Thursday, August 09, 2007

Your fat friends can make you fat? III

The full text for the New England Journal of Medicine, "fat is contagious" study is available for free at this url:

http://content.nejm.org/cgi/content/full/357/4/370?query=TOC

The collection of the data seems hit or miss - individuals in the cohort were apparently asked to identify spouse, sibs and at least one close friend. Over the years, some individuals choose a different close friend.

The closest ties of friendship were viewed when two individuals chose each other as close friends. This was identified by the researchers as a "mutual friendship".

The problem with this is the "social network" of any given individual includes far more than merely ONE close friend but rather up to 30 or more close friends. Thus it might be said that the original information which the researchers fed into the computer in order to derive this analysis was erroneous and thus this would make the resulting analysis, invalid. As we say in the computer field "garbage-in/ garbage out".

Also, their results suggested that sibs were LESS likely than friends to "cause" or "influence" obesity in other sibs and spouses were the LEAST likely to influence their spouses as far as weight.

From observation, this result seems very erroneous... spouses OFTEN influence each other's weight as they share mutual cooking, often recreate together and enjoy similar interests... and sibs more often than not, show the same body type as other sibs due to shared genetics... so it seems like their computer generated results came out backward from what the reality might be....

Finally the last sentence of the conclusion is most troubling. From this erroneous study based on a false picture of the social network of the cohort (a picture which only depicts members of the cohort as having only ONE close friend at any given time), these researchers concluded that:

"This highlights the necessity of approaching obesity not only as a clinical problem but also as a public health problem."


It's almost as if this "study" builds and analyzes a false picture of the social network of the individuals of the cohort to come to a conclusion with political overtones based on what is a buzz word to many Americans i.e. "public health", the implications of which in the extreme, could lead to treating overweight people like lepers were treated in the past. That is, falsely portraying obesity as a "contagious disease" in order to perpetuate the ostracism of anyone who is overweight. Hopefully society is not that insane - yet, but this type of thing is a bit chilling to think about, to put it mildly.

No comments: