Study analysis · Nature Medicine · 2024
This app beat USDA diet advice—here’s how much weight and triglycerides it dropped in just 18 weeks.
A diet app that uses your blood sugar, fat levels, and gut bacteria gave people better health results than standard government diet advice.
Overview
What the study found
The study in plain English — the bottom line, every takeaway we extracted, and what to do with them.
In simple terms
This study is like a fair race between two diets: one tailored to your body and one general advice. It found that the tailored diet helped people lower their blood fats and lose a little weight better than the regular advice. But it doesn't prove it's the best diet ever—just better than this one kind of advice in this group of people.
What’s the bottom line?
This study tested a diet app that gives you custom food advice based on your blood sugar, fat levels, and gut bacteria — and compared it to regular USDA diet tips.
How strong is this study?
This study did a good job by randomly assigning people to groups so it's fair, and it measured real health changes. But it didn't hide which diet people got, so people might have tried harder if they knew they were getting the fancy personalized plan—which could make the results look better than they really are.
40 / 100
- COI disclosure+40/40
- Data availabilitydata not shared
- Code availabilitycode not shared
77 / 100
- Randomization+20/20
- Blindingnot blinded
- Control group+15/15
- Sample size (n=347)+16.5/20
- Follow-up+10/10
100 / 100
100 / 100
- P-values+15/15
- Effect size+20/20
- Confidence intervals+15/15
- Pre-registration+15/15
Each component is scored out of 100 and then capped by the study design — a case series cannot reach the ceiling a randomised trial can, however well it is reported.
Where it sits
RCT reviewsReviews of RCTs (Meta-analyses)
Max 100Randomized TrialsRandomized Trials
Max 90Reviews of Cohort StudiesReviews of Cohort Studies
Max 85Cohort StudiesCohort Studies
Max 72Reviews of Case-Control StudiesReviews of Case-Control Studies
Max 63Case-Control StudiesCase-Control Studies
Max 58Cross-Sectional & Case SeriesCross-Sectional & Case Series
Max 50Expert OpinionExpert Opinion
Max 580 / 100
Probability of being correct
Participants are randomly assigned to treatment or control groups, minimizing bias. The gold standard for testing whether an intervention causes an effect.
This design can establish causation. Randomization allows causal inference between the personalized nutrition program and observed outcomes, but lack of blinding and unequal intervention intensity may introduce performance bias, limiting the strength of causation.
Major COI
Major conflicts that significantly reduce study credibility
The study was conducted by ZOE, a commercial company offering personalized nutrition services, and the personalized program tested is their proprietary product, creating a clear financial conflict of interest.
Funders
Conflict Details
ZOE: Authors are affiliated with ZOE, the company developing and commercializing the personalized nutrition program being tested.
The study was conducted and likely analyzed by the company whose product is being evaluated. No independent oversight or third-party analysis is mentioned. The trial is registered under ClinicalTrials.gov, but no funding source or conflict of interest statement is disclosed in the provided text. The results strongly favor the company's product, raising concerns about bias.
Key takeaways
- 01
People using the app lost 2.46 kg, reduced waist size by 2.35 cm, improved diet quality by 7 points, and lowered triglycerides by 0.21 mmol/L — all more than those following standard advice.
- 02
These changes are modest but meaningful — even small drops in triglycerides and waist size lower heart disease risk, and better diet quality is linked to long-term health.
Surprising findings
- LDL cholesterol didn’t improve overall—even though triglycerides and weight did.Most people assume weight loss automatically lowers ‘bad’ cholesterol, but this study shows LDL is stubborn and may need longer-term or targeted interventions.
- The control group improved fiber intake and reduced fat—but still lost less weight and had worse microbiome shifts.Even when people followed standard advice ‘correctly,’ they didn’t get the same biological benefits as those using personalized feedback.
- Participants reported better energy, mood, and reduced hunger—without being told to focus on mental health.Most nutrition studies ignore subjective well-being, but here, the app improved how people felt—suggesting biology and mood are deeply linked.
Practical takeaways
Try a personalized nutrition app that tracks your blood sugar and gut health—stick with it for at least 12 weeks and log your meals consistently.
The app used in the study is expensive and tied to a company with financial stakes; results may not be replicable with cheaper tools.
high confidenceFocus on food quality scores (like fiber, processing level, fat type) instead of just calories or macros.
Not all apps are scientifically validated—look for ones that use glucose monitoring and microbiome data, not just generic meal plans.
medium confidenceIf you’re trying to lower triglycerides or waist size, prioritize consistency over perfection—even small daily adherence yields big results.
This study lasted 18 weeks; long-term sustainability and cost-effectiveness are still unknown.
medium confidenceWhy this study matters
Personalized app beats USDA guidelines
Participants using the ZOE app saw a 0.21 mmol/L drop in fasting triglycerides—nearly triple the 0.07 mmol/L drop in the USDA group. They also lost 2.46 kg and reduced waist size by 2.35 cm, while the control group saw almost no change.
