ViVE 2024: How Increased Data Sharing Can Improve Health Outcomes

How Does Patient-Generated Data Fit into Healthcare?

Moderator Ezra Mehlman, managing partner at Health Enterprise Partners, opened the session “Data that’s Getting Personal” by explaining that there’s been a tsunami of healthcare data, a disproportionate amount of which is fueled by patient-generated sources.

“The proliferation of wearables, mobile health apps and patient-reported outcome platforms has dramatically increased the number of nodes by which patient data can be ingested and incorporated into a clinical context,” he said. “This is being driven by meteoric expectations assigned to real-world evidence initiatives, a renewed focus on patient experience — some of that underpinned by financial penalties — and a technological infrastructure that, for the first time, may meaningfully enable value to be unlocked for patient-generated data.”

RELATED: A Banner Health’s leader discusses the organization’s journey toward a unified data model.

PGHD has the potential to improve patient wellness by notifying providers of side effects or trends so they can act upon that data. It also allows for further engagement with patients about their health beyond a visit or even beyond a period of remote patient monitoring.

Adoption is still in its early stages. Dr. C.K. Wang, chief medical officer at COTA, explained that, unfortunately, PGHD is not commonly used in specialties such as oncology, but it’s much relied upon in the clinical trial space.

“Widespread adoption has not truly occurred. It’s too early to tell what impact PGHD will have,” said Wang. “However, I do predict it will be an integral part of patient care and that sponsors will continue to work PGHD into their trial protocols. It is my hope that oncology providers will start adopting these types of technologies so they can take care of patients in a better way and improve their quality of life, thereby reducing the cost of care.”

Becky Fox, chief clinical information officer at Intermountain Health, pointed out that the challenge in adopting PGHD is validating the data and including it in the right clinical workflows so clinicians can take action based on that information.

That data will often be unstructured, and healthcare organizations will need someone to sort through the data to make sure it’s clean and usable within existing workflows. Most providers don’t have the manpower to handle that data ingestion, said Wang.

According to Dr. Ashwini Davison, healthcare executive adviser for Amazon Web Services, a tsunami of data is not something the human mind or any clinician or encounter-based approach will be able to fully interpret. AI can help with the process of pattern identification and presenting those findings to a clinician during an encounter in a way that won’t make them feel drowned in data.

Bradd Busick, senior vice president and CIO for MultiCare Health System, highlighted the need to find a partner that can discuss interoperability in a meaningful way, which he said will be a big challenge over the next four years.

“IT isn’t at the table, IT is the table,” he said. “When IT is at the table, you can have a meaningful conversation, and you have to then go for part two, which is how do you find a dance partner that shares the same hunger for enhanced patient care as you?”

While obstacles exist, Fox said that the organization has better engagement and partnership with patients who share their data.

The next question for healthcare organizations collecting PGHD is how to determine which data is important. Should the care team be concerned by a sudden drop in blood sugar, or should they be more concerned with trending changes? When to intervene becomes an ethical decision, said Fox, who emphasized the importance of being open and transparent with patients about expectations for their level of engagement.

Unlocking Secure Data Sharing to Improve Patient Outcomes

Data sharing is crucial for supporting value-based care models as it enables a more comprehensive picture of a patient’s or population’s health. As organizations look to participate in VBC, they will need to ensure they have clean, organized, quality data. Enterprise data warehouses will be needed to make a healthcare organization’s data accessible from one location.

Kerri Webster, chief analytics officer at Children’s Hospital Colorado, said an enterprise data warehouse is key to leveraging data at her organization. Data can impact kids’ care outcomes by revealing risk factors that could bring them into the emergency room. For that, organizations need to break down data siloes.

