Our healthiest weight is not what we think

It is increasingly evidenced in the medical literature that being ‘overweight’ and ‘obese’ offer a survival advantage for seniors and that “encouraging weight loss at older ages can even have an adverse effect, resulting in shorter life span,” said researchers at the Center for Population Health and Aging at Duke University at Durham, North Carolina. A new study sought to look more closely at this and how disability might impact mortality at various sizes.

Their study, just published in the Journal of the American Geriatrics Society, is another that didn’t make the news but might be of help to retiring Babyboomers and mature adults concerned with being fat.

The researchers examined data on 4,789 people who, in 1994, participated in the National Long Term Care Survey, which was a representative survey of U.S. seniors that examined health and well-being, with a special focus on functional limitations among disabled seniors. The participants were asked their weight and height, diet, smoking, alcohol consumption, health problems, and limitations in activities of daily living. The researchers examined mortality over 9 years, using vital statistics data.

Their findings consistently showed that the recommended BMIs considered, “normal” and “healthy” weights — 18.5 to 24.9 — are not ideal as we age.

First, fewer percentages of the ‘overweight’ and ‘obese’ (even extreme obesity at BMIs >35) died than those of ‘normal’ BMIs and smaller. The lowest overall risks of dying (p<0.01) were among those who were ‘obese,’ with BMIs 30-34.9 — with risks 23% to 33% lower than those of ‘normal’ BMIs. But there was a wide range of BMIs which shared nearly similarly low mortality rates, from the most extremely obese (BMI>35) through the ‘overweight’ (BMI 25-29.9).

Achieving thinness after retirement age appears an especially unsound goal. Mortality risks steadily increased as weights dropped, from ‘normal’ BMIs through underweight — this was after adjusting for age, smoking, drinking, cancer, heart problems, heart attacks, strokes, race and any weight changes. They did not even adjust for “obesity-related” health problems of diabetes and hypertension, which would have skewed the results even more so in favor of higher weights.

“Heart problems were found uniformly distributed across all categories of BMI,” they noted. “Prevalence of stroke and cancer decreased with increasing BMI. As expected, hypertension and diabetes mellitus increased with increasing BMI.”

Going to their models of relative risks for mortality among seniors with and without disability, BMIs were associated with nominal differences. Among the disabled and nondisabled, there were no tenable correlations with BMI and mortality. This held when adjusting for age, smoking, drinking, cancer, heart problems, strokes, race and weight change.

Splitting hairs: Among disabled and nondisabled, being ‘overweight’ and ‘obese,’ even extremely obese, had slightly lower risks (up to 20%) than those with high ‘normal’ BMIs>22. Being heaviest was especially protective for women. Risks for both men and women increased as BMIs fell below 22 — even though 22 is within the range recommended as ‘normal.’

These rising risks only approached tenable correlations among those with disabilities and BMIs <18.5. Only then did risks jump up to double.

As they concluded: “The analyses support early findings that older persons can be more tolerant of being overweight than their younger counterparts....For larger BMI, the risks were smaller except for with a BMI of 35.0 and higher, for which the risk was the same as for the reference group [‘normal’ range BMI 22-24.9]. Therefore, the analyses suggest a U-shaped BMI pattern of the RR of death for the entire sample, although skewed toward larger risk at lower BMI.”

“Potential protective effect of excessive weight explain these findings well,” they said. At ages 65 and older, disabled individuals with excessive weight have more reserves and are better able to cope with the additional stressful factors that accompany disability. At the very oldest ages, however, age becomes the biggest factor and the differences in mortality risks between disabled and nondisabled, as well as BMI categories, diminish.

Recommended ‘normal' BMIs do not equal “healthy” for everyone. When ‘overweight’ and ‘obese’ have lower or similar risks as those of ‘normal’ weight, they can clearly be “healthy,” too. There is no evidence that being fat should be a concern for mature adults.

© 2007 Sandy Szwarc