A third of adults in Wales (34%) have been revealed to be living with obesity in 2022-23, according to new analysis by Nesta - potentially more than any other nation in the UK - with older generations much more severely affected than previously thought.

The methodology for generating official government statistics on obesity prevalence within the Welsh population is different from that used in the other three nations. Whereas in England, Northern Ireland and Scotland obesity figures are mainly based on measurements taken by health professionals — or when this has not been possible self-reported data has been adjusted to approximate professionally measured data — in Wales the numbers rely on unadjusted self-reported height and weight measurements. 

However, it’s well known that in self-reported cases, people tend to underreport their weight and overreport their height. This skews the body mass index (BMI) calculations used to figure out how widespread adult obesity is across a whole population.

But by applying the same prediction equations used in England and Scotland to correct for bias in self-reported height and weight data, Nesta has found that the Welsh Government has been dramatically underestimating the extent of obesity in Wales.

Making these corrections resulted in estimated adult obesity rates in Wales jumping from 25% in 2021-2022 to 32%. This is considerably higher than figures for that year in both England and Scotland where 26% and 29% of adults were living with obesity respectively. It also means that potentially Wales has higher obesity rates than in Northern Ireland, where 27% of adults were living with obesity in 2019-20 (when height and weight were last asked as part of Health Survey Northern Ireland).

Using the same approach to correct data from 2022-23 — the most recent National Survey for Wales — we found that approximately 34% of adults in Wales were estimated to be living with obesity (compared to 26% on the uncorrected measures). The data for 2022-2023 is not currently available for the other UK nations.

Our re-examination of data from the 2022-2023 National Survey for Wales also revealed that when the self-reporting data is corrected, the distribution of obesity in the population also changed significantly - showing over 55s to be at greatest risk. 

Obesity prevalence based on corrected height and weight was found to be 10 percentage points higher for those aged 55-64 years, 12 percentage points higher for those aged 65-74 years and 11 percentage points higher for those aged over 75 years, compared to uncorrected figures. This suggests that in fact those in the 55-64 years age group are at the highest risk of obesity, rather than the 45-54 year age group, as unadjusted figures would suggest.

The corrections to Welsh adult data mean that it now aligns with the patterns of childhood obesity seen across the UK nations.

That Wales previously had lower adult obesity than Scotland and Northern Ireland and similar figures to England had always seemed somewhat puzzling and unexpected, given that obesity rates amongst children in Wales — which for obvious reasons are based on professional measurements — are considerably higher than in both England and Scotland.

Taking reception age children (4-5 years) as an example, 12.6% of reception age children in Wales were obese compared to 10.2% in Scotland and 9.7% in England in 2018/2019 (when national figures for Wales were last reported). (Obesity prevalence is not reported for reception age children in Northern Ireland).

How do ‘prediction equations’ work?

Prediction equations were originally developed by comparing how people reported their height and weight with what was actually measured by interviewers as part of Health Survey for England (HSE) data between 2011 - 2016. 

The chart below shows how the average BMI calculated in the 2011-2016 HSE data differed between self-reported results and an interviewer measured height and weight readings.

The prediction equations subsequently developed can be applied to new datasets by inputting individuals' self-reported height and weight along with their sex and age — which affect the extent to which individuals mis-report —  and generating height and weight estimates which researchers have found to be much closer to the measurements taken by professionals. 

While these prediction equations were generated from English data, the misreporting of height and weight in the HSE sample broadly follows trends observed in older work carried out with Welsh data, suggesting that these equations are likely to generalise well to a Welsh sample.

Our research has found that the extent of the obesity issue in Wales is much greater than previously thought. We need bold solutions to successfully tackle obesity in Wales, and the Welsh Government's Healthy Weight, Healthy Wales strategy outlines a range of potentially impactful policies — such as restrictions on the placement and price promotion of products high in fat, sugar and salt – but it's still not clear which solutions will have the greatest impact.

To ensure obesity data accurately represents the true picture in Wales, and to allow more reliable comparison with other countries in the UK, we encourage the Welsh Government to use the same approach used in this analysis when reporting obesity prevalence in the future. We’d also encourage retrospectively applying this approach to previous years’ data to give a more accurate picture of how obesity rates have changed over time.

This would strengthen the evidence base for interventions and complement the research Nesta is doing, in conjunction with an expert panel, to synthesise peer-reviewed evidence and reach a consensus about the relative impact and cost of a wide range of different obesity policies. Insights from this work will be made publicly available through an online toolkit later this year.

Going beyond those policies already being developed by Welsh Government, we also need bolder policies that will incentivise businesses to prioritise health and deliver the level of impact required. Nesta has recently proposed that the UK Government should implement mandatory health targets for large grocery retailers. Such targets would encourage retailers to improve the healthiness of their offer by increasing the overall nutritional quality of the food they sell — but give retailers the flexibility to choose how they meet the target. Our modelling estimates that setting a health target for the 11 largest grocery retailers, at levels similar to those already achieved by today’s ‘best’ players, could reduce obesity prevalence by approximately 23%. By taking a lead on supporting this proposal, Welsh Government could make real progress towards turning the tide on obesity once and for all.