skip navigation

CHS Research News
Vol 20, Issue 1, Spring 2008

Feature Article

"Content of Care" program seeks standardization to improve outcomes affordably

By Rebecca Hughes

Imagine if you could improve care, while controlling costs. What if health care weren’t way too much, or too little—but “just right”?

That may sound like a fairy tale, but it’s just what the Content of Care program seeks to do every day in the real world of Group Health. How? By exploring what care is made of, then finding unwarranted variations in diagnosis and treatment—and finally reducing those variations.

Launched in 2007, the Content of Care started as a joint initiative of Group Health Center for Health Studies (CHS), Group Health Cooperative, and Group Health Permanente (the Co-op’s medical staff), with support from the CHS Development Fund.

“The Content of Care is like an in-house version of the Dartmouth Atlas Project,” explained Matt Handley, MD, associate medical director for quality and informatics at Group Health.

For over two decades, the Dartmouth Atlas Project has analyzed Medicare and other data about how medical resources are distributed and used throughout the United States. Led by Dartmouth professor John E. (“Jack”) Wennberg, MD, MPH, the Project has shown that the care American patients actually receive tends to vary widely, for no good reason.

Care by ZIP code
“Far too often, U.S. care correlates more with a patient’s ZIP code than with any clinical evidence or patient preference,” said Paul Sherman, MD, Group Health’s medical director of consultative specialty services. Sherman co-chairs the Content of Care oversight group with Handley.

This variation often leads to worse outcomes and higher costs. The drawbacks of lack of access to care are obvious. But more care isn’t necessarily better, either, as described in the Dartmouth Atlas Project’s just-released 2008 report and in Shannon Brownlee’s book Overtreated: Why Too Much Medicine Is Making Us Sicker and Poorer. Overuse of “supply-sensitive” fee-for-service care in tony ZIP codes like 90210 (Beverly Hills) often leads to outcomes as bad as those in poor areas with rampant under-treatment.

The Dartmouth Atlas Project has found that decisions about care are often driven more by physicians’ practice patterns than by patients’ preferences. And across the board, many kinds of effective, evidence-based care are underused, even in the nation’s “best” hospitals.

The Northwest is already providing care that’s generally as good as or better than the nation’s average, said Sherman. And Group Health is already providing care that exceeds our geographic area’s average quality. “Group Health’s incentives are aligned to improve care and, as a result, control costs, by eliminating the “costs of poor quality,” said Sherman.

Room for improvement
“But we still have room for improvement at Group Health,” said Tiffany Nelson, who directs the Content of Care program. “That’s where the Content of Care comes in.” This program doesn't stop at identifying variation at Group Health. “We’re working to ensure that when variation exists in the care we deliver, it provides value and is driven by clinical circumstances and patient preferences.”

CHS Analyst/Programmer Tyler Ross, who has worked on the program since its start, describes how the Content of Care team provides specialty services, one by one, with data from their own clinical practice: That means pulling clinical and cost data on each service’s most common diagnoses and their associated procedures, prescriptions, hospitalizations, and follow-up visits. The upshot, which Ross called the “intersection among business, research, and the practice of medicine,” is a profile of the eventual outcomes and total costs of each episode of care. These measurements across practitioners on the same service are compared, including the number of patients with a condition, and outcomes and costs for different approaches to diagnosis and treatment.

The program started by using data from Group Health’s electronic medical record (EMR) to compare specialty groups to the nation, the Northwest, and Washington state. With colleagues including CHS Senior Investigator Michael Von Korff, ScD, Ross examined variation over time and between Group Health internal and contracted network providers and among regions in the Co-op—and between individual practitioners within group practices.

Ross described the challenges of doing “data dives” for specialty groups within Group Health: For instance, if one doctor has done 10 of a certain procedure, and another 100, this might at first seem unusual. But it may be entirely appropriate, said Ross. It can make good sense for surgeons to specialize in different operations. “The variations that we target are those that are unwarranted according to either the evidence or patient preference” he added.

“The Content of Care team gives providers useful information about what they’ve been doing, and what their outcomes have been,” said David McCulloch, MD, medical director for clinical improvement, an active member of the program’s oversight group. “We help them review the literature. We work with them to decide where they can make a change.” Then, the Content of Care team helps them track it forward to see how their upstream change affects downstream outcomes and costs.

Getting providers engaged
“It’s been a learning process for all of us. We keep seeing how important it is for the providers to be engaged in this work,” said Ross. “Their clinical experience and expertise is crucial to finding ways in which we can implement changes in our medical practice.”

CHS Research Associate Katie Coleman, MSPH agreed. “We’ve cranked through many different ways to identify opportunities for practices to improve,” she said. “We found what matters most is for the medical teams to ‘buy into’ the process as early as possible.” She likened this to her experience with implementing the Chronic Care Model in the Improving Chronic Illness Care program outside Group Health. “We need to involve the teams, and make it a fun process for them,” she said. “Then we can support them as they identify and pursue their own best opportunities for improvement,” she added.

In 2007, the Content of Care team went through this process with five specialty practices. To date, the Content of Care’s best success has been development of new procedures to culture nose swabs from dialysis patients to reduce their incidence of staph infections. And Group Health urologists devised new standards to administer Lupron, an injectable medication for patients with prostate cancer.

“It’s a testament to Group Health commitment to evidence-based medicine that this work is so incremental,” said Coleman.

From research to production
The Content of Care team has moved from research to production mode, said Sherman. “In 2008, we plan to work with eight more specialty services.” Each service will try to identify two or three opportunities for improvement that could potentially save at least $500,000. Eventually, the goal is to work with every specialty service, plus primary care, within Group Health and with contracted providers.

“Like Lean, the Content of Care is a bottom-up process of continuous improvement,” said Sherman. “We’re working with our Lean folks to standardize our processes and make this become a routine part of our care.”

Sherman said he and other Group Health leaders are “jazzed” about the chances to base even more of Group Health’s care on evidence, and improve care and affordability—without, of course, ever rationing care. “We also hope to be able to tell the story,” he said. “I’d like to see this on billboards one day: ‘Come to Group Health, and you’ll live longer, happier, and less likely to die of diseases x, y, and z.’ We won’t just be saving money—but limbs and lives—and improving the quality of patients’ lives.”

Top