A University of Pennsylvania team and Tennessee Oncology have won a $600,000 2-year grant from the Emerson Collective for a project that will leverage behavioral economics and predictive analytics to improve oncology patient health outcomes.
The project will be led by Penn Center for Cancer Care Innovation’s (PC3I) Ravi Parikh, MD, MPP, FACP, (PI) and Justin Bekelman, MD, (Co-Investigator) who are both CHIBE-affiliated faculty members, along with Tennessee Oncology’s site PI Sandhya Mudumbi, MD, and Stephen Schleicher, MD, MBA.
The issue they seek to address is that patients with advanced cancer often experience severe symptoms and poor quality of life, and instead of receiving palliative care that is more in line with their goals, 40% of patients receive aggressive end-of-life care, such as chemotherapy. The investigators will examine whether default algorithm-based referrals and peer comparisons could help increase palliative care visits and decrease aggressive end-of-life utilization.
“Specialty palliative care is one of the most evidence-based practices in all of oncology, yet is underutilized for patients with advanced cancer,” Dr. Parikh said. “Our hypothesis is that default referrals for high-risk patients identified by predictive algorithms can increase access to palliative care in a scalable way. We are excited to partner with one of the largest community oncology practices in the country to test this in a randomized trial.”
The investigators identified three biases from behavioral economics that may be at play in this area of oncology:
- Status quo bias
- “Status quo bias, which predisposes clinicians to continue current practice even if it is not the optimal option, may lead to delayed or missed palliative care referrals,” they stated.
- Optimism bias
- “Optimism bias, the cognitive bias that causes clinicians to believe that their own patients are at lesser risk of negative outcomes, may cause clinicians to underestimate a patient’s mortality risk, thus delaying palliative care referral.”
- Overconfidence bias
- “Overconfidence bias, the propensity to overestimate one’s desired behaviors when it is not objectively reasonable, may lead clinicians to incorrectly believe they are initiating similar or more palliative care referrals than their peers.”
The team anticipates a few ways that behavioral economics tools could potentially help overcome these biases:
- Default, opt-out nudges may encourage clinicians to refer patients to palliative care.
- “Triggered” identification of high-risk patients could help counteract optimism bias.
- Social comparisons – showing how clinicians compare to their peers – could help overcome overconfidence bias.
The randomized trial will involve 400 patients with advanced lung and gastrointestinal malignancies and 40 clinicians at four practices within Tennessee Oncology, which is one of the largest community-based practices in the country and member of OneOncology, a network of community practices nationwide. The investigators will assess completions of a palliative care visit within 3 months among high-risk patients.
“This is a really innovative collaboration between an academic institution and community oncology practice to accelerate real-world results,” Dr. Parikh said. “The majority of patients with cancer receive their care in community oncology settings; yet, most clinical trials are performed in academic settings. We have an opportunity to work with community oncology leaders to co-design a unique program that leverages state-of-the-art algorithms, electronic health record infrastructure, and behavioral economics principles to increase access to palliative care. If successful, this could serve as a truly generalizable model across all of oncology care.”