AnthropoAge, a novel approach to integrate body composition into the estimation of biological age

Abstract Aging is believed to occur across multiple domains, one of which is body composition; however, attempts to integrate it into biological age (BA) have been limited. Here, we consider the sex‐dependent role of anthropometry for the prediction of 10‐year all‐cause mortality using data from 18,794 NHANES participants to generate and validate a new BA metric. Our data‐driven approach pointed to sex‐specific contributors for BA estimation: WHtR, arm and thigh circumferences for men; weight, WHtR, thigh circumference, subscapular and triceps skinfolds for women. We used these measurements to generate AnthropoAge, which predicted all‐cause mortality (AUROC 0.876, 95%CI 0.864–0.887) and cause‐specific mortality independently of ethnicity, sex, and comorbidities; AnthropoAge was a better predictor than PhenoAge for cerebrovascular, Alzheimer, and COPD mortality. A metric of age acceleration was also derived and used to assess sexual dimorphisms linked to accelerated aging, where women had an increase in overall body mass plus an important subcutaneous to visceral fat redistribution, and men displayed a marked decrease in fat and muscle mass. Finally, we showed that consideration of multiple BA metrics may identify unique aging trajectories with increased mortality (HR for multidomain acceleration 2.43, 95%CI 2.25–2.62) and comorbidity profiles. A simplified version of AnthropoAge (S‐AnthropoAge) was generated using only BMI and WHtR, all results were preserved using this metric. In conclusion, AnthropoAge is a useful proxy of BA that captures cause‐specific mortality and sex dimorphisms in body composition, and it could be used for future multidomain assessments of aging to better characterize the heterogeneity of this phenomenon.


Variable transformation
To improve model diagnostics, we conducted variable transformations for anthropometric variables by evaluating logarithmic, squared root, cubic root, inverse, and exponential transformations. Optimal transformations were chosen by minimization of the A-statistic from the Anderson-Darling normality test.
The following anthropometric measurements were transformed using natural logarithms: weight, body mass index, arm length and thigh circumference. Square root transformation was used for arm circumference and cubic root transformation was used for height, triceps and subscapular skinfolds, and waist-to-height ratio. Leg length did not require any specific transformation (Supplementary Figure 1).
All models were fitted using transformed variables unless indicated otherwise. Orthogonal polynomials were fitted to introduce non-linearity to the models using the poly R function and the number of degrees of freedom was arrived at by using BIC minimization as the model selection criterion. The subject is positioned standing with both feet flat on the floor, heels together and toes pointed slightly outwards, looking straight forward and with head, shoulder blades, buttocks, and heels touching the measurement surface. The head is aligned until horizontal line from the ear canal to the lower border of the orbit of the eye is parallel to the floor. Height is recorded to the nearest millimeter with a fixed stadimeter (headboard positioned on top of the head with enough pressure to compress the hair).

Upper arm length (cm)
With the subject standing (right arm flexed at 90 degrees and palm facing up), the length is recorded to the nearest millimeter from the posterior border of the acromion process of the right scapula to the tip of the right olecranon process. A mark is placed at the midpoint to define where arm circumference and tricipital skinfold will be measured.

Upper leg length (cm)
With the subject sitting (lower legs hanging and knees at 90°), the length is recorded to the nearest millimeter from the right inguinal crease to the distal end of the right femur. A mark is placed at the midpoint to define where thigh circumference will be measured.

Weight (kg)
Weight is measured on a Toledo digital scale with the subject standing still in the center of the scale platform facing the recorder, hands at side, and looking straight ahead. The measurement is recorded in pounds and then automatically converted to kilograms.
Body mass index (kg/m 2 ) Calculated by dividing the body weight (in kilograms) by the squared body height (in meters).

Visceral adiposity
Waist circumference (cm) With the subject standing, the measuring tape is placed around the trunk right above the right iliac crest, parallel to the floor. The tape should be snug without compressing the skin. The circumference is recorded at the end of expiration to the nearest millimeter.
Wast-to-height ratio Calculated by dividing the waist circumference by the body height (both in centimeters).

Subcutaneous adiposity
Triceps skinfold (mm) With the subject standing (shoulders relaxed, arms hanging freely), enough skin and adipose tissue is grasped to form a distinct fold that separates from the underlying muscle about two centimeters above the midpoint of the posterior surface of the arm (previously marked). The skinfold thickness is recorded to the nearest 0.1 mm with a caliper.

Subscapular skinfold (mm)
With the subject standing (shoulders relaxed, arms hanging freely), a fold of skin and adipose tissue is grasped right above and medial to the inferior angle of the right scapula.
The skinfold should form a line of about 45 degrees extending towards the right elbow and its thickness is recorded to the nearest 0.1 mm with a caliper.

Primarily lean mass
Mid-upper arm circumference (cm) With the subject standing (shoulders relaxed, right arm hanging loosely without tightening the muscles), the measuring tape is placed around the midpoint of the upper arm (previously marked) perpendicular to its long axis. The two ends of the tape are pulled together without compressing the skin and circumference is measured to the nearest millimeter.
Mid-thigh circumference (cm) With the subject standing (right leg forward, knee slightly flexed), the measuring tape is placed around the midpoint of the upper leg (previously marked) perpendicular to its long axis. The two ends of the tape are pulled together without compressing the skin and circumference is measured to the nearest millimeter. We also provide a table specifying the transformation that was performed for each variable, as well as the A-statistic and p-value from the Anderson-Darling test prior and after transformation. These transformations were the ones we used for the development of AnthropoAge and S-AnthropoAge.