Sometimes a risk factor is just a risk factor

March 18, 2009

Possibly the least controversial statement to have come out of the Fat Acceptance and Health at Every Size movements is the idea that obesity is not a death sentence – in other words, that not every fat person is one calorie away from heart failure, diabetes, and the many other diseases linked (often tenuously) to obesity. Now, mainstream medicine is starting to accept this. As reports,

One of Canada’s top obesity doctors says it’s time to stop recommending weight loss for everyone who meets official criteria for obesity. Dr. Arya Sharma says being obese doesn’t necessarily doom people to poor health and that weight loss recommendations should be targeted at those most at risk because of medical problems.

Many people who meet the body mass index criteria for obesity “are really not that sick at all,” says Sharma, chairman for cardiovascular obesity research and management at the University of Alberta and scientific director of the Canadian Obesity Network. “It’s not unusual to find someone come into your practice whose BMI is 30 or 32 (technically obese). This might be someone who is physically active, who is eating a good healthy diet. If you followed the guidelines to the letter you would be prescribing obesity treatment when there’s really no reason to do that, because they’re not medically obese.” …

His appeal comes as evidence begins to mount that a significant proportion of fat people are metabolically healthy. One in every three people who are obese — and half of those who are overweight — may be resistant to fat-related abnormalities that increase their risk of cardiovascular disease, according to new research from Albert Einstein College of Medicine in New York. … In [that] study, nearly 17 per cent of obese men and women possessed not one of the heart or metabolic abnormalities the researchers considered.

On the one hand, this is fairly obvious stuff. Many fat people remain fat despite leading a healthy lifestyle; and many thin people remain thin despite doing everything “wrong” with their diet and/or exercise. There has never been a perfect correspondence between (over)weight and health, and it’s about time the public discourse on obesity acknowledged that basic fact.

On the other hand, it may be premature to dismiss the effects of obesity on populations’ health. In the Albert Einstein College study mentioned by the article, 83% of obese participants had at least one heart or metabolic “abnormality” that may have been linked to obesity. Now, this absolutely does not imply that these 83% were sick because they were fat, or that the sample is representative of any larger population. However, it does raise the question of whether fat people (not all of whom are in poor health) are disproportionately sicker than thin people.

Instead of a conclusive answer, I have some tangentially-related old data to share. In 1993, the CDC’s Behavioral Risk Factor Surveillance System survey asked a large, nationally-representative sample of American adults to report their general health, height, and weight, among many other things. This crosstabulation shows the relationship between respondents’ classification as obese (by their BMI) and respondents’ self-reported general health status.

Health Crosstab

A vast majority of obese (and non-obese) respondents reported their health as “good” or better. However, comparing the two BMI categories suggests a strong correlation between obesity and worse self-reported health. For instance, obese respondents were twice as likely as non-obese ones to report their health as “poor,” and half as likely to report their health as “excellent.” This relationship persisted in three-way crosstabs controlling for sex, race, education, and income.* While this analysis was carried out on unweighted cases, weighting the data set by a product of poststratification and design weights did not alter or weaken this relationship.**

The table raises as many questions as it answers. It appears true that in 1993, obese people were more likely to report being in poor health than non-obese people. However, one must ask:

  • Has this relationship persisted over time?
  • Does this relationship persist under different statistical methods?
  • To what extent does this relationship exist because obese respondents perceive their obesity as a health problem, independent of any diseases it may cause?
  • By extension, does this relationship persist when controlling for body image?
  • If the relationship is robust in various years, under various methods of analysis, and while controlling for body image, then what causes obese respondents to be more likely to self-report poor health?

    As this (overly) simple analysis suggests, the effect of obesity on the public health is not a closed case. While many people classified as obese lead healthy lives and suffer from no diseases, it remains to be seen whether the obese are still more disposed to be in poor health than the non-obese, and what (if any) maladies of the former are actually caused by their obesity.

    * The main relationship was statistically significant in all subsamples. The relationship was substantively weaker in the following subsamples: Respondents with a household income of less than $20,000 (gamma=.262); respondents who did not graduate from high school (.202); and male respondents (.271).

    ** You can access the original dataset and codebook here.

  • Leave a Reply

    Fill in your details below or click an icon to log in: Logo

    You are commenting using your account. Log Out /  Change )

    Google+ photo

    You are commenting using your Google+ account. Log Out /  Change )

    Twitter picture

    You are commenting using your Twitter account. Log Out /  Change )

    Facebook photo

    You are commenting using your Facebook account. Log Out /  Change )


    Connecting to %s

    %d bloggers like this: