How Healthcare Organizations Are Understanding Data Better to Drive Change

A growing number of organizations, especially those serving distinct communities, are using data and modern analytics to drive operational and clinical decisions and, ultimately, improve patient care. The initial impetus for most was to achieve the value-based care goals set forth by the federal government, but for many, the pandemic validated and fueled data analytics initiatives, says Laura Craft, expert partner for data and analytics at Gartner.

“Organizations needed to report on what was going on. They wanted to know best practices, to understand which treatment to use based on all these comorbidities,” Craft says. “All of a sudden, there were predictive models and algorithms to help support this.”

Some healthcare organizations are early in terms of analytics maturity, while others are already innovating and using artificial intelligence and predictive modeling. “Their mindset is, ‘Let’s have a much more intelligent ecosystem where we can take larger sets of data and get better personas of our patients,’” Craft says.

Growing Demand for Operational Metrics by Healthcare Clinics

Before implementing Tableau, Romo says, clinics kept requesting operational metrics. He was collecting such data but not displaying it anywhere — and when he finally did, the tools weren’t user friendly. “People didn’t understand how to navigate it, or it would take too long for a report to run,” he says.

Romo decided to move to a self-serve analytics model. “The goal was to build an automated pipeline so users could get new data every day,” he says. “They wouldn’t have to wait on us to send them a report via email. They could go to it when they were ready.”

CMC chose Tableau because of its reputation, cost and user interface. The organization then began using the platform in mid-2018.

WATCH: Learn how Providence uses data analytics and the cloud to enhance patient care.

That experience proved to be invaluable during the pandemic. Suddenly, CMC had to gather and analyze new types of data in real time and report it within and outside the organization.

“That prep work gave us the ability to respond quickly to those demands. We were able to deploy and automate it so we could move on to other urgent projects instead of running daily reports,” says Romo, whose team has grown from one to eight since he started.

The pandemic made a big difference in the data culture at many organizations, says Craft. “While the pandemic continued to hit, people got more sophisticated about developing those kinds of models. That required a cultural transformation in the workforce, to say everybody needs to understand what the currency of data really means to them in their jobs.”

Data-Driven Transformation with Operational Data

CMC also needed operational data to steer the organization. The health system has 28 sites serving 100,000 patients annually, including agricultural workers and low-income residents. But its revenue came solely from face-to-face visits.

“Before the pandemic, we were still growing significantly,” Romo says, “so we didn’t have the bandwidth to say overnight that we were going to, for instance, switch from in-person care to telehealth.”

But amid the shutdown, they didn’t have a choice. “We needed to change dramatically just to survive financially and provide services to our patients,” Romo says. When CMC began offering telehealth, his team compiled figures on in-person, video and phone visits.

“We said, ‘Let’s give the executives a window into whether this is working,’” he says. “Because that was another issue: Were we going to do all of this just for our patients to say, ‘Oh, I’ll just wait until this is all over,’ and then it would be a wasted effort?”

CMC has recovered financially, and telehealth has been a key driver, Romo says. Data has fueled the program’s success, providing clinicians with dashboards during virtual exams offering holistic views of patient data and risk protocols to determine which patients need to follow up with in-person care. “Telehealth is here to stay,” he says.

Starting Points for Data-Driven Decisions in Healthcare

When Karl Hightower joined Novant Health as chief data officer in 2018, he spent his first three months asking business leaders how they make decisions. Many said they spent a lot of time searching for answers or going off intuition as opposed to making data-driven decisions, Hightower says.

But there was a hunger for change at Novant Health, which has 800 locations in North Carolina, South Carolina and Georgia. “You go around enough times before saying, ‘We’re getting the same results that we’ve always gotten, and we’re not improving at the rate that we want to improve,’” says Hightower, who is also senior vice president of data products and services.

Knowing how data analytics can help an organization is key. “People tend to throw that aside. But that’s your roads, your power, your water, your sewer,” Hightower says. “If that doesn’t happen, you’re not building neighborhoods.”

He created operations, patient and clinical groups to drive discussion. “I like to say I’m Harry Potter,” he says. “Assume the technical challenges are gone; I’ve wizarded them away. What decisions will you make? How are they going to make your world better?”

Click the banner below to discover this year’s 30 health IT influencers to follow.

Novant Health built its analytics infrastructure on the Microsoft Azure cloud, which didn’t limit users to one analytics tool. But because of the training and resources offered with Power BI, people naturally gravitated toward it. Novant Health uses Azure Databricks for advanced data processing and Teams to facilitate ­analytic product groups.

“It’s not, ‘Here’s your report,’ and then you go from there,” Hightower says. Employees work together as teams to create and utilize reports. “Make everyone a part of solving the problem.”

Building upon that collaborative structure, Novant Health’s core philosophy is “all data, all the time,” Hightower adds. “I don’t know what the questions will be, but the platform should have all clinical, operational and supplemental data.”

For instance, social determinants of patient health can include transportation options, jobs and proximity to grocery stores. “Most of the things that impact a person’s health and wellness are outside of a healthcare setting,” he says. “It gets into how to best use data to understand the person and deliver the right type of health and wellness care.”

The challenge is getting people to commit long-term, Hightower adds. He likens enterprise analytics to dieting. Losing a few pounds is a good start, but the challenge is sticking with it. “It is a discipline to continue to get better — it’s not the end of the journey.”