Obesity associated with fewer deaths than previously thought (updated Friday, April 22, 2005)

A new study from the U.S. Centers for Disease Control estimates that obesity is associated with 111,909 early deaths in the U.S. each year as opposed to a previous estimate of 400,000 deaths per year.

The biggest reason for this dramatically lower estimate is from them including data from two other U.S. nationally representative surveys conducted by U.S. National Center for Health Statistics.

The previous estimate of 400,000 extra deaths due to obesity was based on data from only one survey, NHANES I which was conducted in 1971 to 1975, and follow-up data was taken until 1992.

This study, however, included data from not only NHANES I (1971-1975), but also NHANES II conducted in 1976-1980 with follow-up through 1992, and NHANES III conducted in 1988-1994 with follow-up through 2000.

The addition of data from the two newer studies, NHANES II (1976-1980) and NHANES III (1988-1992), added to data from NHANES I (1971-1975), reduced the estimated number of deaths by 63 percent.

This estimate should not be taken at the “final truth”

The authors of this study note that a small change in the numbers, that is if there was a relatively small increase in the number of obese people who had died, the estimated number of excess deaths associated with obesity could increase from 111,000 extra deaths per year to 214,000 or even 305,000, an increase of 100,000 to 200,000 additional deaths.

Therefore, this estimate of 111,000 additional deaths should not be looked at as the final truth.

Risk factors for heart disease have decreased recently

Part of the reason for this—the decreased number of early deaths due to obesity—“cardiovascular risk factors have declined at all BMI levels in the US,” according to the study.

Lifespan has increased from 74 to 77 years since 1980

Another reason for the decrease in estimated excess deaths from obesity is that the average lifespan has increased from 73.7 years in 1980 to 75.4 years in 1990 to 77 years in 2000.

Deaths from ischemic heart disease (heart attacks) has decreased from 345 per 100,000 people per year in 1980 to 250 in 2000.

Three-fourths of excess deaths from BMI greater than 35

Nearly three-fourths of obesity-related deaths (an estimated 82,066 of the 111,909 excess deaths, or 73 percent) were associated with severe obesity, that is when body mass index (BMI) was greater than 35.

Three-fourths of excess deaths in people under 70

Three-fourths of the obesity-related deaths (an estimated 84,145 of the 111,909 excess deaths, or 75 percent) were associated with people under the age of 70.

This suggests, as previously studies have also found, that being obese increases the risk of death more in younger people, and less in older people.

Definitions of underweight, normal weight, overweight, and obesity

Underweight is considered having a BMI less than 18.5.

Normal weight is considered having a BMI of 18.5 to 25.

Overweight is considered having a BMI of 25 to 30.

Obesity is considered having a BMI of 30 or more.

While severe obesity is considered having a BMI of 35 or more, and extreme obesity is having a BMI of 40 or more.

BMI Table

A BMI Table can be found here.

Overweight people less likely to die than normal weight people?

The study even suggests that people who are overweight, but not obese, that is having a BMI of 25 to 30, are less likely to die than people who are normal weight, that is have a BMI of 18.5 to 25.

Among 25-59 year olds, overweight 17% less likely to die than normal weight

They found that for people 25- to 59-years-old, people who were overweight (BMI 25-30) were 17 percent less likely to die than people who were normal weight (BMI 18.5 to 25), however, the difference was not statistically significant, which means that there is more than a 5 percent chance that this difference could have been due to random chance.

Among 25-59 yr olds who had never smoked, overweight 34% less likely to die than normal weight

The effect was even greater when only including people who had never smoked.

Among 25- to 59-year-olds who had never smoked, those who were overweight (BMI 25-30) were 34 percent less likely to die than people who were normal weight (BMI 18.5 to 25). However, this difference was not statistically significant either, meaning it could have been due to random chance.

Among 25-59 yr olds, obesity increased risk of death by 20%, but reduced risk of death by 23% among those who had never smoked

Among 25- to 59-year-olds, being obese (BMI of 30-35) increased the risk of death by 20 percent.

When only including people who had never smoked, 25- to 59-year-olds who were obese (BMI 30-35) were 23 percent less likely to die than people who were normal weight (BMI 18.5-25).

Among 25-59 yr olds, severe obesity increased risk of death by 83%, or 25% among those who had never smoked

Among 25- to 59-year-olds, those who were severely obese (BMI greater than 35) were 83 percent more likely to die than people of normal weight (BMI 18.5-25).

When only including people who had never smoked, 25-59 year olds who were severely obese (BMI greater than 35) were 25 percent more likely to die than people who were normal weight (BMI 18.5-25).

Among 60-69 year olds, overweight 5% less likely to die than normal weight

They found that for people 60- to 69-years-old, people who were overweight (BMI 25-30) were 5 percent less likely to die than people who were normal weight (BMI 18.5 to 25), however, the difference was not statistically significant, which means that there is more than a 5 percent chance that this difference could have been due to random chance.

Among 60-69 yr olds who had never smoked, overweight 19% less likely to die than normal weight

Again, as with the younger group, the effect was even greater when only including people who had never smoked.

Among 60- to 69-year-olds who had never smoked, those who were overweight (BMI 25-30) were 19 percent less likely to die than people who were normal weight (BMI 18.5 to 25). However, this difference was not statistically significant either, meaning it could have been due to random chance.

Among 60-69 yr olds, obesity increased risk of death by 13%, or 21% among those who had never smoked

Among 60- to 69-year-olds, being obese (BMI of 30-35) increased the risk of death by 13 percent.

When only including people who had never smoked, 60- to 69-year-olds who were obese (BMI 30-35) were 21 percent more likely to die than people who were normal weight (BMI 18.5-25).

