Simple food-based score can predict long-term obesity risk in healthy adults

You are here

Excess weight is the largest cause of ill health and early death in Queensland, causing 8.5% of the total health burden. That's hardly surprising, with two thirds of Queensland adults being overweight or obese. Poor diet, physical inactivity, overweight and obesity all significantly contribute to poor health, chronic diseases and reduced life expectancy in Queenslanders (Overweight and Obesity Prevention Strategy). Now a simple score based on the food people eat could help clinicians predict patients’ risk of becoming overweight and ensure they get the right support and advice, say Spanish researchers who undertook a major long-term study, presented at this year's European Congress on Obesity (ECO) in Vienna, Austria (23-26 May).

The evidence-based Dietary Obesity Prevention Score (DOPS), created by researchers from the University of Navarra in Spain, is based on information about diet and the amount of different types of food people consume.

To create the score, the intake of 14 food groups that have been shown to either promote or protect against obesity were adjusted for total energy intake. The DOPS positively weighted the consumption of vegetables, fruits, legumes, yogurt, nuts, fish, and vegetable to animal protein ratio; while red meat, processed meat, saturated animal fat, refined grains, ultra-processed food, sugar sweetened beverages, and beer and spirits were negatively weighted.

A validated food-frequency questionnaire was used to measure adherence to the DOPS at the start of the study and again 10 years later. Adherence to the DOPS ranged from 14 (non-adherence) to 42 (complete adherence). A higher score indicated a protective diet—one with a higher consumption of mostly plant-based foods and a lower consumption of risk-promoting dietary components such as red and processed meat, ultraprocessed food and sugar sweetened beverages.

Under the system, people score more on a one to 24 scale for eating plenty of healthier foods like fruit, vegetables, nuts and fish. They score less well for eating lots of red meat, processed foods and drinking sugary drinks and alcoholic beverages like beer and spirits.

Its developers say their research, which followed more than 11,300 people over about 10 years, shows it can be used fairly accurately to predict whether or not someone is likely to become overweight or obese. Spanish university graduates who signed up to take part in a long-term health study in 1999, complete a detailed questionnaire about their eating habits. All were of a normal weight at the time and their responses were used to generate their DOPS. When they were followed up around 10 years later – now middle-aged – nearly 2,200 had become overweight or obese.

Researchers found a clear link between their initial score and whether or not they had maintained a healthy weight. Those with a high score indicating a healthier diet were found to have a 42% lower risk of being overweight or obese, compared to those with a low score indicating an unhealthy diet. Analysis found that the risk of becoming overweight or obese went down roughly 7% for every two additional points on the scale.

They say this shows the score could be a highly useful tool to help clinicians give advice on healthy lifestyles and refer “high risk” individuals to receive extra support. Study author Professor Maira Bes-Rastrollo, of the University of Navarra, said the score had the potential to be developed into a quick and easy-to-use assessment tool.

“We hope that our easy-to-implement preventive score might be developed into a rapid assessment tool to provide both individuals and doctors with helpful dietary guidance-raising awareness of diet susceptibility to weight gain in the long-term, and encouraging healthier eating patterns,” she said.

However, she said the performance and reliability of the score had yet to be tested among different groups. “Future research will increase the accuracy of outcome predictions in different populations and age groups,” she said.

The authors acknowledge that their findings show observational differences rather than evidence of cause and effect. They point out some limitations inherent to the methodology, including that the results are based on self-reported dietary intake and that, although they performed multivariable analyses adjusting for a wide variety of confounders, some degree of residual confounding cannot be discarded. They also note that the study includes mainly Caucasian adults so the generalisability of the findings to other ethnicities are uncertain.