Exploring associations between presence of skeletal indicators of disease and sub-adult mortality

Skeletal indicators of disease are frequently used in bioarcheological research as a proxy for the frailty of an individual preceding death. Whilst certain indicators have been established as being associated with excess mortality in adults, these relationships are not necessarily replicated in sub-adults, nor are they consistent in strength of association across sites or time periods. Further, the influence of co-morbid indicators of disease have rarely been assessed in a bioarcheological context. This study addresses this research gap by examining the relationship between skeletal indicators of disease and mortality in sub-adults whilst controlling for observable co-morbidities in a medieval Irish population from Ballyhanna (Co. Donegal). Multivariable logistic regression on a subset of the Ballyhanna population reveals significant associations between linear enamel hypoplasia and stunting with increased likelihood of mortality in sub-adult years, therefore representing reliable indicators of frailty in this population. However, other examined skeletal indicators either had no association, or were more likely to occur in individuals who died in adulthood. Complex relationships between cribra orbitalia, porotic hyperostosis, and reactive periosteal lesions by severity or state of healing were observed via univariable analysis, suggesting that presence of these skeletal indicators are not reliably representative of frailty. More broadly, these combined findings suggest that skeletal indicators of disease, especially non-specific indicators such as periosteal lesions, may be more informative of population-level exposure to stressors than of variation in individual frailty. These findings reiterate the need to quantify the relationship between skeletal indicators and measurable outcomes if they are to be used as generalised indicators of frailty in a population.