Blood Metabolism Changes Better Predict Diet and Disease Risk

Therefore, researchers need tools that are more reliable and accurate and easy to use for everyone. Using metabolite signatures and data science, they can improve understanding of how much people are ingesting, as well as the risks they may face for cardiometabolic disease affecting millions of people.

Develop algorithms to predict food and dietary habits

Researchers followed more than 2,200 white and black adults in the Young Adult Coronary Arterial Risk Development study, which used blood samples and food surveys to determine metabolic markers of the disease. diet and subsequent disease risk for 25 years.

Through a machine learning model, the investigators were able to generate a blood-based dietary signature to more accurately predict a person’s entire diet across 19 food groups from 10- 20%.

In addition, blood-based signatures are often superior to the healthy eating index, a standard measure of diet quality, to determine who is more likely to develop both diabetes and cardiovascular disease based on on each food group.

In addition to understanding which types of nutrients are better or worse for our health, the methods herein can enable food science researchers to grasp the metabolism of nutrition and food preparation. diets to better understand their impact on health.

Blood-based signature techniques need to be tested in prospective, controlled studies of different diets. Knowing exactly how well people adhered to the diet using blood-based signatures would produce even stronger results.

This is an important step and a toolkit for carrying out nutritional research with greater accuracy and efficiency. Ultimately, such work could allow us to better understand the optimal diet for our patients.

Source: Eurekalert