The prevalence, predictors, and clinical impact of postoperative sleep disturbance: A prospective cohort study

ABSTRACT Background: Postoperative sleep disturbance (POSD) is a pervasive and an overlooked complication; however, the relative importance of its modifiable drivers and their direct impact on clinically relevant outcomes remains inadequately quantified. This study aimed to determine the prevalence of POSD, identify its key modifiable predictors, and investigate its association with opioid consumption and hospital length of stay (LOS). Methods: A prospective observational cohort study was conducted at a tertiary care center. We enrolled 143 adult patients undergoing elective surgery under general anesthesia. Sleep quality was assessed on the first postoperative morning using a modified version of the Pittsburgh Sleep Quality Index (PSQI). Multiple linear regression identified predictors of sleep duration. Post-hoc analyses included a Pearson correlation between global PSQI score and 24-hour opioid consumption (Morphine Milligram Equivalents, MME), and a Mann–Whitney U test to compare LOS between good (PSQI ≤ 5) and poor sleepers (PSQI > 5). Results: The prevalence of poor sleep quality was 89.5% (128/143). Regression analysis revealed that sleep disturbance (β = 0.702, P < 0.001), sleep efficiency (β = 0.288, P < 0.001), and sleep latency (β = 0.224, P = 0.005) were significant predictors of sleep duration (R² = 0.348, P < 0.001). A significant positive correlation was found between global PSQI score and 24-hour MME (r = 0.41, P < 0.001). Patients with poor sleep had a longer median LOS of 4 days (IQR: 3–6) compared to 3 days (IQR: 2–4) in patients with good sleep ( P = 0.018). Conclusion: POSD is highly prevalent and primarily driven by modifiable, environment-related factors. Its strong associations with increased opioid requirements and prolonged hospitalization underscore the critical need to implement nonpharmacological sleep promotion protocols as a standard component of postoperative care.