Tuesday
February 15, 2005
A few years ago, primary care doctors at tiny Peace Harbor Hospital on the Oregon coast were handed a report stuffed with bar charts rating how well they were managing their diabetic patients. One ranked the 10 physicians against their peers on the percentage of patients who had received a blood-sugar test. Another showed how well each doctor made sure patients were tested for protein in their urine.
For some physicians, there was uncomfortable silence.
"I think some people were unpleasantly surprised at how they were doing," says Sherry Catlin, M.D., medical director for the Siuslaw region of PeaceHealth, a six-hospital health system in the Northwest that includes 21-bed Peace Harbor. "Physicians think they're working hard and conscientiously trying to give their patients good care. But they don't think of everything, every time for every disease."
PeaceHealth was out to change that. The report was among the first fruits of its new disease registry for diabetes patients, implemented in 1999. Today, physicians aren't measured simply on whether they remembered to order the appropriate tests--they're ranked on how well their patients' diabetes is controlled. And Catlin says there are signs the registry has helped: The percentage of patients with blood-sugar readings within the recommended range (a hemoglobin A1c of less than 7 percent) increased from 44 percent to 60 percent in the three-year period ending January 2004.
Chronic disease registries were virtually unknown to all but a handful of pioneering providers as recently as five years ago, says Jane Metzger of Boston-based First Consulting Group, author of a report on disease registries released in early 2004 by the California HealthCare Foundation. But a confluence of factors--including the emergence of pay-for-performance and public quality reporting measures--is driving providers to better manage chronically ill patients, who drain the U.S. health care system of nearly 80 percent of expenditures.
"There is a much greater realization that you need to have some way of knowing who these patients are and helping point-of-care practitioners organize what we should be doing with them," Metzger says. "You could never pool your medical records and do that."
Chronic disease registries are clinical information systems that capture and track a relatively narrow range of key data for a specified population, such as diabetes, asthma and congestive heart failure patients. They serve as repositories of recommended clinical interventions, helping clinicians make proactive, consistent care decisions and anticipate problems.
A registry is not an electronic health record, nor is it nearly as complex or costly to implement and manage. Some are stand-alone applications, while others are embedded within an EMR. They range from free database programs to Web-based systems loaded with functionality--automatically alerting a physician during an exam that a pediatric asthma patient has not been prescribed an inhaled steroid, for example.
Registries are used in three ways:
At Peace Harbor, a diabetes wellness visit works like this:
Using an exam-room computer, a health assistant enters vital signs and other data directly into the registry, which is displayed as a separate screen in the EHR. As the physician works through the electronic chart, he or she fills in boxes in the registry screen that are specific to the patient's diabetes--there might be early signs of a foot ulcer, for example. A preprinted list of alerts also identify uncontrolled aspects of the patient's disease. One prompt might say: "Patient's last urine test showed protein in urine, consider a medication."
The ability to capture data electronically is key to a successful disease registry, according to Cambridge, Mass.-based Institute for Healthcare Improvement, which offers a list of tips for implementing registries.
Steps should be taken to shield doctors from having to punch in the same information they just finished scribbling in a chart, IHI says.
"Data entry and maintenance are part of the ... daily operations, not an 'add-on' task at the end of the day," reads a list of recommendations on IHI's Web site. The institute recommends pulling electronic data from as many external systems as possible to minimize data entry.
But computer terminals in the exam room are the exception rather than the rule. More often, the handwritten notes on a visit planner must be manually entered into the registry, usually by a nurse, the medical records department or another data center.
Cambridge Health Alliance in Massachusetts, which includes three hospitals and more than 20 primary care practices, has an adult diabetes registry of more than 4,000 patients, and also tracks patients with pediatric asthma and depression. Because most clinics remain on a paper-based system, the doctors' handwritten notes must be typed into the registry after a visit.
Temporary staff were hired to take the data-entry onus off physicians, says Hilary Worthen, M.D., senior director of clinical informatics. But data entry has been minimized by interfacing the Web-based, homegrown registry with the hospital information system, which automatically streams patient information into the registry every 10 minutes.
Cambridge Health also uses the registry for patient outreach. Clinicians routinely run queries of their patient panels, organized by, say, lipid levels for diabetes patients. Within seconds, a report spits back a list of patients who are overdue for a lipid panel. The list is passed along to a care manager for scheduling.
"To be able to take an aggregate look at their panel of patients helps our clinicians identify gaps in care," Worthen says. "As a primary care doctor, it's impossible to get all these things on your radar screen and arrange them in a 20-minute visit."
Proponents caution, though, that any registry will be rendered useless unless it's part of a broader disease-management program, spearheaded by medical leadership.
"The redesign for a registry is minimal, but the redesign for chronic disease management is radical," says Dennis Pope, vice president of administration at Luther Midelfort, a four-hospital, 15-clinic health network in west-central Wisconsin. "Most people get lost in thinking that the computer that tracks the stuff is the program; it's not, it's a supporting player in the program."
Indeed, registry implementation typically progresses much as it did at Luther Midelfort, where the board decided in 2002 that it wanted to improve chronic care. A subcommittee of about 15 people met weekly for 18 months to shape the registry and other disease-management components. The group was led by Pope and the chief of cardiology, and included physicians, nursing managers, IT personnel, a behavioral health psychotherapist and leadership from medical records departments.
