Unless you have been living off-planet for the last 20 years or so, you will be aware that obesity prevalence has risen sharply all across the world. We now view obesity as a greater threat to health than undernutrition and it is implicated in the development of heart disease, type-2 diabetes, several types of cancer and a wide spectrum of other health problems.

The standard approach to tackling obesity in individuals is to introduce an energy restricted diet to induce a loss of fat mass. To do this we depend upon personalised calculations of actual energy requirements so that we can ensure that the restricted diet delivers energy below requirement, but whilst still maintaining physiological functions.

Determining energy requirement can be done in a variety of ways, but mostly clinicians rely on predictive equations to determine energy expenditure. These include the Mifflin-St Jeor, Henry, Scholefield, Harris-Benedict, Owen and Katch-McArdle equations. These generally take into account height, weight, age and physical activity levels. All of the equations make assumptions about the individual and are prone to error associated with ethnicity, age and, most importantly, the BMI of the subject. Accuracy tends to be better in lean individuals and the equations lose accuracy with overweight and obesity. Given the importance of estimating energy requirements in order to achieve weight loss in obese individuals, there is a need to identify which equations are most useful for this population.

Angela Madden (University of Hertfordshire) and colleagues carried out two systematic reviews of the literature (http://onlinelibrary.wiley.com/doi/10.1111/jhn.12355/full ) to consider which predictive equations were the most accurate and precise in estimating resting energy expenditure (REE) and total energy expenditure (TEE) in healthy obese adults. They considered 25 studies which were able to directly compare calculated REE and TEE with measurements made using doubly-labelled water and other techniques. The reviews showed that none of the available equations for measuring TEE were accurate in obese people and as such they cannot be recommended for this purpose.

Importantly the reviews found that there was considerable variation in accuracy of equations to measure REE, with BMI as the main determinant of that accuracy. The important outcome of the work is that we have some guidance on which equation to choose based upon the body composition of the individual.

Although a single equation for all individuals is not ideal, Madden and colleagues recognised that in clinical practice, having just one equation for all obese people has greater utility. On this basis they suggest that the Mifflin equation is most appropriate for patients with BMI>25 kg/m².