Thanks for all the thoughtful responses to my previous post – they are worth a follow up. I think part of the reason for the vigorous response is that I attempted in the post to turn the critical eye of skepticism back on skeptics. Often skeptics spend their time (myself included) shooting fish in a barrel – analyzing the worst of the absurd claims in the public. This is extremely useful, in my opinion, but is often akin to a prize-fighter street fighting with weak and hapless thugs. We need to balance this with mainstream issues and taking on more serious thinkers with whom we disagree. Further, we need to spar with each other to keep our skills honed.

Beyond mere intellectual exercise, criticism and argument are priceless tools for improving our own opinions and understanding. One of my primary goals with this blog, as with the SGU and the skeptical movement as a whole, is to make skeptics better skeptics (and that includes me). My last post was definitely designed to serve this purpose.

There are a few specific issues that were raised that I would like to address directly. Daedalus2u correctly pointed out that faulty reasoning can sometimes lead to a conclusion that happens to be correct – but the reasoning is still faulty and deserves to be criticized. I agree, and I tried to express this when I wrote about Huckabee that: “He was one of three Republican candidates who indicated that they do not accept the scientific consensus of evolution – and this was reflected in his statements and should be soundly criticized.”

Daedalus2u, if I read him correctly, however, seemed to contradict himself by first saying that Huckabee used faulty reasoning to arrive at his conclusions then saying that he probably did not do this, but rather arrived at his healthcare conclusions through conventional evidence and logic and then reverse engineered a justification couched in his religious beliefs. The latter is exactly the point I was making. Huckabee made some specific claims – diet and exercise are important to health, and 1/3 of cancers can be prevented by good diet and eliminating obesity. These claims are reasonable, backed by logic and evidence, and cited by legitimate sources that came to their conclusions by having experts review all available evidence (like the ACS and the NCI).

Huckabee appears to have then embedded these conclusions into his creationist world view, and he did so for obvious political reasons, positioning himself as the traditional conservative candidate. As I mentioned, I could not find anything written by Huckabee that would shed further light on his beliefs (whether, as daedalus2u assumes, he thinks lifestyle causes all illness). It is possible he thinks this, and might even harbor more bizarre health notions, but it is not reasonable to make such assumptions, and these conclusions cannot be confidently extrapolated from what Huckabee did say.

To be clear, I think we need to sometimes separate a claim from any specific argument used to support it. In this case, the specific claims made by Huckabee are reasonable, and since he used a specific number (1/3 of cancer) it is likely that he was basing these claims on the sources that cite the same number. We can simultaneously criticize the creationist context in which he expressed these claims, but they seemed to be ultimately irrelevant (nothing but window dressing).

It is also possible that, even though Huckabee arrived at his conclusions in a valid manner, he then attempted to justify them with his creationist logic Again, separating the claims from the argument, we can criticize this justification. (This may be the point daedalus2u was making, and this is legitimate.) From reading the transcript it was not my sense, however, that Huckabee was actually trying to justify his position with a creationist argument – he was just using the opportunity to gratuitously mention his faith.

It would have been nice if Chris Matthews pushed him on these points and forced him to clarify what he meant. I could not find anything written by Huckabee that would clarify them. To be intellectual honest I think that we need to remain agnostic about Huckabee’s beliefs until he clarifies them. Annoyance with Huckabee’s openly creationist beliefs does make it tempting to assume the worst, but I think we need to resist such temptation.

I have kept looking for more on Huckabee and healthcare – to see if we can infer anything from his specific policy recommendations. I discovered that he has supported in his state and now as a candidate:
– a smoking ban in public places (you have to give him props for doing this in Arkansas)
– mandating coverage and reducing co-pays for colonoscopies, prostate exams, and other cancer screening
– using food stamps to favor healthy food choices

Politics aside (I mean this, please do not try to infer my political views from this blog entry), these are all sensible evidence- based recommendations. I believe they support my hypothesis that Huckabee’s healthcare views are based on scientific consensus, not on his personal religious faith. But, as always, I am open to new evidence.

Rene raised the separate question about whether or not the belief that diet and weight can influence cancer risk has been adequately established scientifically. This raises some very interesting points.

First let me say that I definitely approached this claim very skeptically. My first reaction was that the 1/3 figure seemed too high, and I wondered how solid the evidence for this was. Further, I am not a cancer expert nor do I have a mastery of the relevant literature. I am therefore open to new evidence and opinions on this topic.

My goal was not to conduct an exhaustive review of this question but to find out what Huckabee’s sources were and if they were reasonable. What I found is that the American Cancer Society and the National Cancer Institute both endorse the 1/3 figure. The ACS specifically states that they put together a panel of experts to review the literature and this was their conclusion. I also found many published papers, include the NEJM review article, that support the connection between diet and cancer and obesity and cancer and put the size of this combined effect in the ballpark of 1/3 of all cancer. I could not find any reason to seriously doubt the connection or the 1/3 figure.

Rene’s point that the evidence is largely based upon epidemiological studies is valid, but I disagree with her apparent conclusion that this calls into question the claim and that only controlled trials are acceptable. This point comes up frequently in skeptical issues. It is true that prospective controlled intervention trials are the most reliable type of clinical data and, if robust and well designed and executed, would trump lesser forms of evidence, such as epidemiological evidence. But this does not mean that epidemiological evidence cannot be robust and reliable as well.

If, for example, multiple independent epidemiological measures all point in the same direction and attempts to thoroughly account for confounding factors are done, then epidemiological evidence can be highly reliable. It is necessary to put this in perspective – the data from which the scientific community claims that there is a link between smoking and cancer is also largely epidemiological. There has not been, and never will be, a study in which people are randomly assigned to smoking and non-smoking groups.

I still have many questions about the link between diet/obesity and cancer, and this is now a question on which I will seek to better inform myself. For example, perhaps there are genetic predispositions to both obesity and cancer, or perhaps the poor lifestyle choices that lead to obesity also lead to cancer, even without a direct connection. I am curious to what extent these alternate interpretations have been addressed in the literature (and someone better versed than me in this particular literature would likely know).

What is clear is that there is a strong epidemiological link between obesity and cancer (I still reserve judgment on diet alone), there are plausible mechanisms for this link, and this is the current consensus of opinion. In medicine we often have to make recommendations with imperfect data. At this point it seems that the evidence and logic for a connection is strong enough that it is reasonable to base both personal and public health decisions on this – especially since there is no health risk or downside to having a healthy diet and engaging in regular exercise. (I might not accept the same level of evidence to recommend a highly invasive procedure or a toxic drug.