Health Letter, November 2013
Have you ever wondered how doctors decided what it means to be “normal” weight? The average weight of Americans has gone up over time, and about three-quarters of Americans today are considered overweight, obese or extremely obese by government standards. Who decided that the average American weight is not “normal” weight, and how did they know?
Although the official line between “normal” and “overweight” has remained largely unchanged since the mid-1980s, scientists continue to debate whether being a few pounds above normal is bad for one’s health. In fact, some recent studies suggest that being slightly overweight may even be beneficial – particularly for people over 65. Understanding these new findings may help Americans focus less on achieving some poorly defined ideal weight and more on making smart choices to stay healthy.
The earliest connection between weight and poor health outcomes came from analysis of insurance claims data during the first half of the 20th century. Insurance companies working in the 1940s and 50s developed tables to estimate “desirable” weights, taking into account gender, height and overall build. Early on, researchers discovered that mortality risk followed a U-shaped curve: Both extremely thin and extremely heavy people seemed to die at higher rates than people in the middle.
In the 1980s, the U.S. government developed a standardized definition for “overweight” based on height and weight, two factors scientists now use to calculate body mass index, or BMI. The table below presents some examples of BMI values for various heights and weights. Most family doctors keep a more detailed table in their offices to help patients determine their BMI more precisely.
Examples of BMI Values
|Height||Body Weight (pounds)|
|5 ft (60 in)||102||128||153||179|
|5.5 ft (66 in)||124||155||186||216|
|6 ft (72 in)||147||184||221||258|
The government initially settled on values corresponding to a BMI of about 25-26 for men and 24-25 for women as the line between “normal” weight and “overweight.” But the insurance data the government used to develop this standard had many drawbacks, including the fact that people tended to self-report weight and height inaccurately to insurance companies (for example, men tended to overestimate height and women tended to underestimate weight). The data also generally only applied to people between the ages of 25 and 59, so it was hard to say whether the figures held true for older people.
Did the desirable weight change as people aged? In the early 1980s, a group of researchers led by Dr. Reubin Andres, the director of the National Institute on Aging, part of the National Institutes of Health, used life insurance data to try to answer this question. In one study, Dr. Andres’ team computed an entire set of curves plotting mortality rates at different BMIs for people in different age ranges, looking separately at men versus women. What they found was remarkable: There was no clear pattern of differences between the genders, but a strong pattern emerged when looking at age. Younger adults still appeared to be better off with a BMI of around 21, well within the government range for normal weight. But people in their 50s and 60s seemed to be at their healthiest when their BMI was approximately 25 — right at the point at which government officials would label the person as overweight.
Over the next several decades, Andres’ work was roundly criticized by the mainstream medical community. One problem was that no one could explain why being overweight could be good for you, when excess fat also raised your chances of life-threatening diseases such as hypertension and diabetes. Some critics claimed that the so-called obesity paradox was just an artifact of poor study design, with studies showing better outcomes among overweight people simply failing to take into account the people who were thin because they smoked cigarettes or suffered from some disease that resulted in weight loss and an increased risk of death. This made it look as though being thin itself led to higher mortality. Some studies also over-adjusted by controlling for conditions that were associated with obesity, such as hypertension and diabetes. Researchers who adjusted for these conditions may have failed to detect some of the negative consequences of being fat, making excess weight seem healthier than it really was.
These criticisms appeared to be confirmed in 1995 by an important analysis, funded by the National Institutes of Health, that relied on data from a survey of more than 115,000 female nurses. The NIH-funded researchers excluded data from smokers and women who died early in the study (who may have been thin due to disease), and found that there was a direct linear relationship between BMI and mortality, with the lowest mortality at below BMI 19.
Today, the government’s definition of overweight remains remarkably similar to the figures first developed in the 1980s. Americans, both men and women, are considered overweight when their BMI exceeds 25 and obese at a BMI greater than 30.
Yet the question of what weight is “normal” is by no means settled. Recently, new evidence emerged suggesting that Dr. Andres may have been close to the truth after all. In early 2013, the Journal of the American Medical Association published findings from a particularly thorough study carried out by a research team at the Centers for Disease Control and Prevention (CDC). The CDC researchers examined close to 100 studies assessing the effects of weight on mortality, focusing on studies that took into account smoking, age and gender, and did not overadjust for other obesity-related health conditions. When they did this, they saw the same pattern that had been described by Dr. Andres’ research: People who were over 65 and overweight but not obese (a BMI between 25 and 30) were about 10 percent less likely to die than people categorized at normal BMI, a statistically significant result.
Furthermore, the CDC’s results seemed to hold true among the population as a whole (all ages), with overweight people appearing to be at about 6 percent lower mortality risk than their normal weight peers. In contrast, people with BMI values between 30 and 35, considered obese by today’s standards, had no decreased mortality risk relative to normal-weight peers, and people with a BMI of 35 or more had significantly higher mortality.
How do I know if my weight is “healthy”?
Researchers generally agree that obese people with a BMI of 35 or more should lose weight, as having this much extra weight poses clear health risks. People with hypertension and diabetes who are overweight also should try to lose weight, as there is good evidence that losing weight can help bring these conditions under control.
For otherwise healthy people carrying around just a small amount of extra weight (a BMI between 25 and 35), particularly older adults, it is not clearly understood whether their weight is “healthy.” If you are generally healthy but slightly overweight, you may be better off focusing less on figuring out which weight is “normal” and more on making healthy lifestyle choices. Quit smoking; eat a balanced, nutritious diet; and stay active by incorporating regular exercise into your weekly routine. Avoid crash dieting, diet pills or other schemes designed to achieve rapid weight loss. Remember that being thin is not the same as being healthy, and weight loss should never come at the expense of your overall health.