Mackenzie Shanahan Dissertation
Older adults with persistent pain experience reduced physical functioning, increased disability, and higher rates of depression. Previous research suggests that different types of positive and negative expectancies (e.g., optimism and hopelessness) may be associated with the severity of these pain-related outcomes. Moreover, different types of expectancies may interact with perceived control to predict these outcomes. However, it is unclear whether different types of expectancies are uniquely predictive of changes in pain-related outcomes over time in older adults and whether perceived control moderates these relationships. The primary aims of the current study were to 1) examine how the shared and unique aspects of optimism and hopelessness differentially predict changes in pain-related outcomes (i.e., pain severity, pain interference, disability, and depressive symptoms) in older adults experiencing persistent pain over a 10-year and 2-year timeframe and 2) examine whether perceptions of control over one’s health moderate these relationships. The present study sampled older adults with persistent pain who participated in a nationally representative, longitudinal study (i.e., The Health and Retirement Study) at three timepoints across a 10-year period. First, confirmatory factor analyses (CFA) were conducted to determine appropriate modeling of expectancy variables. Second, mixed latent and measured variable path analyses were created to examine the unique relationships between expectancy variables and changes in pain-related outcomes over both a 10- year and 2-year period. Finally, mixed latent and measured variable path analyses and PROCESS were used to test perceived control as moderator of the relationships between expectancy variables and changes in pain-related outcomes over time. CFA results suggested that measures of optimism and hopelessness were best understood in terms of their valence, as positive (i.e., optimism) or negative (i.e., pessimism and hopelessness) expectations. Results from path analyses suggested that only negative, not positive, expectancies were significantly associated with worsening pain severity, pain interference, disability, and depressive symptoms across both 10-year and 2-year periods. Moderation analyses demonstrated inconsistent results and difficulties with replication. However, post-hoc path analyses found that perceptions of control over one’s health independently predicted some changes in pain-related outcomes over time, even when controlling for expectancies. Altogether, the current findings expand our knowledge of the associations between expectancies and pain by suggesting that negative expectancies are 10 predictive of changes in mental and physical pain-related outcomes across years of time. The current study also suggests that positive and negative expectancies may be related, but distinct factors in older adults with persistent pain and that health-related perceived control may be predictive of changes in pain over time. The current discussion reviews these extensions of our current knowledge in greater detail, discusses the potential mechanisms driving these relationships through a theoretical lens, and identifies the implications of this work.