Arthritis in adults, socioeconomic factors, and the moderating role of childhood maltreatment: cross-sectional data from the National Epidemiological Survey on Alcohol and Related Conditions

S. L. Brennan-Olsen, T. L. Taillieu, S. Turner, J. Bolton, S. E. Quirk, F. Gomez, R. L. Duckham, S. M. Hosking, G. Duque, D. Green, T. O. Afifi

These data present associations between socioeconomic status (SES), different types of childhood maltreatment (CM) history and family dysfunction, and arthritis in men and women across a wide age range. Arthritis was less likely among those with higher SES, regardless of CM history. CM has been associated with increased risk of adult-onset arthritis; however, little is known about whether socioeconomic status moderates arthritis risk in those with CM history. We investigated arthritis across education, income, and race/ethnicity and whether CM moderated associations between SES and arthritis. Data were drawn from Wave 2 (2004-2005) of the nationally representative (USA) National Epidemiological Survey on Alcohol and Related Conditions (NESARC, n = 34,563; aged ≥ 20 years). Self-reported CM history included physical abuse, sexual abuse, emotional abuse, emotional neglect, physical neglect, and exposure to intimate partner violence (IPV). We used descriptive statistics and logistic regression to determine relationships between SES, CM, and arthritis. Interaction terms were used to test if CM moderated relationships between SES and arthritis. Arthritis prevalence was 21.1% (n = 3093) among men and 30.1% (n = 6167) among women. In unadjusted analyses, women (p ≤ 0.001) and older age (both sexes, p ≤ 0.01) were associated with increased odds of arthritis. All CM types were associated with increased odds of arthritis, except exposure to IPV among women. In sex-stratified, age-adjusted analyses, lower education and income, family dysfunction, being Hispanic or Asian/Native Hawaiian/Pacific Islander, and ≥ 1 physical comorbidity were associated with increased odds of arthritis among those with and without CM: trends were similar for both sexes. In age-adjusted two-way interaction terms, CM did not moderate associations between SES and arthritis. Although CM was associated with arthritis, associations between SES and arthritis were not amplified. Arthritis was less likely among those with higher SES, regardless of CM history.