Allina’s Pioneering Move Forward on Population Health Risk Stratification and Management | Healthcare Informatics Magazine | Health IT | Information Technology Skip to content Skip to navigation

Allina’s Pioneering Move Forward on Population Health Risk Stratification and Management

August 20, 2012
by Mark Hagland
| Reprints
How a multidisciplinary team helped the Minnesota integrated system create a groundbreaking dashboard for assessing readmissions risk for inpatients—before they’re discharged

The Healthcare Informatics Innovator Awards Program reaped a cornucopia of outstanding submissions in 2012. Indeed, the editors at HCI concluded that every semi-finalist entry was worth coverage, given the very high level of quality of the entries in this successful program, whose goal has been to highlight team-based achievements in informatics work in patient care organizations nationwide.

One of the semi-finalist teams this year was the Patient Census Dashboard Team at the Minneapolis-based Allina Health. The 11-hospital, 90-plus-clinic, 24,000-employee, 5,000-physician Allina organization is already nationally known for its innovations in many areas of patient care delivery and operations.

With regard to the organization’s Innovator Awards program submission, in 2011, staff from Allina’s Performance Resources department convened an interdisciplinary team from 10 hospitals to understand the barriers to coordinating care for patients with complex needs. The team learned of a desperate need to assist clinicians in the identification of patients at higher risk for readmission.

As it turns out, clinicians were spending hours finding these patients, who require an interdisciplinary approach in order to achieve a successful transition from the hospital to the post-acute setting. The team began work on designing a case-finding tool to quickly identify these patients and assist in convening the care team to create a transition plan. What was needed, the leaders of the team decided, was a tool that could quickly identify these patients while still in the hospital and could provide instant access to real-time clinical information and summarized key healthcare resource utilization measures for each patient, including indicators such as how frequently a patient has visited a hospital in the past 30 days, a count of total emergency department visits in the prior year, and how many different medications a patient might be taking.

After months of work on this project, the ultimate result was the Patient Census Dashboard, a business intelligence application developed internally at Allina using QlikView business intelligence technology, and providing users with an intuitive, interactive environment in which to explore these types of data points. The tool leverages established and prior investments in the organization’s electronic health record (EHR), enterprise data warehouse (EDW), and business intelligence tools. Given those advantages, the Allina team members were able to produce the tool with nominal effort and no cost in additional hardware, software, or professional services. This tool is now being used widely by Allina clinicians.

HCI Editor-in-Chief Mark Hagland recently interviewed members of the team that developed the Patient Census Dashboard. Present at the interview were the following Allina leaders: Michael J. Doyle, manager of the Enterprise Data Warehouse; Penny Wheeler, M.D., chief clinical officer; Susan Heichert, R.N., vice president, health information systems and CIO; Karen Tomes, R.N., director of quality improvement and care management; and Jason Haupt, Ph.D., senior statistician. Below are excerpts from that interview.

What was the strategic impulse behind this initiative?

Penny Wheeler, M.D.: We are very much on the path of, and feel that it’s our civic duty to, transform healthcare for our community. So all of this is about getting away from the traditional revenue generation, providing the most care to the most people, and getting towards quality-based care delivery. It’s really about pushing that Queen Mary of healthcare into a more positive direction, to demonstrate that we can deliver superb outcomes to our community. For example, we’re a pioneer ACO [accountable care organization, as part of the federal Medicare Shared Savings Program] organization. And data is really something that needs to be aggregated to help us do the right thing; and we’re really on the forefront of using some of those tools to accomplish those things.

Susan Heichert, R.N.: We’ve made significant investments not only in our EHR, but also in data warehouses and other tools, and our goal is to use those tools to improve care delivery and outcomes.

Karen Tomes. R.N.: The exciting outcome of this project is that the clinicians designed it, and also gave us feedback into what needs to happen in the current state, to help them succeed in terms of care transitions. So the exciting part of this project was listening to clinicians, including physicians, nurses, pharmacists, case managers, and social workers.

Michael J. Doyle: The investment that Susan talked about that Allina has made in infrastructure for performance measurement has really made this possible in a way that wouldn’t have happened otherwise. Second, it was an extremely gratifying project; we worked very closely with our clinicians, and it helped the front-line clinicians understand that we’re not too far removed from their concerns. It was really neat; it was very much not a “request a dashboard and we’ll build it” kind of situation; it was a collaboration.

Wheeler: We’ve got a ways to go, like everyone else. But when people visit and see these tools, their jaws drop; and we’ve got something very precious here that we’re using on behalf of our community.

