As the search to quantify healthcare delivery intensifies, the outcomes data movement appears to be gaining greater physician acceptance. Key to effective performance-measurement, experts believe, is physician participation in the data-collection process.
"In many groups, outcomes data collection is done by nurses and others who get the data from patient charts," says Rocelyn Lonborg, director of the office of value assessment at Seattle's Virginia Mason Medical Center.
"Physicians just aren't involved," Lonborg says. "If physicians' ideas haven't gone into what's being collected, if they don't get their own data back, and if the information is treated as a high-level report, there's bound to be resistance. If you want physician acceptance, you let your group participate, let them own the process, affect it and alter it."
The 370 physicians who practice at Virginia Mason are among six medical groups that in 1990 helped launch the American Medical Group Association's national outcomes database. Today, participants in the AMGA Outcomes Management Consortia collect data on patients with eight targeted health conditions: asthma, cataract extraction, diabetes, total hip replacement, total knee replacement, hypertension, sleep disorders and low back pain. Patients are tracked and, at standard intervals, comparisons are made about treatments, sources of care and results.
From the initial six clinics, the consortia has expanded to include 95 medical groups, which treat about 3,000 patients. Participation gives groups access to the other groups' data as well as their own.
Lonborg views the AMGA data as essential to establishing "benchmark" quality levels. "The significance of the AMGA data is that we can see ourselves in comparison with other organizations like us delivering the same type of treatment . . . . We might be satisfied with what we're doing but when we discover some other group is obtaining some better or different outcomes, it's part of our improvement process."
It's also a chance for others to benefit from what Virginia Mason is doing, for example, using outcomes data to improve total hip replacement care.
The first step in the improvement process was finding a way to standardize the selection of prostheses, which numbered half a dozen.
The surgeons studied the different prostheses and, in the process, became aware of some of the differences in the ways they practice. It provided an opportunity for them to question what they usually do, Lonborg says.
Comparing practice decisions propelled the surgeons to make qualitative changes, she says. "Physicians have a positive reaction to real data that tells them something. So when you're involved in outcomes measures and management, they believe this kind of information and therefore feel free to act on it. These doctors realized maybe there is a reason to not use just one type of prostheses. They could track patients through the clinical process and see differences in complications and failure of prostheses based in part on which prosthesis was being used."
A good example of the ramifications of outcomes data concerned a doctor using a new type of prosthesis that he believed was going to be better than the older models.
But after looking at his outcomes data, Lonborg says, he discovered certain complications and patient pain that hadn't occurred with older prostheses. He concluded that the new prosthesis should not be used and then he passed his findings along to his peers at a hip society meeting. A clear advantage to being the one to generate the data is that the results of the data didn't come to him anecdotally, and he had access to the results before any other surgeon in the country.
Such information is valuable ammunition in the arsenal of outcomes management, which seeks to perform systematic measurement of patient care and providers on a national basis. Today, the effort to provide outcomes data has been given a boost by the spread of managed-care organizations that seek to know not only how much they are spending but also the quality of care being given to their enrollees.
"We've been looking at outcomes for research purposes for 40 or 50 years in clinical trials, but we've only been focusing on patients' experiences for the past 10 or 15 years," says David Lansky, president of the Foundation for Accountability, a two-year-old Portland, Ore.-based organization aimed at helping patients make quality-based healthcare choices.
Among FACCT's goals is a plan to present outcomes data sometime next year to the public in a form consumers can easily use and understand.
"I see FACCT as a catalyst for a number of voices that haven't been very well expressed in technical and political databases," Lansky says. "We listen to consumers and patients, translating their experiences into quality measurements that physicians, hospitals and health plans can understand as expressions of the public's concern. I think we have begun to speak up about that perspective. It's not the only perspective, but it seems to be needed now to help reassure the public that it's being heard."
Physicians and patients alike benefit from outcomes data, Lansky believes. "I think it's fundamentally important for physicians to know how they're benefiting patients in a scientifically valid way. I think that for the most part, physicians are very supportive and enthusiastic about outcomes data because they're trained to help patients be healthy and achieve optimal health, and this is a tool that can help," he says.
Yet Lansky says many doctors are anxious about outcomes measurement. "They're not convinced that these measures are sound and reliable. And because it's not the kind of measurement they've been trained to use, they're uncomfortable these measures will be used to judge them and influence their pay and practice autonomy."
Lansky suggests physicians fears can be reduced by getting them involved in the outcomes-data process. "The worst scenario," he says, "is when doctors stand on the sidelines and complain instead of getting involved."
So far, the outcomes movement has had only a minimal impact on quality of care. "The primary impact has been psychological, not technical, because the information has not been standardized and is not widely used yet, despite efforts of the AMA and others," Lansky says.
"Few doctors are systematically taking steps to manage their patients to achieve improved outcomes," he says. "Few follow patient results and can tell you beyond their own recollection what the level of health is for patients they've been serving, or how fast patients with back pain are resuming work activities, or what the level of blood sugar control is for patients with diabetes."
Julie Sanderson-Austin, director of quality management and research for the AMGA, says there is a rational reason that outcome-based healthcare is still in its infancy.
"Healthcare is in such a state of flux today that most organizations can only worry about the bottom line," she says. "Most quality activities are focused on lowering cost and reducing resource utilization. I don't think we'll have a universal approach to medicine based on outcomes until we can get beyond focusing on cost--make it a national priority--and implement a strategy to compete on quality instead of costs."
One step the AMGA is taking to improve the outcomes-data process is to add risk adjustment measurements, Sanderson-Austin says. "Individual patients come into the healthcare system at different levels of illness," she says. "What they bring to the healthcare system impacts on their outcomes. You have to be able to measure what patients bring with them before you can measure the kind of outcomes they're going to have.
Despite slow progress, Sanderson-Austin remains optimistic about the value and application of performance measurements. "There are a lot of healthcare organizations that truly believe you can't quantify outcomes," she says. "But I think it's more accurate to say it's not easy to quantify outcomes when you haven't defined what outcomes you wish to create."
Cheryl A. Sweet is a freelance writer based in Phoenix.