As we have discussed elsewhere, the association of obesity to increased mortality is complex. Studies vary. The “obesity-paradox” reflects studies showing that higher BMI was associated with lower mortality in hospitalized patients after a heart attack. Body Weight After Stroke: Lessons From the Obesity Pa… [Stroke. 2011] – PubMed – NCBI

Obesity-deniers exploit these studies to argue obesity is no big deal.

Now comes research out of Israel of older persons, age 75-94.  They found that obesity was significantly predictive of higher mortality for persons aged 75-84 but from 85 onward, obesity had a modest, non-statistically significant protective effect.

The authors suggest that the obesity-paradox may be disentangled in that there is a shift from detrimental to favorable influence of obesity from ages 80 to 90.  They speculate that persons with obesity lower rates of osteoporosis and extra energy reserves available in times of illness, stress and trauma may contribute to higher survival.  They write, “Decreased mortality among persons with obesity in very old ages may reflect a selective survival effect whereby persons who are more prone to the adverse health outcomes of obesity due to the effect of genetic or environmental factors suffer from higher mortality in middle age, which leaves a more resilient overweight older population. Another possible explanation is that of a ceiling effect, as absolute mortality long-term risk increases with age and eventually converges, regardless of any health-associated risk levels. Body Weight After Stroke: Lessons From the Obesity Pa… [Stroke. 2011] – PubMed – NCBI

Why is this important? First, it is a contribution in resolving the obesity-paradox and subsequent public confusion about obesity’s impact on survival. Second, it may impact current and future clinical trials. The Food and Drug Administration (FDA) sees obesity primarily (if not exclusively) as a cardiovascular risk factor. To determine if an anti-obesity drug reduces or increases cardiovascular risks, it is asking that clinical trials address this problem. However, for a clinical trial to study cardiovascular risks, it has to look to an older population than the typical weight-loss seeking patient, such as in the SCOUT trial. It has to go to an older population because cardiovascular disease is primarily a disease of an older population.Prevalence of Coronary Heart Disease — United States, 2006–2010

So, what if, as in this study, obesity is protective for some of the older population but not for others? How does this affect the outcome of the trials? Recall that in the SCOUT trial, there were relatively few cardiovascular events and the absolute number between those with events on and off the drug was quite small. So, this age difference could be quite significant.

Finally, consider this. In a study such as this, participants in their 80s were born in the 1920-30s. Odds are pretty good that they became obese during their adult years. Now, we have to consider that the population is becoming  obese at far younger ages and we do not know what effect the extra decades of higher weight will have on their mortality.