What have been the revelations and learnings so far around the leveraging of IT?

Tomes: I think the patient story is so intuitive in this tool, that the way that it was designed has meant that it takes very little training to use. It actually takes more training just to make a few clicks to get to the tool—and once you open it up, it is so intuitive in terms of accessing data on the patient. The other great benefit is that this crosses the entire continuum of care. And we started new relationships, quite honestly, using this tool. We still have opportunities to build on that concept. Formerly, it had been relatively hard to find information in the record; now, the entire care team across the continuum can easily find that information.

Heichert: And I would add, from an organizational perspective, I think it’s important that, as we’re putting in our electronic record, that’s not going to be the be-all and end-all for everything. You try to use the functionality and work with the vendor and all that, and that’s great. But at some point now, we’re starting to have to go faster and faster and faster. But it’s been kind of a revelation that we can kind of figure out some of this ourselves. And if you can get these fantastic people together with the clinicians, they can figure things out, and operationalize things. We’re learning that we need to be a little bit open-minded about what tools to use and how to use them, and we need to be willing to fail a bit in order to learn. So we have to allow ourselves to be a little bit more adventurous in that.

What kinds of lessons learned could you share with fellow CIOs?

Heichert: Well, it does help that I have a clinical background. But it’s been Karen and Mike who’ve been doing the work. And they come to me and say, here are some tools and technologies that we can work with, and what do you think? Certainly, we say, you have to make the business case for what we’re doing. And we don’t want to harm the infrastructure with this. And Mike pushes the envelope a bit, but these are the challenges we need to take on, or we can’t change the way we do things.

What do non-clinician CIOs need to do?

Heichert: Well, it’s not earth-shattering, it’s something we’ve known for a long time: that CIOs need to know the business that their organization is in. And CIOs need to knock on people’s doors and follow them around, and learn.

Wheeler: I want to underscore something that Susan said, and something that Mike and Karen said. And that is that they got very deeply involved with the clinicians from the outset, and that’s very important. Another element is that all this clinical IT infrastructure should report up through the clinical reporting structure in the organization. That keeps it aligned to what Susan said is our core business. And make sure that the people using the tools are co-designing those tools. And a big lesson learned, too, is, meet the clinicians where their passion lies. So if you have a group of clinicians ready to run on, say, reducing heart failure readmissions, then dig in deep with them.

Tomes: And I would add that a CIO should bring clinicians in and shoot for the sun; give them that permission to dream. Sometimes, they’ve been working in an environment where they’ve been adapting to the current environment. So allow them to dream.

What has been most challenging for you so far in all this?

Heichert: Actually, what’s most challenging is the frustration with having all of this data and not having great tools in the past; I’m thinking maybe five years ago. We didn’t have those tools. And it would be great if things were a lot more intuitive. And we do have some limitations in that regard. And frankly, things are very expensive; it does require an investment on the part of the organization. We’ve got a lot of clinicians around here, and if you think about the ability to put all this information and all these tools in every clinician’s hands, it does require an investment. And ultimately, we want to put these tools even into the hands of patients. So it does require a very large investment, and not every health system has as much capital to invest as we do.

Doyle: And we couldn’t have done the work we’ve done without Susan and Penny setting up the culture to innovate towards success. Let me give you an example: a lot of this started because we had a data feed that went to the EDW, on a shared drive, a couple of times a day. And we realized that with a little more effort, we could get that feed going every hour, but we had to work more closely with IT. And once we got that developed, we were on the path.

Another possible lesson here is that one’s organization has to be willing to invest in the technology, too, correct?

Doyle: Yes, and these really are great multi-purpose tools. A great carpenter has to invest in a great table saw, and a great set of wrenches, for example. In a similar way, Allina made the choice to invest in a great BI technology; it’s not a healthcare-specific BI technology. But our organization’s leaders also were willing to invest in general tools at a time when we couldn’t find the healthcare-specific tools we were looking for at that time, so we re-purposed broader tools.

How hard is this, on a scale of one to 10?

Doyle: If you had no EHR, it would be a 10; if you had no enterprise data warehouse, it would be a 9. But given the fact that we have both means that it’s not technically difficult; it’s something like a 2 or 3, if you have people trained on the tools like QlikView, Cognos, or BusinessObjects. And you get down to that 2 or 3 because of the investment you’ve made in the culture, the EHR, and the data warehouse.

The Health IT Summits gather 250+ healthcare leaders in cities across the U.S. to present important new insights, collaborate on ideas, and to have a little fun - Find a Summit Near You!


See more on