Quality improvement initiative: use of the STOP-BANG score and monitoring to reduce adverse events in hospitalised patients at risk of obstructive sleep apnoea

Abstract

Background: Obstructive sleep apnoea increases risk of respiratory depression with administration of sedatives, narcotics or anxiolytics. To reduce adverse events during hospital admission, we implemented STOP-BANG screening to prompt respiratory monitoring for inpatients receiving these medications. This study reports on protocol development, implementation and an initial analysis over 5 years to evaluate implementation success and outcomes.

Intervention: The STOP-BANG measure was embedded in the nurse navigator at admission. If the score was ≥3 and sedatives, narcotics and/or anxiolytics were ordered, the provider was prompted to monitor patients with continuous pulse oximetry and/or capnography.

Methods: We assessed the impact of the intervention using a retrospective pre-post design. Preprotocol data from all adult inpatients over a 2.5-year period, and postprotocol data from all adult inpatients from over a 5-year period, were extracted from the electronic health record. Outcomes included use of monitoring; adverse events during hospitalisation were included to evaluate the effects of the intervention: mortality, rate of rapid response team events, reversal and/or rescue, intensive care unit admission and orders for positive airway pressure equipment.

Results: The combined preprotocol and postprotocol sample included 254 121 patients. After protocol implementation, overall mortality for patients receiving sedatives, narcotics or anxiolytics decreased slightly from 2.1% to 1.9% (p<0.001). In the postprotocol cohort only (n=193 744), monitored patients had a higher probability of experiencing all adverse events. Among monitored patients, mortality was lowest in the high-risk group (STOP-BANG≥5).

Discussion: Triaging by STOP-BANG coupled with monitoring appeared to be helpful for patients at highest risk of obstructive sleep apnoea. Given the complexity of obstructive sleep apnoea, further pursuit of subphenotypes is warranted.

Keywords: Anaesthesia; Chronic disease management; Quality improvement; Quality improvement methodologies.

Figure 1. The dashed vertical line indicates not monitored patients as the reference group. The figure shows that the difference between monitored and not monitored patients for each STOP-BANG stratum is reduced when adjusted for general sickness (ROM). These data suggest that differences across STOP-BANG levels and monitored versus not monitored may be more due to some other factor (captured in the ROM) than to OSA risk or to monitoring alone. STOPBANG score of 0-2 was low risk, 3-4 is moderate risk and 5+ is high risk.OSA, obstructive sleep apnoea.