Arthritis Burden Severity As a Predictor of Anxiety and Depression in U.S. Adults

Background/Purpose: Research suggests arthritic adults commonly experience comorbid anxiety and/or depression. Evidence linking arthritis to anxiety and depression, coupled with the high volume of adults affected by arthritis has treatment implications for physicians and social workers. Significant attention has been paid to depression; however, evidence suggests anxiety is more strongly correlated to arthritis than depression. Disorders of anxiety and depression differ regarding pharmacological and psychotherapeutic treatment. Additionally, treatment outcomes vary for those with anxious depression versus those without anxious depression (Fava et al, 2008). Understanding the unique and specific relationship between arthritis and co-occurring mental health issues is imperative to providing and advocating for competent, quality care for arthritis patients. The present study aims to first replicate findings conducted in the small scale clinical trials that found arthritis to be more significantly associated with anxiety than with depression and second, to investigate whether severityof arthritic burden (as opposed to whether someone has an arthritis diagnosis) is associated with anxiety and/or depression. Finally, this study aims to investigate whether a moderating effect of race/ethnicity exists between arthritis burden and anxiety and arthritis burden and depression. Implications for physicians and social workers regarding the need for collaboration to deliver quality client care will be presented.

Method: The Behavioral Risk Factor Surveillance System (2015) was used to conduct analyses. This study utilized a subsample of adults 18 and over who have been diagnosed with arthritis and reside in Mississippi, North Dakota, Tennessee, and West Virginia (8,765). The sample was limited to these states because only these states adopted the optional anxiety/depression module needed to address the study objectives. Two separate logistic regression models were used to analyze the relationships between arthritis burden and anxiety and depression and to test for a moderating effect of race/ethnicity.

Measures: The primary independent variable, arthritis burden, consists of the sum of four arthritis related survey items which asked respondents to report on how much arthritis negatively impacted their lives (0-6). The dependent variables, anxiety and depression, each consist of one item in which respondents report whether they have been diagnosed with anxiety or depression. The moderating variable of race/ethnicity consisted of self-reported non-Hispanic Whites, non-Hispanic Blacks, and Hispanics.

Results: Results of logistic regression models showed a similar relationship between arthritis burden and anxiety (effect size = 1.36) and arthritis burden and depression (effect size = 1.35). Race/ethnicity was found as moderating in the depression model, but not in the anxiety mode. The moderating effect was found between Hispanics and Whites (b = -.49, SE b = .23, p =.05). As arthritis burden increases, the probability of depression decreased for Hispanics whereas the probability of depression increased for Whites. 

Conclusions and Implications: Findings suggests that a significant number of arthritic adults experience anxiety, depression, or both and varies significantly by race/ethnicity regarding depression. Because mental health functioning has the potential to impact physical health and vice versa, a need for collaboration between the medical and social work fields is warranted.