Abstracts Anthropology

Estimating Adult Age Using Radiographs of the Proximal Femur

by Jong In Kim

Most skeletal age estimation techniques are based on assessing wear on joints or bone involution. Bone involution refers to microscopic bone loss. Pryce (2012) proposed a five-trait proximal femoral bone involution method to estimate age. The purpose of this research was to test Pryce’s (2012) age estimation method and to determine whether a Bayesian approach improves the method, using radiographs from the University of Toronto at Mississauga’s radiograph database. The objectives were to determine Pryce’s (2012) inter-observer reliability, to modify the method with transition analysis, and to test the new method. The revised method using 95%-CI and 50%-CI were created. Pryce’s (2012) method had a poor inter-observer reliability, but for the revised method, four of the five traits had an accuracy rate greater than 90%. The revised method can be invaluable when it is part of a multifactorial approach; however, further research is required to investigate potential limitations.