While it's unquestionably good new that the cancer _death_ rate from cancer is going down, the broader picture is not positive at all. That is because the overall cancer _incidence_ rate is not going down, i.e. even adjusted for age/population growth/demographics, a higher % of the population are getting cancer than ever before. The incidence rate for smoking related cancers like lung cancer and breast cancer are falling but the overall trend is not favourable.

Reducing the death rate is good news, but we're actually doing worse until we begin to reduce the incidence rate.

When the rate of death from cancer declines, the rate of death from other causes - heart disease, stroke, etc., go up. At that point we will hear demands for new crash programs and more research dollars for cardiovascular research. It is a vicious circle where everyone eventually dies of something, and no matter the disease, those on the recieving end of grants will tell you "This death could have been prevented if only we had been given sufficient resources."

That is because the overall cancer _incidence_ rate is not going down, i.e. even adjusted for age/population growth/demographics, a higher % of the population are getting cancer than ever before.

Is this a recent change? According to the NIH, " The "Annual Report to the Nation on the Status of Cancer, 1975-2001*" finds overall observed cancer incidence rates dropped 0.5 percent per year from 1991 to 2001[.]" Not as fast as the 1.1% annual decrease in the death rate, but still a hopeful trend. (Especially given that our detection technology has become more sensitive, so we're presumably identifying cancer incidences that wouldn't have been noticed till they were further along-- something that probably contributes to the improved survival rates.)

Bah, all this is completely uninterpretable without age-related statistics. As pointed out, if you work to improve survival from one sort of death (e.g. heart disease), the death rates for other sorts of death (e.g. cancer) necessarily go up. Even mortality rates from things like cancer are meaningless unless age and other things are factored in.

Life expectancy excluding accidents would be a better indicator of how health-related mortality is doing.

That is because the overall cancer _incidence_ rate is not going down, i.e. even adjusted for age/population growth/demographics, a higher % of the population are getting cancer than ever before.

Hmm, the next poster seems to contradict this claim, and has citations to back it up. Also, what about adjusted for better detection techniques? The same better diagnostic techniques that catch cancers that would spread early before they are deadly also catch benign cancers that wouldn't have caused problems. Better safe than sorry, but better detection can save lives and increase the incidence rate.

Clark, man, read the excerpt again.

"Dr. Thun (pronounced tune) said the death rate from cancer had been falling by slightly less than 1 percent a year since 1991, but even so, the actual number of deaths kept rising because the population was growing and aging.

"The decrease from 2002 to 2003 means that the decline in death rates had become sufficiently large that it was bigger than the aging and growth of the population," Dr. Thun said."

I think it is pretty clear that the death rate is age adjusted. And I don't know if Mike S' statistics are also age adjusted, but since the trend is down and the average age trend is up, the direction won't change even if they aren't.

Mea culpa, serves me right for not checking CDC and relying on 3rd party info before posting.

Here are the CDC National Program of Cancer Registries Invasive Cancer Incidence Results figures for the 1999-2002.

Year Count Population Age-Adjusted Rate Per 100,000
1999 1,159,615 246,191,733 478.1
2000 1,146,059 242,884,515 476.8
2001 1,194,907 253,010,517 475.2
2002 1,174,191 253,546,412 463.0