Semester of Graduation

Spring 2026

Degree

Master of Arts (MA)

Department

Geography and Anthropology

Document Type

Thesis

Abstract

Age estimation is a key component of the forensic anthropological profile used in the identification of unidentified human remains. Adult age estimation methods typically assess degenerative changes of skeletal joint surfaces, such as the pubic symphysis, iliac auricular surface, and sternal rib ends, which exhibit relatively predictable changes over time. However, variation in physiological aging, biomechanics, and lifestyle introduces uncertainty to age estimates. One lifestyle factor that may affect the utility of age estimation methods is substance use, as both prescribed and recreational substance use reduces bone quality and density. Despite high rates of alcohol, tobacco, and drug use in the United States, little research has examined how substance use may influence forensic skeletal age estimates. This study evaluates whether reported substance use impacts the classification accuracy, inaccuracy, and bias of skeletal age estimation.

Age estimates were derived using Hartnett’s revised methods for the sternal end of the forth rib and the pubic symphysis, and Transition Analysis 2 (TA2) for the pubic symphysis and iliac auricular surface, both individually and combined. The sample included 151 skeletons from the Southeast Texas Applied Forensic Science (STAFS) Donated Skeletal Collection, 74 with histories of reported substance use and 77 without. Estimated ages were compared to chronological ages to assess overall performance and subgroup variation based on sex, ancestry, age group, and substances used.

Results indicated that while substance use did not significantly reduce overall accuracy, substance users tended to have age estimates with lower classification accuracy, that were further from true age-at-death, and more frequent underestimation, than controls. Subgroup analysis revealed significant effects. Pearson chi-square results indicated that individuals aged 66 years and older had significantly lower classification accuracy compared to controls (p = 0.033) when using Hartnett’s sternal rib end method. Using Hartnett’s pubic symphysis method, Pearson chi-square testing revealed that individuals who reported tobacco use showed a significant reduction in classification accuracy compared to controls (p = 0.050). Additionally, two-way ANOVA and subsequent Tukey’s post-hoc testing results showed that individuals reporting tobacco use were significantly under-aged with TA2’s pubic symphysis method compared to controls, who were slightly over-aged (p = 0.037).

These findings suggest that while substance use does not broadly reduce method accuracy, it contributes to greater error and bias. This study underscores the importance of considering individual life history factors, such as substance use, in forensic anthropological analysis. Future research incorporating larger samples and additional aging techniques may further clarify the relationship between substance use, skeletal degradation, and age estimation.

Date

3-10-2026

Committee Chair

Listi, Ginesse

LSU Acknowledgement

1

LSU Accessibility Acknowledgment

1

Download

Available for download on Wednesday, March 10, 2027

Share

COinS