Sleep disturbance in orthopaedic trauma patients

Objective: To evaluate the prevalence and severity of sleep disturbance experienced by patients who have sustained a traumatic orthopaedic injury, how sleep disturbance affects the patient's perceived health quality, and identify factors associated with sleep disturbance. Design: Cross-sectional cohort study. Setting: Urban Level I trauma center. Patients/Participants: Three hundred thirty-five nonconsecutive patients who presented to clinic at various stages of treatment for their traumatic orthopaedic injuries. Main Outcome Measurements: Pittsburgh sleep quality index (PSQI) and 36-item short form-36 (SF-36) survey questionnaires; injury severity score (ISS). Results: The average PSQI score was 10.3 (64.8). Two hundred eighty-eight (86.0%) patients had a PSQI score $5, indicating the presence of sleep disturbance. The PSQI score was $10 in 183 (54.6%) patients, which is sleep disturbance similar to the level seen in clinical depression. Patients reported an average sleep latency of 38.9 (637.5) minutes, with a total nightly sleep time of 6.3 (61.9) hours. Univariate statistical analysis demonstrated that age, time since injury, and all components of the SF-36 were significantly associated with increased PSQI scores. When these variables were assessed with multivariate analysis to control for confounding variables, the bodily pain, vitality, and mental health components of the SF-36 remained independently associated with PSQI (P ≤ 0.001, 0.002, and 0.001, respectively). ISS measurements at the time of presentation were not associated with PSQI scores. Conclusions: Our findings suggest that sleep disturbance is both highly prevalent (86% PSQI $5) and severe (54.6% PSQI $10) in patients recovering from a traumatic orthopaedic injury. The bodily pain, vitality, and mental health components of the SF-36 were independently associated with worse sleep quality. The average orthopaedic trauma patient presents with a sleep score similar to that seen in clinical sleep disorders and clinical depression. Interestingly, in our study, the severity of the overall injury burden as measured by ISS and time since injury were not independently associated with the severity of sleep disturbance, as one might expect.