BNN provides English-language US and world news, analysis and opinion from all over the Internet. We strive for high standards, ethical behavior, and the presentation of multiple responsible points of view.
|
Get More Traffic For Your Blog! Blog Explosion brings hundreds of interested visitors to your blog - without costing you a cent. BNN News Archive Page |
       |
Tuesday, August 29, 2006
Serious flaws in the latest big obesity study The latest study has received what appears to be totally uncritical mention in the press so I think it is time I pointed out just some of the glaring problems with it. For starters, however, I reproduce the original journal abstract below: Overweight, Obesity, and Mortality in a Large Prospective Cohort of Persons 50 to 71 Years Old 1). What most glaringly identifies the sample concerned as not a random one is of course the percentage of women. A random sample would comprise about 50% women but there were actually twice as many men as women in this sample. So the population to which the findings may be generalized is essentially unknown, though a guess that it is a population who were worried about their health would probably not be too far astray. 2). The results were "adjusted" for physical inactivity. That is entirely inappropriate. Overweight people undoubtedly exercise less so the adjustment in effect creates an artificial population with no relevance to the real world. It is also possible that the adjustment for alcohol intake was inappropriate. 3). The overall results were, as usual, that people of middling weight lived longest. It was only in selected subsets of the sample that people of middling weight died somewhat younger. It is those subsets, however, that have attracted most media attention. If it were my practice in my own research to generalize from arbitrary subgroups of non-samples, I could prove anything too. 4). BMI is now in any case a rather contentious index of "obesity", for the amusing reason that in some populations it shows that overweight people live longer, as indeed it did in the present study. 5). The article looks at obesity at only one point in the lifespan. Weight tends to increase unevenly with age so that some people become overweight in later life who were not previously so. So what is true of those who are overweight in later life may not at all be true of (say) childhood obesity, and vice versa. This is a lacuna rather than a flaw in the study but it is yet another reason why the results of the study should not be generalized. Since the conclusion given in the journal abstract is wildly inappropriate to the data, however, the media can hardly be blamed for their dramatizations, for once. It would seem that the prestigious academic journal -- NEJM -- in which the study appeared has gone the way of the BMJ in becoming a largely politically correct organ. (For more postings from me, see TONGUE-TIED, EDUCATION WATCH, GREENIE WATCH, POLITICAL CORRECTNESS WATCH, FOOD & HEALTH SKEPTIC, GUN WATCH, SOCIALIZED MEDICINE, AUSTRALIAN POLITICS and DISSECTING LEFTISM. My Home Page. Email me (John Ray) here.) Blogger News Network is advertiser-supported, and your visits to our advertisers help BNN to meet its expenses. Help keep us afloat! posted by JR at 6:22 PM |
       |
Subscribe to BNN and get a daily bulletin of all our news postings. Interested in writing for BNN? Want information on our news service? Contact The Editor Writing for BNN BNN Editorial Policies Previous Posts
|
0 Comments:
Post a Comment
Links to this post:
Create a Link
<< Home