One of the places to see (and be seen?) in the public health blogosphere these days is the new group blog, The Pump Handle. Among the many terrific posts there recently was one by Boston University School of Public Health epidemiologist Richard Clapp, "Drop in Cancer Deaths hype - what's behind the numbers?" The numbers he refers to are the 0.5% drop in US cancer deaths, from 556,902 in 2003 to 553,888 in 2004.
The story made the newswires and big media outlets as good news. It certainly isn't bad news, but as Clapp points out, the drop in cancer deaths lags behind the drop in deaths from other major causes such as cerebrovascular disease (strokes), chronic lower respiratory disease, influenza and pneumonia, septicemia, hypertension and diabetes and all have some common risk factors (smoking, obesity, lack of exercise, access to health care). He then asks:
The question then becomes, why are these other diseases declining more than cancer. The headlines might have been, "Cancer lags behind other diseases in decline from 2003 to 2004."
Nor is the picture all rosy:
Another piece of the cancer story has to do with childhood cancer incidence and mortality. Cancer is, thankfully, a rare illness in children, and mortality rates have gone down for many childhood cancers over the past two decades. Childhood cancer incidence has been steadily going up, however, and this cannot be due to some of the factors cited in adults (smoking, obesity, lack of exercise, etc.). It's also not that genetic susceptibility is increasing in children because heritable factors would not be likely to change in one generation. Here, most people would gladly accept the burden of keeping a child alive who has been diagnosed with cancer. But wouldn't we all rather that the children not be getting cancer in the first place? The explanation for the steady increase in childhood cancer incidence is not at hand, but at least one place to look is prenatal and early childhood environmental carcinogenic exposures. A report by Tami Gouveia-Vigeant and Joel Tickner published by the University of Massachusetts's Lowell Center for Sustainable Production (PDF summary here; PDF report here) finds that "evidence increasingly indicates that parental and childhood exposures to certain toxic chemicals including solvents, pesticides, petrochemicals and certain industrial by-products (dioxins and polycyclic aromatic hydrocarbons) can result in childhood cancer."
This is good stuff and this is a blog to bookmark and visit often. The Reveres have contributed there and will do so in the future. Most posts are signed by well known and authoritative public health scientists like Clapp. They (we) are off to a great start.
Go prime The Pump.
- Log in to post comments
Very interesting, topic, but your link to the article is 404.
This one should work.
Question: is the increase in childhood obesity not a possible contributor because obesity is more of a long-term cancer risk? Or because obesity is not positively correlated with cancer in children?
caia: Thanks. correcting the link now. Obesity is not linked to childhood cancer.
This got me speculating - is population size linked to cancer incidence.
From the naive (physics) viewpoint, if the probability is independent per cell, then more cells means higher incidence per person per unit time.
Is the fact that people are bigger, and are bigger younger, a factor in cancer incidence?
http://scienceblogs.com/catdynamics/2007/02/cancer_incidence_wild_physi…
I should clarify - I mean size of the individuals in the population, not the number of people. Clearly more people means more cancer, but not a higher incidence.
Steinn: Not too likely that the number of cells you have is the issue. Certainly not for childhood cancer.
Should not, or does not?
If the straw hypothesis of "independent triggers per cell" is correct, then the incidence ought to depend on the number of cells in the tissue, and IF kids grow bigger earlier, then the incidence should track that.
Arguably, incidence could also plausibly depend on cell division, not just cell count, which would be a double strike for rapdily growing kids.
Potentially disturbing, but as with all "fool conjectures", not trivial to test.
At least it is statistically testable if anyone could be bothered to do the work.
Isn't it so that about a little less than 25% of ppl in the US die of cancer (all ages together)?
The article only treats 'drops in death due to x' following the report quoted and points out that cancer is a weak contender. Far weaker, in fact than the article states, if one takes longer term trends into account (which they don't, ok, but it is not obvious to the skimmer....) - the percent decline for heart disease, is not just greater but so many magnitudes larger one cannot really compare. Diabetes is on the rise longer term, and the article leaves out the biggest drop: accidents. (But I speak of long term trends, add in Alzheimers, incredible rise for the elderly...)
So what does a decrease in .05% for 'cancer' really mean?
3 104 fewer deaths (04 comp to 03 - US pop 300 million...), that number is so small it cannot be considered meaningful in any terms whatsoever, it is just fuzz, unless it is part of a very long trend, which it is not. (Not really.) Taking it into account is pseudo science. Debunking what others make of is just as bad... unless a larger pov is adopted (political particularly, long term, thoughtful, etc. children's cancer doesn't hack it) in which case one would come to different conclusions ...
Revere, that article was not impressive at all, a lot of guff that took a 'debunking line' while not tackling the main issues in any reasonable, decent, informative way. All it does is obfuscate, glorify trivia, offer more spin to confuse, mixing putative causes (obesity..) to explain insignificant numbers.
Steinn: Replied as a comment on your post (which I missed first time around). Here's the link for others: http://scienceblogs.com/catdynamics/2007/02/cancer_incidence_wild_physi…
Steinn's blog is highly recommended.
Ana: It means cancer deaths are not increasing as they have for decades and it doesn't mean we are "winning the war on cancer," which I think was the point of the post at TPH. This wasn't debunking as much as pointing out some important elements that were overlooked in the MSM reporting. Proportional mortality from cancer in the US is about 22%. But it is a dread disease and affects public policy. That's why the wire service reports are important. They are an argument that it is not the environment or the workplace that cause cancer. Thus I disagree about the interest and value of the post, but that's what makes the world go around. That and money.