Accenture: Tell us about your role and responsibilities at Penn Medicine.

Roy Rosin: As chief innovation officer, I co-lead the Center for Innovation with Dr. David Asch, a respected researcher and clinician. We’ve built a team that incorporates different skills and perspectives, including clinicians who are entrepreneurial in their approaches to changing care delivery, designers, engineers, project managers, data analysts and others who are well versed in innovative methods. We test novel ideas quickly at low cost to overcome inertia and see if these ideas will deliver value.

We do all of this in conjunction with partners that range from information services, who have provided a foundation of systems and patient data, to legal, privacy, our CMIO, front line care providers and clinical operations. We’re making good progress towards our goal of a culture of rampant, rigorous experimentation. Our team looks out for where the new ideas and approaches get stuck and we methodically identify and eliminate points of friction.

A: Your role centers on innovation, but innovation means different things to different people. What does innovation mean to you?

RR: I define it in contrast to creativity and invention. Creativity is coming up with new ideas. Invention is bringing new ideas into the world. Innovation captures value from taking new ideas and approaches and casting them in a way that moves the needle, such as reducing readmission or infection rates. Having a new idea alone is not innovation.

Creativity is coming up with new ideas. Invention is bringing new ideas into the world. Innovation captures value from taking new ideas and approaches and casting them in a way that moves the needle, such as reducing readmission or infection rates. Having a new idea alone is not innovation.

A: Describe the innovation process at Penn Medicine. Are there fundamental principles you apply to each project?

RR: It’s a methodological approach with extremely fast experimentation or rapid validation of a belief or hypothesis at the core. When people have ideas, but little time or money, they might not get started. But if I can do something quickly to generate evidence, I can get started and overcome inertia.

Most of what we do is in the realm of contextual interaction testing. We ask questions like, ‘how would people use or interact with a given design – one that’s maybe just on paper for now - if it really existed?’ or ‘if people take this action we want, will it change the outcome the way we expect?’ We create prototypes that people play with. For instance, we’ll build a fake front end on paper or a fake back end where there’s actually a human being replicating what would happen if a real system were working automatically, such as a person pretending to be an automated robot texting back answers to a patient to test an AI bot concept. This allows us to test cheaply because it’s not built to scale and can be changed quickly multiple times each day—or week—as necessary. The goal at this point is still to figure out what works and get it right.

Most innovation efforts fail because they were scaled prematurely. The innovation started in the wrong direction and the team then executes by scaling in the wrong direction. Of innovations that are successful, 9 out of 10 of those started in the wrong direction. Of the ones that had potential to succeed, deliberate execution may have sent them off the cliff. You have to iterate quickly to find the right direction at a small scale, then scale what works.

A: What are the biggest healthcare challenges you are trying to solve for?

RR: So many areas need redesign. Readmission is one. We signed a contract with Independence Blue Cross where we’ve essentially warrantied 30-day readmissions, so Penn is now at risk if a patient comes back to the hospital within 30 days. A patient does not want to be back in the hospital, and both payers and providers want them to have better outcomes. Sick patients may be with us a few hours or days a year, but they spend another 5,000 hours awake and living their day-to-day life. If we don’t know what they are doing during that time, it minimizes our opportunity to intervene.

One of our core approaches to rethinking health care delivery involves what we call automated hovering, or remote surveillance, which gives us insight into how people are doing beyond the hospital or clinic’s walls and the chance to intervene. It might be a scale, blood pressure cuff or pillbox sending us data or the patient texting with us, but we’re seeing and finding out about things we never used to know, allowing us to change a patient’s experience or outcomes.

We’ve helped incubate several programs based on this concept of remote sensing. Towerview Health is a startup taking this approach, achieving some of the highest adherence rates we’ve seen. Through remote sensing and automated hovering, we can know when a patient is taking his or her pills, communicate with them or a spouse if necessary, and drive higher rates of medication adherence. That, in turn, leads to better health outcomes.

A: Are there particular innovation outcomes you are most proud of?

