Applicability, reliability, and accuracy of age-at-death estimation methods on a contemporary Italian population

This research tested the applicability, intra- and inter-observer reliability, and accuracy of nine macroscopic methods for estimating age-at-death from skeletal elements. The sample included 400 individuals from the contemporary CAL Milano Cemetery Skeletal Collection, equally divided by sex assigned at birth and with age-at-death ranging from 20 to 104 years. Statistical analyses used standard measures of reliability and validity. The most applicable methods assessed the auricular surface of the ilium (92%), whereas the preservation and/or identification of the fourth rib was rare (16%). All methods proved repeatable, except for the assessment of the obliteration of palatine sutures, which showed greater subjectivity to the observer's experience. The Rougé-Maillart (2009) method exhibited low bias and inaccuracy across all age groups in both sexes and the strongest correlation with chronological age in males. In females, the Buckberry and Chamberlain 2002 method showed the strongest correlation, but it tended to overestimate the age of younger individuals in both sexes. Age-at-death estimation methods revealed varying accuracy across age groups. The Suchey-Brooks 1990 method was effective for younger individuals, while the Rougé-Maillart (2009) and Falys and Prangle 2015 methods showed potential for the estimation of older adults, with lower bias and better precision. However, no approach was entirely satisfactory for older adults. The increasing life expectancy and the likelihood of encountering older adults’ remains highlight the need for refined techniques that better address age diversity in contemporary and ancient populations.