Clinical decision support tool reduces routine nighttime vital sign checks but does not affect clinical outcomes

Référence

Najafi N, Robinson A, Pletcher MJ, Patel S. Effectiveness of an analytics-based intervention for reducing sleep interruption in hospitalized patients. JAMA Intern Med 2022;182(2):172-177.

Sommaire

Sleep deprivation due to interruptions such as routine vital sign checks is a common phenomenon for hospitalized patients. In this single-center trial, investigators created a predictive algorithm to identify hospitalized patients who were at low risk of having abnormal nighttime vital signs. The algorithm was based on patient age, day of hospitalization, laboratory values, daytime vital signs, and prior nighttime vital signs. Patients admitted to a general medicine service who were identified by the algorithm as being at low risk of having abnormal nighttime vital signs were randomized into an intervention group (n = 966) or a usual care group (n = 964). In the intervention group, primary care team members received clinical decision support in the electronic health record to discontinue nighttime vital sign checks. In the usual care group, there was no such clinical decision support. Patients in the 2 groups had similar demographics. Overall, almost twice the number of patients in the intervention group received an order to discontinue nighttime checks than in the control group (80% vs 45%; P < .001). Although this resulted in a 31% reduction in nighttime checks in the intervention group, there was no significant change in the primary outcome of delirium. There were also no significant differences detected in secondary safety outcomes, including rapid response calls, transfers to the intensive care unit, or death. Of note, despite an order to discontinue nighttime checks, 35% of the encounter-nights with discontinuation orders had nighttime vital sign checks performed anyway.