SOCIAL VULNERABILITY, STRESS, AND PAIN: MULTILEVEL DETERMINANTS OF PAIN SEVERITY AND PAIN MANAGEMENT IN THE UNITED STATES
Date Issued
August 2026
Author(s)
Pehrson, Aimee
The University of Tennessee
Advisor(s)
Laurie Meschke
Additional Advisor(s)
Stephanie Eucker
Louis Rocconi
Brittany Shelton
Jennifer Russomanno
Abstract
Pain is a leading cause of disability in the United States, affecting more than 50 million adults leading to social, economic, and healthcare burdens. Increasing evidence shows that pain experiences and management behaviors are shaped by biological pathology and psychosocial conditions. Guided by the biopsychosocial model of pain and allostatic load theory, this dissertation examined how social vulnerability and perceived stress influence pain severity and engagement opioid sparing pain management strategies across population, individual, and community levels. In doing so, this dissertation integrates three empirical studies using to evaluate the multilevel determinants of pain severity and treatment behaviors.
Study 1 used cross-sectional data from 2002, 2012, and 2022 National Health Interview Survey (NHIS) to examine trends in complementary and integrative health approaches (CHA) for pain management. Survey-weighted logistic regression models with multiple imputation evaluated whether CHA utilization differed by insurance coverage and sociodemographic characteristics over time. CHA utilization increased between 2002 and 2022; however, differences in use by insurance type remained stable. Nonetheless, disparities in CHA use persisted, with individuals with private insurance and higher educational attainment having greater odds of using CHA therapies, highlighting inequities in utilization of non-opioid pain management.
Study 2 examined individual-level pathways linking social vulnerability, perceived stress, and pain severity among patients presenting to the emergency department with musculoskeletal pain. Multiple imputation addressed missing data, and Least Absolute Shrinkage and Selection Operator regression identified key social vulnerability indicators associated with perceived stress and pain severity. Results indicated that greater perceived social support was associated with lower pain severity at triage, while perceived stress did not significantly mediate these relationships. Study 3 extended this analysis by examining whether community-level social vulnerability indices were independently associated with pain severity and if they moderated the relationship between perceived stress and pain severity. Linear regression models did not identify significant moderation effects, suggesting that community-level vulnerabilities may influence chronic pain trajectories rather than acute pain severity in emergency settings.
Together, these findings demonstrate that pain disparities arise from interacting social, psychosocial, and structural influences, underscoring the importance of integrating biopsychosocial determinants into equitable, multimodal pain management strategies.
Subjects
Degree
Doctor of Philosophy
Major
Public Health Sciences
Embargo Date
August 1, 2029