>Instead, he says, the statistics demonstrate clearly that
>while THE VERY FAT ARE GETTING FATTER, THINNER PEOPLE HAVE
>REMAINED PRETTY MUCH THE SAME."
>
>He's saying that the explanation for the problem is not that a large
>proportion of USers are way fatter than they've ever been. (And btw,
>the UK and Australia have seen similar increases in the obesity
>rate.) What he's complaining about is the use of a _mean_ to suggest
>that everyone in the country is carrying around 6-10 lbs more than
>their counterparts in 1991. Rather, he's saying that the research is
>showing that most "normal" people weigh about the same or just
>slightly more. The big increase came from gains seen when comparing
>those at "mid-range" in 1991 to those at "mid-range" today, and when
>comparing the obese now and then:
>
>Dr. Friedman points to careful statistical analyses of the
>changes in Americans' body weights from 1991 to today by
>Dr. Katherine Flegal of the National Center for Health
>Statistics. At the lower end of the weight distribution,
>nothing has changed, not even by a few pounds. AS YOU MOVE
>UP THE SCALE, A FEW ADDITIONAL POUNDS START TO SHOW UP, BUT
>EVEN AT MIDRANGE, PEOPLE TODAY ARE JUST 6 OR 7 POUNDS
>HEAVIER THAN THEY WERE IN 1991. ONLY WITH THE MASSIVELY
>OBESE, THE VERY TOP OF THE DISTRIBUTION, IS THERE A
>SUBSTANTIAL INCREASE IN WEIGHT, ABOUT 25 TO 30 POUNDS, DR.
>FLEGAL REPORTED.
Like I said before, 6-7 pounds on the median ain't birdfeed, nor is an increase in the share of the population that's obese from 23% to 31%. His own stats undermine his argument.
According to a summary <http://www.wlscenter.com/ResearchArticles/ObesietyReview.htm> of the Flegal paper from which he draws his numbers:
>The report analyses data collected by the National Health and
>Nutrition Examination Survey (NHANES) since 1960. It shows that the
>level of obesity was stable at about 14% through the first survey
>(1960-1962), the second survey (1971-1974) and the third survey
>(1976-1980). Obesity then spiked up to 23% for the survey conducted
>1988-1994 and continued the upward trend to 30% for this most recent
>survey, 1999-2000. The most recent survey collected data from 4,115
>adults.
By the way, 2000 isn't "now." The increases he's downplaying happened in 9 years.
Another set of stats <http://www.cdc.gov/nchs/data/hus/tables/2003/03hus068.pdf>:
percent of population aged 20-74 years, age adjusted, that's...
1960-62 1971-74 1976-80 1988-94 1999-2000 overweight 44.8 47.7 47.4 56.0 64.5 obese 13.3 14.6 15.1 23.3 30.9
Something happened starting in the late 1970s. I blame Reagan.
Doug