"Saving Lives with Data."
It is a bold vision statement. Truveta, a health system-led company, says the integrated health data platform it is building will help save lives. It will enable researchers to find treatments faster, clinicians to ferret out root causes of disease and
families to make better health care decisions.
Dr. Nick Stucky, Truveta vice president of research, says that the company and its 20 member health systems have prioritized health equity as a top focus area. Stucky says Truveta's data bank and data analysis tools hold the potential to give researchers
a solid grasp of why certain marginalized communities have poorer overall health and health care outcomes as compared to other groups, and to help reveal the most effective ways to tackle those inequities.
The inspiration for Truveta came from Providence St. Joseph Health around 2018 as leadership sought a way to harness the power of big data in plumbing pressing questions in health care delivery. Providence, Trinity
Health, Advocate Aurora Health and Tenet Health joined to launch Truveta as an independent, for-profit company in September 2020.
Truveta refers to participating health systems as "members" and says its member health systems deliver over 16% of clinical care in the U.S.
Trinity Health, CommonSpirit Health, Bon Secours Mercy Health and Providence St. Joseph Health each have a representative on the Truveta board of directors.
The health system members and Microsoft are the investors in Truveta. All Truveta member systems contribute their patient data to the pooled data platform.
Dr. Daniel Roth, executive vice president and chief clinical officer of Trinity Health, explains the usefulness of the pooled data this way: "Now that we have this big data set, we can do a better job of understanding what drives different health conditions,
of elucidating risk factors for certain diseases and predicting who is at higher risk for those diseases. And we'll get a better understanding of what contributions social influences and racial influences, for example, have on health inequities.
"If we are to attack racism as a public health crisis, we need to have a better understanding of it and we need to maximize the investments we make" to address it, Roth says.
Deidentified and normalized
In January, Truveta entered a strategic partnership with the health care business of LexisNexis Risk Solutions. The Truveta data platform includes the member systems' deidentified clinical data on morbidity
and mortality. Lab results, diagnosis codes, procedure codes, clinical notes and pathology reports are part of the cache. The agreement with LexisNexis allows the integration of aggregated data from member systems with information harvested by LexisNexis
from 40% of all Medicare and Medicaid insurance claims and 70% of all commercial medical insurance claims.
Mining Truveta's data platform may yield a better understanding of how social, economic and racial factors influence the onset and trajectory of disease.
The LexisNexis data set includes information from that company's access to public and proprietary data, such as property records, vehicle title records, address changes/location, auto and home claim records, public education records, census data, death
records and more.
According to information that Truveta provided by email as well as information that appears on the Truveta website, this data set enables access to comprehensive socioeconomic data on every adult American. This deidentified data enables researchers to
understand the impacts the type of transportation, housing and education patients have access to, as well as their income level, stressors they may be experiencing and the number of people living in their household might have on patient outcomes.
LexisNexis says its patent-pending technology called Gravitas provides the ability to join that information with "deidentified patient records that would otherwise go unmatched, thus providing consumers of this data with a more comprehensive and complete view of an individual, while maintaining strict compliance
with security and non-reidentification requirements."
The Truveta platform with the integrated LexisNexis data contains deidentified patient data for about 50 million people in the U.S. — and counting. Stucky says that many health care databases oversample white patients because people of color often have
worse access to health care. He says the combined Truveta and LexisNexis data will be more representative of the racial composition of the U.S. population.
In a March 8 blog post, Truveta Chief Executive Terry Myerson says the company's data platform "is updated daily and continuously flowing to be responsive in pandemics, quickly staff clinical trials, and to provide the most up-to-date picture of U.S.
Answering complex questions
The main user groups for the data platform are data scientists and other researchers employed by Truveta, the member health systems and life sciences companies. According to information from Truveta, when
health systems and life sciences companies are members of or clients of Truveta, they have full access to the Truveta data platform to conduct their own research and benchmark performance against a national average in the data platform.
