The Nudge Unit Prompts Clinicians to Align with Patient Goals

A physician receives a text alert to zero in on patient goals to better provide care.

A University of Pennsylvania Health System pilot program called, “The Nudge Unit,” is increasing patient-centered conversations to improve their healthcare outcomes. Electronic nudges based on prognosis and strategic goals are sent via text or email to medical staff. These subtle triggers are proving to be successful. From emergency room treatment to end-of-life decision making, nudges are quickly becoming a valuable healthcare tool in 21st century medicine.

The Nudge Unit is the world’s first behavioral design team embedded within a health system. Computer-calculated prompts provoke clinicians to initiate patient dialogue as means to achieve optimal medical outcomes. The unit is a prime example of a growing body of research in behavioral economics: a field of inquiry that uses principles of economics and psychology to understand how individuals make decisions and then use those insights to help people make choices consistent with their own long-term interests.

The Nudge Unit’s algorithm is designed to guide good decision-making toward understanding and meeting patient healthcare objectives.

The Nudge Unit is designed to identify potential patient targets based on algorithms gleaned from their clinical data and electronic health records. The goal is to trigger the medical team to engage in serious illness conversations, or patient driven dialogues because these don’t happen as often as they should. These types of exchanges are integral for clinicians to understand a person’s values, priorities, and preferences, so best decisions can be made to meet patient goals.

A randomized study of the unit supported by the National Institutes of Health and the Penn Center for Precision Medicine proved to triple rates of serious illness conversations for more than 20,000 cancer patients. These conversations were shown to reduce the use of aggressive treatments and unnecessary hospitalizations for patients with end-stage disease of any kind. Instead of treatments aimed at extending life, the focus became reducing symptoms like pain and nausea, ultimately improving patient quality of life.

Co-lead author of the study, Ravi B. Parikh, MD is confident the technology will prove to be valuable in a myriad of clinical care settings. Dr. Parikh states, “Within and outside of cancer, this is one of the first real-time applications of a machine learning algorithm paired with a prompt to actually help influence clinicians to initiate these discussions in a timely manner, before something unfortunate may happen.” By default, the Nudge Unit is positively affecting clinician behavior in this regard. The study shows that medical personnel become more apt at eliciting these conversations and initiatives on their own after utilizing it.

The Nudge Unit spurs healthcare providers to initiate a patient-centered approach to care

In a more recent study, the Nudge Unit spurred increased hospital emergency department treatment of Opioid Use Disorder. Electronic prompts sent to triage nurses boosted their ability to identify those patients suffering from withdrawal. Subsequently, recovery-focused care was provided to dozens of patients in need. Researchers hope the unit will increase implementation of evidence-based treatment in emergency department settings for those with the disorder.

The Penn Medicine Center for Health Care Innovation hosts its fourth “Nudges in Health Care Symposium” in Philadelphia, Pennsylvania in September 2023. Their event page states that stakeholders from health systems nationwide will come together to share insights about creating behaviorally enabled organizations. Keynote addresses, research presentations and panel discussions will highlight how machine-learned nudges and more can maximize positive impact on patient care. Dr. Parikh holds hope for future mainstream use of this type of automation: “Because of its success, I think we’ve provided of a road map for other institutions that may be thinking of using analytics to drive important behaviors.”