A recent study by Afzal et al., published in JAMA, concluded that the BMI associated with the lowest all-cause mortality (death from any cause) has shifted upwards over the course of three decades, and that optimal body weight for decreasing mortality currently falls in the overweight category (BMI 25-29.9) (1). This finding has received its share of media attention—garnering confusing headlines such as, “Does Putting On a Few Pounds Help You Cheat Death?,” or, “How healthy is thin? The big BMI debate.”
This finding should not provide cause for debate, and The Nutrition Source has addressed misleading results on higher BMI and mortality before.
Vastly larger studies, including those with millions of participants (around 120 thousand people were involved in the JAMA study) and much longer follow-up, have repeatedly confirmed that a BMI in the normal weight range—between 18.5 and 25—is associated with lowest mortality rates, especially for people younger than 70 years of age (2)(3). This is important to note, as the relationship between body weight and mortality is much stronger before age 65 or 70, due to a greater prevalence of underlying disease and loss of muscle mass at older ages.
So from where did this most recent conclusion originate? In the JAMA study, researchers used data from three Danish cohorts, enrolled in 1976-1978, 1991-1994, and 2003-2013, to examine the relation between BMI and all-cause mortality. Among participants in the first cohort, lowest mortality was reported at a BMI of 23.7, which falls within the normal weight range. In the second cohort, lowest mortality was again associated within the normal range at a BMI of 24.6. However in the final 2003-2013 cohort, a BMI of 27—which falls into the overweight range—was associated with the lowest all-cause mortality. Cue the sensational headlines.
Although many people would like to believe that “putting on extra pounds” has health benefits, these findings are likely the result of bias known as reverse causation, which affects the cohort most severely.
For example, the average follow-up for the 2003-2013 cohort was less than five years, whereas follow-up for the earlier two cohorts was much longer at 11 years and nearly 20 years. With such short follow-up, individuals in the 2003-2013 cohort who died would have been more likely to have existing disease at baseline or be elderly and frail. Many of these participants who died would have experienced weight loss due to the condition that resulted in their death, i.e., reverse causation, making it appear that a higher BMI would be protective against death.
Further, along with age or the presence of chronic disease at the time of study enrollment, the association between BMI and mortality can also be affected by factors such as sex and smoking. The JAMA authors recognized this but analyzed these factors individually, whereas they should be evaluated simultaneously (4).
When creating weight guidelines for healthy people and adequately accounting for reverse causation, it is essential to analyze individuals younger than 60 or 70 years old who have never smoked, had no presence of chronic disease at the start of the analysis, and exclude the first five years of follow-up—which was not done in the JAMA study, and was probably not possible in the most recent cohort because of the short follow up time.
Further, weight guidelines should go beyond evidence of BMI and mortality, considering the relationship between body weight and incidence of chronic diseases like cardiovascular disease, diabetes, and cancer. The many studies of these relationships have documented that the healthiest BMI is below 25, and, most importantly, that weight gain after age 20 has many adverse effects on health.
For more guidance surrounding misinformation on BMI and mortality, see our Ask the Expert with Dr. Walter Willett: “Does Being overweight really decrease mortality? No!”
- Afzal S, Tybjaerg-Hansen A, Jensen GB, Nordestgaard BG. Change in Body Mass Index Associated With Lowest Mortality in Denmark, 1976-2013. JAMA, 2016; 315(18), 1989-1996.
- Aune D, Sen A, Prasad M, et al. BMI and all cause mortality: systematic review and non linear dose-response meta-analysis of 230 cohort studies with 3.74 million deaths among 30.3 million participants. BMJ, 2016; 353:i2156.
- Berrington de Gonzalez A, Hartge P, et al. Body-mass index and mortality among 1.46 million white adults. New England Journal of Medicine, 2010; 363(23), 2211-2219.
- Manson JE, Stampfer MJ, Hennekens CH, Willett WC. Body weight and longevity. A reassessment. JAMA, 1987; 257(3):353-358.