pageok
pageok
pageok
Scientists vs. Activists on the Causes of Cancer:

The New York Times has an interesting article on the current conventional wisdom within the scientific communityt on pollution's relative role (or lack thereof) in U.S. cancer rates. Contrary to some claims, there is little evidence that pollution and exposure to toxic chemicals are significant contributors to cancer rates.

pinning cancer on trace levels of poisons in the environment or even in the workplace is turning out to be a vexing task. There has been recent progress in addressing the issue, but the answers that many people believe must be out there remain elusive.

"It's an area where there's certainly been a lot of heat and not a lot of light for some time," said Robert Hoover, director of the epidemiology and biostatistics program at the National Cancer Institute. For the most part, Dr. Hoover said, "we are down to speculations based on some data but without having the information we need."

In 1993 the National Institute of Environmental Health Sciences and the Environmental Protection Agency began a major study of chemical exposures and cancer rates of 55,000 farmers and their spouses in Iowa and North Carolina. Farmers are a good group to study because they are regularly exposed to pesticides. Thus far, the study has yet to find any "definitive" connections, just what one researcher calls "interesting leads." Overall, cancer data show no evidence of a "cancer epidemic" brought about by industrial pollution.
Rates of cancer have been steadily dropping for 50 years, if tobacco-related cancers are taken out of the equation, said Prof. Richard Peto, an epidemiologist and a biostatistician at Oxford University.

What appear as increases in cancers of the breast and prostate, Dr. Peto added, are in fact artifacts of increased screening. When healthy people are screened, the tests find not only cancers that would be deadly if untreated, but also a certain percentage of tumors that would never cause problems if let alone.

His analysis of cancer statistics leads Dr. Peto to this firm conclusion: "Pollution is not a major determinant of U.S. cancer rates."

The story does contain some contrary views -- but not from scientists. Rather, various activists and cancer survivors are quoted on their beliefs about the causes of cancer.

Dave Hardy (mail) (www):
Lost my ex to cancer (www.franceshardy.com) and in that process talked a lot to a biochemist who works in the area (and is on the NRA board). His thought was, essentially: (1) we all "have cancer"-- we are trillions of multiplying cells, and there is a good probablility that at any one moment at least one of them has the two defects required for cancer (rapid division and "immortality"--normal cells are programmed to die off after so many divisions, cancer cells are not). (2) Most defective cells either die of their own, or are spotted and killed by your immune system. But every case of cancer starts with one lousy cell that gets thru. (3) "Getting cancer" is mostly bad luck. The most you can do to reduce the odds is to keep your immune system up. You can live in a glass bubble and drink only distilled water and get it, or shower in Agent Orange and never get it. The most that can be said is that exposure to *some* carcinogens shifts the odds to some small degree.
12.13.2005 9:15am
Richard Bellamy (mail):
The most that can be said is that exposure to *some* carcinogens shifts the odds to some small degree.

Well, assumedly all carcinogens shift the odds to some degree, or else they wouldn't called carcinogens.

Speaking as a conservative: The problems in most studies is that they look at exposures to large doses over small periods of time, and extrapolate to exposure to small doses over larger periods of time. If exposure to 1000 units in one day leads to cancer, that does NOT mean that exposure to one unit a day for 1,000 days leads to cancer.

Speaking as a liberal: The problem is that the companies that use these "potential carcinogens" are not required to prove that they are safe before they use them. What if the same process needed to approve a cancer drug from Merck was needed to approve use of an industrial solvent? When the activists are ahead of the scientists, that could either be because the activists are crazy, or because there simply is not enough science going on.
12.13.2005 9:57am
Freder Frederson:
Umm, cancer rates have actually decreased over the last 50 years, and this can't be linked to the increase in industrial pollution in the United States.

Excuse me, but can somebody point out the problem with that statement? Over the last fifty years we have made vast strides in cleaning up the environment and controlling pollution in this country as well as reducing exposure to chemicals into the workplace. Asbestos related cancers, black lung disease, lead and mercury poisoning are not just the fantasies of some whacked out environmentalists. The Cuyhauga River in Cleveland burning is not just some Fairy Tale told by environmentalists to scare their children (I am old enough to remember and lived in Cleveland when it was an annual event during the summer). It isn't a coincidence that bald eagle populations recovered after DDT was banned in this country.

