Do we really understand what a healthy weight looks like for all ethnicities?

Population health and prevention are top of the agenda for the government and the NHS. To understand whether progress is being made, we need benchmarks for health that can track health across the diversity of our population.

For example, we need to be able to measure levels of obesity in the population, as this is a major public health concern. Currently, BMI benchmarks are widely used to do this. However, BMI benchmarks are based on White European or American populations and were not designed for people from global majority ethnic backgrounds (Black, Asian, Brown, dual heritage and indigenous ethnicities, otherwise described as ethnic minority groups in the UK). At the same time, the 2021 census showed that the global majority ethnic population is growing in England and Wales, and reached 9.6 million in 2021. So health benchmarks need to evolve quickly to keep pace with these shifting demographics.

We know there are differences in health outcomes for people from global majority backgrounds. Some of these differences exist because people have different, and often unequal, experiences of the world around them (eg, experiences of poverty, education and racism) and because people exhibit different health behaviours. And some of these differences are inherited – for example, genetic differences between ethnicities. BMI benchmarks need to account for these inherited ethnic differences. One solution is to create ethnicity-specific benchmarks, which NICE (which produces guidelines for the health system in the UK) created for BMI in 2023.

However, is it even possible to create ethnicity-specific BMI benchmarks for everyone? In 2021, in England and Wales, there were 1.3 million people in the ‘Other’ ethnic group – one of the fastest growing populations – and there were 1.7 million people in the ‘Mixed or Multiple’ ethnic group. But when updating their BMI guidelines, NICE identified zero scientific studies on the BMI of people from Mixed ethnic backgrounds. The studies used to create the new ethnicity-specific BMI thresholds actually excluded people from Mixed ethnicity. Also, NICE does not mention ‘Other’ ethnic categories, which could include categories commonly used outside the UK, such as Latino or Native American in the US.

These evidence gaps likely exist because ethnicity-specific benchmarks rely on ethnicity being an objective and unchanging characteristic of an individual. The reality is very different – ethnic identity is subjective and determined not just by an individual by also by the social and political context in which that person lives. This fluidity of ethnic identity is particularly evident for people who have multiple heritages or are from ethnicities less common in the population in which they live, as these people might define their ethnicity in different ways in different contexts. Ethnicity-specific thresholds also rely on ethnic groups being somewhat homogeneous, but people of Mixed and Other heritages could have a range of ethnic backgrounds and have anywhere from 1% to 99% White heritage. So where does that leave the increasingly large number of people in these groups?

The current NICE guidelines take a broad-brush approach by recommending lower BMI thresholds for anyone of ‘Black, Asian or Middle Eastern background’, which effectively includes almost everyone who is not ‘White’. This creates some knotty issues. First, because it leaves some people wondering if they are ‘white enough’ to use the old thresholds, which is a subjective and problematic way to categorise people from global majority ethnic backgrounds. For example, does this include people who identify as Latino, White Middle Eastern, or Hispanic? Or does this include someone who is Mixed and has one grandparent who is Black or Asian and the rest White?

Second, it raises some questions about how scientific evidence is used to create ethnicity-specific BMI thresholds. The NHS BMI calculator lowers BMI thresholds for people from a Mixed background, despite the fact there are no scientific studies that include people of a Mixed background. There are also no scientific studies on people from a Latino background, but if someone from this background (who might have similar amounts of Black heritage as someone who is Mixed) choses the ‘Other ethnic group’ on the NHS BMI calculator, it’ll automatically give higher thresholds, as if they were White.

These different BMI thresholds have real-world consequences. At a population level, including or excluding an ethnic group from the threshold adjustment could significantly impact national obesity figures, and change how we measure health inequalities between different ethnicities. For individuals, lower BMI thresholds could be life changing. For the average-height adult, the cut-off point for being obese for people from a White ethnicity is more than a stone heavier than for people from a ‘Black, Asian or Middle Eastern background’. That could be significant if that person was sat in a GP appointment asking whether they qualify for bariatric surgery or weight loss drugs, or if they are healthy enough to receive NHS-funded IVF.

Ultimately, this means the current BMI thresholds potentially do not fully work for up to 3 million people in the UK. Maybe one day, with advances in individual genome research, it will be possible for everyone to get an individual BMI threshold appropriate for their unique ethnic background. Or perhaps, in the future, there will be a new way to study the risk factors for people of Mixed or Other background. Or maybe BMI thresholds need to be less definitive and more of a conversation starter between clinicians and patients, particularly for patients from a Mixed and Other background. Because if improving the health of the population is going to be a priority, the NHS should aim for health benchmarks that embrace the beautiful complexity of defining ethnic identity and are appropriate for the increasing ethnically diverse population.