National board starts wheels rolling to develop standards for cancer care
IOM panel seeks better data, improvements in access
Overuse and underuse of procedures lead to inappropriate care for some cancer patients. But without more research and a system of quality measurement, it is impossible to know how often the country’s cancer care system fails patients, according to a report by the National Cancer Policy Board of the Institute of Medicine and the National Research Council in Washington, DC.
Lack of data and quality studies on cancer care frustrated the panel’s 20 members as they sought to assess the state of cancer care, says Maria Hewitt, DrPH, senior program officer at the Institute of Medicine. "Why is it that we don’t really know what’s going on now with people with cancer?" she says. The board was to convene a workshop in October to address the need for national data.
But the board did find enough evidence for one care-related recommendation: Patients undergoing technically difficult procedures that have been associated with higher mortality in lower-volume settings should receive their care from high-volume facilities
Nine other recommendations addressed the need for treatment guidelines and quality measures and improvement in such areas as end-of-life care and access. (For a list of recommendations, see p. 123.)
"We do know there is wide variability in quality in the small samples that have been taken around the country [in studies]," says Joseph Simone, MD, vice chairman of the panel and medical director of the Huntsman Cancer Foundation and Institute at the University of Utah in Salt Lake City. "But we don’t know the general level of cancer care across large populations. There is no mechanism for measuring quality."
Some efforts are emerging to define both process and outcome indicators. The Foundation for Accountability in Portland, OR, developed a breast cancer measurement set with both patient-based and clinical measures. The National Comprehensive Cancer Network, which is composed of 17 leading cancer centers, has developed treatment guidelines and is phasing in a breast cancer measurement set that assesses adherence to the guidelines. The National Committee for Quality Assurance in Washington, DC, has an oncology panel that is considering new measures but currently focuses on cancer screening.
"There are beginning to be some suggestions [from clinical studies] that it may matter quite a bit where you get your cancer care. And that’s a scary statement," says Jane Weeks, MD, associate professor of medicine at the Dana Farber Cancer Institute and Harvard Medical School in Boston. Weeks is also principal investigator of the cancer center network’s outcomes database project.
"We need a process that establishes whether that’s true and what the nature of the differences are, and begins to work on getting rid of those differences," she says.
When it comes to measuring outcomes, cancer care has lagged far behind other medical conditions such as cardiovascular disease and diabetes. In part, that is due to the complexity of cancer and its treatment, says Simone.
A cancer patient may receive care from a primary care physician, surgeon, medical oncologist, radiologist, and pathologist. "The quality of each of their inputs is a factor in outcomes," says Simone.
In addition, characteristics of the patient and the tumor greatly affect outcomes. Patients may choose more extensive surgery than recommended or refuse additional radiation therapy or chemotherapy. And for some cancers, the incidence rate is too low to allow the accumulation of enough cases for comparison at all but the highest-volume centers.
Despite those difficulties, the National Cancer Policy Board and other leading oncologists point to possible markers of care as a starting point for data collection. Complete documentation in the medical record is essential even for adequate clinical studies, says Simone. The board is made up of consumer representatives, health care providers, and researchers.
"If you randomly review charts in some studies, you will find out that the pathology report doesn’t mention whether the margins [of the tumor] were clean or not," he says. "This is a key ingredient in trying to plan future care for patients.
"We don’t have any mechanism for assessing what proportion of chemotherapy is actually given to the patient," he adds. "It does make a difference in some forms of cancer if they receive a full dose or consistently low doses."
At the same time, cancer care has a rare repository of data in these two national databases:
The Surveillance, Epidemiology, and End Results (SEER) cancer registry is maintained by the National Cancer Institute and covers about 14% of the U.S. population in certain geographic locations.
"They capture almost all the cases in the areas in which they serve," says Hewitt. "Those data are used to generate a lot of the numbers you see in the press about cancer prevalence." However, the regions included in the SEER database aren’t representative of the entire U.S. population, she notes.
The National Cancer Data Base (NCDB) is a joint project of the American College of Surgeon’s Commission on Cancer and the American Cancer Society and contains information on more than half of all newly diagnosed cases of cancer nationwide. One weakness of the NCDB is a lack of information about outpatient care, says Hewitt.
Meanwhile, most hospitals maintain tumor registries with some basic information such as stage of cancer at diagnosis and mortality rate, and some states mandate registries. By linking such data with Medicare files, peer review organizations have conducted quality improvement projects in cancer care.
"About 60% of people with cancer are 65 or older. There’s tremendous opportunity to improve quality through HCFA’s [Health Care Financing Administration] efforts," says Hewitt.
The most detailed effort to define and measure quality is emerging from the National Compre-hensive Cancer Network (NCCN), which is developing process and outcomes measures based on an array of treatment guidelines. The first such measurement program began two years ago and focuses on breast cancer care.
"The guidelines are quite specific about how patients in a given category should be treated," says Weeks. For example, the breast cancer guidelines address such issues as breast-conserving surgery and radiation after surgery. They state what kind of chemotherapy should be used in certain circumstances and the duration of the treatment.
Centers collect data that show their level of adherence to the guidelines. Weeks acknowledges that in some cases, patient preference affects the care. For example, a physician may recommend breast-conserving surgery, but the woman may feel a mastectomy gives her greater peace of mind.
The network decided not to include information on patient preference in the data analysis. "There’s no reason to think that patients at one institution would be more likely to want a particular therapy as opposed to another," she says. "If we find three times the rate of mastectomy at one institution vs. another, it’s pretty unlikely that that’s due to patient preference."
The NCCN program also illustrates the resources required for effective measurement. Each of the cancer centers has a full-time data manager. Centers must follow a basic time line of reporting, collecting information at the patient’s first visit, four- and nine-month follow-up, and annually. Some centers update the database in a real-time fashion at every routine visit.
The NCCN collects data on "every patient who has a new diagnosis of breast cancer and is receiving some or all of primary care at the institution," says Weeks.
Results are reported at a facility — not physician — level. The report is coded so one facility learns its own rates and the range among other centers but not the high and low performers.
The NCCN project is funded by member fees and grants from the pharmaceutical and biotech industries. The network is also seeking foundation and federal grants.
"We’re trying to demonstrate that it’s possible to move this whole field forward, building on a tumor registry, and to move [outcomes management] to the next century," says Weeks.
While the National Cancer Policy Board advocates standards-based measurement, setting up such a system nationwide would be a complex and costly task, she notes.
"What I really hope will come out of this report [by the cancer board] is some hard thinking about where the resources ought to come from and how much as a society we can afford to invest in doing this sort of data collection," she says. "There are probably lots of people out there getting suboptimal care, and that’s not acceptable.
"A lot of places, if they knew they weren’t doing a great job, would be willing to change what they’re doing," she says. "But first they need that information."
[Editor’s note: A copy of the report, "Ensuring Quality Cancer Care," is available from the National Academy Press ($34.95 plus $4.50 shipping and handling for the first copy and $.95 for additional copies) at (800) 624-6242 or (202) 334-3313 or at the Web site: www.nap.edu.]
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content