Body Mass Index as a Predictor of Mortality in Elderly People in Taiwan

Abstract

An interesting original article was reported by Hwang et al.1 recently. The authors found that body mass index (BMI) was a predictor of elderly mortality in Taiwan. Their conclusion was that obesity (BMI> 25 kg/m2) was a significant independent predictor of all-cause mortality, whereas being overweight (BMI> 23 kg/m2) increased the risk of mortality from cancer, cardiovascular diseases, and diabetes. That was a cohort study. The researchers measured the BMI values of the participants at baseline and followed up for 23 years. They excluded subjects who died during the first 3 years of followup. However, they did not consider to evaluate weight changes according to their baseline BMI values of the participants during the follow-up period; in other words, they did not have the premortality BMI values of the participants. Thus, we thought there was a question about the causal relationship between BMI value and mortality. Because the body weight dynamically changed over time2,3, could the baseline BMI values of the participants represent the BMI values during the follow-up, especially 3 years later? In addition, body weight loss of 5% or more in a 3-year period was an independent risk factor of mortality in older adults3. In Hwanget al.’s study,we found that beingoverweight increased the risk of mortality from cancer, cardiovascular diseases, and diabetes, but not from respiratory diseases. Mortality factors of being underweight (BMI < 18.5 kg/m2) 1 increased mortality from cancer and respiratory diseases but not from cardiovascular diseases and diabetes in the elderly review of IJGE. Both over and under BMI changes increase the risks from cancers. It is interesting to consider whether the abnormal BMI value leads to different preexisting chronic disease mortalities or whether different preexisting chronic