Opportunity 

Patients with cancer often receive care that is discordant with their wishes, resulting in inappropriate and costly resource utilization. Evidence demonstrates that proactive serious illness conversations (SICs) can help clinicians align care with patient goals. However, many patients with advanced cancer die without ever having a SIC.   

Intervention 

We piloted an intervention that paired nudges with a machine learning algorithm that predicts the risk of six-month mortality to see if we could increase SICs between clinicians and cancer patients. Each week, clinicians were given a list of patients they were scheduled to see who had a high mortality risk.  

Then, using an opt-out approach, we asked clinicians to pre-commit to having SICs and sent text message reminders on the day of the visit. Clinicians also received weekly emails comparing their SIC rates to their peers' rates. To assess the impact of the intervention, we designed a stepped-wedge cluster randomized controlled trial. 

Impact 

Combining machine learning mortality estimates with behavioral nudges led to a fourfold increase in SICs. Among subgroups, Medicare beneficiaries had the greatest increase in SIC rates. We also observed an increase in the overall rate of SICs among all patients during the intervention period.

Systemic therapy, which has been linked to lower quality of life, was less common among patients in the intervention group. Other end-of-life outcomes such as inpatient death, ICU admission, and hospice enrollment, were unchanged.

The intervention from this study has been implemented across all oncology practices in the University of Pennsylvania Health System to enhance experience and quality of life for cancer patients at high risk for mortality and better align care with patient values and goals.