U.S. cancer deaths rose by more than 5,000 in 2005, a somewhat disappointing reversal of a two-year downward trend, the American Cancer Society said in a report issued Wednesday.

The group counted 559,312 people who died from cancer.

The cancer death rate among the overall population continued to fall, but only slightly, after a couple of years of more dramatic decline.

In 2005, there were just under 184 cancer deaths per 100,000 people, down from nearly 186 the previous year. Experts said it wasn't surprising that the rate would stabilize.

The cancer death rate has been dropping since the early 1990s, and early in this decade was declining by about 1 percent a year. The actual number of cancer deaths kept rising, however, because of the growing population.

So it was big news when the rate dropped by 2 percent in both 2003 and 2004, enough to cause the total number of cancer deaths to fall for the first time since 1930.

President Bush and others hailed that as a sign that federally funded research was making strides against the disease.

But now the death rate decline is back to 1 percent. And the 2005 numbers show annual cancer deaths are no longer falling, but are up more than 5,400 since 2004.

"The declining rate was no longer great enough to overcome the increase in population," said Elizabeth Ward, a co-author of the cancer society report

Officials with the organization say they don't know why the decline in the death rate eased.

It may be that cancer screenings are not having as big an effect as they were a few years ago, said Dr. Peter Ravdin, a research professor in biostatistics at the University of Texas M.D. Anderson Cancer Center in Houston.

One possible example: In 2004, the largest drop in deaths among the major cancers was in colorectal cancer. Experts gave much of the credit to colonoscopy screenings that detect polyps and allow doctors to remove them before they turn cancerous. They also mentioned "the Katie Couric effect" _ a jump in colonoscopy rates after the "Today" show host had the exam on national television in 2000.

In the new report, the colorectal cancer death rate decreased by about 3 percent from 2004 to 2005, after plunging 6 percent from 2003 to 2004.

Colorectal cancer screening rates through 2003 did not show a decline. But it's possible they have fallen since then, Ravdin said.

Cancer society officials have also voiced concern that cancer deaths may increase as Americans lose health insurance coverage and get fewer screenings.

The good news is the cancer death rate is still declining, and that since the early 1990s is down more than 18 percent for men and more than 10 percent for women. Those reductions translate to more than half a million cancer deaths avoided, according to the cancer society.

Experts attribute the success to declines in smoking and to earlier detection and more effective treatment of tumors.