Sedative-naive older adults who receive a new sedative prescription after an ICU stay often continue taking the drugs, according to a recent study.
Researchers in Ontario performed a population-based cohort study of ICU survivors at least 66 years of age who had not filled sedative prescriptions within six months prehospitalization from April 2003 through September 2019. They used multilevel logistic regression to determine the association of demographic, clinical, and hospital characteristics with a new sedative prescription within seven days of discharge. The researchers quantified between-hospital variation using the adjusted median odds ratio and examined factors associated with persistent prescriptions, defined as those continued up to six months, using a multivariable proportional hazards model. The study results were published Jan. 4 by CHEST.
Overall, 250,428 patients were included in the study. Mean age was 75.8 years, and 61% were male. During hospitalization, 63.6% of patients had surgery, 26.3% received invasive mechanical ventilation, and 14.8% had sepsis. A total of 15,277 patients (6.1%) filled a new sedative prescription within seven days of discharge, with rates varying from 2% (95% CI, 1% to 3%) to 44% (95% CI, 3% to 57%) across hospitals. Benzodiazepines were most common, filled by 8,824 (3.5%) patients, followed by a nonbenzodiazepine (n=2,749 [1.1%]), an antipsychotic (n=2,745 [1.1%]), and multiple sedative classes (n=959 [0.4%]). Persistent sedative prescriptions were filled by 8,458 patients (3.4%).
Patients were more likely to receive a new sedative prescription if they were discharged to a long-term care facility (adjusted odds ratio [aOR], 4.00; 95% CI, 3.72 to 4.31), received inpatient geriatric or psychiatric consultation (aORs, 1.95 [95% CI, 1.80 to 2.10] and 2.76 [95% CI, 2.62 to 2.91], respectively), received invasive ventilation (aOR, 1.59; 95% CI, 1.53 to 1.66), and stayed seven days or longer in the ICU (aOR, 1.50; 95% CI, 1.42 to 1.58). These factors were also associated with persistent sedative use, as were female sex (subdistribution hazard ratio, 1.07; 95% CI, 1.02 to 1.13) and pre-existing polypharmacy (subdistribution hazard ratio, 0.88; 95% CI, 0.80 to 0.93). Residual heterogeneity between hospitals (adjusted mean odds ratio, 1.43; 95% CI, 1.35 to 1.49) was more strongly associated with new sedative prescriptions than with Charlson Comorbidity Index score or sepsis.
The researchers said the most important limitation of their study was the inability to link postdischarge prescriptions to sedative exposure during ICU admission. They noted that variability in prescribing practices across hospitals suggests the potential for interventions, such as targeted discharge medication review involving community-based physicians and pharmacists. “Detailed review of sedative pharmacotherapy after an ICU admission can enhance postdischarge pharmacotherapy regimens and communication for survivors, their families and multiple care teams to support recovery,” they wrote. “Given the many critically ill patients discharged every year, even small changes in prescribing practices may yield large benefits across a healthcare system.”