Francis X. Solano, M.D. and Jim Venturella, of the 20-plus-hospital University of Pittsburgh Medical Center (UPMC) health system participated in a virtual roundtable discussion convened by HCI Editor-in-Chief Mark Hagland this summer, as part of the October HCI cover story. Dr. Solano is vice president in the Physician Services Division at UPMC, and president, Community Medicine, Inc., and is also medical director of the Donald D. Wolff, Jr. Center for Quality, Safety, and Innovation, all at UPMC. Jim Venturella is CIO, Physician and Hospital Services, for the UPMC organization. Solano and Venturella gave Hagland an extensive interview as part of the process of the creation of the roundtable article. Below are excerpts from that longer interview.
Dr. Solano, your umbrella organization, UPMC, has long been a pioneer in leveraging clinical IT to address core issues in patient care delivery. Where do you see the nexus between performance improvement in care delivery right now in the medical group setting, and what IT can do to improve that performance?
Solano: Getting rid of the variability in care is what’s keeping me awake at night. I just don’t realize why so many physicians are so variable in their care in so many ways, or why so many physicians are still not optimally using some of the tools in the electronic record. Some of it has to do with such elements as best practice alerts, preventive maintenance screens, and allergy-allergy checking: if you ignored them in the paper world and you ignore them in the electronic world now, you’re going to be mediocre. And the thing is, this transparency train is coming, and physicians have to be ready for it and be aware that it’s coming.
Also, it’s not automatically true that quality care will cost less. And that conundrum is out there, and it bothers me if we’re going to be benchmarked based on quality and cost. If you happen to work in a quaternary care center as I do, your costs will be a lot higher, because you’re supporting a teaching institution. We actually did an interesting study: we took our top 50 providers and bottom 50 providers in primary care, and found that it actually cost more money to provide higher quality in a care group; so that’s out there. And the newest conundrum that we face is, how do you go from a volume-based payment system, to doing the right thing in a quality-based system? And that will be our biggest challenge. How do you start to align yourself to set up what you have to do. You’ll have to make some changes around utilization, around the use of diagnostic tests and drugs.
And how does that translate into what IT implementation can do to support performance improvement?
Jim Venturella: It gets to the workflow and efficiency issues that Dr. Solano talked about. In some of the practices, the doctors are well-educated, and our tools may not be as efficient as they could be, so we’re trying to figure out how to make the tools more efficient; because some of the tools actually make their day less efficient. The other area is looking at devices: how do we use devices more efficiently? How will we use the new mobility tools more efficiently in the clinic setting?
And Dr. Solano talked about reporting and report cards. And in fact, we’re still in the very early stages of working with those elements. There’s still a lot we can do to pull out data and report it to the physicians.
How are you working with the medical specialists to help bring them along on the need to improve care quality performance?
Solano: About three years ago, we set up a physician service division quality leadership group, made up of about 30 key physicians from all the subspecialty departments, from primary care, from our health plan leadership, and from key clinical informaticists like Dan Martich [G. Daniel Martich, M.D., UPMC’s CMIO]. It started out with PQRI, now PQRS [the federal Physician Quality Reporting Initiative, now the Physician Quality Reporting System], and now meaningful use. The other thing is getting specialists to think about elements involved. Cardiothoracic surgery, for example, has a specialty-wide database to draw on. Transplantation has a national database as well. Bariatric surgery has its own key quality metrics; and OB/gyn has key quality measures as well.
You have to begin asking specific questions in every specialty area. For example, among your diabetic patients, what percentage have had microvascular or macrovascular complications? Using tools like Archimedes [the data modeling and analytics solution from the San Francisco-based Archimedes Inc.] is important. So we’re struggling in some subspecialty areas where they really don’t have defined metrics, other than the crude things. So when you take out a gallbladder, what are you looking at in terms of outcomes measures? Well, you hope the patient can eat without intestinal issues. And in knee or hip replacement, when did the patient return to work, and when did the patient return to full functionality?
So we’re all just beginning to ask these kinds of outcomes-related questions, obviously. But you have to find ways to create the foundational technology for this work, correct?
Solano: Well, Jim has put together a team to do this work. But the meaningful use dollars don’t begin to pay for the creation of quality reporting systems in the electronic record. It takes a tremendous amount of resources. But we’re developing a program in Epic.
Venturella: We’re working with Radar. It’s a dashboard within Epic that you can use to present reports from either the revenue cycle side or the clinical side. It presents it in an easier fashion. We’re going to bring it out next month, and we’re going to start going live next month [September]. It’s actually a dashboard within Epic itself.
Solano: When you log in, you’ll see this dashboard within Epic. So, how many encounters, and how many were closed, etc.? And if you’re a CT surgeon, it will probably give you how you’re doing with your SpS [spinal stenosis] benchmarks, and so on. The biggest challenge is, you hope that once you build it, they will come, but that’s not always the case. And so getting this so that it becomes part of their standard workflow, is important.
How long will it take for even large medical groups to do this?
Solano: Well, at least large organizations have the resources for this. And there is some out-of-the-box reporting, but we’ve had to spend a lot of time building our own reports so they look the way you want them to look. And ensuring attribution—that even within my own practice, my patients are correctly attributed to me and not to one of my partners—is a challenge. We’ve just purchased a tool called Crimson that helps with quality and cost measures. And we have nine physicians in my group; and is it the attending of record, or the person who does the initial history and physical, or the physician who discharged, who should be held responsible?
And hospitals are going to get a lot of pushback from docs, because of the way the hospital reports the data. If Fran Solano is the PCP, even if Fran Solano didn’t see the patient, Fran Solano is the patient to whom the patient is attributed. So attribution is a huge issue moving forward, especially in the electronic world. And it’s even a bigger headache when profiling specialty care. Say a patient goes to a cardiology clinic, and sees three or four doctors over the course of a few months, which guy gets the reports?
Venturella: I think the big takeaway from that is that there’s a ton of work to do from a technology standpoint, to get the systems in and established, but the operational and workflow issues that follow are tremendous. And there’s a tremendous amount of work involved. There are multiple ways to do things in these electronic systems.
