In his mission to build the world’s largest clinical data platform with health systems across the U.S., Truveta CEO Terry Myerson is constantly asking himself the same question—or hearing that question from others within the industry.
“Why hasn’t this been done before? The pandemic provided a moral imperative: We need to learn faster,” says Myerson. “We need to learn how to respond. We need a tighter learning loop within health care on what’s working, what’s not working, and how do we take care of patients more effectively and efficiently? For that, you need big data.”
Truveta has brought together 17 major health systems across the country representing 15 percent of all patient care to create a de-identified data platform that will aim to gather insights for researchers, clinicians, and other stakeholders. The goal, Myerson says, is to enable comprehensive research around any drug or disease using a representative, diverse patient population data set to account for a wide variety of demographics and conditions.
“A lot of diseases today are studied with unrepresented data sets because no one has enough data in one place to get an accurate representation of who has that disease. You’ll always have biases and fewer insights with that kind of smaller data set. It’s exciting to think about the research that will be possible with a statistically representative data set,” Myerson says.
While the pandemic created the impetus for health care organizations to collaborate in unforeseen ways, particularly for research, Myerson says these stakeholders have recognized that creating shared data platforms is just a better way of taking care of patients, period. More than ever, there is an opportunity for organizations to break down data silos and create a more effective learning health system, he says. In other words, the collaborative mindset is here to stay when it comes to data insights and research.
This emerging mentality within the U.S. health care system mirrors a trend that has been happening internationally, according to an article in Nature, which found an increase in the number of global research collaborations in the last 10 years. Nearly one-in-four research papers are written by authors from more than one country, which is up from around 15 percent in 2010. COVID-19 has spurred the trend even farther along, as the Organisation for Economic Co-operation and Development found that a quarter of COVID-19 papers were co-authored by researchers in different countries and three quarters of scientific papers on COVID-19 were published with open access.
Cancer care leads the way
One area of medicine that has long since benefitted from research collaborations both at home and internationally is oncology. Collaborative research for cancer treatments is necessary because the multi-disciplinary field has complex clinical challenges requiring a diverse set of expertise, backgrounds and tools, according to a 2018 article in Nature.
These collaborations come in various forms. Major cancer centers, like MD Anderson, Memorial Sloan Kettering, Johns Hopkins and others have engaged and started numerous research collaborations on various elements of cancer care with a number of partners. The American Association of Cancer Research has raised hundreds of millions for various research collaborations. At Miami Cancer Institute, Leonard Kalman, MD, chief medical officer, sees many reasons to share the organization’s clinical data in research platforms.
“It gets back to the core of what we do. It helps us treat our patients better if we know how we are actually treating them,” says Kalman. “If we say on January 1, we are going to treat breast cancer patients a certain way for the next 12 months, well on the following January 1, we can look at the extracted data and find out what we did right and wrong.”
Kalman says the company is leveraging a cancer data research platform, created by a company called COTA, that ties in data from other cancer institutes and life science companies, which generates insights on treatments and medications. The platform also allows Miami Cancer Institute to do clinical trial matching, treat patients more cost efficiently and be a good shepherd of managed care dollars, he says. It’s also a competitive advantage, Kalman says, particularly when it comes to moving towards a value-based payment model.
“Managed care is tough these days. Managed care organizations are cost conscious. You can’t just say, ‘We’re the best. You should pay us the most.’ You need to come to them with data and there are many companies that will partner with you on that front,” says Kalman. “We are looking for partnerships that are nimble, multi-faceted and allow us to act on the data in facing some of the challenges we have.”
COTA CEO Mike Doyle says that the company created a data classification system because oncologists and patients were frustrated with the lack of comparative information available on treatments and medications. He says that academic medical institutions are increasingly recognizing that we have only reached the tip of the iceberg in how to properly treat cancer patients and the components of the disease. This is especially the case, Doyle says, with the emergence of genomics. That’s why they’ve been able to recruit organizations including Hackensack Meridian, University of Chicago, MedStar and others.
“We’ve had a who’s who of academic medical centers that have come forward and say they want to join the research collaborative we are building,” says Doyle.
Trust is important
Research collaborations, in cancer care and elsewhere within medicine, are certainly not without challenges. Kalman at Miami Cancer Institute says that it takes time to extract data from an EHR to a shared data platform and there must be a level of trust between the two entities. In that regard, Truveta’s Myerson says that earning the trust of all the constituencies is both the biggest challenge and opportunity in creating a shared data platform.
“With patients, it’s creating a commitment to security, privacy, ethics and equity. With researchers, it’s data quality. Can this data be trusted to make real scientific decisions? Earning trust requires consistency and high-quality work overtime. It’s not something that will come through a press release or a feature in the software,” Myerson says.
An added layer of trust, Kalman says, is ensuring whatever company you are partnering with on this initiative has staying power. He says organizations don’t want to wonder if the company they have partnered with on a research collaboration will have the funds, the staff, and resources to keep going. The company has been evaluating potential partnerships in early cancer detection research and has a genomic partner that helps them with next generations sequencing. The same thought process has gone into these collaborations.
“If we are looking at these new partnerships, they require considerable investments in money, time and personnel. That’s one of my biggest challenges is finding the people who are willing to do it or hiring someone new to work specifically on this project. If the work is that important than you may have to do that. We’re getting to that point with our cancer data research work,” Kalman says.