If cancer rates had continued to rise in spite of all the environmental and strict worker exposure laws we have enacted, then the statement would make some kind of sense.
12.13.2005 10:13am
JosephSlater (mail):
I'm waiting for all the critics of the scientific method that chimed in on the intelligent design/creationism threads to question whether the "science" paradigm can really answer our questions here.

On the merits, ditto what Freder F. said.
12.13.2005 10:22am
Daniel Chapman (mail):
What do bald eagles and DDT have to do with this topic? Were they dying of cancer?
12.13.2005 11:30am
Freder Frederson (mail):
What do bald eagles and DDT have to do with this topic? Were they dying of cancer?

No they were dying from the bioaccumulation of DDT in fish. Contrary to their noble visage, bald eagles mainly feed off of dead and dying fish. DDT bioaccumulated in the eagles' food source from runoff. Although the DDT did not directly kill the eagles, it did cause them to lay eggs with very thin shells. Consequently, the eggs were crushed in the nest and chicks never hatched. And that is a scientific fact. It is just a very good example of direct cause and effect damage to the enviroment that is well-documented and is demonstrably a success story.

We managed to come up with a bunch of pesticides that were much less persistant in the environment. DDT last forever (or at least for a very long time).

It's just freaking amazing how these things happen.
12.13.2005 11:58am
JDS:
Banning DDT may have helped the bald eagle, but other factors like the Endangered Species Act and, as pointed out above, initiatives that cleaned up rivers and reduced industrial pollution, and better counting may explain their rebound.

But what is certain is that banning DDT killed the chance we had in the 1960's of eliminating humanity's second greatest scourge (after government-directed killing [1]): malaria. This disease kills more people than any other and, more than any other single factor, keeps the third world (and thus all humanity) poor. At least one million Africans die of malaria every year, more than 90% of them children under five.

Furthermore, because we failed to eradicate malaria when we had the chance, we are now faced with a disease that continues to increase its range and its resistance to known treatments.

See, for example,
http://www.economist.com/displaystory.cfm?story_id=1748455

Finally, in one interesting published study, workers who had substantial exposure to DDT were studied... No cases of neoplastic disease were found, suggesting that DDT prevents cancer in humans.

[1] http://www.hawaii.edu/powerkills/WSJ.ART.HTM
12.13.2005 12:20pm
Freder Frederson (mail):
The most that can be said is that exposure to *some* carcinogens shifts the odds to some small degree.

This is absolutely one of the silliest statements I have ever read on the web--and believe me I have read a lot. Would you deny that there are no chemicals in the world that could kill you outright? That would just be stupid.

I could provide you a list of chemicals and minerals that if you were exposed to in sufficient quantities (assuming I gave you a sub-lethal dose) would guarantee you cancer some time in the future, and some of them in the very near term. Now, granted the list of absolutely, positively, known carcinogens among the millions of chemicals we have discovered and invented is a relatively small number, but they do exist and they are backed by ample scientific evidence. But there are thousands more where there is a link between human cancer and the chemical but cancer cannot be tied to specifically to the agent.

We all know smoking "causes" cancer, but science still can't tell us how it does, which is why for so many years the tobacco industry was able to say with a straight face that there was no proof smoking caused cancer. In a strictly scientific sense that statement is still correct. We still don't understand the mechanism that causes the cigarette smoke to turn lung tissue cancerous. But that doesn't mean that we aren't certain that the link is there, we just can't identify the specific carcinogen or carcinogens.

Benzene, Boron, Chromium VI, Asbestos, Uranium and Plutonium(which will give you cancer through radioactivy and just because it is toxic), Radium and all the other radioactive elements, just off the top of my head are known carcinogens. The more you are exposed to them, the higher your cancer risk. At some exposure level you are guaranteed cancer (provided you don't die from the immediate toxic effects or get hit by a bus before the cancer devolops).
12.13.2005 12:20pm
Freder Frederson (mail):
But what is certain is that banning DDT killed the chance we had in the 1960's of eliminating humanity's second greatest scourge (after government-directed killing [1]): malaria.

