Parasites and skeletal indicators of anemia in the eastern United States

Abstract

The goal of this research is to examine the influence of parasitic infection and diet in the etiology of anemia in prehistoric human populations of the eastern United States. Prehistorically, anemia is often attributed to a nutrient-deficient diet, while parasite infection is discussed as a secondary cause if at all. However, parasite infection is a leading cause of anemia in the developing world today. Modern epidemiological studies have demonstrated that parasites thrive or perish under particular environmental conditions, and risk for parasite infection can be predicted based on environment using GIS. Here I apply this method to see whether environmental conditions, acting as a proxy for parasite infection risk, can predict prehistoric skeletal lesion rates for porotic hyperostosis and cribra orbitalia, lesions thought to reflect acquired anemia. Rates of porotic hyperostosis and cribra orbitalia in the skeletal remains of children and adults were collected from published data for 22 sites in the eastern United States. GIS was used to gather comparable environmental data. Soil drainage, elevation, precipitation, temperature and the surface area of bodies of water were recorded within a 15 km radius of each site. Carbon isotope data deriving from bone collagen and historic hookworm infection rates were also collected when available. Multiple linear regression was used to test how well environmental variables could predict lesion rates. Statistically significant correlations were found for both adults and children, but the strength and direction of relationships with environmental variables were inconsistent. It is possible that the correlations were related to parasite infection, but it is also possible that the skeletal 'lesions' may result from post-mortem bone degeneration rather than anemia. The correlations for porotic hyperostosis and cribra orbitalia were stronger when examined separately than when examined together, suggesting that the two conditions may have separate etiologies; however, the sample sizes were too small to provide the statistical power required for drawing strong conclusions. Comparison of children and adults showed stronger correlation for children, though when observing the lesions separately this pattern was not consistent. Collagen carbon values and historic hookworm infection rates correlated with lesion rates in children but not adults, perhaps because of differential healing in adults. These results demonstrate that environmental conditions and skeletal lesions are correlated, but the underlying mechanism for this remains unclear. Larger sample sizes would allow for more robust statistical analyses of the trends observed here. Nevertheless, these results do confirm that porotic hyperostosis and cribra orbitalia cannot be assumed to be the result of nutrient-deficient diets. Interpretation of skeletal data for assessing health in the past must also consider the natural and social context in which individuals lived.