The Effects of Health Behaviors and Perceived Discrimination on the Trajectory of Depressive Symptoms Among Middle-Aged and Older Adult Cancer Survivors

Background/Purpose: About 90% of cancer survivors were aged 50 years and older in 2019 (Miller et al., 2019). Research has shown that health behaviors and perceived discrimination are associated with depressive symptoms of cancer survivors (Andersen, Davidson, & Baumeister, 2014; Link & Phelan, 1995). However, few studies have analyzed this association longitudinally. This study described the trajectory of depressive symptoms among middle-aged and older adult cancer survivors and examined the effect of health behaviors and perceived discrimination on this trajectory.

Methods: The Health and Retirement Study (HRS) is a nation-wide panel study that uses a multi-stage area probability sample that surveys adults 50 aged or older across the US. Data from 1,961 cancer survivors aged 50 or older were drawn from the three waves (2010, 2012, and 2014) of the HRS (public and restricted records). Descriptive analyses were performed to summarize the sociodemographic characteristics of the sample. Next, latent growth modeling analyses were conducted in two steps. First, unconditional modeling was processed to investigate the pattern and variances of the trajectory of depressive symptoms. Second, a conditional modeling including health behaviors (e.g., physical activities), perceived discrimination, and covariates (e.g., age) was examined to investigate if health behaviors and perceived discrimination significantly affect the variances in the depressive symptoms trajectory.

Results: The respondents’ mean age was 71.7 years. About 52% were male, 74% were White, 18% were Black, and 7% were Hispanic. In unconditional modeling, the linear model fit well to the data (χ²=0.051, df=1, p>.05; RMSEA=.000, CFI=1.000, TLI=1.001). The overall depressive symptoms trajectory increased over a 4-year period and the variances in the intercept (B=2.287, SE=0.167, p<.001) of the trajectory was significant while variances in the slope (B=0.005, SE=0.021, p=.800) was not. The conditional growth model fit well (χ²=36.928, df=27, p>.05; RMSEA=.030, CFI=0.975, TLI=0.925). Middle-aged and older adult cancer survivors who were female (p<.001), had a lower annual income (p=.001), had a lower sense of control (p<.001), had poor self-reported health (p=.017), experienced cancer recurrence (p<.001), reported higher perceived discrimination (p<.001), and engaged in fewer physical activities (p=.008) were more likely to have a higher depressive symptoms at baseline, but the rate of increase in depressive symptoms was non-significant. Older adults (p=.021) as well as breast (p=.049) and lung (p=.030) cancer survivors were more likely to have higher increasing rates of depressive symptoms compared to their counterparts, while baseline depressive symptoms were not significantly different.

Conclusions and Implications: This study underscores the importance for public health social work interventions that increase physical activities for adult cancer survivors. It also suggests that the influence of perceived discrimination on health outcomes needs to be integrated into routine assessments of clients and patients, especially for people who experience discrimination based on race/ethnicity.