Wednesday, August 24, 2011

Obesity is a public health problem, not a genetic one

Everybody knows that obesity is a risk factor for many diseases, as well as decreased life span -- except that what everybody knows isn't always what science shows.  While some studies have shown obesity to be a significant factor in risk of disease and/or mortality -- the higher the body mass index, the higher the risk -- some have found that obesity is not correlated with increased morbidity or mortality after all.  (That is not the same as impaired lifestyle such as inability to go on nature hikes or sit in airline seats, nor morbidity such as joint disease.)  Now a paper in the International Journal of Epidemiology may explain why some studies may have shown little to no excess risk.

Oddly enough, it seems that the length of time people are obese (BMI, or weight to height ratio over 30) -- that is, how long people are exposed to the risk factor -- has never previously been taken into account.  Abdullah et al. looked at number of years lived with obesity, and its association with mortality from any cause, as well as selected causes, such as cardiovascular disease or cancer.  The researchers used data from the Framingham Cohort Study, which has followed up a population of about 5000 individuals every 2 years for about 50 years. (Framingham is a project that will never die, but that's another story).  

It turns out that for every additional 2 years of being obese, risk of mortality increases. 
The adjusted hazard ratio (HR) for mortality increased as the number of years lived with obesity increased. For those who were obese for 1–4.9, 5–14.9, 15–24.9 and ≥25 years of the study follow-up period, adjusted HRs for all-cause mortality were 1.51 [95% confidence interval (CI) 1.27–1.79], 1.94 (95% CI 1.71–2.20), 2.25 (95% CI 1.89–2.67) and 2.52 (95% CI 2.08–3.06), respectively, compared with those who were never obese. A dose–response relation between years of duration of obesity was also clear for all-cause, cardiovascular, cancer and other-cause mortality. For every additional 2 years of obesity, the HRs for all-cause, cardiovascular disease, cancer and other-cause mortality were 1.06 (95% CI 1.05–1.07), 1.07 (95% CI 1.05–1.08), 1.03 (95% CI 1.01–1.05) and 1.07 (95% CI 1.05–1.11), respectively.
The association remained significant even after adjusting for factors like smoking, age, current BMI, cardiovascular disease, cancer and diabetes.  This seems to mean that there's an effect on length of life even beyond accounting for these other diseases.  The authors propose a number of possible pathways through which obesity duration may have an effect on risk of mortality, including increased exposure to "endogenous production of reactive oxygen species and oxidative DNA damage, alterations in carcinogen-metabolizing enzymes, alteration in endogenous hormone metabolism and partial exhaustion of beta cells, with the resultant insulinopenia causing depressed glucose oxidation and impaired glucose tolerance."
Consequently, any association between duration of obesity and mortality might be expected to be partially explained by intermediate variables in the causal pathway to mortality such as blood pressure, serum cholesterol and serum glucose, with a longer duration of obesity potentially linked to an increased risk of chronic diseases such as diabetes, cardiovascular disease (CVD) and cancer.
The authors conclude that, because onset of obesity is happening at earlier and earlier ages, obesity prevention strategies need to be stressed.

We would agree with this conclusion.  Millions of dollars have been spent on the genetics of obesity.  Hundreds of candidate genes have been suggested, and yet, obesity rates are still increasing.  It is grossly clear that lifestyle makes the difference that matters by far the most, not genetic susceptibility (in earlier times, when obesity was much less common, many diseases related to obesity, and hence death due to them, were much less frequent).  Obesity is a public health problem, not a genetic one. If we slimmed down -- or better yet, prevented obesity in the first place -- those who for genetic reasons simply can't, would have truly genetic problems that would then be much easier to understand at the gene level, and perhaps even to design gene-informed therapies and interventions.

And the resources spent on so much genetic research on obesity, largely an excuse for not tackling the real problem, could have had far more impact.  The first studies to show clearly that obesity was a genetically complex trait may have been worth their cost.  But we know that now.  Now it's time to stop the epidemic, and genetics isn't going to do it. 


Nate Davis said...

Have there been cross-cultural studies on this topic? A cursory glance at the internet shows me that much of the research has been focused on western nation-states (with S. Korea having the lowest obesity rate, at 3.2%, from OECD data) or indigenous societies that have been largely subsumed into western lifestyles, such as Hawaiians and Aboriginal Australians.

But, of course, there are many societies around the world that don't fit the environmental trajectory of the western lifestyle. It's assumed that they have some rate of obesity that might be revealing.

Anne Buchanan said...

Are you asking whether the effect of obesity duration has been studied in a non-western context? Or other than in the US? Obesity rates are known throughout the industrialized world, and the association between obesity and morbidity and mortality has been studied in many places, though the IJE study we discuss here is apparently the first to look at the effect of obesity duration (or "obese-years" as they are now calling this risk factor) anywhere. So, this has only been done with the Framingham data, and can only be done with a similar kind of long established data set, with regular follow-up.

But, if you're asking whether obesity has been found to be associated with disease in other contexts, yes, it has been. As you mention, adoption of the 'western' diet and lifestyle is correlated with diseases like diabetes and heart disease around the world. Each population seems to become more obese in somewhat different ways (American Indians have less heart disease than European Americans, e.g., for a given obesity level -- but perhaps duration of obesity will make that look different now).

Ken Weiss said...

Many other formerly 'indigenous' societies have experienced a rise in obesity and related disorders, often at a pandemic level, when exposed even to a small amount of westernization or 'modernization'. So this is a general human response to whatever the actual factors are (dietary excess or imbalance, lack of exercise and so on). Living on reservations is also not good for people in this regard, presumably for similar reasons.