Luther Midelfort now has registries for diabetes, congestive heart failure and preventive health across three ambulatory sites, and was scheduled to roll them out systemwide in January. The preventative health registry will include all patients--even healthy ones, Pope says. The idea: Visit planners will remind physicians to order preventive interventions such as a mammogram or colonoscopy.
It's up to providers to choose which data elements they want to include in a registry, but there's no shortage of resources to tap. Disease-specific associations such as the American Diabetes Association are a good starting point. Quality improvement organizations also are good sources, such as IHI, the American Medical Association's Physician Consortium for Performance Improvement, Bridges to Excellence and the federal Agency for Healthcare Research & Quality.
The real challenge, experts say, is not to get bogged down with too much data. IHI recommends starting small; secondary data elements, such as a patient's self-care goals, can be added later.
Advocate Health Centers in suburban Chicago, which now has registries for five chronic conditions, started its first registry, for asthma, in the late 1990s with only the patients' names and contact information. The goal was simply to figure out who the asthma patients were, says Pankaj Patel, M.D., medical director for quality improvement. The registry expanded over the years, including the addition of a patient survey aimed at identifying which asthma patients were most likely to end up in the hospital.
So is there evidence that chronic disease registries improve care?
Many users say measuring effectiveness is tricky. One reason: New patients added to the registry can dilute results. Also, it can be difficult to distinguish between the improvements resulting from the registry and those stemming from other components of the disease-management program, such as patient education.
But some providers cite anecdotal evidence that registries vastly improve their ability to manage chronic conditions. The federal Bureau of Primary Health Care, which has helped more than a thousand community health centers establish multiple disease registries as part of its Health Disparities Collaboratives, has seen the average hemoglobin A1c level for more than 135,000 diabetes patients drop from 8.5 percent to 7.9 percent, according to Suzanne Feetham, a registered nurse and director of the bureau's division of clinical quality.
Since creating a mammography registry a few years ago, Advocate Health Centers has cut in half the average time between when a woman's mammogram shows an abnormality and when she receives a biopsy, Patel says. At Cambridge Health Alliance, the registry has helped slash emergency room visits and hospitalizations among pediatric asthma patients by 50 percent.
IT costs for a registry are modest--much less than an EHR. In fact, free registries developed by quality improvement organizations are the most common form of disease registries now in use, according to another report released by the California HealthCare Foundation in May.
The report, which evaluates 16 free and commercial registries, says public domain registries generally lack technical support, flexibility and the ability to import electronic data from external systems such as the lab. Commercial registries are more robust, featuring tech support and point-of-care tools such as automatic reminders. One vendor, DocSite, charges around $500 per physician per year for its Web-based registry, which allows for users across multiple sites.
The labor costs associated with registry maintenance are more significant yet tougher to peg. The California HealthCare Foundation report says that upkeep such as data entry and patient outreach "requires a substantial labor effort," although it says those tasks typically are absorbed by existing staff. One of the most time-consuming aspects of implementation: selling the concept to "independent-minded" physicians.
Providers should not expect any near-term return on investment. "For clinical systems such as chronic disease registries, there are rarely enough financial benefits to offset the costs," the report reads. But proponents say a registry is a mission-driven initiative that can't be viewed through an ROI lens.
"If you're going to start out looking to get green dollars in return, you'll never get anywhere," says John L. Haughom, M.D., senior vice president of the Healthcare Improvement Division at PeaceHealth. "You do end up saving money, but you can't start out that way."
John Haughton, M.D., CEO of DocSite, concedes that the lack of a clear financial incentive likely is the largest barrier to wider acceptance of disease registries. "Most people adopting them aren't looking at them for an ROI," Haughton says.
Even so, many early adopters say they're convinced that registries produce a financial return--or will someday. Pope says better chronic care might prevent revenue-generating hospitalizations, but it also frees up time and space for other acute encounters. "It's a double positive," he says. "It allows us to take in new patients because we have the room, and we're taking much better care of our chronic patients."
And there are rare cases in which successful registries have helped produce tangible financial benefits. A large independent physician association in New Jersey was able to negotiate better rates with local employers and health plans after showing improvement in controlling patients' asthma, according to the California HealthCare Foundation.
Advocate's Patel believes a robust ROI will accrue over time through the growth of pay-for-performance and increased recognition, which can boost volume and provide leverage in contract negotiations.
"If you put the whole package together, the case for disease registries is very compelling," Patel says. "It isn't that complicated or expensive, and returns are so big. I can't imagine doing any clinical initiative in any area without having a disease registry as the centerpiece of the program."
Mike Colias is a freelance writer based in Chicago.
GIVE US YOUR COMMENTS! Hospitals & Health Networks welcomes your comments on this article. Simply click on the computer mouse next to the appropriate headline, then type in your comments. Alternatively, you can e-mail your comments to hhn@healthforum.com, fax them to H&HN Editor at (312) 422-4500, or mail them to Editor, Hospitals & Health Networks, Health Forum, One North Franklin, Chicago, IL 60606.