This of course, is nonsense, perpertrated for some reason by people who want to paint environmentalists as some kind of evil Luddites who hate people in Africa and want them to die of malaria.

If we really cared about eradicating malaria we would have spent more money and research dollars developing better anti-malarial drugs. Of course because malaria is a disease of people in poor countries there is very little profit in it so the big pharmaceutical companies can't be bothered, they would rather sink their research dollars into the next Viagra. Secondly, the only advantage of DDT is its persistence in the environment. So we have to spray the huts twice a year with less persistent pesticides rather than once with DDT. If we refuse to spend that extra money then I doubt we would be behind the DDT effort either.

Finally, in one interesting published study, workers who had substantial exposure to DDT were studied... No cases of neoplastic disease were found, suggesting that DDT prevents cancer in humans.

One interesting study suggesting? Oh boy, where can I get my super-duper DDT pills to prevent cancer. I bet Kevin Trudeau will be hawking them as the next great cure "they" don't want you to know about.
12.13.2005 12:32pm
Mikeyes (mail):
The reason the article does not contain a surfeit of opposing views is that there are no credible opposing views.

There are well known toxin/cancer relationships, smoking and lung cancer being one of them, but if you compare smoking (which involves constant exposure over a period of decades) to the occasional inhalation of a diluted chemical, you can see the problem in determining if A causes B in the real world.

More people are getting cancer because more people are living longer and there is a greater chance of a cell (we don't really know the mechanism) turning to cancer and growing. Cancer is a multifactorial illness: genetics, exposure to whatever and bad luck, not in that order and not in equal parts. There are a zillion things that can be demonstrated to "cause" cancer in strains of mice that get cancer 90% of the time anyway. There is virtually no proof that most of these agents cause cancer in humans in the types of exposure that most of us have.

Millions of dollars are spent each year on projects that clean up rivers or the air in which there is no medical demonstration of a problem based on the belief that cancer (or birth defects or poor parent child relationships, etc., your choice) might be caused by the agent in question. If you challenge the project by asking for proof, you will be met with anecdotes and ad hominem remarks.

Don't get me wrong, the environment should be clean and we should enforce the rules that are in place and probably continue to develop methods for clean air and water. But this is because of acute illnesses which are demonstrable in an unclean environment and which cause untold suffering and cost for all involved (with the possible exception of the perpetrators, but that is what laywers are for.)

If I am wrong, please cite the study that proves me wrong, I'd like to see it.
12.13.2005 12:35pm
anonymous coward:
Oh god, I am so sick of the DDT-as-panacea crowd. See Lambert for a refutation of many pro-DDT myths.
12.13.2005 12:40pm
Freder Frederson (mail):
There are a zillion things that can be demonstrated to "cause" cancer in strains of mice that get cancer 90% of the time anyway. There is virtually no proof that most of these agents cause cancer in humans in the types of exposure that most of us have.

The problem is the "proof" that would convince you is unavailable. It would involve raising generations of genetically identically humans in a sterile environment. Exposing them to different levels of suspected carcinogens, killing some of them at early stages, letting the disease run its course in the rest. Even if such things were practical, some people might consider it unethical.

That is why there are relatively "known" human carcinogens. Only when the connection is glaring and obvious can we say Chemical X causes Cancer Y.
12.13.2005 12:45pm
anonymous coward:
"There is virtually no proof that most of these agents cause cancer in humans in the types of exposure that most of us have."

You need to balance this with an understanding that it is pretty hard to get that proof, even if there's really a significant relationship.

How would you construct an experiment showing that decades of low levels of exposure to Dread Chemical X increase the odds of developing cancer to a satistically significant degree? How about in combination with a few dozen other chemicals or lifestyle factors? Well, it ain't easy, unfortunately.

