An Analysis on the Impact of Childhood Adversity, Anxiety, and C-Reactive Protein on Adult Chronic Pain in the Midlife in the United States (MIDUS) Study

Objective: This study used the Midlife-Development in the United States (MIDUS) dataset to a) examine relationships between reported childhood adversity (CA), anxiety, and pain; b) assess associations between CAs, anxiety, C-reactive protein (CRP) levels, and pain; and c) explore how CAs, anxiety, and CRP are associated with pain medication consumption.

Methods: Data were from Project-4 of MIDUS-II (n = 1225), which utilized Project-1 demographics and supplemental chart review. For objectives 1-2, structural equational modeling (SEM) followed by general linear modeling (GLM) regression was conducted. For objective 3, all variables from the objective 1-2 dataset were used as possible independent variables for the exploratory regression.

Results: For objectives 1-2, CRP was significantly correlated with anxiety, emotional abuse, physical neglect, and chronic pain (n = 1173). The SEM (n = 1173) indicated that CAs, anxiety, and CRP all played a role in predicting chronic pain. Regression results (n = 1173) indicated that gender, total income, and highest education were significant predictors of chronic pain. Significant interactions to explain chronic pain included physical abuse/emotional neglect, emotional abuse/physical abuse, physical abuse/minimization, physical neglect/education, CRP/income, and CRP/education. For objective 3 (n = 600), there were no significant main effects, but a large variety of interactions contributed to predicting pain medication consumption. CAs interacting significantly to explain this included emotional abuse/physical abuse, physical abuse/emotional neglect, physical abuse/minimization, and sexual abuse/minimization. There were also significant interactions between CRP/income and CRP/education.

Conclusions: Based on a large US sample, sociodemographics played a meaningful role in predicting chronic pain in adults, and CRP was significantly correlated with anxiety, emotional abuse, physical neglect, multiple sociodemographic variables, and chronic pain. The influence of CAs on predicting long-term medication use for chronic pain was complex and warrants further study.