Pooja Babbrah, Practice Lead, PBM and Pharmacy Services, Point-of-Care Partners; Children's Hospital Colorado Chief Analytics Officer Kerri Webster; CAQH CEO Sarah Ahmad; Todd Crosslin, Industry Principal for Healthcare & Life Sciences at Snowflake; and Ashok Chennuru, Global Chief Data & Insights Officer at Elevance Health, discuss the importance of data privacy.,

There is a desire to access genomic and digital pathology data cross the healthcare and life sciences spectrum, said Todd Crosslin, industry principal for healthcare and life sciences at Snowflake. He added that pursuing this depth of data is not fit for an API.

“As we move toward precision healthcare, you now have cloud technologies that allow you to take a person’s genetic variations and use that to have a better outcome for that patient. That collaboration requires us to stop copying data,” he said.

Webster added that Children’s Hospital Colorado leaves data at rest in its source systems and uses technology to bring that data together — a powerful tool, she said, because it reduces technical debt and enables broader sharing and combining of meaningful data.

Sarah Ahmad, CEO of CAQH, a company focused on enabling information exchange among health plans and providers, said provider data quality is one of her company’s main priorities as it takes in and pushes out data on both the provider and payer side.

“When it comes to provider data quality, it causes tons of downstream problems, from claims being paid accurately to people being able to find care in the moment they need it,” she said, adding that it’s important for patients to be able to find in-network providers without being surprised by out-of-network charges.

DISCOVER: Modern data platforms advance the quintuple aim.

From the payer perspective, Ashok Chennuru, global chief data and insights officer at Elevance Health, said the accuracy of the company’s data on providers improved from about 60 percent to close to 95 percent. Much of this data is focused on provider addresses and whether they’re accepting new patients.

Crosslin pointed out that another important aspect of data quality is knowing where the data came from, who touched it and when, why they touched it, and the data’s current state. Who is validating that the data is accurate? With the embrace of PGHD and generative AI solutions in healthcare, it’s key that organizations are able to trace data lineage.

Webster said she doesn’t believe healthcare is doing a good job of putting patients or parents at the center of managing and owning their own data, or saying who can use it and how it can be used. Providers enter data into the data warehouse. They own it, know what’s in it and are responsible for its accuracy — the patient does not. Different hospitals or insurers may have different sets of data for that patient, meaning they won’t have a consistent care experience across organizations.

“We treat the data as a commodity, whether it’s the hospital, insurer or the government. We have to start thinking differently,” she said. “The data is there, but whose data is it?”

Data privacy is another issue the industry needs to tackle, especially with increasing adoption of generative AI. Chennuru reminded the audience that data security and privacy are foundational to everything healthcare does with data. Because generative AI uses both structured and unstructured data as its sources, he said, Elevance Health has expanded its governance around how its managing data from a privacy standpoint. In its patient-matching initiatives, the payer has focused on ensuring good identity management, going beyond traditional matching algorithms to probabilistic and deterministic matching.

To secure protected health information, Snowflake is tokenizing data natively, Crosslin said, creating a de-identified version of data that’s been tokenized and made shareable, rather than copying the original data source.

“It is our sacred calling to hold this data as precious assets and to be stewards of this data. I speak for children specifically; they are extremely vulnerable from a data stealing perspective because they might have 18 years before they know their identity has been stolen,” said Webster, adding that genomic data in the wrong hands can lead to care being withheld or premiums going up.

Chennuru highlighted the importance of payers being completely transparent with patients about how their data is used and secured, and limiting their sharing and use of it based on health plan member preferences.

While healthcare still has more work to do to successfully navigate the sharing and protection of patient data, panelists agreed that collecting data is key to whole-person health.

Crosslin hopes that healthcare providers will soon be empowered to generate their own data and have access to that data to better help patients and prevent poor outcomes with earlier actions.

“I’m hopeful that we figure this out and become healthier communities and citizens, and that healthcare can leverage the data to support them,” said Webster.

“I’m hopeful about achieving whole-person health and having that be the foundation of everything we do,” added Chennuru.

Keep this page bookmarked for our ongoing coverage of ViVE 2024. Follow us on X at @HealthTechMag and join the conversation at #ViVE2024.