Phase of Care Mortality Analysis According to Individual Patient Risk Profile.

Background: Phase of care mortality analysis (POCMA) is a quality improvement tool categorizing triggers for mortality into phases of patient care. However, the relationship between a patient's risk profile and the triggers for mortality is incompletely understood.

Methods: POCMA was implemented for cases with available Society of Thoracic Surgeons (STS) risk models. Risk-adjusted rates were obtained from the STS database. Mortality triggers were categorized by the phase of occurrence (preoperative, intraoperative, intensive care unit [ICU], postoperative floor, and discharge). Patients were then stratified by STS risk score: low risk (<4%), intermediate (4% to 8%), and high risk (>8%).

Results: A total of 3,919 isolated coronary artery bypass grafting (CABG), 901 isolated valve, and 321 CABG plus single-valve procedures were performed from 2012 to 2018, with 4.6% crude mortality and a median STS risk score of 5.8%. POCMA was performed on 67 patient mortalities, with triggers occurring in the following phases of care: 49.3% preoperative, 13.4% intraoperative, 23.9% ICU, 3.0% postoperative floor, and 10.4% discharge phase. Mortality distribution was bimodal, occurring mostly in low-risk (37.3%) and high-risk (38.8%) patients. For low-risk patients, the trigger for mortality most frequently occurred during the postoperative ICU phase, while for high-risk patients, the trigger for mortality most frequently occurred during the preoperative phase.

Conclusions: Mortality had a bimodal distribution with respect to patient risk profile. Phase-of-care triggers for mortality differed according to patient risk profile: low-risk triggers during the postoperative ICU phase versus high-risk triggers typically during the preoperative phase. Specific focus on phases according to the patient's risk profile represents an opportunity to improve quality and outcomes.