The importance of being normal weight

I wrote just yesterday about Body Mass Index and the increased risk of death and the value of prospective studies as opposed to retrospective studies. Most studies of BMI have been conducted in Western populations, but recently scientists in Korea designed a prospective study to ascertain if Body Mass Index was correlated with risk of death for Asian populations (N Engl J Med 2006;355(8):779-87).

Over 1.2 million persons between the ages of 30 and 95 were enrolled in this Korean Cancer Prevention Study. They excluded those persons who died before the end of the first year of the study, those who initially reported having cardiovascular disease, cancer, liver disease, diabetes, or a respiratory disease, and those who were unusually thin (under a BMI of 16.0) or unusually short (under 51 inches tall).

After twelve years of followup, Dr. Sun Ha Jee and colleagues analyzed the results according to 10 levels of Body Mass Index, sex, age at enrollment, alcohol intake, whether the subject exercised, and whether they smoked cigarettes. Major causes of death were categorized into cancer, cardiovascular causes, respiratory causes.

Their findings are similar to other studies of Body Mass Index: a normal BMI (23.0 - 24.9) and never smoking resulted in the lowest risk of death from any cause for both men and women. Being overweight (BMI 25.0 - 29.9), however, increased the risk of death only slightly (by about 10% for men and 4% for women). Being obese (BMI > 30) carried an increased risk of death for men of 71%, while for women that increased risk was 20%.

What's really interesting is the results for those subjects who were underweight: those men with a BMI of 18.5 or less who had never smoked had an increased risk of death from any cause of 29%. Similarly, women who were underweight saw an increased risk of 17%. (Guess you CAN be too thin!)

The researchers point out that Asian populations tend to have a higher percentage of body fat at the same BMI than Western populations. Therefore the World Health Organization has recommended that for Asian populations, the cutoff values for overweight and obesity should be lower than for Western populations. This only strengthens the observed relationship in this study between BMI and risk of death.

What this means for you:

It's important that research be replicated and this study certainly does that. This is a well-designed, large, prospective study that confirms the results we've seen time and again: that BMI is a good indicator of your overall health risks.

First posted: August 29, 2006