Each of 382 healthy adults had their responses to nine foods predicted just by scaling their own glucose reference test by the food's glycemic index. The error was 0.78 mmol/L, smaller than their own test-to-retest wobble of 1.02, which argues against person-specific food effects for glucose.
the research library the whole person
precision wellness
The same banana does not do the same thing in every body.
I kept watching clients follow the same food rules and land in completely different places, and I wanted to know whether that was real or an excuse. It is real, and it has been measured. This page shows both how much people differ and how small the benefit of personalized advice has been in trials so far.
12papers
11studies in people
3counterpoints
this page, by tier
- 8 human evidence
- 3 early human
- 0 lab + animal
- 1 background
the short version
what the research found
the headline numbers
-
Glucose monitors on 800 adults across 46,898 meals showed wide differences in response to identical meals, and a model using blood tests, habits, activity and gut bacteria predicted each person's response better than the food itself did.
-
In 1,002 UK adults eating identical test meals, the variation between people was 103% for blood fats, 68% for glucose and 59% for insulin.
-
In 225 adults with prediabetes, a personalized diet cut time above 140 mg/dL by 1.3 hours a day, versus 0.3 hours on a Mediterranean diet, and the gap held at 12 months.
-
In a randomized trial of 347 US adults over 18 weeks, personalized advice lowered triglycerides by 0.13 mmol/L more than standard advice, and LDL, blood pressure, insulin and glucose did not differ.
-
In 382 healthy adults, scaling each person's own glucose test by a food's glycemic index predicted their responses within 0.78 mmol/L, tighter than their own test-to-retest wobble of 1.02.
my honest bottom line
The person-to-person differences in blood sugar and blood fats after identical meals are well documented, across tens of thousands of meals in several countries. What is not settled is whether acting on them changes health much, and the newest analysis argues that a lot of that variation is day-to-day wobble in each person's own glucose rather than a stable personal signature.
evidence at a glance
12 papers
- human evidence8
- early human3
- lab + animalnone yet0
- background1
Includes 3 counterpoints.
every title links to the original
the papers
Sorted from the strongest human evidence down to the lab work and background reading. Counterpoints are marked, and they stay in on purpose.
347 US adults were randomized to a personalized nutrition program or standard advice for 18 weeks. Triglycerides fell 0.13 mmol/L more with personalization, but the co-primary outcome, LDL cholesterol, did not change, and blood pressure, insulin and glucose did not differ. Weight and waist improved, most in people who stuck with it.
In 225 adults with prediabetes, both diets helped and the personalized one helped more: time above 140 mg/dL fell 1.3 hours a day versus 0.3 hours on a Mediterranean diet, and HbA1c fell 0.16% versus 0.08%. The difference was still there at 12 months.
A review of 11 randomized trials found personalized advice improved what people ate more than general advice did. Only 5 of the 11 used any behavior change technique, and the reviewers called for better designed trials.
The PREDICT study fed 1,002 UK twins and adults identical test meals. Variation between people was 103% for triglycerides, 68% for glucose and 59% for insulin. Personal factors like the gut microbiome explained more of the fat response than the meal did, but for glucose the meal itself mattered more.
An independent US replication in 327 adults without diabetes. Adding personal characteristics and the gut microbiome to food features predicted post-meal glucose with a correlation of 0.62, versus 0.34 for calories alone and 0.40 for carbohydrate.
show all 12 papersshow fewer
Food4Me randomized 1,269 adults in seven European countries to personalized or general online advice for six months. Personalized advice nudged diets a little (5.48 g less red meat and 0.65 g less salt a day), and adding blood markers or genes did not make it work better than diet information alone.
Glucose monitors on 800 adults captured 46,898 meals, and responses to the same meal varied widely between people. A model combining blood tests, habits, body measures, activity and gut bacteria predicted each person's response better than nutrient content did, and a blinded diet built from it lowered post-meal glucose.
A pilot crossover in 23 adults newly diagnosed with type 2 diabetes. The personalized diet gave mean glucose 7.8 mg/dL lower and 2.42 fewer hours a day above 140 mg/dL than a Mediterranean-style diet. Six months later HbA1c had fallen 0.39%, with 16 people left and no control group.
Blinded glucose monitors on 153 healthy people aged 7 to 80 set a reference for normal. Average glucose was 98 to 99 mg/dL (104 over age 60), and the typical person spent 96% of the day between 70 and 140 mg/dL, about 30 minutes above 140 and 15 minutes below 70.
108 adults had their genomes sequenced plus blood metabolites, proteins, gut microbes and clinical labs measured three times over nine months, with daily activity tracking. The data formed connected clusters, and coaching based on each person's data was followed by changes in their clinical markers.
The methods case for testing personalization one person at a time, with repeated crossovers in the same individual, instead of guessing individual responses from group averages. It also explains why these designs need more measurements and more careful statistics.
keep reading
more from the library
food, supplements and pregnancy eating for metabolism Meal timing, protein, fiber, a walk after dinner, and the trials that found less than the headlines. the whole person mindset and the body The milkshake study, the hotel housekeepers, open-label placebos and the growth mindset debate. food, supplements and pregnancy therapeutic supplements Vitamin D, omega-3, magnesium, creatine, probiotics and iron, sorted by the strength of the human data.
This page summarizes published research for education. It is not medical advice, and nothing here describes what any nanonourish™ product does. Talk with your care provider before changing your diet or starting a supplement, especially if you are pregnant, nursing or taking medication. Summaries by Dr. Kaylan Jackson, PhD, a chemist, not a physician. Citations checked against PubMed and Crossref, September 2026.