(This upsets certain econometrics types who feel that big datasets are required to yield answers.)
12.13.2005 12:49pm
DC Lawyer (mail):
More on the DDT issue. Not only are many mosquitos now immune to DDT, but the pro-DDT crowd misses other important issues that law waste to their politically motivated campaign. For example:

In the 1960s in Borneo public health workers tried DDT to control mosquito-borne malaria. The result? The local lizard population was decimated after eating contaminated insects. This in turn lead to decreases in the local cat population (which ate lizards as a main part of their diet). That led to a population explosion in caterpillars and rats, with the caterpillars destroying the thatched roofs and the rats causing increases in disease in the village. [Source: Richard Lazarus, The Making of Environmental Law].
12.13.2005 1:23pm
David Matthews (mail):
I'm an ex-smoker, and believe me, I'm glad I quit, but there's an issue that I haven't seen addressed by the medical/scientific community, perhaps because I don't know where to look, and likely because even discussing it seems to send the wrong message. And that is the fact that the majority of smokers DON'T get cancer. (For example, go to the Sloan-Kettering Cancer Center's website and put in some specs for the LungCancerRiskAssessment. I tried a 55 year old who has been smoking 2 packs a day for 35 years and doesn't quit. The risk of of developing lung cancer in the next 10 years is a surprisingly low 5%) Don't get me wrong. I'm not denying that smoking causes lung cancer (among many other diseases.) But I would think that a detailed study of the smokers who don't develop cancer (or significant heart disease, or emphysema, or whatever) might give useful insight into the disease. Does anyone know of any research in this area?
12.13.2005 2:02pm
Mikeyes (mail):
I'm not an absolutist in this proof business, but the examples cited so far are either persistent exposures (such as the asbestos/mesothelioma connection which occurred in workers spraying asbestos daily in and enclosed environment with no protection for over a year) or acutely deadly such as the plutonium. You can add smoking, waving your hand in front of an x-ray machine to see it it works on a daily basis, and living in Pittsburg in 1900 to those examples if you want to. These kind of relationships cry out "I cause cancer!"

What is at question is the premise that any chemical that might cause cancer because some mouse died of a carcinoma after being exposed to it should be regulated to the point of extermination. I'm not really asking for the kind of scientific proof that is available for smoking and cancer, just something beyond the illusion of smoke coming out of the gun. So far no one has cited a study on the level of cigarettes, asbestos, benzene or the radioactive substances that even comes close to making a hint at causation for the zillion molecules that are in the air as a result of both natural and industrial pollution.

Even a study that hinted that there was an increase in cancer in areas that have had the kinds of pollution mentioned would be nice even though it would probably not pass the editorial board of any reputable medical journal. I don't doubt that there are chemicals which will cause cancer in humans if they are exposed to it long enough, but do you want to ban well done steaks because one of the products in the crust is "carcinogenic"?

The asbestos mess is a good example of this because it involves a known carcinogen. Physicians noticed that a large number of workers in shipyards that made the Liberty ships in WWII had lung disease and a number of them had mesotheliomas which were otherwise very rare. On autopsy, asbestos granules were found in the cancers and in the lungs of the workers. These workers were exposed daily to very high concentrations of asbestos particles just the right size to get into the lungs and breathed in the particles for years.

Now there are rules mandating the removal of fixed asbestos shingles and fire walls, etc. based on the fear that the mere presence of these objects will cause cancer. The ironic part is that the removal of them will cause small particles to be freed which, if there was enough exposure for a sufficient amount of time, might duplicate the Liberty ship scenario, but otherwise they are safe. So there has to be a sequence of events to result in cancer and that sequence is often very hard to duplicate in the real world, especially if there are warnings and common sense precautions taken. The same is true of radioactive substances and a lot of the solvents. You have to protect yourself or avoid them.

