Target-Based Care: An Intervention to Reduce Variation in Postoperative Length of Stay

Abstract

To derive care targets and evaluate the impact of displaying them at the point of care on postoperative length of stay (LOS). A prospective cohort study using 2years of historical controls within a freestanding, academic children's hospital. Patients undergoing benchmark cardiac surgery between May 4, 2014, and August 15, 2016 (preintervention) and September 6, 2016, to September 30, 2018 (postintervention) were included. The intervention consisted of displaying at the point of care targets for the timing of extubation, transfer from the intensive care unit (ICU), and hospital discharge. Family satisfaction, reintubation, and readmission rates were tracked. The postintervention cohort consisted of 219 consecutive patients. There was a reduction in variation for ICU (difference in SD -2.56, P<.01) and total LOS (difference in SD -2.84, P<.001). Patients stayed on average 0.97 fewer days (P<.001) in the ICU (median -1.01 [IQR -2.15, -0.39]), 0.7 fewer days (P<.001) on mechanical ventilation (median -0.54 [IQR -0.77, -0.50]), and 1.18 fewer days (P<.001) for the total LOS (median -2.25 [IQR -3.69, -0.15]). Log-transformed multivariable linear regression demonstrated the intervention to be associated with shorter ICU LOS (β coefficient -0.19, SE 0.059, P<.001), total postoperative LOS (β coefficient -0.12, SE 0.052, P=.02), and ventilator duration (β coefficient -0.21, SE 0.048, P<.001). Balancing metrics did not differ after the intervention. Target-based care is a simple, novel intervention associated with reduced variation in LOS and absolute LOS across a diverse spectrum of complex cardiac surgeries.