Bringing Consistency to the 9-Month Episode of Care

By redesigning the care model using uniform best practices borrowed from surgical procedure, leaders at Geisinger Health System reduced perinatal variability.

This article appears in the July/August issue of HealthLeaders magazine.

The women's health physician and leadership team at Geisinger Health System started with a challenge: Could they take the core fundamentals of the health system's ProvenCareTM model of best practices from a surgical procedure such as a coronary artery bypass graft and apply those to perinatal care?

Geisinger's model—which combines a uniform set of best practices with accountable systems to make certain the right care is followed at the right time—had demonstrated results in surgical procedures. But could it work in an episode of care that is nine months long?

Harry O. Mateer Jr., MD, director of obstetrics and gynecology at Geisinger Medical Center, says despite some obvious differences between perinatal care and a surgical procedure, delivering consistent care involves the same elements.

"The basis behind ProvenCare is that we know that for certain procedures, whether it's a surgical procedure or a complex nine-month ordeal such as pregnancy, that there are certain aspects of care that should be offered to all patients at various points during the procedure," Mateer says. "So if it's a surgical procedure, all individuals should be offered certain things prior to the surgery, during the surgery, and then after the surgery, and those are usually called best practices in most modern health literature. Most physicians know what they have to do. It's just really making sure that it does get done."

In prenatal care, the best practices can sometimes be lab work, education, family history, social history, or radiology studies, Mateer says. What matters is that "all of those best practices are offered at the appropriate time for each and every individual, and that people don't fall through the cracks because you think somebody else did it and you don't have a good way of documenting and making sure it was done."

So how does a clinical team make certain it gets done? Hans P. Cassagnol, MD, associate chief quality officer of Geisinger Health System and director of obstetrics and gynecology for Geisinger-Northeast, says implementation is built around two primary tools.

"We actually used evidence-based medicine and the electronic medical record to come up with a set of best practice measures that we were going to hold all providers to," Cassagnol says. "What we have done over the past several years is use those two components with different ways of actually guiding the providers into delivering the evidence-based medicine at every single opportunity."

The first speed bump that many health systems may face is to create the initial set of best practices. Geisinger began with a set of 103 distinct best practices for perinatal care. "Some of those best practices can be as easy as something like taking vital signs," Cassagnol says, "recognizing that a blood pressure has to be taken at each and every visit. And some of those best practices are things that may just have to be offered once but at a specific time during the pregnancy. So in those 103 best practices, there are usually between about 240 to 300 times when those particular best practices need to be validated during a normal pregnancy."

Even with its history of integration and consistency of care, when the Geisinger team started the exercise it found "unnecessary variation" in how care was delivered across its 25 clinic locations in the region. The leadership team made certain to communicate that the goal of the program was not to restrict physicians but to give them a defined framework, Cassagnol says.

"One of the biggest misconceptions is that whenever we actually go through the process of delivering the best practice measures, people tend to think we are restricting providers from practicing a certain way," he says. "The goal is to provide overall guidance of what's the expected level of care in a particular situation. There's always going to be deviation from the guidelines. We just want to make sure people are thinking about the guidelines and what should be done within the best practice measure. If there's a good reason to deviate from that, we just want to make sure there's an active thought process behind
the deviation."

Any set of best practices has to be fluid to embody the latest evidence. One example is when growing evidence from the American College of Obstetricians and Gynecologists suggested that women who delivered via elective cesarean prior to 39 weeks saw an elevated risk of complications.

John Nash, MD, chairman of the department of obstetrics and gynecology at Geisinger Health System, says the physician team quickly moved to adopt procedures to avoid elective cesareans prior to 39 weeks.

"All we had to do as a group was say, 'This is now what [ACOG] says is the standard. It is the best practice.' How can we justify putting babies and moms at risk?" Nash says. "That risk is fairly small, but why put them at any risk? How could we defend ourselves if we got a bad outcome? So our Geisinger docs got together and said, 'We are not going to accept anybody that schedules a C-section prior to 39 weeks.' "

To hardwire this particular best practice, the electronic health record tracks all C-sections throughout the system, along with the gestational age of the baby and the reason for the C-section. If by chance an indication shows up in the system for an elective C-section prior to 39 weeks, a red flag alerts the senior obstetricians to contact the delivering physician in real time to review, Nash says. For the past two years, Geisinger's elective delivery rate prior to 39 weeks has been zero.

The order sets embedded into the health record are meant to support and remind physicians about the benchmarks of care, but not to bog down workflow. One revamp they call the "Result Consult" allows the physician to divide all of the different lab and radiology work by weeks of pregnancy. So a physician can simply check if a patient at 20 weeks has had all of the recommended studies, and if not to order them according to the timeline. If an opportunity is missed, then the alert reminds the physician at the next visit that the recommended care is overdue but still within the time window to correct, Cassagnol says.

"Hard stops"—orders that a physician has to click though in the course of the patient encounter—are built into the health record selectively, Mateer says.

"We also have what are called best-practice alerts for certain high-importance areas, such as receiving Rh immune globulin for our pregnant patients who are Rh-negative," Mateer says. "If something has not been completed at a particular point in the pregnancy, an alert will come up at the very top of the patient encounter. Then a drop box will allow you to complete that in a very easy, timely fashion. We didn't want to have best-practice alerts for every component because that can get overwhelming if you have a hard stop to every component of care, but for certain crucial areas of care we felt that hard stops were beneficial and would then be harder for a physician to miss."

HealthLeaders Media LIVE from Geisinger: Women's Health Leadership Discover how Geisinger Health System used its ProvenCare model to streamline perinatal care, created stronger patient engagement and communication, and drove better outcomes for both safety and quality.

Reprint HLR070813-10

This article appears in the July/August issue of HealthLeaders magazine.