In Pittsburgh, a Big Data Alliance Looks to Transform Healthcare

April 13, 2017
Rob Hartman, program manager, UPMC Enterprises, discusses what’s new with the Pittsburgh Health Data Alliance and what lessons have been learned as it enters its third year.

Next week, the Pittsburgh Health Data Alliance (PHDA)—a collaboration among the University of Pittsburgh Medical Center (UPMC) , the University of Pittsburgh and Carnegie Mellon University that is focused on building new companies that create data-intensive software and services focused on healthcare and wellness—will host an event, “The Next Big (Data) Thing,” which will cover the role of Pittsburgh’s universities, leading health systems, and the tech industry in creating value from data.

As the Pittsburgh Health Data Alliance (PHDA) enters its third year, researchers, scientists, clinicians and executives from these organizations will get together to discuss the progress and promise of the PHDA in creating data-focused solutions to some of the biggest challenges in healthcare.  Indeed, announced in March 2015, this alliance aims to leverage massive amounts of data—from electronic health records (EHRs), imaging, prescriptions, genomics, insurance records and wearable devices—to improve healthcare and wellness. Rob Hartman, program manager, UPMC Enterprises (the venture arm of UPMC), and who oversees the relationship within the Pittsburgh Health Data Alliance, recently spoke with Healthcare Informatics about what the alliance is working on, what have been the biggest lessons learned in the past few years, and what’s on the docket for the next year. Below are excerpts of that interview.

As the alliance enters its third year, what are some of the core lessons that have been learned?

In this alliance, UPMC funds early-stage research projects and commercially-oriented research projects at both universities [CMU and University of Pittsburgh], and we provide not only the funding for this research, but also access to clinicians, the living lab to UPMC, access to healthcare data, and commercialization advice for the healthcare sector so we can steer projects to high-impact targets in healthcare.

One primary lesson we have learned is that UPMC needs to engage early on in the process and be prescriptive about what our priorities are in healthcare so that we can help the investigators who are working on world-class technologies, help point companies to high-value and high-impact priorities and opportunities in healthcare. The secondary lesson is that we have created the right organizational and data governance structures to facilitate the speed at which proposals are sourced, evaluated and then managed and selected.

What projects are the alliance working on now?

As a reminder, the work is with researchers that are in pre-company formation in the PHDA, so these are teams of investigators that hopefully will become companies. Some of the priorities we have, not just in UPMC, but that are shared in healthcare, are making precision medicine tangible or real; providing consumer-focused care, so how we can construct models for healthcare that put the consumer or patient at the center of care; also, telehealth is very important; and behavioral medicine is too, so understanding how lifestyle behaviors contribute to health, wellness, and disease. These are the broader areas.

In the precision medicine [bucket], we have a couple of technologies that we have funded the development of to help bring precision medicine out of the cloud and down to earth. The big vision for precision medicine is being able to provide personalized treatment, so understanding the characteristics or features of an individual and [then] prescribe certain treatments that would be effective for that individual. So how do you do that? One of the key pieces you need is individualized diagnostics so you understand what’s going on in that person.

One area we are focusing on in the alliance is precision medicine in oncology, and one such project we funded is tumor driver identification. This means taking the genetic information from an individual tumor and using casual inference modeling to provide an interpretation of that genetic and other data to interpret the biological changes and mutations in that tumor to identify which genetic mutations are actually driving that individual tumor. Most precision medicine approaches that analyze the genetic sequence are general and population-based. But this approach is providing individualized genetic interpretation, so for a patient’s tumor, which genetic alterations are driving that tumor? So it allows you to then prescribe a therapy to address the underlying cause of that particular tumor.

Overall, what are the biggest challenges in creating these data-focused solutions?

Getting access to the data you need on the right scale is challenging, but that also points to a strength of the PHDA, in that we are helping the researchers get that data. Data access is quite important from the researcher’s perspective, and then also matching that data with clinical outcomes is important too, and that speaks to the strength of the PHDA as well, where we can provide the data and match that up with the progression or how that patient did or is doing.

Also, moving these projects into the commercial space is also challenging since you need to demonstrate the value of these approaches to payers. These projects would shape new ways that healthcare could be delivered, but in order to be realized they have to be paid for. We try to leverage the difference facets of UPMC to design the work we are doing in the PHDA to lay the groundwork for convincing payers of the value for these kinds of approaches. It’s a little early now, and we think this is one of the final steps since the projects are more at the proof of concept/clinical data stage, but we do have the ability to work with our health plan [internally at UPMC] and interact with leaders there who can help us design what we are doing now.

What do you think will be the key to even more successful research in this initiative?

One key will be to double down on getting involved and collaborating early across the three organizations. The PHDA brings together three different enterprises that have complementary strengths, so in a thoughtful way we can put those pieces together. You have the leading clinicians, the data and commercialization expertise of UPMC; the computer science and research expertise of CMU, and also their machine learning and AI expertise; and top-tier biomedical and clinical researchers at the University of Pittsburgh. Putting the three together in a thoughtful way will move this to the next level.

One project we have that’s funded that does do this is called “readmissions rhythms,” which combines a lead human computer interactions (HCI) researcher at CMU who works with someone from the cancer institution at the University of Pittsburgh, and the two of them are working together to better understand the behavioral patterns of oncology patients who are at high risk of readmissions. So they are using social data, medical record data and activity data to understand behavioral patterns, and then they are using machine learning to understand which patterns are predictive of readmissions and which are predictive of healthier outcomes.

We always knew, as any venture capitalist or other investor would know, that not every project we invest in will be a homerun, but we are excited about several of these getting to the next level and receiving more funding or become startups of some kind.

What qualities/characteristics have made this region, and these organizations in particular, such a hotbed for innovation in healthcare?

The bedrock of the PHDA is that it brings together three preeminent institutions. Pittsburgh is a hidden gem for healthcare innovation and technology. There is a perception that great innovation takes place only on the coasts, but Pittsburgh has the ingredients to do it all.

Sponsored Recommendations

Enhancing Remote Radiology: How Zero Trust Access Revolutionizes Healthcare Connectivity

This content details how a cloud-enabled zero trust architecture ensures high performance, compliance, and scalability, overcoming the limitations of traditional VPN solutions...

Spotlight on Artificial Intelligence

Unlock the potential of AI in our latest series. Discover how AI is revolutionizing clinical decision support, improving workflow efficiency, and transforming medical documentation...

Beyond the VPN: Zero Trust Access for a Healthcare Hybrid Work Environment

This whitepaper explores how a cloud-enabled zero trust architecture ensures secure, least privileged access to applications, meeting regulatory requirements and enhancing user...

Enhancing Remote Radiology: How Zero Trust Access Revolutionizes Healthcare Connectivity

This content details how a cloud-enabled zero trust architecture ensures high performance, compliance, and scalability, overcoming the limitations of traditional VPN solutions...