Getting Creative with Data to Improve Outcomes

By Matt Phillion

Healthcare is awash in data—but without quick access to high-quality data, it’s impossible for clinicians to get the most from that information and effectively treat their patients.

Further, there are specific population health initiatives that benefit particularly from data analytics that identify population needs and measures for the care provided to ensure that the right care is being delivered to the right patients.

How can the industry help providers use quality data to identify social determinants of health affecting their patients to optimize preventive care and get ahead of health issues their patients may experience rather than waiting for those issues to surface?

Data analytics platform MDClone has helped one health system save millions by leveraging high-quality patient data for several population health initiatives. Intermountain Healthcare, a Utah-based not-for-profit health system, looked at the impact of chronic kidney disease in their patient population using medication, lab values, coding, geospatial mapping, and social determinants of health to identify patients most at risk and to connect them with services they need for coaching or interventional work to help address their condition.

“They were able to connect them with nephrology nurse navigators, kidney protection education, and personalized care management programs to help keep their numbers in check,” says Brandi Meyers, vice president of revenue operations with MDClone. “They’ve been working on this for several years and every year they make further advancements on it. I love this as a use case because it really does impact the whole population. There’s a serious quality of life improvement when you don’t have to go on dialysis.”

One of the challenges to making a case like this work is asking the right questions and looking at your data from the right angles, Meyers says.

“Most health systems’ data is going to be extremely messy. It’s just a fact of life. Even if you design your system well, it’ll have clean data for about a year and then you’ll have turnover, processes will be forgotten, and then it’s messy again,” says Meyers. “So it’s important to focus on data quality. At the end of the day, you need to be able to look at problems from multiple angles, even comparing your data to another entity’s.”

MDClone uses synthetic data, Meyers explains, but there are other layers that can be peeled back to look at your patient and population data.

“If you want look at things like the financial layer, you can do so using assumptions from your own data,” she says. “Look at your revenue systems to learn from what the payer receives from you and how you are reimbursed.”

A potentially overlooked layer, though, comes from social determinants of health, Meyers suggests.

“Look at your call center,” she says. “How many times has a patient called in for a refill on a medication or to talk to a nurse? There are so many other potential behavioral data sets. Think about population management and what they do when they’re outside a healthcare facility—then apply that to the data you know is getting collected.”

It takes some creativity, Meyers says, but it’s possible to follow a patient through their process and understand how they interface with your organization even when not physically visiting.

“From a methodology perspective, it’s not some secret sauce. It’s planning and knowing the question you want to answer,” she says. “When you know the question, you want to ask you can plan for it.”
Meyers points to a success case with one client, who have made rapid use of their data by targeting certain use cases. MDClone is still in the process of a multi-month data load effort, but the client was immediately able to start seeing value it because they knew the questions they wanted answers for and prioritized the data loading accordingly.

“They’d sent out a request to all department leaders: what is the tough data you can’t get a hold of?  The departments had to answer the questions: why do you want it and what would you do with it?” Meyers says.   This allowed for leadership to prioritize data quality efforts based on anticipated value to the system.

If you want to understand, for example, how often patients over 40 contact the hospital for a medication refill, you can then also identify moments to engage with those same patients about things like screenings they are due for.

Connecting the dots

A significant opportunity here, Meyers notes, is a chance to connect data to value.

“It’s hard to not only to connect the data, but also it’s really hard to dedicate resources to a project if you’re not sure it’s going to be valuable,” she says.

To demonstrate that value, Meyers talks about the MDClone philosophy of empowering the front line.

“We believe the people on the front line, those department heads and doctors, have very cool ideas, and often when they raise their hand and ask the question, nobody has the answer. That’s what we try to do,” she says. “We try to connect the technical solution to the front line so they can ask their question quickly as a litmus test of: should I continue with this line of reasoning? We want to make sure we’re listening to those frontline questions and ideas and put a plan in place to connect that data to them.”

The speed with which the data is available makes a huge impact on how valuable, and how impactful, that data can be.

“It’s easy now to get data, put it in a warehouse, and say we have access to it, but to do anything with it you need the experts in that subject matter—and they don’t always have the technical skills to pull it by themselves,” says Meyers.

Many times there’s a disconnect between the team members who manage the data and the clinicians who need it to answer their own questions. This is where leadership’s role comes to the forefront.

“It’s about leadership, prioritization, and change management,” says Meyers. “Start with an expectation you will finish a project. So many times someone will ask the fun, interesting question but won’t have a desire to see it through. That’s where you need your CMOs, chief quality officers, or CEOS to push the effort.”

Meyers points to one client whose CEO had a plan: give each department a specific amount of money to pursue creative ideas of value, and then prove the financial value by saving 10 times what the initial cost was. He created a committee and had it report into the CFO to demonstrate from a data standpoint the value of the initiatives. While not every organization could go this route, that combination of empowerment and accountability to use the data in impactful ways is an example of how leadership can influence change.

“When you start empowering the front line, so many cool things happen. You can find a competitive edge, or a financial or operational improvement,” says Meyers. “The funny part is you’d think clinicians don’t want to dig through tons of data but it’s not that you have to force anyone to do this. The ones who want to really enjoy it, and that can help keep them engaged and prevent burnout. These clinicians didn’t go to school to enter information into a form, they went into medicine to solve problems.”

The first part, empowering and pushing people toward change, can be a heavy lift, Meyers says, but with the industry changing so fast, there’s a competitive edge by providing ideas an opportunity to shine.

“It really comes down to people, process, and platform,” she says. “On the people side, it’s creating a mechanism that empowers people who have a good idea. But also, find a way to free up time and invest in your people so you don’t burn them out.”

On the process side, Meyers says organizations need to make sure there is both a bottom-up and top-down structure, where leadership at the top helps bring up and elevate those on the front line.

“The platform shouldn’t be the hardest part, but so many data platforms are very heavily reliant on technical IT staff, and that can sometimes cause friction,” she says. “When it works seamlessly it’s a beautiful thing.”

It’s a balancing game, Meyers says.

“I would love to see the CMOs and CQOs of the world be given more latitude to make investments that drive overall outcomes and operational and financial improvements,” she says. “While many clinicians don’t consider themselves business leader types, CMOs often are, and I’d love to see those people empowered to get more creative and lead toward a more altruistic industry.”

Matt Phillion is a freelance writer covering healthcare, cybersecurity, and more. He can be reached at matthew.phillion@gmail.com.