Histomorphometric age estimation from the femoral cortex: A test of three methods in an Australian population

Histological methods can be used forensically to estimate age-at-death based on patterns of change in osteon shape, size, and population density, all of which result from the continuous process of bone remodelling. The present study examines the applicability of three existing histological age-at-death estimation methods as applied to an Australian population of known age and sex. Microradiographs from 50 mid-shaft femora thin sections, equally divided by sex, were obtained from the Melbourne Femur Reference Collection (MFRC); stated chronological age-at-death is 18 to 88 years. Osteon shape metrics are measured using ImageJ and the age-at-death prediction formulae of i) Singh and Gunberg, ii) Keough et al., and iii) Goliath et al. are applied. The relationship between estimated and actual age-at-death is then statistically quantified. All three formulae demonstrate pooled and sex-specific SEE values in excess of 20 years: i) pooled ±22.92 (♂±20.91, ♀±25.20); ii) ±20.79 (♂±20.96, ♀±21.05); and iii) ±35.43 (♂±32.68, ♀±38.66). When individuals under 40 years of age were excluded from the analysis, only two of the methods demonstrated increased accuracy: i) pooled ±20.87 (♂ ±17.47, ♀ ±23.70); ii) pooled ±18.21 (♂±16.51, ♀±19.90); and iii) pooled ±41.18 (♂ ±40.12, ♀ 43.05). The present study represents a preliminary investigation of the accuracy of existing histological age-at-death standards applied in an Australian population of known age.