I don't think you have to use the most rigid of proofs to determine if something is bad for your health. And you can't have it both ways i.e. if you think it is bad, it is bad, but I have to prove it in such a rigorous manner that it can't be proven without being "unethical" (which reminds me of the ad hominem arguments I mentioned above), when a dose of common sense can help parse out what is an immediate danger and what is not. All of the substances cited were run through that filter, including DDT (although DDT has never been a big problem for humans, you should have mentioned clordane for that one)and there is no reason not to keep looking for pollutants and other types of exposures that cause harm that can be quantified and then regulate it.
12.13.2005 2:16pm
Mikeyes (mail):
David Matthews:
"The risk of of developing lung cancer in the next 10 years is a surprisingly low 5%) Don't get me wrong. I'm not denying that smoking causes lung cancer (among many other diseases.) But I would think that a detailed study of the smokers who don't develop cancer (or significant heart disease, or emphysema, or whatever) might give useful insight into the disease. Does anyone know of any research in this area?"

David, there is a lot of research in that area, in fact treatment aside, the causes of cancer have obsessed the researchers for years. The consensus is that there are a number of factors including genetics, cell change, age, and environmental exposures that contribute. There are a few cancers that are the result of a congential or genetic cause alone and there are some that are due to exposure, but the majority of cancers are a result of a confluence of factors. You are much more likely to get lung cancer if you smoke, but if you don't smoke, it is still possible. Women who have family histories of breast cancer are much more likely to get breast cancer, but women who have no family history have a significant chance too. Certain diets are seen in cohorts that have specific cancers. If you live long enough and are of Irish descent, you will get a skin cancer. Etc.

I'd much rather die of a heart attack, it is quick and you usually don't know what is happening. I'm off to Burger King for my sixth hamburger of the day.
12.13.2005 2:27pm
anonymous coward:
From Tyler Cowen, a link to a defense of using massive lab-rate doses to find the bad evil chemicals.

(It's always fun to assume that people in areas other than yours are morons blind to the obvious flaws of their field's methodology. Economists and lawyers--and also engineers--are the most frequent offenders.)
12.13.2005 2:41pm
Mikeyes (mail):
The issue is not whether a chemical causes cancer, but whether persons exposed to the chemical will get cancer. That exposure alone is not enough to cause cancer in most cases (I'd say in very few cases depending on the person's vulnerability to the chemical) and other factors such as genetics, etc. are involved (see above note).

Frankly it becomes a public policy question once the obvious cancer causing chemicals are eliminated. Even then it is still legal to smoke, just not legal in most public places anymore. There are a lot of molecules that don't pass the extreme exposure test. My question remains, is it reasonable to ban all carcinogenic substances?
12.13.2005 2:55pm
anonymous coward:
Are there rigorous criteria for how much, how long, in what circumstances a chemical is "safe" for any given individual? Of course not. There aren't and won't be except in very unusual cases. That's reality, not "junk science." Uncertainty should affect how much $ we spend reducing carcinogens, obviously, but it's a bad reason to do nothing.

Of course the use of such data in a courtroom to deliver the big bucks is a separate issue.
12.13.2005 3:17pm
frankcross (mail):
People are looking at details, but Peto's big picture is clearly right. Look at the government regulations. First, they assume the most conservative highest risk from the data, extrapolating from animal studies. Then they measure exposure and they find risks at levels like 1 in 100,000 for those with the highest exposure and an upper bound estimate of the risk. That's a pretty trivial risk level compared to everyday living. It's not nothing, it may be worth attention, but there's no evidence that it has any noticeable effect on overall public health.
12.13.2005 3:17pm
Dick King:
"Even a study that hinted that there was an increase in cancer in areas that have had the kinds of pollution mentioned would be nice even though it would probably not pass the editorial board of any reputable medical journal."

One problem with that sort of thing is that if you do 20 such studies, one of them will show a positive result to the 5% significance level. Publication bias will do the rest...

This sort of problem brought us the powerlines-cause-cancer and the cellphones-cause-cancer scares.

-dk
12.13.2005 4:00pm
Mikeyes (mail):
"One problem with that sort of thing is that if you do 20 such studies, one of them will show a positive result to the 5% significance level. Publication bias will do the rest... "

I meant that remark in a semi-sarcastic tongue in cheek way. If that is the level of proof that drives the call for a public policy change, then it is not a very strong argument.
12.13.2005 4:37pm
Rich (mail):
Well, the bottom line is that you are going to die of something at sometime.
12.13.2005 5:03pm
Freder Frederson (mail):
The risk of of developing lung cancer in the next 10 years is a surprisingly low 5%

Compared to the chance of a person who has never smoked, this is shockingly high. It all depends on how you look at it. Lung cancer, other than that caused by smoking or occupational exposure to chemicals, is a very rare cancer.Look at the government regulations.

