Scientist training program aimed at improving patient-centered research and care gets $5 million boost

PHILADELPHIA – A $5 million grant seeks to develop new, more efficient approaches to learn from the care delivered to patients across the University of Pennsylvania Health System and to train local scientists in the principles of “Learning Health System science.” The Agency for Healthcare Research and Quality (AHRQ) and the Patient-Centered Outcomes Research Institute (PCORI) announced the funding in January. The University of Pennsylvania and 15 other research centers are part of this five-year investment in Learning Health System Embedded Scientist Training and Research Centers across the United States.

AHRQ defines a learning health system as a health system in which internal data and experience are systematically integrated with external evidence, and that knowledge is put into practice. Doing so requires having infrastructure in place to implement research findings and optimize care as quickly, broadly, and equitably as possible.

The principal investigators for the project are Scott D. Halpern, MD, PhD, the John M. Eisenberg Professor of Medicine, Epidemiology, and Medical Ethics and Health Policy and founding director of the Palliative and Advanced Illness Research Center (PAIR); Meghan Lane-Fall, MD, MSHP, the David E. Longnecker Associate Professor of Anesthesiology and Critical Care, an associate professor of Biostatistics, Epidemiology, and Informatics, and the executive director of the Penn Implementation Science Center; and M. Kit Delgado, MD, MS, an associate professor of Emergency Medicine and director of the Penn Medicine Nudge Unit.

“As electronic health records have become ubiquitous, and machine learning and other statistical methods have been developed to harness their potential, we now have enormous opportunity to use these data to rapidly find new ways to deliver care that fairly promotes the goals of all patients,” Halpern said. “This requires a different approach and scale than traditional clinical research, so we need to both expand training for this type of work and ensure that it’s integrated with clinical operations teams from the outset in order to realize this full potential.”

Learning health system studies often entail large-scale, randomized clinical trials, or carefully designed observational studies comparing different ways of delivering care. Because most such studies either compare approaches that doctors already use or examine the extra benefits patients might experience if standard care practices were used more often or supplemented with new approaches, they are typically considered low-risk – which is different, for example, from Phase I clinical trials of new drugs, where the first goal is to establish safety.

Learning health system studies can also generate a lot of information in a relatively short period of time because they occur in the context of regular clinical care and use routinely collected data, such as information from electronic health records. These types of studies often focus on the quality, delivery, and/or equity of care. For example, recent projects led by Penn experts that follow a learning health system model have included:

• COVID Watch, a system designed to monitor COVID-19 outpatients using automated texts backed by a small team of nurses. Patients who enrolled in the program were 68 percent less likely to die from COVID. A follow-up trial found adding home pulse oximetry did not improve outcomes further, challenging conventional intuitions and saving enormous resources. Both studies were also funded by PCORI.
• Heart Safe Motherhood, a remote postpartum blood pressure monitoring program that reduced readmission rates for postpartum hypertension and eliminated racial disparities in postpartum blood pressure assessment.
• The largest-ever study of palliative care, which found that “default orders” in the electronic medical record nearly tripled palliative care consultation rates for seriously ill patients in the hospital, and enabled patients to receive such care sooner.

“We look forward to building on our collective experience with integrating learning health system efforts at Penn Medicine and continuing to train more scientists and clinical leaders to carry out this type of work,” said Lane-Fall, who co-led a prior training program at Penn supported by AHRQ and PCORI.

With the new funding, the team will establish the Penn PORTAL (Patient-Oriented Research and Training to Accelerate Learning), which will unite behavioral, data, and implementation scientists from across Penn, including experts from the Perelman School of Medicine, School of Nursing, College of Engineering, and Wharton School of Business; leaders in the design, analysis, and ethics of randomized trials; clinical operations leaders; longstanding community partners; and Accelerate Health Equity, a city-wide initiative aimed at addressing health equity and racism that involves Philadelphia health systems and Independence Blue Cross. Penn PORTAL aims to train a diverse group of scientists who go on to become learning health systems leaders at Penn and elsewhere and to support the implementation of unbiased, patient-centered care delivery at hospitals throughout Philadelphia.

Penn PORTAL will test interventions that have been prioritized among the project stakeholders for their potential impact to the patient population in Philadelphia. Priority focus areas will include maternal morbidity, primary care, and care for patients with multiple chronic conditions.

“With Penn PORTAL, we’re tapping into the incredible range of expertise available both within the Penn community and across our city,” Delgado said. “Ultimately, we want to improve patient outcomes in Philadelphia—and the nation—no matter where those patients are treated. By partnering with a broad coalition of stakeholders and using disciplined, user-centered design and innovation methods from the beginning, we hope to quickly make an impact on patient outcomes far and wide.”

Other key members of Penn PORTAL include Jaya Aysola MD, DTMH, MPH, core leads Michael Harhay, PhD, MPH; Kevin B. Johnson, MD, MS; Jennifer Myers, MD; and Rachel M. Werner, MD, PhD; and steering committee members Kathryn Bowles, PhD, RN; Shivan Mehta, MD, MS; Raina Merchant, MD, MS; Matthew Press, MD, Marylyn Ritchie, PhD; and Sindhu Srinivas, MD, MS.

This work is supported by the AHRQ and PCORI (P30HS029745).