Depressive Symptom Trajectory of Adults with Arthritis: Using Andersen’s Behavioral Model and Theory of Fundamental Causes

Background and Purpose: About 24% of adults in the US have been diagnosed with arthritis, and the prevalence of depression among middle-aged and older adults with arthritis is high. Although the literature has reported that discriminative experiences can exacerbate mental health conditions in adults with arthritis, studies analyzing this association longitudinally in this population are sparse. Grounded in Andersen’s behavioral model and the theory of fundamental causes, this study aimed to investigate the depressive symptom trajectory of middle-aged and older adults with arthritis and examine the effect of perceived discrimination on this trajectory.

Methods: This study employed the Health and Retirement Study (HRS) nationwide panel data from a multi-stage area probability sample of adults aged 50 or older. The 10,923 respondents who had been diagnosed with arthritis were drawn from three waves (2014, 2016, and 2018) of the HRS data. Descriptive analyses were performed to summarize the sample’s sociodemographic characteristics. An unconditional latent growth modeling analysis was initially performed to describe the depressive symptom trajectory. Next, conditional modeling was conducted to examine whether perceived discrimination was significantly associated with variances in the depressive symptom trajectory.

Results: The mean age was 70.6 years; 36.1% were men, 66.8% were White, 19.2% were Black, and 11.2% were Hispanic. In the unconditional modeling (χ2 = 2.248, df = 1, p <.05; RMSEA = .011, CFI = 1.000, TLI = 1.000), the depressive symptom trajectory declined over the 4-year period (p <.05), and the variances of the intercept (p <.001) and the slope were significant (p <.05). The conditional model (χ2 = 54.604, df = 24, p <.05; RMSEA = .023, CFI = 0.990, TLI = 0.970) indicated that adults with arthritis who were younger (p <.001), female (p <.05), and unmarried (p <.05), and those who had lower annual household incomes (p <.01), who reported poor/fair health statuses (p <.001), who experienced higher levels of perceived discrimination (p <.001), and who did not engage in physical activities (p <.001) were more likely to have higher depressive symptoms at baseline than their counterparts. Adults with good health statuses (p <.05) were more likely to have steeper decreasing rates of depressive symptoms over time, compared to those with poor/fair health statuses.

Conclusions and Implications: These findings can be employed in developing health behavior management interventions and assessment protocols to reduce depressive symptoms in adults with arthritis. Health care professionals (e.g., social workers, nurses, and physicians) should be informed of the factors that could reduce depressive symptoms in the target population. Service providers in healthcare settings should pay particular attention to adults who may experience discrimination based on race, sexual orientation, language, and religion. These findings will inform policymakers of the influence of sociodemographic characteristics, health behaviors, and perceived discrimination on the mental health status of adults with chronic diseases by encouraging local, regional, and national legislative efforts to improve their mental health.