Roth says the Truveta platform could provide insight into a wide variety of vexing questions. For instance, he says, health care providers assume that screening for the social determinants of health and then responding to the unmet needs for food, housing
and other life essentials will have a positive impact on patients' health and well-being, including for members of minority populations. But it's been hard to establish and document a causal relationship and a measurable impact in part because it's
difficult for researchers to tease apart variables undermining health and longevity. Roth says the data platform will allow researchers to answer complex questions such as the extent to which housing insecurity and food insecurity impact the ability
of diabetic patients to manage their disease.
Roth says Trinity Health and many of the other Truveta member systems are acutely interested in delving into such questions as they pertain to racial minorities. "We know outcomes are worse for Black people for many conditions, and this platform will
help us understand the key drivers and address them," he says.
Roth sees potential for community benefit staffers to partner with data scientists and use Truveta data to identify the most effective ways to deliver community health programming. "We'll be able to identify who are the first people we should reach out
to. In our communities where people are experiencing poverty and are under-resourced it's hard to understand their needs — they are so complex. This will help us understand the contributors better."
The big picture
In addition to his role with Truveta, Stucky is a practicing infectious disease physician and researcher at Providence Portland Medical Center in Oregon. He agrees that having access to a large database that draws
connection points between health conditions and demographic background and socioeconomic status could give researchers and clinicians valuable insights. He sees much promise in better comprehending some of the socioeconomic drivers of cancer, maternal
and child health, cardiovascular disease and kidney disease, among many other illnesses. He believes the learnings can inform health facilities' interventions with patients.
"Actually, I think every study is a health equity study," says Stucky. "We're getting at root causes — the socioeconomic factors that underlie disease."
More resources Catholic Health World:
• De-identified medical records data is big asset for health systems
• Systems say records platforms have ethical, security safeguards
• View a video on COVID-19 research using the Truveta data base
Early queries on the Truveta platform
Although the Truveta database is fairly new, data scientists at the company already are sharing some of their research findings on the Truveta blog and submitting some of their work for publication in academic journals.
Some of the studies conducted and publicized so far include:
- A data analysis that Truveta put on its blog on June 21 examines which segments of the U.S. population are more prone to long COVID-19. People with long COVID have returning or sustained symptoms after contracting the virus. Truveta researchers
said their analysis revealed females were nearly two times more likely to be diagnosed with long COVID than males. They also found that while people aged 45 to 64 make up 27% of the population in the data set, that group had 42% of long
COVID diagnoses. The data accessed was through June 17.
- Research posted by Truveta on April 22 explores whether emergency department visits for asthma increase when the air quality index shows that the air quality is poor. The Truveta research team looked at a subset of Truveta data on first asthma
diagnosis in Oregon and Washington between Jan. 1, 2018, and June 30, 2021, against air quality index information from the Environmental Protection Agency for the same period. While the researchers did not find a direct correlation between
emergency department visits and poor air quality, they did see a delayed peak in visits between one and three days after poor air quality peaks. The researchers recommend further research on the connection.
- Truveta on April 1 posted research on the impact on colonoscopy rates of actor Chadwick Boseman's Aug. 28, 2020, death from colon cancer. The research team wondered whether more men would get colonoscopies after learning that the actor had
died of this type of cancer. The group looked at data from July 1, 2018, to March 10, 2022, and broke out its results by race. The Truveta database did not show an increase in the number of colonoscopies sought after the actor's death.
- On March 8, Truveta posted findings from research that Truveta data scientists undertook on COVID breakthrough infections. The team compared the rate of breakthrough infections for fully vaccinated people who were not boosted with the rate
for people who were fully vaccinated and boosted. They found that the infection rates were lower for those who were vaccinated and boosted. Truveta notes that its researchers had posted online insights from early versions of this study
in November 2021, announcing their findings prior to the release of a similar research report that the Centers for Disease Control and Prevention issued in January. Truveta notes that its study is reproducible and its data is transparent.
According to Truveta, any Truveta member or customer could reproduce its study with access to the full notebooks, code and data. The company also has made the code available on GitHub, a public platform for software developers. Truveta
says the comparable CDC study is not similarly transparent.
Copyright © 2022 by the Catholic Health Association of the United States
For reprint permission, contact Betty Crosby or call (314) 253-3490.