Examination of Secular Change in the Vertical Head Diameter of the Human Femur in American Males and Females

Secular changes in various aspects of the human skeleton can provide indications of environmental effects on growth and development. Temporal trends, as a reflection of fluctuations in health of a particular population over time, can provide evidence of how that population responds to episodic environmental transitions. Cohort comparisons of trends can reveal differing degrees of change experienced among groups within a single population or between different populations. The purpose of this study is to examine secular changes in the femoral maximum vertical head diameter of Americans male and females, and to investigate whether changes in head diameter size could be a corollary of decreased physical activity.

The maximum vertical head diameters of 19^th Century and modern Americans were examined for secular changes utilizing a sample (n = 1,217) of measurements taken from the Robert J. Terry Anatomical Collection, the M. F. Ericksen Femur Collection, and the University of Tennessee Forensic Data Bank. Two cohorts categorized by sex were evaluated for normality and autocorrelation prior to performing additional statistical analyses. Statistical evaluation of a secular trend for the male sample was executed via an Autoregressive Integrated Moving Average (ARIMA) model of the average maximum vertical head diameter variable on the year-of-birth variable, and a subsequent regression analysis of lag first-differences on five year-of-birth variable. The Hotelling's T-square test was employed to test for differences among group head diameter size means for males in first fifty years of the study and the last fifty years of the study. For the female cohort, piecewise regression analysis was employed to examine the trend in this sample.

Results of the ARIMA statistical analyses indicate that among American males the average maximum vertical head diameter of the femur appears to fluctuate in size over time. For the female sample, employment of the piecewise polynomial curve fitting in conjunction with the multiple regression indicate that female femoral head diameters were increasing in size until approximately 1910 when head diameters begin to rapidly decrease in size. The increase in average femoral head diameter was not statistically significant; however the decrease was determined to be statistically significant.