NEW YORK - As the United States enters the fifth decade of its "war on cancer," deaths continue to decline, according to an exhaustive report based on official data released on Monday.

But that doesn't tell the whole story, say experts not involved in the report from the National Cancer Institute, the American Cancer Society and other groups on progress against cancer since the 1970s. The improvements reflect such lifestyle changes as not smoking more than they do the billions of dollars spent to discover and implement advanced cancer treatments.

"We don't look at this as progress," said Fran Visco, president of the National Breast Cancer Coalition, of the new numbers. "This is such incremental improvement, when you look at the decades of investments, the cost of treatments, the number of researchers and journals, and then at the number of people who die ... well, we are clearly doing something wrong," said Visco, who was not involved in the study.      

The decrease in deaths from all cancers - 1.8 percent a year for men and 1.5 percent for women from 2005 to 2009, the last year with enough data to analyze - while steady, is disappointing to many experts because it is no greater in the most recent five-year period than in the previous one, and because it has hardly been affected by supposed advances in detection and treatment.

"The decrease in cancer mortality is driven largely by the decrease in cancer incidence, which is mostly because of the decrease in smoking," said Dr. Otis Brawley, chief medical and scientific officer of the American Cancer Society (ACS).  

Smoking can cause more than a dozen cancers, including lung, head, neck, bladder and mouth.

Although improvements in treatment for breast and some other cancers have cut death rates, Brawley said, headline-making new drugs have contributed little. "Most of the expensive new drugs prolong survival for no more than three or four months," on average, he said.

INCIDENCE OF CANCER FALLS

At first glance, the findings in the report, published in the Journal of the National Cancer Institute, are worth cheering.

In men, the incidence of cancer fell an average 0.6 percent a year from 2000 to 2009, find the researchers, who include scientists from the NCI, ACS and other federal agencies and private groups. In women, cancer incidence was flat for those years but rose 0.6 percent a year from 2005 to 2009. (The statistics are adjusted to account for the aging of the population, since cancer is largely a disease of the elderly.)

Cancer incidence among women fell when fewer post-menopausal women opted for hormone-replacement therapy following a 2002 report that linked it to breast cancer and other disease, said NCI's Brenda Edwards, the lead author of the report.

But no other factor occurred to cause another dip after the mid-2000s. Instead, although the incidence of breast cancer fell from 2000 to 2009 by an average 0.7 percent a year, it rose 0.9 percent a year from 2005 to 2009.

The trend in childhood cancer is also going in the wrong direction. From 2000 to 2009, cancer incidence among children 19 and younger rose 0.7 percent per year, on average.