How does carrying around excess poundage on one's person impact mortality? That question has occupied the minds of many as the worldwide obesity epidemic continues, and the doomsayers everywhere predict a host of dire consequences. Some of the data is unassailable: there's no question that osteoarthritis, type 2 diabetes, high blood pressure and other chronic conditions enjoy a linear relationship with increasing body mass, as Rick and I have described so often on PodMed. Now comes a study in the Journal of the American Medical Association concluding that overweight and so-called grade 1 obesity do not negatively impact mortality; in point of fact overweight seems to confer a survival benefit. Yikes! Let's look more closely at this study.
The first thing to note about this study is it's a meta-analysis. A total of 97 studies reporting hazard ratios for all cause mortality using standard body mass index (BMI) categories were included. Further study characteristics included the fact that all of the studies were prospective and enrolled adults. In total the sample included almost 3 million people and over 270,000 deaths.
Rick points out in the podcast that one reason this analysis was valuable is because obesity has been subcategorized based on BMI. Thus overweight is a BMI between 25 and 30, grade 1 obesity between 30 and 35, grade 2 as 35 to 40, and grade 3 as greater than 40. While it's sad that we have such prevalence of obesity and profound obesity that it must be stratified even further, the health impact of the condition is better defined using these criteria.
So what did this analysis tell us? Surprisingly, it showed that folks who fell into the overweight category had significantly lower all cause mortality than those in the normal weight group or any group of obese individuals. People who fell into the class 1 obese category enjoyed the same mortality risk as those who were of normal weight, while those in categories 2 and 3 experienced higher all cause mortality. Well. Does this mean that people who are wandering around with an extra 10 or so pounds relative to their height are actually better off than those supposed normal weight individuals?
One hypothesis that has been advanced in explaining the results suggests that people who are already on the cusp of illness, such as those with cancer, would have been included in the normal weight category and therefore would skew the results toward a higher risk of death. This and other potential causes of bias in the analysis are rejected by the authors, although they do identify potential errors as a result of self-report of height and weight data.
Rick and I conclude that for now, we're not prepared to say that being overweight is likely to be a more healthful state than being of normal weight, and we are convinced that greater levels of obesity are most definitely deleterious. We also are both fans of modification of BMI tables to somehow adjust for greater amounts of muscle mass. As Rick reveals in the podcast, according to current tables he's on the cusp of overweight, which he most certainly is not, as a result of extensive lower extremity musculature developed over years of cycling.
Other topics this week include transfusions in people with gastrointestinal bleeding in NEJM, fructose, glucose and brain activity in JAMA, and in the same journal, SSRI use during pregnancy and risk of stillbirth and neonatal death. Until next week, y'all live well.