Biological age is a construct that seeks to evaluate the biological wear and tear process of the organism that cannot be observed by chronological age. We estimate individuals’ biological age based on biomarkers from multiple systems and validate it through its association with mortality from natural causes. Biological age was estimated in 12,109 participants (6621 women and 5488 men) from the first visit of the Brazilian Longitudinal Study of Adult Health (ELSA-Brasil) who had valid data for the biomarkers used in the analyses. Biological age was estimated using the Klemera and Doubal method. The difference between chronological age and biological age (Δage) was computed. Cox proportional hazard models stratified by sex were used to assess whether Δage was associated with mortality risk after a median follow-up of 9.1 years. The accuracy of the models was estimated by the area under the curve (AUC). Δage had equal mean for men and women, with greater variability for men. Cox models showed that every 1-year increase in Δage was associated with increased mortality in men (HR (95% CI) 1.21; 1.17–1.25) and women (HR (95% CI) 1.24; 1.15–1.34), independently of chronological age. Results of the AUC demonstrated that the predictive power of models that only included chronological age (AUC chronological age = 0.7396) or Δage (AUC Δage = 0.6842) was lower than those that included both, chronological age and Δage (AUC chronological age + Δage = 0.802), in men. This difference was not observed in women. We demonstrate that biological age is strongly related to mortality in both genders and is a valid predictor of death in Brazilian adults, especially among men.

生物年龄是一种旨在评估无法通过实际年龄观察的生物体的生物磨损过程的构造。我们根据多个系统的生物标志物估计个体的生物年龄，并通过其与自然原因死亡率的关联来验证它。巴西成人健康纵向研究 (ELSA-Brasil) 首次访问时对 12,109 名参与者（6621 名女性和 5488 名男性）进行了生物年龄估计，这些参与者拥有分析中使用的生物标志物的有效数据。使用 Klemera 和 Doubal 方法估计生物年龄。计算实际年龄和生物学年龄之间的差异（Δage）。在中位随访 9.1 年之后，使用按性别分层的 Cox 比例风险模型来评估 Δage 是否与死亡风险相关。模型的准确性通过曲线下面积（AUC）来估计。男性和女性的Δ年龄平均值相同，男性的变异性更大。 Cox 模型显示，Δage 每增加 1 年，男性 (HR (95% CI) 1.21; 1.17–1.25) 和女性 (HR (95% CI) 1.24; 1.15–1.34) 死亡率就会增加，与时间顺序无关年龄。 AUC 结果表明，仅包含实足年龄（AUC 实足年龄 = 0.7396）或 Δage（AUC Δage = 0.6842）的模型的预测能力低于同时包含实足年龄和 Δage 的模型（AUC 实足年龄 + Δage = 0.802），男性。在女性中没有观察到这种差异。我们证明，生物年龄与两性的死亡率密切相关，并且是巴西成年人（尤其是男性）死亡的有效预测因素。