Most people follow USDA guidelines but still struggle with weight and heart health—this shows personalized biology-based advice may work better than one-size-fits-all recommendations.
Gut bacteria changed—here’s why it matters
The personalized group saw a significant shift in gut microbiome composition, with 11 out of 15 ‘favorable’ bacterial species increasing—while the control group saw none increase. These microbes are linked to lower inflammation and better metabolism.
Your gut isn’t just for digestion—it’s a metabolic organ. This study proves your diet can actively reshape it for better health, not just ‘eat more fiber’.
Adherence = massive results
Highly adherent users lost 4.7% of their body weight (nearly 5 kg) and cut waist size by 6.31 cm—nearly 3x more than low-adherence users. HbA1c dropped 0.06% only in those who stuck with the plan.
It’s not the app that’s magic—it’s sticking to it. This flips the script: the tool is only as good as your consistency.
No calorie counting—yet still lost weight
Participants weren’t told to count calories or restrict portions. Weight loss came from personalized food scores guiding them toward lower-energy-density meals and better satiety—without hunger.
You don’t need to starve yourself to lose weight. This suggests smarter food choices, not fewer calories, can drive results.
Want the whole report?
Detailed mode opens the full scientific breakdown — every score component, the methodology, conflicts of interest, the evidence analysis behind each claim, and the raw study data.
Overview
What the study found
The study in plain English — the bottom line, every takeaway we extracted, and what to do with them.
Not medical advice. For informational purposes only. Always consult a healthcare professional. Terms
This study tested a diet app that gives you custom food advice based on your blood sugar, fat levels, and gut bacteria — and compared it to regular USDA diet tips.
Research results
People using the app lost 2.46 kg, reduced waist size by 2.35 cm, improved diet quality by 7 points, and lowered triglycerides by 0.21 mmol/L — all more than those following standard advice.
What this means - more context
These changes are modest but meaningful — even small drops in triglycerides and waist size lower heart disease risk, and better diet quality is linked to long-term health.
This study tested whether a personalized nutrition program (PDP) using glucose, triglyceride, microbiome, and health data improves cardiometabolic health more than standard USDA dietary advice in generally healthy adults.
An 18-week randomized trial found that a personalized nutrition app significantly reduced fasting triglycerides, weight, waist circumference, and HbA1c, and improved diet quality and gut microbiome composition compared to standard USDA dietary advice, with greater benefits in highly adherent participants.
Methods Used
347 adults aged 41–70 were randomized to a personalized nutrition app (PDP) using postprandial glucose, triglycerides, microbiome, and health history data, or to standard USDA dietary advice via online resources. Primary outcomes were fasting triglycerides and LDL-C at 18 weeks; secondary outcomes included weight, waist circumference, HbA1c, diet quality (HEI), and microbiome beta-diversity. Intention-to-treat analysis was used.
Main Finding
The personalized nutrition program reduced fasting triglycerides by 0.21 mmol/L vs. 0.07 mmol/L with standard advice (mean difference: −0.13 mmol/L, P=0.016), and led to significantly greater weight loss (−2.46 kg), waist circumference reduction (−2.35 cm), diet quality improvement (+7.08 HEI points), and favorable microbiome shifts, with stronger effects in highly adherent participants.
Confidence Level
High — randomized controlled trial with intention-to-treat analysis, pre-registered protocol, effect sizes and confidence intervals reported, and multiple validated biomarkers measured.
Study Flags
Red Flags
- •Control group received less intensive support (leaflet/video) than PDP (app + coaching)
- •High dropout rate (35% of participants excluded from per-protocol analysis)
- •Funding and authorship tied to ZOE Ltd, the company behind the app
Surprising Findings
LDL cholesterol didn’t improve overall—even though triglycerides and weight did.
Most people assume weight loss automatically lowers ‘bad’ cholesterol, but this study shows LDL is stubborn and may need longer-term or targeted interventions.
Practical Takeaways
Try a personalized nutrition app that tracks your blood sugar and gut health—stick with it for at least 12 weeks and log your meals consistently.
RCT reviewsReviews of RCTs (Meta-analyses)
Max 100Randomized TrialsRandomized Trials
Max 90Reviews of Cohort StudiesReviews of Cohort Studies
Max 85Cohort StudiesCohort Studies
Max 72Reviews of Case-Control StudiesReviews of Case-Control Studies
Max 63Case-Control StudiesCase-Control Studies
Max 58Cross-Sectional & Case SeriesCross-Sectional & Case Series
Max 50Expert OpinionExpert Opinion
Max 580 / 100
Probability of being correct
Participants are randomly assigned to treatment or control groups, minimizing bias. The gold standard for testing whether an intervention causes an effect.
Human RCT
Subject
High probability
on the GRADE evidence scale
This study is like a fair race between two diets: one tailored to your body and one general advice. It found that the tailored diet helped people lower their blood fats and lose a little weight better than the regular advice. But it doesn't prove it's the best diet ever—just better than this one kind of advice in this group of people.
