Since 2016, Philip R.O. Payne, PhD, has served as the director of the Institute for Informatics at Washington University in St. Louis. In this role, he’s grown the university’s biomedical informatics and data science capabilities and expanded its educational portfolio with new certificate, master’s and doctoral programs.
Now, as he adds the title of chief data scientist to the list (the university’s first), he’ll build on these successes to oversee the data architecture, infrastructure, and governance models that will accelerate the School of Medicine’s research, education, and clinical care missions.
Here, Payne discusses the new role and its impact on the School of Medicine and the university at large.
What does it mean to be the first chief data scientist of Washington University?
One of the reasons we created this new role was because we recognized that now, more than ever, access to and analysis of data across all of the mission areas of our medical school was foundational, whether that be in the context of research, education, or clinical care. This new role builds on the contributions of leaders in our organizations that focus on how we deliver technology infrastructure, or how we organize and govern access to data, bringing additional focus to the expertise and methods that we need to make sense of that data.
When we created this role, we knew that it was important for our organization to build a comprehensive portfolio of data science initiatives that will allow us to deliver on the promise of precision healthcare, to generate value from our investment in the EHR, and to make sure we’re training a workforce of providers, researchers, and leaders who are able to navigate this new digital health ecosystem.
What does this role mean for the School of Medicine?
It certainly means that we’re recognizing the central nature of data and data science in informing all of our strategies and activities. It also points to the fact that we are committed to building a learning healthcare system, so that every time we generate data, we use that data to make better decisions, whether that be in the hospital, clinic, laboratory, or classroom.
Further, it speaks to the fact that the technology and the methods needed to build a smart, data-driven organization are changing so rapidly that we need a strategic approach to ensure we’re using the best possible tools and methods now and in the future to understand and generate value from all of our data, consistent with our mission.
Is the formalization of this chief data scientist role part of a broader trend you’re seeing across the country?
There are an increasing number of academic health centers and medical schools that are creating similar roles, which reflects the commitments of those organizations to be on the leading edge of informatics and data science.
To this end, Washington University and its medical school have always been on the leading edge in terms of biomedical research; training physician scientists and physician leaders; and translating research into the delivery of care at the individual and population levels. Being on this leading edge means we are now embracing the idea that informatics and data science are pervasive and competitive capabilities that we need as an organization, and that is consistent with our history of leadership.
I see discussions happening almost every day on our campus concerning how we can better collect, manage, analyze, and understand a variety of data types and assets. These conversations are indicative of a really important evolution for us because, to date, it’s not entirely clear that academic health centers, and healthcare in general, have truly generated all the value that can and should be generated from the data that we produce every day. The recent emergence of chief data scientist roles across the country, and here locally, are a recognition that in addition to collecting data, we need to do something meaningful with it. We need to use it to improve research, education, and care delivery.
What does a chief data scientist do?
First and foremost, the chief data scientist needs to be able to work with research and educational and clinical leaders and understand what their data and information needs are, and then align those with the art of the possible. What can we do in response to those needs given the best available methods and technologies? In many ways, the chief data scientist is a data translator that can work with those key leaders and turn their data information needs into something that’s actionable.
The second thing that the chief data scientist does is describe and oversee a strategic approach to organizing the technology, data assets, and people that will allow us to act on those information needs.
The third role for the chief data scientist is to be the spokesperson that explains the outcomes of these data science endeavors to key stakeholders, providers, researchers, educators, community members, and executives, making sure that those individuals have a comprehensive understanding of the outcomes of modern data analytic initiaitives.
How will this role impact the School of Medicine’s data architecture, infrastructure, and governance?
An important aspect of this position will be making sure we have the right data architecture at all levels and the right policies, procedures, and governance models in place to operationalize that architecture. These are not always exciting areas that people like to talk about, but they are essential to our success. This is an area where I work in partnership with our dean and the dean’s staff, the CEO of our faculty practice plan, our department heads, as well as our technology leaders and our partners at BJC Healthcare, in order to understand their needs and translate those into data architectures and governance models that don’t get in the way of arriving at conclusions or insights in a timely manner, but rather, that allow us to do that in a more efficient, responsible, and rigorous manner. It requires a lot of confidence building and active listening to understand how those needs translate into functional governance structures, because governance is a scary word. Many people think of it as an impediment, but if done right, it’s actually an accelerant.
Do you think there are specific fields of study or research that will benefit from this role?
I am very hard pressed to identify an area of research or innovation in a modern academic health center that is not data intensive, whether that be at the bench, in the clinic, or at the population level. If we do our job well, all fields of research and study will benefit.
The real question is, are we truly providing our researchers and innovators with the tools and methods needed to capture all of that voluminous data and make sense of it? Our traditional mode of operation in an academic health center often involves a timeline of weeks and months spanning the identification of a problem, to producing data that is responsive to that problem. If we want to be a high-performing and learning healthcare system, and to enhance our research excellence, we have to bring that timeline down to minutes, maybe days. We have a lot of work to do in order to achieve such goals, but if we do it right, it’ll make all areas of research and innovation in our enterprise better as a result.
What experience do you bring to this position?
I started my career in a role that was a mixture of serving as a principal investigator running my own research lab, but also having an operational leadership role in the IT department of an academic health center. I was focused on how we could leverage data assets to support clinical and translational research. That early experience was very formative because I began to identify the synergy between my own research interests as an informatician and the operational needs where I could deliver solutions at scale that met the needs of large numbers of investigators.
At Washington University, I’ve had the opportunity to engage across all three of our mission areas and bring that same scientific acumen that I have as an informatician to the table to help people navigate these complex environments of data and data analytic methods, in order to arrive at a conclusion that’s meaningful and impactful.
I think the most important aspect of the chief data scientist role is to be a translator and connector, and that’s the expertise that I’ve been building for the vast majority of my career. I would love to say that that was by design, but a lot of my experience was serendipitous. I’m very grateful for all those opportunities because they brought me to the point where I feel that I’m able to make the types of contributions that this new role will enable me to make to Washington University and the communities we serve.