Great article about the "obesity epidemic"

[firecat]

And not only because it uses my favorite example of confusing correlation with causation:

Killing Turkeys Causes Winter by Sandy Szwarc

(Those of you participating in the current LJ fat flame war, feel free to use this as ammo.)

Comments

NYTimes article

[tedesson.livejournal.com]

What did you think of the article in the NYTimes which suggested that the obesity epidemic is related to the 500 calories per person per day which the US agricultural industry is growing now which they didn't used to, because of the 1972 Nixon administration change to direct cash farm subsidies? The article said that of the 500 extra calories per person per day grown, Americans are eating an extra 200.

I also found the analogy between the increase in consumption of whiskey in the 1800s due to low prices, and current food prices quite interesting.

Re: NYTimes article

[firecat]

I didn't read the NYT article, but I think the theory is crap. If reading the Szwarc article doesn't help you understand why I think it's crap, let me know and I'll expand.

premise

[tedesson.livejournal.com]

Would you grant me the premise that North Americans weigh more now than they did in the 1970s?

That is, my point is not about any health effects correlated with body size, but merely about cheaper food and body size.

Re: premise

[sistercoyote.livejournal.com]

I think you'd need to prove that North Americans actually weigh more now.

For one thing, they've lowered the numbers for "obesity" several times in the interim; that 50% of Americans who are now considered obese would most likely not have been so considered in the 1970s.

I don't have the citation with me at work, sorry. I'm actually not sure it's at home - it might be in my folks' home in a box. I'll check.

premise

[tedesson.livejournal.com]

According to this:

http://www.ajcn.org/cgi/content/full/72/5/1074/T3

a greater percentage of North American adults are fat. Notice this study uses a consistent bme>30 measure, not a 'flexible definition' of obesity. So, in 1960 43% of US adults aged 20-74 had bme>30, in 1984-1994, it's 54%.

This table does not indicate if bme's < 30 increased proportionately.

The trend from 1960-1980 was slight, with the bme>30 percentages going from 43% to 46%. The 1984-1994 number was twice the rate with an increase of 8% over the 10 years, compared to 3 percent over the previous 20 years.

Re: premise

[firecat]

I'll grant you the premise, but you need to look back farther than the 1970s to determine if your theory is correct. The article says this:

As Richard Klein chronicled in his book, Eat Fat (Vintage Books, 1996), fatness is often a cultural trend with cycles of fat and thin occurring throughout history. Typically people get fatter -- and healthier -- during times of prosperity. [...] In the past century people in industrialized nations have grown taller by about 4 inches on average. In Holland, in just four generations the average height for men grew 8 inches. (Burns, 1998; DeGregori, 2003)

If weights have been increasing for the last 120 years, and if people in other countries have been getting bigger as well, then the US government's farming subsidies in the 1970s can't account for the increase. They might be a contributing factor, but I think there are lots of other contributing factors.

The article also says this:

Increasing body weights during the 1990s, while statistically resulting in a national 61% increase in obesity, haven't been a health crisis for individuals whose weights have actually increased on average a mere 7 to 10 pounds, according to research by Jeffrey M. Friedman, M.D., Ph.D., at Rockefeller University.

Note that the statistical increase in obesity resulted partly from the government's changing the definition of obesity one or more times during the 1990s.

premise

[tedesson.livejournal.com]

This:

http://www.ajcn.org/cgi/content/full/72/5/1074/T2

has a description of the ways in which obesity has been defined since 1942. In my reply to sistercoyote, I posted evidence of a trend which uses a consistent definition, and indicates that a higher proportion of US adults have bme>30 now, than they did 30 years ago.