Eddy Edson
Well-Known Member
- Relationship to Diabetes
- In remission from Type 2
I found this fascinating study from 2015: https://www.cell.com/cell/pdfExtended/S0092-8674(15)01481-6
People eating identical meals present high variability in post-meal blood glucose response. Personalized diets created with the help of an accurate predictor of blood glucose response that integrates parameters such as dietary habits, physical activity, and gut microbiota may successfully lower postmeal blood glucose and its long-term metabolic consequences.
...
Here, we set out to quantitatively measure individualized PPGRs [post-prandial glucose response], characterize their variability across people, and identify factors associated with this variability. To this end, we continuously monitored glucose levels during an entire week in a cohort of 800 healthy and prediabetic individuals and also measured blood parameters, anthropometrics, physical activity, and selfreported lifestyle behaviors, as well as gut microbiota composition and function. Our results demonstrate high interpersonal variability in PPGRs to the same food. We devised a machine learning algorithm that integrates these multi-dimensional data and accurately predicts personalized PPGRs, which we further validated in an independently collected 100-person cohort. Moreover, we show that personally tailored dietary interventions based on these predictions result in significantly improved PPGRs accompanied by consistent alterations to the gut microbiota.
They developed a model for predicting individual PPGRs based on a "pool of 137 features representing meal content (e.g., energy, macronutrients, micronutrients); daily activity (e.g., meals, exercises, sleep times); blood parameters (e.g., HbA1c%, HDL cholesterol); CGM-derived features; questionnaires; and microbiome features..."
On a validation test, the model correlated against actual outcomes with R = 0.68, about as high as theoretically possible, given that the correlation between outcomes for the same individual eating the same meal at different times was about R = 0.71.
By comparison: " ... the ‘carbohydrate counting’ model, as it is the current gold standard for predicting PPGRs (American Diabetes Association., 2015b; Bao et al., 2011)" delivered R = 0.38.
Anyway, the paper is a data/diabetes/nutrition geek's picnic & I hope is a sign that real, useful, convenient personalised nutrition advice will become a matter of course before too long.
People eating identical meals present high variability in post-meal blood glucose response. Personalized diets created with the help of an accurate predictor of blood glucose response that integrates parameters such as dietary habits, physical activity, and gut microbiota may successfully lower postmeal blood glucose and its long-term metabolic consequences.
...
Here, we set out to quantitatively measure individualized PPGRs [post-prandial glucose response], characterize their variability across people, and identify factors associated with this variability. To this end, we continuously monitored glucose levels during an entire week in a cohort of 800 healthy and prediabetic individuals and also measured blood parameters, anthropometrics, physical activity, and selfreported lifestyle behaviors, as well as gut microbiota composition and function. Our results demonstrate high interpersonal variability in PPGRs to the same food. We devised a machine learning algorithm that integrates these multi-dimensional data and accurately predicts personalized PPGRs, which we further validated in an independently collected 100-person cohort. Moreover, we show that personally tailored dietary interventions based on these predictions result in significantly improved PPGRs accompanied by consistent alterations to the gut microbiota.
They developed a model for predicting individual PPGRs based on a "pool of 137 features representing meal content (e.g., energy, macronutrients, micronutrients); daily activity (e.g., meals, exercises, sleep times); blood parameters (e.g., HbA1c%, HDL cholesterol); CGM-derived features; questionnaires; and microbiome features..."
On a validation test, the model correlated against actual outcomes with R = 0.68, about as high as theoretically possible, given that the correlation between outcomes for the same individual eating the same meal at different times was about R = 0.71.
By comparison: " ... the ‘carbohydrate counting’ model, as it is the current gold standard for predicting PPGRs (American Diabetes Association., 2015b; Bao et al., 2011)" delivered R = 0.38.
Anyway, the paper is a data/diabetes/nutrition geek's picnic & I hope is a sign that real, useful, convenient personalised nutrition advice will become a matter of course before too long.