RR: Preeclampsia – or high blood pressure related to pregnancy – is very difficult to manage, as it is a silent disease. Katy Mahraj, Dr. Sindhu Srinivas and Dr. Adi Hirshberg developed an effective program in this space called Heart Safe Motherhood, addressing this severe condition when others weren’t sure it was solvable. Health systems had tried a number of solutions from walk-in clinics to follow-up phone calls that didn’t enable keeping women safe, as clinicians weren’t able to get the required insight into women’s blood pressures in time to avoid bad outcomes, and in fact, postpartum preeclampsia was the top driver of morbidity and 7-day readmissions for this maternal population. The standard of care is having two blood pressure readings within one week of discharge. Historically clinicians haven’t had that data for any patients.

Working with at-risk women, the team learned that their preferred method of communication was texting. We explained the program and sent them home with a blood pressure cuff. Then we began a texting protocol where instead of a technology system, Dr. Hirshberg manually acted in the role of being the system we might eventually build. Using this ‘fake back end’ approach we were able to test different communications quickly at low cost to identify the right interaction with these women. We learned both how to engage this patient population, but also how to design the care model to respond to the blood pressure information they were now texting to us, allowing early interventions.

Once we got it working, we went from having the target two blood pressure values that first week for zero patients to getting values from 82 percent of the women, and the newly defined protocols did allow our care team to respond effectively to the information coming in from this remote monitoring. The high readmission rate dropped to almost zero. Heart Safe Motherhood has since become a best practice care model to keep women safe, now being licensed to other systems beyond Penn’s walls.

A: Are there any current or historical innovators that you admire, and why?

RR: Scott Cook, co-founder of Intuit, has been an inspiration and a mentor. He’s not only an entrepreneur who founded a multi-billion-dollar enterprise helping tens of millions of people around the globe, but is a lifelong learner who built a culture of continuous learning at Intuit. Scott is an expert at getting teams embedded for deep, contextual learning so they can see what others have missed and have the empathy to design experiences that delight customers. I admire him for both his focus on leading innovation that really matters in consumers’ and small business owners’ lives and for his passion for always finding better, more effective methods. Finally, both Scott and Intuit’s CEO Brad Smith have created a high integrity culture with diversity, transparency and personality. We need more companies like that!

Scott always says, “Fall in love with the problem, not the solution.” That mantra reminds me to keep learning and iterating even when executing against deadlines, as getting embedded in the life of the people we’re trying to help, and trying to walk in their shoes to understand what they’re experiencing, will continue to evolve and refine the solutions we’re designing. He taught me methods like contextual inquiry – following a consumer into their home or to work to see a problem in context. You learn so much more by being embedded, including the fact that what people say they do often diverges from what they really do; so contextual observation beats interviews and focus groups any day. Falling in love with your solution and executing against an initial direction fails more often than not. More than 9 out of 10 successful innovations started in the wrong direction. Remember, YouTube was originally going to be an online dating service.

A: What do you like most about your job?

RR: I love working with mission-driven clinicians and others in the health system who are so motivated to help people and make the world a better place. I truly value what I learned in Silicon Valley and enjoy applying those methods to make a meaningful difference. It’s a pleasure to work with people who care so deeply about doing things better to make people better off. Thinking about the problems we’re working on to alleviate suffering and allow people to live healthy, full lives to achieve their goals helps me get out of bed in the morning.

A: What do you like to do outside of work? Do you have any special hobbies or interests?

RR: My favorite thing to do is spend time with my family. I have three children, two in high school and one in middle school. I thoroughly enjoy watching them explore the world and develop opinions about what needs to change. Attending their sporting events and walking with my wife and our dog in the woods near our home keep me happy.

My new passion is supporting work toward sustainable models for local journalism. I recently joined the board of the Lenfest Institute for Journalism. Good, independent journalism is the underpinning of a strong democracy. It pains me to see journalists under attack and to see the business model struggling to support newsrooms. The Lenfest Institute is doing great work identifying and supporting programs that reimagine journalism in ways that are sustainable and high impact for society.

Roy Rosin

Chief Innovation Officer - Penn Medicine

Subscription Center
Stay in the know with our newsletter Stay in the know with our newsletter