cox proportional hazards (PH) models and competing risk models to estimate the associations of FI, PhenoAge, and their interactions with all-cause mortality and cause-specific mortality. The C-statistic, continuous net reclassification index (NRI), and integrated discrimination improvement (IDI) were used to evaluate model performance. After an updated median follow-up of 11.8 years among 20, 089 adults, a total of 3778 deaths were documented. Each 1 standard deviation (SD) increment in FI and PhenoAgeAccel was associated with 47% (42%-51%) and 41% (38%-45%) higher risk of all-cause mortality, respectively. However, their multiplicative interaction on death risk was not found (<i>P</i> = 0.054). Compared with the reference group (robust-decelerated aging), the risk of all-cause mortality was the highest in the frail-accelerated aging group with a hazard ratio (HR) (95% confidence interval [CI]) of 5.32 (4.50-6.29). Notably, adding both FI and PhenoAgeAccel to the traditional model significantly enhanced predictive accuracy (C-index 0.882 <i>vs</i>. 0.866, NRI 46.18%, and IDI 4.10%). Individual death risk could be better predicted by a joint inclusion of both FI and PhenoAge, indicating that targeting their components may reduce death risk and prolong health span.