Cookies on this website

We use cookies to ensure that we give you the best experience on our website. If you click 'Accept all cookies' we'll assume that you are happy to receive all cookies and you won't see this message again. If you click 'Reject all non-essential cookies' only necessary cookies providing core functionality such as security, network management, and accessibility will be enabled. Click 'Find out more' for information on how to change your cookie settings.

Body mass index (BMI: weight [kg]/height[m]2) is commonly used to measure general adiposity. However, evidence of its appropriateness for males and females remained inconsistent. This study aimed to identify the most appropriate sex-specific power value that height should be raised to in the formula and that would make it achieve height independency and body fatness dependency. We randomly assigned UK Biobank participants recruited between 2006-2010 in the UK (n=489873; mean age 56.5 years; 94.2% White) into training and testing sets (80%:20%). Using height raised to the power of -50.00 to 50.00, the optimal power value that either minimised correlation with height or maximised correlation with body fat percentage were identified using age-adjusted correlations. The optimal power values for height were 1.77 for males and 1.33 for females. The new formulas resulted in 4.5% of females and 2.4% of males being reclassified into a different BMI category, and did not show significant improvement (in area under Receiver Operating Characteristics Curve, sensitivity and specificity) in identifying individuals with excessive body fat percentage, or in risk prediction of all-cause mortality. Therefore, the conventional BMI formula is still valuable in research and disease screening for both sexes.

Original publication




Journal article


Am J Epidemiol

Publication Date



UK Biobank, body composition, body mass index, correlation, fatness, height