A Comprehensive Pain-Related Risk And Protective Index And Machine Learning Brain Age Gap

Allostatic load is a measure of individual health status which can be assessed from brain measures, a clinical biomarker composite, and telomere length among others. We reported that greater chronic pain severity and socioenvironmental risk and lower biobehavioral/psychosocial resilience are associated with higher allostatic load across multiple measures. The purpose of this study was to examine the relationship between a combined measure of chronic pain, socioenvironmental status, and biobehavioral/psychosocial resilience with a measure of allostatic load, predicted-chronological brain age gap (BAG). Non-Hispanic black (NHB) and non-Hispanic white adults between 45-85 years old with/without knee pain and brain imaging were included. Chronic pain stage was determined by knee pain persistence, intensity, duration and total body pain sites. Socioenvironmental factors included education, poverty level, Area Deprivation Index, and marital, employment, and insurance status. Resilience factors consisted of tobacco use, waist circumference, optimism, positive and negative affect, perceived stress, social support and sleep impairment. Each variable was assigned a 1 for protective or 0 for risk based on evidence-based ranges and summed. High risk (0-15) and high protection (21-36) were determined by tertile splits. Predicted brain age was calculated using a validated machine learning technique, DeepBrainNet. A total of 197 adults, 58.2+8.3 years old, 66% women and 44% NHB were included. With sex, study site and comorbidities in the analysis, individuals in the protective group had significantly younger BAG (p<.0001). Findings indicate a potentially clinically useful tool for assessing risk and protection and differentiating individual differences contributing to disparities in pain-related outcomes. Funding: NIH/NIA Grants R01AG054370 (KTS); R01AG054370-05S1 (KTS & JJT), R37AG033906 (RBF) and UF CTSA Grant UL1TR001427 and UAB CTSA Grant UL1TR001417 from the NIH Center for Advancing Translational Sciences.