Configurations, comparisons, and challenges calculating allostatic load

Robert-Paul Juster,Enzo Cipriani,Philippe Kerr,L. Zachary DuBois

DOI: https://doi.org/10.1016/j.psyneuen.2023.106787

IF: 4.693

2024-02-01

Psychoneuroendocrinology

Abstract:Background Allostatic load algorithms have provided a comprehensive way to assess the causes and consequences of chronic stress. In general, allostatic load is calculated using a count-based index representing the number of biomarkers attaining cut-offs. These cut-offs are traditionally derived from the sample’s biomarker distributions that can create challenges for researchers trying to select the most appropriate algorithm. Method Using different datasets, we explore various configurations of allostatic load algorithms based on (1) group-based distributions without consideration of any other variables, (2) clinically-based distributions derived from population norms, (3) comparison-based anchoring distributions on a separate control sample, and (4) sex- and gender-based considerations. These different allostatic load formulations are next tested in association with various social determinants of health as well as psychiatric symptoms with a statistical concentration on effect magnitudes. Results While the traditional count-based approach of calculating allostatic load is the most common approach in the literature and the simplest, it does correlate as strongly with social determinants of health as do alternative allostatic load formulations. Moreover, associations with psychiatric symptoms were stronger when using cut-offs based on clinical norms or comparison groups. Lastly, sex- and gender-specific profiles of psychiatric symptoms align better with allostatic load formulations that are calculated disaggregated by birth-assigned sex or gender identity. Conclusion To date, there is no consensus on the best way to calculate a “standard” allostatic load index. We contribute to this debate by showing the importance of modifying allostatic load calculations based on your sample and your research questions.

neurosciences,endocrinology & metabolism,psychiatry