Tom Sullivan | June 25, 2020
COVID-19 has called to attention many or the promises and problems facing the health care industry and the collection, bi-directional sharing and use of relevant and timely information is among those.
Leveraging the right data at the right time is not an easy matter, making it all the more important for chief executives leading provider, payer and life sciences organizations to understand what other CEOs are learning now and how it can be applied to shape a better system for everyone.
“These are particularly difficult issues and they exist at the local level, then in cities, also at the state level and, of course, at the federal level,” said Neil de Crescenzo, CEO, Change Healthcare.
de Crescenzo hosted the Health Evolution Executive Briefing Battling COVID-19 with Surveillance Data and IT Capabilities virtual gathering. Joining de Crescenzo were New York City Department of Health and Mental Hygiene Commissioner Oxiris Barbot, MD, UCSF Chief Data Scientist Atul Butte, MD and National Coordinator for Health Information Technology Don Rucker, MD.
The experts discussed key learnings about data and the nation’s technological capabilities, including:
Here is a closer look at each of those lessons.
1 U.S. health care is beginning to understand new ways to leverage existing data infrastructure.
With health information exchanges now covering approximately two-thirds of the country and maintaining patients’ identities and providing patient matching services, Rucker explained that many of the public health questions with COVID can be more effectively answered than they could be with just one-way reports that are not used for care provision.
“We’re starting to see early parts of that transition with longitudinal data that’s complete over different classes of providers, sites of service,” Rucker said. “We’re starting to see the ability to actually have a totally different way of thinking about health information reporting with the infrastructure we’ve built already.”
In New York City, for example, the Department of Health and Mental Hygiene has been utilizing its data infrastructure during the pandemic to conduct syndromic surveillance daily, according to Oxiris Barbot, MD, Commissioner, New York City Department of Health and Mental Hygiene
“We get electronic data transmitted to us securely from all of our emergency departments throughout the city. It’s roughly 30 hospitals,” Barbot said. Public health officials analyze that data to look at the volume and the number of visits to emergency departments for symptoms potentially consistent with COVID-19 and track that over time. And as the COVID-19 outbreak began, which was at the very beginning of flu season, they used that data to determine that the increasing volume was not the result of influenza. “It was due to something else and that was our first indication that we were starting to see COVID-19 here in New York City,” Barbot said.
Because of post-9/11 investments, New York City also the continuum of data to trace from the time someone gets tested to when they become symptomatic enough to present at the emergency department, become sick enough to be hospitalized, proceed to the ICU and for the most unfortunate cases, until they die.
“That has helped increase the completeness of our data,” Barbot added. “Moving forward, we are leveraging the data that we have to start doing forward planning for the second wave.”
2 Executives need to lead on transparency and data liberation. It’s well understood at this point that a number of large-scale research projects are happening on the academic side and in consortiums, at organizations including the NIH, Harvard, Johns Hopkins, Mayo Clinics, Penn and many others.
But because coronavirus is essentially an invisible disease for most people, the only way they even know it exists is through the media and what they can see of empty shelves, masks and unemployment, said Atul Butte, MD, Chief Data Scientist, UC Health; Priscilla Chan and Mark Zuckerberg Distinguished Professor, UCSF. “There’s a lot of trust needed,” Butte added. “And that trust is really stretching thin right now.”
Butte recommended that health care leaders demonstrate COVID-19 is a real condition, infecting real patients and causing legitimate problems, whether via hospital videos, statistical curves or other means. UCSF, for its part, has reached more than 100,000 people by circulating statistics via social media every day.
“The idea is to liberate as much of your data as it’s safe and responsible as possible,” Butte said. That includes flowing the data to federal agencies, not just the public, so UCSF is constantly pointing to the CDC, FEMA, FDA, the White House taskforce as well as county health departments in California.
While it can be difficult to stand up robust data sharing operations quickly, there are simple web-based tools that can be used to disseminate the information. “The more we can show this is a real condition, and there are real patients with real problems, the better it is to get population–level compliance,” Butte said.
3 Local knowledge is critical.
While public health and academic medical center data is essential, it cannot be used to shape a complete picture of what is happening in various geographical regions. The outbreak epicenter has shifted once already, from Seattle to New York, and it’s likely that another city or state will follow.
While UCSF and others had previously been talking about coding diabetes and hypertension or certain medications, for instance, and all the data around doing those, the pandemics suddenly made harmonizing ventilator and ECMO settings something that needed to be done quickly — and that required executives having a strong relationship with those teams.
It’s also important to understand basics, such as where the COVID patients are being placed or which sites exactly have been transformed into newfangled ICUs.
“All that local knowledge is critical,” Butte said. “We have to count PPE, we have to figure out where the surges are, if we have enough ventilators, if we can map against capacity. We had to get the operational work done first.”
Butte cautioned that the buzzword real-time data, even pertaining to local information, merely refers to daily data, not up-to-the-second.
“We were happy with a monthly data dump,” Butte said. “But we had to switch to a daily COVID dump.”
4 Cutting edge technology, medicine and equitable health care require data reuse.
Interoperability. COVID-19 and the aforementioned data lessons of the pandemic, have exposed many reasons why the Holy Grail of health information sharing is more important during this emergency than it ever was — and history is peppered with years of discussion, debate, consensus and bi-partisan work, even legislation, designed to enable patients, providers, payers to exchange information.
What did it take for the New York Department of Health and Mental Hygiene pulled what Barbot said was a continuum of data? The ability to tap into electronic medical records and regional health information exchanges as well as administrative data, and information from outside sources, notably the U.S. Census Bureau’s American Community Survey.
Having that spectrum of data and also using it in conjunction with information labs, academic centers, a robust surveillance system and survey data from the community enables New York City health officials to overlay social determinants of health on areas with clusters of positive tests.
“When you look at the cutting-edge work that we’ve seen both in New York City and with the University of California system, the takeaway is the importance of operational data,” Rucker said.
Rucker continued that some of the money flowing into the system because of COVID should be used to build operational tools via APIs, rather than one-off reporting.
“You can call it interoperability, but ‘data reuse’ gives it a crisper focus. So that’s what I think needs to happen,” Rucker said. “The critical thing is data reuse.”
5 What the US needs next: Rethinking the power structure in public health reporting.
Describing what exists today as an old school model, Rucker said that the US health care system tends to conduct a lot of one-way reporting, even with health information exchanges, instead of true data exchange.
“I want people to rethink what the continuum of care is and to include long–term care in that as well as, social services,” Barbot added. “And I really want to put an emphasis on the value of relying on webs as opposed to point–to–point services. Especially for those that have a presence or footprint in more than one state, I think that the value could be really multiplicative if we think about it from that perspective.”
Butte said that US health care needs consortia to work together to figure out where we go with employment, testing, employee law, students when considering interoperability across health systems.
“There’s more to this than just the health systems. So we all need to work together, think even broader than just what’s happening inside our hospital and clinic walls,” Butte continued. “We need to work better with employers to figure this out.”
Data and the infrastructure to enable collecting and sharing health information will invariably become even more important in the post-pandemic world — wherein health care executives, public health officials and policymakers will look to surveillance and analytics to inform strategies for keeping Americans as safe as possible until the country achieves herd immunity.
“I think we’d all acknowledged that public health, it now seems obvious, had less investment than maybe it could have over recent years,” de Crescenzo said. “So we’re more resilient for these circumstances in the future.”
Register for an on-demand recording of this webcast.