Predictors of oversedation in hospitalized patients

Abstract

Purpose: Results of a study to determine demographic and clinical characteristics predictive of oversedation and potential opioid-induced respiratory depression (OIRD) in hospitalized patients are reported.

Methods: In a retrospective case-controlled study, an incident reporting database was searched to identify cases of in-hospital oversedation; to form the control group, patients who did not experience an oversedation event while hospitalized were sampled in reverse chronological order until the desired total sample size (n = 225) was obtained. An allocation ratio of 2:1 was specified to adjust for case variability. Binary logistic regression was employed to identify factors predictive of oversedation.

Results: Female sex (odds ratio [OR], 2.41; 95% confidence interval [CI], 1.05-5.50), comorbid renal disease (OR, 4.22; 95% CI, 1.66-10.70), untreated sleep apnea (OR, 32.32; 95% CI, 2.72-384.72), receipt of long-acting oxycodone (OR, 4.76; 95% CI, 1.70-13.33), and as-needed use of hydromorphone (OR, 2.73; 95% CI, 1.19-6.27) were significant predictors of oversedation; as-needed analgesia administered by the oral route (OR, 0.16; 95% CI, 0.07-0.36) or i.v. route (OR, 0.33; 95% CI, 0.14-0.80) had a significant protective effect. The final prediction model explained 47.8% of variance in oversedation risk and was found to have strong discriminatory performance.

Conclusion: The identified risk factors for oversedation and potential OIRD in hospitalized patients can form the basis of quality-improvement initiatives to prevent oversedation through improved prescribing and patient monitoring.

Keywords: logistic regression; opioids; oversedation; patient safety; respiratory depression.