Do posttraumatic stress symptoms predict trajectories of sleep disturbance and fatigue in patients with breast cancer? A parallel-process latent growth model.

Abstract

Using a parallel-process latent growth model (LGM), this study examined whether posttraumatic stress symptoms (PTSS) are associated with the trajectory of sleep disturbance (SD) and fatigue and whether the SD trajectory mediates the PTSS-fatigue relationship. Data were from 215 patients with breast cancer recruited from a tertiary hospital in South Korea. A self-report survey was administered at four time points during the course of adjuvant chemotherapy. The mean age of the participants was 46.69 (SD=9.08) and the majority was at stage I and the average months since diagnosis was 1.33 (SD=1.43). Unconditional parallel-process LGM indicated that SD and fatigue were positively associated with each other, both in terms of initial status and growth rate. Then, the conditional parallel-process LGM with baseline PTSS (i.e., avoidance, intrusion, and hyperarousal) as predictors were examined and anxiety, depressive symptoms and chronotype were entered as covariates in the model. Results indicated that a higher initial status and faster growth of SD were associated with a faster increase in fatigue. Greater baseline hyperarousal was directly related to a higher initial status and a slower increase in SD, and higher initial fatigue. Furthermore, a higher hyperarousal was associated with a greater initial SD, which was related to a faster increase in fatigue. Additionally, the late chronotype was related to a faster increase in fatigue through its impact on the initial SD. The detrimental impact of hyperarousal on the SD trajectory and fatigue suggests the need to intervene in PTSS and SD early and throughout the course of cancer treatments to prevent fatigue.