Among 60-69 yr olds, severe obesity increased risk of death by 63%, or 130% among those who had never smoked

Among 60- to 69-year-olds, those who were severely obese (BMI greater than 35) were 63 percent more likely to die than people of normal weight (BMI 18.5-25).

When only including people who had never smoked, 60-69 year olds who were severely obese (BMI greater than 35) were 130 percent more likely to die than people who were normal weight (BMI 18.5-25).

Another way of stating this is that severely obese people (BMI greater than 35) were more than twice as likey—2.3 times more likely—to die than normal weight people (BMI 18.5-25).

Among people 70 or older, being overweight reduced the risk of death by 9% compared to being normal weight

For people 70 or older, people who were overweight (BMI 25-30) were 9 percent less likely to die than people who were normal weight (BMI 18.5 to 25), however, the difference was not statistically significant.

Among people 70 or older who had never smoked, overweight 10% less likely to die than normal weight

Among people 70 or older who had never smoked, those who were overweight (BMI 25-30) were 10 percent less likely to die than people who were normal weight (BMI 18.5 to 25). However, this difference was not statistically significant.

Among people 70 or older, obesity increased risk of death by 3%, or 13% among those who had never smoked

Among people 70 or older, being obese (BMI of 30-35) increased the risk of death by only 3 percent.

When only including people who had never smoked, people 70 or older who were obese (BMI 30-35) were 13 percent more likely to die than people who were normal weight (BMI 18.5-25).

Among people 70 or older, severe obesity increased risk of death by 17%, or 12% among those who had never smoked

Among people 70 or older, those who were severely obese (BMI greater than 35) were 17 percent more likely to die than people of normal weight (BMI 18.5-25).

When only including people who had never smoked, people 70 or older who were severely obese (BMI greater than 35) were 12 percent more likely to die than people who were normal weight (BMI 18.5-25).

Being underweight increased risk of death by 38% to 130% compared to being normal weight

Being underweight (BMI less than 18.5) increased the risk of death compared to being normal weight (BMI 18.5-25) in all age groups and when only looking at people who had never smoked.

Among people 25 to 59-years-old, being underweight (BMI less than 18.5) increased the risk of death by 38 percent compared to people of normal weight (BMI 18.5-25).

Among people who were 60- to 69-years-old, being underweight (BMI less than 18.5) increased the risk of death 130 percent compared to people of normal weight (BMI 18.5-25).

In other words, they were 2.3 times more likely to die than people of normal weight.

Among people 70 or older, being underweight (BMI less than 18.5) increased the risk of death by 69 percent compared to people of normal weight (BMI 18.5-25).

Being underweight among people who had never smoked increased risk of death by 25% to 197% compared to normal weight

When only including people who had never smoked, being underweight increased the risk of death by 25 to 197 percent compared to people of normal weight.

Among people 25 to 59-years-old who had never smoked, being underweight (BMI less than 18.5) increased the risk of death by 25 percent compared to people of normal weight (BMI 18.5-25).

Among people who were 60- to 69-years-old who had never smoked, being underweight (BMI less than 18.5) increased the risk of death 197 percent compared to people of normal weight (BMI 18.5-25).

In other words, they were nearly 3 times more likely—2.97 times more likely—to die than people of normal weight.

Among people 70 or older, being underweight (BMI less than 18.5) increased the risk of death by 50 percent compared to people of normal weight (BMI 18.5-25).

79% of excess deaths in underweight people occur in people more than 70-years-old

There are an estimated 33,746 excess deaths per year associated with being underweight, however, nearly four-fifths of them (79 percent) occur in people over the age of 70.

Comment: This fact is very important to take note of.

This suggests to me that excess deaths in underweight people are associated with underlying, perhaps undiagnosed, disease or depression.

Accounting for underlying disease did not change results

The best study I have seen to date on the effect of body weight on the risk of death, tried to account for underlying diseases by excluding data from anyone who died in the first few years.

In this other study, after excluding deaths in the first few years, they found no increased risk of death from being underweight. They found that the thinner people were, the longer they lived, and the heavier people were, the greater their risk of dying.

Why didn’t this study do the same, that is, exclude data from anyone who died in the first few years?

They did, but said it had little change on the results.

“To examine whether the increased relative risks at lower BMI levels might be related to possible weight loss associated with illness and increased mortality, which could also have decreased the relative risks associated with overweight and obesity, we repeated analyses excluding the first 3 or the first 5 years of deaths and found little change in the relative risk estimates (data not shown),” the study notes.

Comments: Something seems wrong

The study found that

• weight loss due to illness—for example, people who were underweight because they had cancer—did not have a major impact
• people who were underweight due to smoking did not have a major impact, and
• gaining or losing large amounts of weight did not have a major impact.

This is not to say that cancer or smoking or gaining or losing large amounts of weight does not reduce lifespan.

Instead, what they are saying is that there were not enough people with these conditions in the normal or underweight groups to change the results of the study.

Frankly, I am surprised by the results of this study.

The results of this study do not agree with any other study I have seen.

I am not sure why this would be.

It does not make sense to me that being normal weight would increase your risk of death compared to being overweight. I don’t understand why this would be.

REFERENCE

Flegal KM, Graubard BI, Williamson DF, Gail MH. Excess Deaths Associated With Underweight, Overweight, and Obesity. JAMA. 2005 April 20;293(15):1861-1867.

AUTHOR’S CORRESPONDENCE

Katherine M. Flegal, PhD
National Center for Health Statistics
Centers for Disease Control and Prevention
3311 Toledo Rd, Room 4311
Hyattsville, MD 20782
.(JavaScript must be enabled to view this email address)

Articles on the same subject can be found here:

• Mortality associated with obesity

• Deaths from obesity