I'm in favor of articles
like these, wherein a major news outlet is criticized for poor journalism. On the other hand, it seems
Reason itself needs to reassess its commitment to, er, reason. I'm not trashing the article as a whole, as in fact the internal logic and overall quality of evidence seemed to me convincing enough in general, at least at a glance. However, the section on obesity doesn't measure up to its peers. It's flawed, and obviously so, in ways I intend to prove.
"But the science proved hard for the panic to overcome. In 2005, a team of CDC researchers published a study finding significant flaws with the 400,000 figure. The real number, they said,
was closer to 112,000." [link in original]
Yes, overestimating the figure by a factor of four is a serious problem. Still, 112,000 is still quite a few individuals, wouldn't you say? What might be the context for this sort of figure? In other words, what is the threshold for an acceptable level of corpulence? To put the idea of perspective into, uh, perspective, the United States suffers about 30,000 suicides every year, and about 40,000 deaths due to automobile accidents every year. So now the 112,000 figure looks enormous. Yet a minute ago
Reason made it look small by comparing it to a poorer estimate of itself.
"And when you add in the
protective effects of being mildly overweight, the number drops to 26,000."
There
two unsubstantiated claims here, both of which are remarkable. Remarkable claims demand proof with evidence and reason, and none are offered. First of all, what exactly are the "protective effects" of being overweight, aside from, I suppose, resistance to starvation? The clinical definitions of terms such as "overweight" and "obese," as I understand them, are carefully chosen by medical science. They are unhealthy conditions
by definition. If there are also
benefits to being overweight, of such efficacy that they offset the deathcount by 75% (112K down to 26K), one would think that there would be more information somewhere. If there is,
Reason has unreasonably chosen to withold it. Moreover the 26,000 figure itself is entirely unsubsantiated.
"As for medical costs,
a 2008 Dutch study suggests what would seem to be intuitive: People who live longer tend to incur more lifetime medical expenses. Meaning that if obesity does modestly shorten lifespans, it does so at a
savings to taxpayer-funded programs like Medicare and Medicaid."
Misuse of context! Ten yards and loss of down! The Dutch study's cost-benefits model is entirely based on being Dutch. From that study: "Except for relative risk values, all input parameters of the simulation model were based on data from The Netherlands."
Reason is attempting to tear down
Time's claim that obesity is a problem
in the United States. The hilarious thing is, I don't even have to prove that the Netherlands is substantially different from the US in relevant ways, as I suspect it is; the serious flaw in their reasoning is that they
assume the study's conclusion is as valid here as it is there.
Feh,
Reason. They make a mockery of the word.