Thanksgiving Special: Personalized Nutrition?

A very interesting new study shows that for their sample population, there was a wide variation in post-meal blood glucose form person-to-person even when eating the same food; conversely, what was best for one person was not good for another. An algorithm predicted post-meal blood glucose levels. The authors claim that this type of personalized nutrition will be better than a less efficient “one size fits all” generalized set of dietary recommendations:

  

 Highlights

• High interpersonal variability in post-meal glucose observed in an 800-person cohort
• Using personal and microbiome features enables accurate glucose response prediction
• Prediction is accurate and superior to common practice in an independent cohort
• Short-term personalized dietary interventions successfully lower post-meal glucose

Summary

Elevated postprandial blood glucose levels constitute a global epidemic and a major risk factor for prediabetes and type II diabetes, but existing dietary methods for controlling them have limited efficacy. Here, we continuously monitored week-long glucose levels in an 800-person cohort, measured responses to 46,898 meals, and found high variability in the response to identical meals, suggesting that universal dietary recommendations may have limited utility. We devised a machine-learning algorithm that integrates blood parameters, dietary habits, anthropometrics, physical activity, and gut microbiota measured in this cohort and showed that it accurately predicts personalized postprandial glycemic response to real-life meals. We validated these predictions in an independent 100-person cohort. Finally, a blinded randomized controlled dietary intervention based on this algorithm resulted in significantly lower postprandial responses and consistent alterations to gut microbiota configuration. Together, our results suggest that personalized diets may successfully modify elevated postprandial blood glucose and its metabolic consequences.

Note the graphical and especially, the video abstracts.

 A link to the full paper is here.

From a scientific standpoint, this is a very good, comprehensive, and interesting study, and it highlights the large amount of inter-personal variation in the metabolic response to diet.

Three concerns:

1. From the standpoint of practicality, is it currently realistic to test a significant portion of, for example, the American population in this manner, and generate personalized nutrition for them?  For the more well-off, certainly, but for the masses, we already are in a situation that the health care system is stretched, including financially.  Without having the ability to test the general population, studies like this may be used by some to justify bad eating habits : “I’m likely one of those people who thrive on ice cream!”

2. If a person goes on a personalized diet based on these data, their gut microbiota, as well as aspects of their metabolism, may change over time, which would also change to some extent the degree to which that personalized diet is useful. Will they need to be periodically re-tested, and how would that impact the issues brought up in concern #1?

3. For those people “lucky” enough to get a good response to a “bad” diet, diet compliance may not be an issue, but for many people, compliance will be an issue; people today have difficulty adhering to dietary recommendations. Will they listen better if their personalized profile requires a diet heavy in broccoli?  If they have to make big changes in their diet?  Should this testing therefore be reserved only for those people who exhibit a strong commitment to engage in healthy eating, whatever the outcome of their personal dietary profile is?

In the long run, this study needs to be repeated with a larger, more diverse population. One question to be answered with a larger follow-up study is: for what percentage of the population is a generalized “healthy diet” good?  If a large majority of people – in a very large and diverse population sampled - would get results indicating that the typically recommended “healthy diet” is best, then it may not be necessary to test most people in this manner. Perhaps only the most high-risk individuals would need to be tested, including those who demonstrate high blood glucose even after switching to a “healthy” diet.