Daytime symptom trajectories among adults with stable heart failure and insomnia: evidence from a randomised controlled trial of cognitive behavioural therapy for insomnia.

Summary: People with heart failure (HF) experience a high symptom burden and prevalent insomnia. However, little is known about daytime symptom trajectories after cognitive behavioural therapy for insomnia (CBT‐I). In this study we describe: (1) daytime symptom trajectories among adults with insomnia and stable HF over 1 year, (2) how symptom trajectories differ between CBT‐I versus HF self‐management interventions, and (3) associations between demographic, clinical, and sleep characteristics, perceived stress, health‐related quality of life (HRQoL), functional performance and daytime symptoms trajectories. We retrospectively analysed data from a randomised controlled trial of CBT‐I versus HF self‐management (NCT0266038). We measured sleep, perceived stress, HRQoL, and functional performance at baseline and symptoms at baseline, 3, 6, and 12 months. We conducted group‐based trajectory modelling, analysis of variance, chi‐square, and proportional odds models. Among 175 participants (mean [standard deviation] age 63.0 [12.9] years, 57.1% male, 76% White), we found four daytime symptom trajectories: (A) low improving symptoms (38.3%); (B) low psychological symptoms and high improving physical symptoms (22.8%); (C) high improving symptoms (24.0%); and (D) high not improving symptoms (14.9%). The CBT‐I versus the HF self‐management group had higher odds of belonging to Group A compared to other trajectories after controlling for baseline fatigue (odds ratio = 3.27, 95% confidence interval 1.39–7.68). The difference between the CBT‐I and the HF self‐management group was not statistically significant after controlling for baseline characteristics. Group D had the highest body mass index, perceived stress, and insomnia severity and the lowest cognitive ability, HRQoL, and functional performance. Research is needed to further evaluate factors contributing to symptom trajectories. [ABSTRACT FROM AUTHOR]