The death toll of the United States' obesity epidemic may be much higher… (Spencer Platt / Getty Images )
The death toll of the nation's obesity epidemic may be close to four times higher than has been widely believed, and all that excess weight could reverse the steady trend of lengthening life spans for a generation of younger Americans, new research warns.
Some 18.2% of premature deaths in the United States between 1986 and 2006 were associated with excess body mass, according to a team of sociologists led by a Columbia University demographer. That estimate, published online Thursday in the American Journal of Public Health, is far higher than the 5% toll widely cited by researchers.
The new figures do not reflect newly discovered facts about obesity's effects on health. Rather, they emerged after the researchers applied a finer-grained approach to examining obesity across the U.S. population.
Using historical survey data, the study authors toted up differences in excess weight status across different gender, ethnic and age groups. They combined that data with existing "mortality risk" statistics to estimate how many Americans over age 40 who died during that 20-year period did so because of weight-related causes.
The study makes clear that as obesity has become more widespread across successive waves of American generations, it has the momentum to reduce the average life expectancy of an entire population for many years to come.
Barring dramatic changes, "obesity is going to account for a rising share of mortality," said study leader Ryan K. Masters.
Americans who became overweight or obese as children and remained so into adulthood "have borne the greatest brunt of the obesity epidemic," Masters said. The evidence suggests that adults born in the 1970s and 1980s — a generation for whom excess weight has been widespread and lifelong — will suffer higher premature death rates than have older Americans, he added.
Though the current study may detect the leading edge of that trend, the full effect remains to be tallied by later research, Masters said. And some premature deaths could still be prevented by public campaigns or medical therapies that drive down obesity or its effect on health.
The study found that weight-related early mortality had struck American women harder than men, and that African American women had suffered the most. The premature deaths of 21.7% of white women between 1986 and 2006 could be attributed in part to excess weight, as could 26.8% of early deaths among African American women.
Among white men, 15.6% of premature deaths in that period were linked to excess weight. Among black men, the figure was only 5%.
Though African American men have high rates of obesity, they are also more likely than all other groups to die prematurely of other causes, such as injury or violence, Masters noted.
The latest calculation also calls into question the emerging belief that obesity in old age confers some protection against premature death — the so-called obesity paradox that has given comfort to many older adults struggling to shed weight. In fact, the study concluded, the probability of death among those carrying excess weight continued to rise after age 60, and did so steeply.
Masters said the obesity paradox was based on surveys that were fundamentally flawed. Only "non-institutionalized" respondents were counted, overlooking seniors in hospitals, skilled nursing homes and assisted living facilities, who may disproportionately suffer weight-related health consequences and early death.
The new calculations were met with some skepticism among those who study the cost, health and mortality effects of disease across broad populations.
Roland Sturm, a senior economist at the Rand Corp. in Santa Monica, said that if the number of weight-related deaths because of obesity had grown by a couple of percentage points, that would have seemed reasonable. But a more-than-threefold increase?
"Not likely," he said.
Still, Sturm praised the study for drawing needed attention to the effect of generational changes in the prevalence of excessive weight and obesity.
"Taking this into account can improve the accuracy of estimates," he said. "So while the idea is right, the execution has problems."