Big Drop in Colon Cancer Attributed to Colonoscopy
By Rita Rubin
WebMD Health News
Reviewed By Louise Chang, MD
Colorectal cancer cases and deaths have been falling for decades, with the most recent decline likely due to screening tests that enable doctors to detect and, if necessary, remove precancerous growths, researchers write in the journal Gastroenterology.
Still, the disease kills more Americans than any other cancer except lung cancer, and half of all Americans over 50 aren't getting any screening for it, according to the American Cancer Society.
Since Medicare and private insurers began covering screening colonoscopy for average-risk people in 2001, colonoscopy has become the main screening tool. But some recent studies have questioned whether it is any better at reducing cancers in the upper part of the colon than sigmoidoscopy, the scientists write.
Colonoscopy involves inserting a flexible lighted tube tipped with a camera through the entire rectum and colon. Sigmoidoscopy involves inserting a flexible camera-tipped tube through the rectum and only into the lower part of the colon.
The new study analyzed hospitalization data from the largest inpatient care database in the U.S., which includes patients covered by Medicare, Medicaid, and all private insurance companies. It was developed by the federal Agency for Health Care Research and Quality.
Researchers compiled the rates of all hospitalizations for colorectal cancer surgery from 1993 to 2009. Most people diagnosed with the disease undergo at least one operation, called a resection, so the number of resections for colorectal cancer closely reflects the number of cases, the scientists say.
Overall, the colorectal cancer surgery rate, expressed as the number per 100,000 people, dropped from 71.1 in 1993 to 47.3 in 2009. Most of the decline occurred in the latter half of that period, which correlates with the expansion of Medicare and private insurers' coverage of colonoscopy.
“The curves are very dramatic,” says researcher Uri Ladabaum, MD, associate professor of gastroenterology and hepatology at Stanford University. “Once we got the data and looked at it, we said, 'Wow, this is really quite a marked change here.'”
The rate of operations in the lower part of the colon fell from 38.7 per 100,000 people in 1993 to 23.2 in 2009. While the resection rate in the upper part of the colon fell from 30 per 100,000 people in 1993 to 22.7 in 2009, it declined significantly only after 2002.
Ladabaum's team attributes the decline to the wider use of colonoscopy.
“[It] is fairly logical,” says Brenda Edwards, PhD, a senior advisor for cancer surveillance at the National Cancer Institute in Bethesda, Md. She wasn't involved with the study. Still, “as they point out, this is not a cause-and-effect kind of thing,” Edwards says, because patients weren't randomly assigned to colonoscopy or another screening test.
High Tech vs. Low Tech
Colorectal cancer cases and deaths have declined because of screening, but “the question is, could we do it cheaper with stool blood testing?” asks Otis Brawley, MD, chief medical officer for the American Cancer Society.
Testing for microscopic blood in stool costs only $30, compared to $3,000 for a colonoscopy, Brawley says. Back in 2000, researchers reported that screening with the stool blood test every one or two years cut the risk of colorectal cancer by about 20%. That finding stemmed from 18 years of follow-up of more than 46,000 people, ages 50 to 80, who'd been assigned to screening with the stool blood test every year or every two years, or to their doctor's usual care, which typically was no screening.
“We don't have science that good [with] colonoscopy, which may surprise a lot of people,” Brawley says.
Uri Ladabaum, MD, associate professor of gastroenterology and hepatology at Stanford University.
Brenda Edwards, PhD, a senior advisor for cancer surveillance at the National Cancer Institute. Otis Brawley, MD, chief medical officer for the American Cancer Society.
© 2012 WebMD, LLC. All rights reserved.
Get the latest treatment options.