The push towards creating true clinical integration across a health system that encompasses a hospital or hospitals, and physicians and physician groups, inevitably requires a great deal of strategizing and a lot of agile executing. David Higginson, senior vice president and CIO, and Bob Campbell, senior vice president of business development and chief strategy officer, are among those leading their colleagues on a journey towards clinical integration, at Phoenix Children’s Hospital.
Inevitably, developing the first-of-its-kind clinically integrated pediatric care network, has presented somewhat unique challenges, as well as opportunities. But Campbell and Higginson and their colleagues have moved forward confidently on their path. As it stands, the Phoenix Children’s Care Network, which is both a network and an accountable care organization (ACO), encompasses 824 pediatric care providers (physicians and allied health professionals) in the Phoenix metropolitan area, 50 percent of whom are located in Maricopa County, where the city of Phoenix is located (though the network spans five counties in central Arizona), and those providers care for over 750,000 unique patients in active in the network’s population health database. That database is tracking 14 primary care outcomes measures and 34 specialty measures, and providers are accessing its physician portal to work with primary care measures.
Examples of some primary care and utilization measures include Chlamydia screening, diabetes hemoglobin a1c tracking, body mass index assessment, asthma management, immunizations, well-child visits, ED utilization, and hospital readmissions rates.
They and their colleagues have been partnering with the Chicago-based Valence Health, which specializes in clinical integration and population health development, leveraging IT.
Higginson and Campbell spoke recently with HCI Editor-in-Chief Mark Hagland regarding their journey towards clinical integration. Below are excerpts from that interview.
Let’s talk about your organization’s journey towards clinical integration.
David Higginson: Phoenix Children’s started looking at clinical integration a little over two years ago, and obviously, with an accountable care organization initiative moving forward and the need to unify pediatricians in the valley and the attempt to control costs and improve care, we partnered with Valence. That company’s origins were as a claims processor but has leveraged their knowledge into collecting data from many providers and then providing great actuarial tools and other knowledge tools, to make it easy for providers to learn from it.
As you know, one of the greatest challenges in accountable care is in health information exchange. And what Valence came up with and is fairly unique, in my view, is that they really take it upon themselves to connect all the providers. So rather than a large population health tool or HIE [health information exchange] tool where the host organization will be tasked with building all the interfaces and it may take 6 months and $50,000 per connection, they quickly connected 800 providers and began pulling lab and clinical data into the database within a very short time; so six months within signing that contract, we had data flowing into the database. So we were able to show early wins. Also, some of those were members of other ACOs that hadn’t achieved this.
Tell me about your accountable care organization?
The Phoenix Children’s Care Network is a private ACO. It is a non-exclusive (providers can be in other networks) pediatric-focused network with general pediatricians and hospital-based pediatricians, including specialists, including a lot of subspecialists. I last heard we had 60 percent of the pediatric providers in the area in our network; and we have 832 providers submitting their data to the Valence database. We have more providers than that in the ACO, but not everybody’s connected yet. But the database represents 80 percent of the practices submitting their data.
When did you go live with the database?
We signed the contract, and then it took six months to get scheduled and started; and then in quarters two and three of 2014, we implemented and collected the tools. And then with regard to the use of the tools, we began fully rolling it out in the first quarter of this year.
What kinds of data are involved?
What has really transpired at a fairly high level is that, rather than using canned sets of reports or measures, the leaders of the ACO had to agree what their quality measures would be and tightly align those with nationally accepted measures—so, we had to define pediatrically relevant measures. So there was a whole cross-team group—where they agreed on those quality measures.
That was our first-prong, short-term approach, to get up and get going. It’s critical for ACOs to gain credibility by bringing some things live. And now we’re focusing on care coordination and population health management, so we’re beginning to work with valence and other vendors to support that. So we’re now building on those tools, to manage that population.
For children, population health inevitably involves the participation of parents for it to work, right?
Yes, obviously, the child will have little to do with the factors around getting consistent care.
What have the biggest challenges been overall?
The care coordination aspect—and that we’re just beginning to solve. But sometimes simple things like getting the patient to the appointment, can prove to be challenging to overcome. If the parents have to take the bus or something, what can we do to make the scheduling work for them?
Bob Campbell: Most of our challenges at the ACO or integrated network level have been in aggregating and integrating data to establish baselines, and then using the data to affect the care model. Our overall challenge is the implementation of a more effective care model to improve quality and reduce costs; that remains the biggest challenges.
What kinds of measures are involved?
Campbell: We’ve developed measures for each specialty as well as general pediatric measures.
Tell me about what the process of getting clinical consensus has been like.
When we set up our clinical integrated organization, we set up a quality committee through the board of the ACO. And the task of that committee is to define the merits that will be used. The committee consisted of 12 participating physicians. A number of them were general pediatricians, and then a cross-section of specialists as well. So we got input from payers in terms of what they were interested in; we had interest from our state Medicaid program around integrating metrics that as of this year are now mandated to track. And it was very challenging in terms of where to start and it wasn’t clear, in terms of pediatrics. In the adult world, you have the MSSP ACO metrics; we didn’t have that, so we looked at our greatest opportunities and priorities, and developed our metrics and dashboards out of that. The other challenge we’ve had is that there is the tendency of payers to focus on measures of individual interest to them. But we wanted to standardize across multiple payers.
What have been the biggest IT challenges so far?
Higginson: There really hasn’t been a significant amount of IT challenge, because our vendor, Valence, part of the value they provided was to do most of the legwork to get providers connected and get the data flowing. And because they had been a claims processing organization in the past, they were all set up for that. Of course, there’s the general struggle of making sure all our contact information is updated for all providers. And in the past, we didn’t have to have that contact information 100-percent updated in the past, but now we do. Other than that, from an IT perspective, there hasn’t been that much groundwork to do, which has been great. But we did create a single log-in for PCH or the ACO. We effectively devised a single log-in to the portal. So as you can imagine, providers already have enough user IDs and passwords and don’t want additional ones. We also developed mobile applications; the physicians in the ACO wanted a directory through an app, and they have that now through an app on Apple and Android. And so they could share their information privately.
What have been the biggest lessons learned so far?
Campbell: From my perspective, one of the biggest lessons learned was to make sure we have a well-developed plan for the start-up of the organization. In many cases, there’s a tendency to try to develop too much too quickly. So having a well-designed process—developing consensus among the physicians—and investing time and effort in developing a shared understanding and a plan that can be incrementally implemented, have proven to be very important. And David was saying we also tried to take a practical view of what was possible in terms of data collection. One lesson learned is selecting a single vendor with a track record, to get the process up and running; that’s very important. Also, you constantly have to make refinements as you move towards a more integrated system of care, but getting all that stuff up first is critically important.
Higginson: We’re really been in the learning phase, and were fortuitous in picking a product that worked well. Don’t start with a grand design and spend three years on it. Start with a fairly attainable goal, get boots on the ground, and move forward. We continue with our approach of continually adding to the solution. And from a data perspective, what worked for Phoenix Children’s Hospital in terms of data collection didn’t necessarily work well for the ACO. For years and years, we’ve been thinking somewhat narrowly only about Phoenix Children’s in terms of data and IT, so there’s had to be a mindshift from a narrow corporate perspective, to a mindset of thinking about the ACO going forward.