First, they assume the most conservative highest risk from the data, extrapolating from animal studies. Then they measure exposure and they find risks at levels like 1 in 100,000 for those with the highest exposure and an upper bound estimate of the risk.

Actually, they do nothing of the sort. Most of our environmental limits are based on occupational exposure limits. Occupational limits are set based on the best available scientific evidence and epidemilogical studies as well as political and industrial consensus. They are also based on a forty hour workweek and the assumption that those exposed are healthy adults taking reasonable precautions to limit exposure. Obviously, a lot of these assumptions are just not applicable to the public at large and makes the task of setting lower bounds or trying to extrapolate the hazards to children, pregnant women, or old people very problematic. And to contend that enviromentalists have more of a say in setting exposure limits than industry is just denying political reality.
12.13.2005 6:08pm
vic:
I am betting that most of the people making a big brouhaha make a substantial proportion of their income from the cancer scares

some obvious categories
1. Trial lawyers: enough of them here
2. Professional activists: who dip into their not-for-profit's " profits" to make a decent living
3. Government regulatory bureucrats: who owe their existance to the constant churning out of new regulations so that they can justify their exixtance.

Piecing together data and coming to an understandable working explanation requires technical education, experience, dilligence, the scientific method, scepticism and background in epidemilogy and biological staistics.

for example the statement above: "One problem with that sort of thing is that if you do 20 such studies, one of them will show a positive result to the 5% significance level. Publication bias will do the rest" is counterintutive to most, including most physician colleagues.

Maybe before we allow any environmentalist to pontificate about his/her pet peeve of the day we need to give him a little test in epidemilogy and biological statistics.

Yes, proof or at least good data and common sense would be a necessary prerequisite to regulations or tort.

On the question of tort, how in the world does the world expect 12 laymen to make an approprite judgement when the data is so complex and requires so much statistical understanding to be coherent. I guess it is a great setup for the snake oil salesmen to make a quick buck.
12.14.2005 10:47am
David Matthews (mail):
Mikeyes:

Thanks

Freder:

The point is that 19 out of 20 get off scott-free (well, at least as far as lung cancer in the next 10 years goes.) My question is, why? And, if we can find out why, can we use that knowledge to prevent it in the other 5%?

(The reason that I maintain that it is "surprisingly low" is that most anti-smoking campaigns leave the impression that if you smoke, you will probably get lung cancer, and that those who don't are definitely the exception, when it turns out that they are, in fact, the norm.

The only reason to compare rates of smokers to that of non-smokers is to establish a correlation, which is now pretty much beyond dispute. A statement like "smoking increases your chance of getting cancer by 5000%" or whatever the actual value, has no meaning, beyond establishing correlation, unless it is converted into actual probabilities for the individual. I buy a Powerball ticket twice a week, so my chances of winning the Powerball are infinitely greater than my brother's chances, since he never plays. But the truth, when we convert it into my actual chances, is that I'm wasting about $104 per year that my brother isn't.)
12.14.2005 12:48pm
Freder Frederson (mail):
Maybe before we allow any environmentalist to pontificate about his/her pet peeve of the day we need to give him a little test in epidemilogy and biological statistics.

Gee, can we do the same on the libertarians and lawyers (both plaintiffs and defense). Above, someone thought that it was amazing that the had only a 5% chance of getting lung cancer after a lifetime of smoking. Can you imagine any other legal product that we would consider acceptable if it only created a 5% chance of causing cancer. And actually, 1/3 of smokers will die as a result of cigarettes, cancer is just one route.

So what's the solution? Assume everything is benign until someone proves it causes cancer. Why should the onus be on the aggreived party. Why not assume everything is dangerous until the party who wants to introduce into the environment proves it is safe. Oh, but that is too big a hurdle. How do you prove such a thing. Science can't prove that absolutely. We're right back where we started.
12.14.2005 4:10pm