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

Abstract

Overweight and obesity are associated with an increased risk of all-cause mortality, though the shape of the dose-response curve and the location of the mortality nadir are significantly influenced by smoking status and preclinical disease. A meta-analysis of 230 cohort studies involving 3.74 million deaths reveals a J-shaped association among never smokers, with the lowest mortality risk occurring at a body mass index (BMI) of 23–24. This nadir shifts to a BMI of 22–23 in healthy never smokers and further declines to 20–22 in studies with at least 20 years of follow-up. In contrast, analyses that include current or former smokers, or those with shorter follow-up durations, tend to produce U-shaped curves with a higher nadir around a BMI of 25. These discrepancies suggest that the apparent survival advantage of overweight individuals in some populations is a result of residual confounding by smoking and weight loss associated with undiagnosed illness. While underweight is also linked to increased mortality, this relationship is likely attenuated by longer observation periods, suggesting the influence of pre-existing disease. When potential biases are minimized by focusing on healthy never smokers over long durations, the data demonstrate a progressive increase in the risk of premature death across the overweight and obese BMI ranges. – AI-generated abstract.