MORE CANCER DEATHS CAN BE AVOIDED BY FOCUSING
NATIONAL QUALITY INITIATIVES ON FACTORS IMPACTING
LONG-TERM SURVIVAL

National Cancer Database Research lends strong support
for utilizing quality measures

CHICAGO: In a first-time study to be published in the August 1, 2008 issue of the Journal of Clinical Oncology (JCO), researchers from the National Cancer Database (NCDB) of the American College of Surgeons (ACS) compared patient mortality in the days and weeks following cancer operations with survival outcomes five years beyond surgical treatment. “We were surprised by the number of potentially avoidable deaths that could be avoided long term if care at lower volume hospitals could be improved,” reported lead author Karl Y. Bilimoria, MD, former research fellow at the American College of Surgeons and general resident at Northwestern University, Department of Surgery, Chicago. Dr. Bilimoria and colleagues compared these outcomes in the highest-volume hospitals with those in the lowest-volume hospitals across the country. 

This study “is not just another article on cancer surgery and patient outcomes,” writes Nicholas J. Petrilli, MD, in the commentary that accompanies the research published in JCO. “Dr. Bilimoria and associates look at this issue from a different perspective,” he writes. “The authors’ objective was to determine whether differences in hospital surgical volume have a larger effect on perioperative mortality or on long-term survival.”

“In some ways, focusing on long-term outcomes is a common sense observation,” Dr. Bilimoria said. “A small percentage of cancer patients die in the hospital just after a surgical procedure. More people die of cancer in the long term. Current national quality improvement efforts primarily address the perioperative period, but few measures address long-term surgical cancer care,” he explained. “We wanted to know whether the focus on the perioperative outcomes is merited or whether quality initiatives should be expanded to address factors affecting long-term outcomes. No one had previously directly compared short-and long-term outcomes to see where the most lives could be saved,” he said.

The study team followed 243,103 patients from the National Cancer Database who underwent surgical procedures for non-metastatic colon, esophageal, gastric, liver, lung, pancreatic, or rectal cancer from 1994 through 1999. The team used two approaches to compare survival during the perioperative period within 60 days of death and five-year survival excluding perioperative deaths. “First, we compared the magnitude of the relative differences between hospitals in both time frames, perioperative and long term. Second, we calculated the number of potentially avoidable deaths if outcomes could be improved at lower volume hospitals to those seen at highest-volume hospitals,” Dr. Bilimoria explained.

The researchers discovered that hospital surgical volume was a factor for both time periods studied for all cancer sites, except for liver resection surgery, where no survival rate differences were noted. Upon evaluating outcomes during the 60-day perioperative period, the researchers found that overall, patients treated at highest-volume hospitals had significantly lower mortality rates compared to patients treated at the lowest-volume centers. Likewise, over a five-year period, patients who were treated at the highest volume centers had significantly higher survival rates compared with patients treated at lowest volume centers. However, this cancer care study was not done for purposes of evaluating surgical volume outcomes. “The purpose of this study was to try to determine the best strategy for directing future quality improvement efforts,” Dr. Bilimoria emphasized.

Therefore, the researchers performed an analysis of the hospital-volume data to see if deaths could be avoided if the low-volume hospitals improved their outcomes to the same levels as the highest-volume hospitals. They discovered that for the seven cancer sites combined, the total number of potentially avoidable deaths in the United States each year was 2,207 during the 60-day perioperative period, and 7,245 for long-term survival under these adjusted circumstances. “We found that the magnitude of the risk of dying is greater early on, but the number of lives affected long term is considerably greater.” reported Dr. Bilimoria.

“Our results indicate that quality initiatives should move beyond measuring factors affecting perioperative outcomes,” Dr. Bilimoria reported. In order to accomplish this goal, he explained, “the cancer care community needs to identify the things that high-volume hospitals do that affect long-term outcomes so that those steps can also be utilized in low-volume hospitals to improve their patient outcomes as well.” An example he cited in the case of colon cancer is that top performing hospitals generally examine 12 or more lymph nodes for about 80 percent of their patients. “In contrast, many low performing hospitals examine a dozen or more nodes in far fewer patients.” he explained.

Based on multiple studies in recent decades, quality improvement initiatives are already being developed to identify the practices high-volume hospitals follow to get good outcomes and transfer these practices to low-volume hospitals in hopes of improving outcomes there. It has been unclear, however, whether quality initiatives in surgical oncology should focus on factors affecting short- or long-term survival.

Based on these research findings, Dr. Bilimoria believes the key is to improve the quality of surgical treatment at all hospitals to potentially affect long-term outcomes. “We can achieve this goal by making sure we remove all of the cancer, take out the right number of lymph nodes, give all appropriate individuals chemotherapy and other treatments, as necessary, and aggressively follow patients for recurrences. We may also encourage some patients to participate in clinical trials,” he said.

Quality measures for cancer treatment have already been developed–specifically for pancreatic cancer and melanomas, Dr. Bilimoria said. “We assembled experts from across the country and from various specialties, and asked them what we absolutely need to ensure quality care for pancreatic cancer patients, as well as melanoma patients. We ranked the number of potential quality indicators and distilled that down to a valid quality measure that most of the experts believed important. For both cancers, we already have a number of quality measures that potentially affect long-term outcomes,” he said. Thus, findings from this study soon will be married to practical guidelines for low-volume hospitals to follow.

“Most patients in the United States undergo cancer resection at low-volume hospitals,” Dr. Bilimoria said. “Moving all patients to high-volume centers is an impractical policy initiative at the national level. Rather, we would like to find what those highest-volume hospitals do to get their better outcomes and transfer those treatment strategies to the lowest-volume hospitals. Small changes could have a big impact,” he concluded, “and quality measures, such as those endorsed by the National Quality Forum, will be a big help. Our goal is to raise the tide to improve the quality of care across the board.”

In addition to Dr. Bilimoria, contributing authors include David J. Bentrem (Northwestern University); Mark S. Talamonti, MD, FACS (Evanston Northwestern Healthcare); James S. Tomlinson, MD (Evanston Northwestern Healthcare); Andrew K. Stewart, MA (American College of Surgeons Commission on Cancer); David P. Winchester, MD, FACS (American College of Surgeons); and Clifford Y. Ko, MD, MS, Mshs, FACS (University of California, Los Angeles (UCLA) and VA Greater Los Angeles Healthcare System).

The study was supported by the American College of Surgeons Clinical Scholars in Residence program and the Department of Surgery, Feinberg School of Medicine, Northwestern University.

The National Cancer Database is recognized as the largest clinical registry in the world. It is a nationwide oncology outcomes database for more than 1,430 Commission-approved cancer programs in the United States and Puerto Rico, and captures approximately 70 percent of all new invasive cancers diagnosed annually.

Online July 24 , 2008

News from the College