Using Data Analytics to Close Care Gaps, Improve Patient Outcomes

August 30, 2021 - Data analytics serves as an essential tool when trying to close gaps in care. By studying data, providers can determine what steps need to be taken to improve patient outcomes.

Assistant Vice President of Analytics Services at Intermountain Healthcare Greg Nelson shared his analytics product development insights.

“One of the things that we’re doing here at Intermountain Healthcare is focusing on how [our patients] experience care, and what does that look like to our patient populations and our community members?”

Nelson explained the healthcare company focuses on tackling what they refer to as “wicked problems” in it comes to analytics. These are complexities that arise when doing analytics in healthcare.

“Healthcare data is hard: The digitization of healthcare data that describes the patient experience is a modern phenomenon that we’ve all learned about,” Nelson explained. “Healthcare is still in large part in its infancy, whether we’re dealing with the lack of a universal identifier, interoperability challenges, incentive structures, compatibility across systems, complex reimbursement systems. All of these point to why we can’t always do our best in analytics and apply novel insights in innovations.”

READ MORE: Precision Medicine Treatments Improve Cancer Patient Outcomes

Generally speaking, providers are being held back based on how organizations think about the role of data and analytics. Data analytics is often thought to be a utility, an enabler, or a driver. However, providers should have the shared goal of innovation rather than thinking about the different roles of analytics.

“What we want to do is get to the point where we engage with our end-users in problem-solving, and drive towards that discovery, being able to automate insights, being able to deliver information that would have never otherwise been found,” Nelson said.

Another challenge of conducting data analytics is the language divide when it comes to how data is processed. “While we all speak the same language, we may not understand each other. I see this happen often, especially in the constructs of data literacy,” Nelson said.

In the analytic space, Nelson said there will be winners and losers. The losers will be those who cannot advance their analytic technology to produce better outcomes for their patients. Therefore, outcomes should be the number one priority for providers looking to improve their analytic systems.

“As we think about how we can modernize our organization, almost analytics reimagined if you will, it’s about realizing an economy of scale with people. It’s insights driving action. It’s about being strategically aligned and value-oriented,” Nelson continued.

READ MORE: Improving COVID-19 Vaccine Outreach with Data Analytics, EHRs

“In product management, particularly, we talk about whole-person thinking. In healthcare, we talk about whole-problem healthcare. In the analytics space, what I want to talk about is whole-person thinking and whole-problem thinking. That means understanding and living in the shoes of those people who are facing the problems. Being able to be positioned right to capitalize on market opportunities is absolutely essential.”

Nelson and Intermountain Healthcare’s Chief Analytics Officer Albert Martinez spent time discussing where the organization should spend its energy and strategic capital. The pair landed on three strategic themes: embed, empower, and engage.

The theme embed is about the organization’s data collection method that creates forecast models, systems of insight, and reports.

“What we discovered was the challenge isn’t so much in the formulation of the data or the story that it tells, but that it takes away people from the decision-making process and puts them into a separate process,” Nelson revealed. “It takes them and relocates their mind into that dashboard versus. What we want to do is embed insights and decisions into their work. By being able to deliver insights into systems of engagement, we suddenly begin to reimagine how our data should be consumed.”

Rather than just data analysis, risk and prediction models can be created to improve patient outcomes further. The second area is an empowerment method that encourages people to use, consume, act on, and measure data. The final piece of the strategy focuses on engagement.

READ MORE: How Big Data, Analytics Drives Population Health, Closes Care Gaps

“Another word for this is co-creation. In analytics product management, we’re designing products and product lines. So instead of solving an individual problem, we’re solving systems of problem,” Nelson said.

With data analytics, providers are better able to identify health disparities using prediction and risk models to address gaps in care. Additionally, the technology can spot at-risk patients who don’t appear to have any risk factors by incorporating artificial intelligence. However, Nelson said good data is critical in doing so.

“Models are only as good as the data, and the design, and the process that we use to create them. What we need to get really good at is engaging people in the co-creation of those models,” Nelson continued.

“It’s going to necessitate different thinking where we augment the clinician’s perspective by bringing in lots and lots of data. The data is the differentiator in healthcare. We have lots of data. We just need to figure out how to harness that to be able to predict high-risk patients for future sepsis, or readmission, or whatever the model happens to be.”