Bypassing the blues: Insomnia in the depressed post-CABG population

Abstract

Background: BACKGROUND: Recovery from coronary artery bypass graft (CABG) surgery often is complicated by depression and insomnia, resulting in poorer health-related quality of life and clinical outcomes. We explored the relationships among depression, insomnia, quality of life, and the impact of a collaborative care strategy on reducing insomnia in patients after CABG surgery.

Methods: METHODS: Patients with a Patient Health Questionnaire score ≥10 were randomized to nurse-delivered collaborative care for depression (n = 150) or their physician’s usual care (n = 152). A convenience sample of patients without depression (n = 151) served as the control group. Using the Hamilton Depression Rating Scale sleep questions, we created an “insomnia index.”

Results: RESULTS: At baseline, 63% of participants who were depressed vs 12% of those who were not depressed reported insomnia. Compared with usual care, fewer collaborative care participants reported insomnia at 8 months, and they tended to have a lower insomnia score (insomnia index change score −0.95 and −1.47, respectively; P = .05) with no time-by- randomization interaction, Cohen’s d = 0.22 (95% confidence interval, −0.001 to 0.43). Participants with baseline insomnia reported greater improvements in mental health–related quality of life (Medical Outcomes Survey 36-item Short Form Mental Component Summary score; −3.32, P = .02), but insomnia was not a significant moderator of the effect of collaborative care.

Conclusions: CONCLUSIONS: This is the first study to examine the long-term impact on insomnia among post-CABG patients treated for depression. Future collaborative care studies could consider including a therapeutic focus for insomnia.

Trial registration: ClinicalTrials.gov NCT00091962.

Keywords: Insomnia; depression; CABG; PHQ-9.

Figures

Figure 1

a. Predicted Probability of Insomnia Over Time by Randomization Status b. 8-Month Mean Change in Insomnia from Baseline by Randomization Status Model-based estimates of change in insomnia index (a: binary; b: continuous variable) by intervention status over 8 months. Point estimates and 95% confidence intervals derived from generalized linear mixed models with study arm, time point, their interaction, gender, education, presence of diabetes, percent ejection fraction, baseline depression, and presence of anxiety as fixed effects.

Figure 2

8-Month Change from Baseline in Mean SF-36 MCS By Baseline Insomnia and Randomization Status Model-based estimates of change in SF-36 MCS by intervention status and baseline insomnia over 8 months. Point estimates and 95% confidence intervals derived from mixed models with study arm, time point, baseline insomnia status, all two- and three-way interactions, gender, education, presence of diabetes, percent ejection fraction, baseline depression, and presence of anxiety as fixed effects. Abbreviation: SF-36 MCS, Short Form-36 Mental Component Summary