The research question
The NIH Nutrition for Precision Health programme, linked to All of Us, studies individual responses to diet. It combines information about food, genetics, the microbiome, lifestyle and social circumstances. Its aim is to develop predictive algorithms. A research programme is not proof that a perfect diet for every person already exists.
What researchers measure
The design includes dietary information and biological samples, meal challenges and, for some participants, periods with researcher-provided diets. Comparing responses under controlled conditions can help distinguish biological signals from errors in food recall. A measurement nevertheless remains connected to the conditions under which it was obtained.
Prediction is not prescription
A useful algorithm must work for people beyond its development sample. Benefits, safety and accessibility need assessment before predictions become advice. Nutryfit does not use experimental NPH algorithms or interpret genetic or microbiome tests. Our calorie calculator uses an explicit estimation formula.
Assessing a personalisation claim
Ask what data a service uses, whether its model has independent validation and which meaningful outcome improved. A precise-looking decimal is not evidence of clinical accuracy. Preferences, budget and cooking time already help organise meals without giving an app diagnostic powers.
