Using Electronic Health Record Mortality Data to Promote Goals-of-Care Discussions in Seriously Ill Transferred...

Area of Science:

• Medical Informatics
• Clinical Decision Support
• Palliative Care

Background:

• Identifying high-risk patients for early goals-of-care discussions (GOCD) is crucial but under-evaluated.
• A novel model predicts 30-day inpatient mortality, potentially facilitating early GOCD.

Purpose of the Study:

• To assess the impact of sharing mortality risk predictions with hospitalists on GOCD planning and timing.
• To evaluate hospitalist agreement with predicted mortality risk.
• To measure the association between risk predictions and patient mortality.

Main Methods:

• A nonrandomized quasi-experimental pilot study with a historical control group.
• Model-generated mortality risk was communicated to hospitalists on day two of hospitalization.
• Data collected via mobile communication system and chart review.

Main Results:

• Hospitalists agreed on mortality risk for all intervention patients.
• GOCD planning was significantly higher in the intervention group (p < 0.001).
• GOCD completion within 72 hours was higher in the intervention group (78%) vs. control (50%).
• Both model-generated and hospitalist risk assessments correlated significantly with mortality.

Conclusions:

• Communicating model-generated mortality risk to hospitalists promotes timely GOCD.
• This approach shows promise for improving end-of-life care discussions in high-risk patients.