Major conflicts — industry funding with significant control over the research. Reporting score and overall score cap have been reduced.
Strengths
- Randomized controlled trial design with intention-to-treat analysis
- Large sample size (n=347) with adequate power for primary outcome
- Use of validated biomarkers (TG, HbA1c, microbiome sequencing)
Weaknesses
- No blinding of participants or personnel
- Control group received less intensive intervention (lower contact time)
- High attrition in per-protocol analysis (65% completion)
Methodology
Evidence Keywords
Statistical Reporting
Not medical advice. For informational purposes only. Always consult a healthcare professional. Terms
This study tested a diet app that gives you custom food advice based on your blood sugar, fat levels, and gut bacteria — and compared it to regular USDA diet tips.
Research results
People using the app lost 2.46 kg, reduced waist size by 2.35 cm, improved diet quality by 7 points, and lowered triglycerides by 0.21 mmol/L — all more than those following standard advice.
What this means - more context
These changes are modest but meaningful — even small drops in triglycerides and waist size lower heart disease risk, and better diet quality is linked to long-term health.
This study tested whether a personalized nutrition program (PDP) using glucose, triglyceride, microbiome, and health data improves cardiometabolic health more than standard USDA dietary advice in generally healthy adults.
An 18-week randomized trial found that a personalized nutrition app significantly reduced fasting triglycerides, weight, waist circumference, and HbA1c, and improved diet quality and gut microbiome composition compared to standard USDA dietary advice, with greater benefits in highly adherent participants.
Methods Used
347 adults aged 41–70 were randomized to a personalized nutrition app (PDP) using postprandial glucose, triglycerides, microbiome, and health history data, or to standard USDA dietary advice via online resources. Primary outcomes were fasting triglycerides and LDL-C at 18 weeks; secondary outcomes included weight, waist circumference, HbA1c, diet quality (HEI), and microbiome beta-diversity. Intention-to-treat analysis was used.
Main Finding
The personalized nutrition program reduced fasting triglycerides by 0.21 mmol/L vs. 0.07 mmol/L with standard advice (mean difference: −0.13 mmol/L, P=0.016), and led to significantly greater weight loss (−2.46 kg), waist circumference reduction (−2.35 cm), diet quality improvement (+7.08 HEI points), and favorable microbiome shifts, with stronger effects in highly adherent participants.
Confidence Level
High — randomized controlled trial with intention-to-treat analysis, pre-registered protocol, effect sizes and confidence intervals reported, and multiple validated biomarkers measured.
Study Flags
Red Flags
- •Control group received less intensive support (leaflet/video) than PDP (app + coaching)
- •High dropout rate (35% of participants excluded from per-protocol analysis)
- •Funding and authorship tied to ZOE Ltd, the company behind the app
Surprising Findings
LDL cholesterol didn’t improve overall—even though triglycerides and weight did.
Most people assume weight loss automatically lowers ‘bad’ cholesterol, but this study shows LDL is stubborn and may need longer-term or targeted interventions.
Practical Takeaways
Try a personalized nutrition app that tracks your blood sugar and gut health—stick with it for at least 12 weeks and log your meals consistently.
RCT reviewsReviews of RCTs (Meta-analyses)
Max 100Randomized TrialsRandomized Trials
Max 90Reviews of Cohort StudiesReviews of Cohort Studies
Max 85Cohort StudiesCohort Studies
Max 72Reviews of Case-Control StudiesReviews of Case-Control Studies
Max 63Case-Control StudiesCase-Control Studies
Max 58Cross-Sectional & Case SeriesCross-Sectional & Case Series
Max 50Expert OpinionExpert Opinion
Max 580 / 100
Probability of being correct
Participants are randomly assigned to treatment or control groups, minimizing bias. The gold standard for testing whether an intervention causes an effect.
Human RCT
Subject
High probability
on the GRADE evidence scale
This study is like a fair race between two diets: one tailored to your body and one general advice. It found that the tailored diet helped people lower their blood fats and lose a little weight better than the regular advice. But it doesn't prove it's the best diet ever—just better than this one kind of advice in this group of people.
Major conflicts — industry funding with significant control over the research. Reporting score and overall score cap have been reduced.
Strengths
- Randomized controlled trial design with intention-to-treat analysis
- Large sample size (n=347) with adequate power for primary outcome
- Use of validated biomarkers (TG, HbA1c, microbiome sequencing)
Weaknesses
- No blinding of participants or personnel
- Control group received less intensive intervention (lower contact time)
- High attrition in per-protocol analysis (65% completion)
Methodology
Evidence Keywords
Statistical Reporting
Scoring
How strong is this study?
This study did a good job by randomly assigning people to groups so it's fair, and it measured real health changes. But it didn't hide which diet people got, so people might have tried harder if they knew they were getting the fancy personalized plan—which could make the results look better than they really are.
40 / 100
- COI disclosure+40/40
- Data availabilitydata not shared
- Code availabilitycode not shared
77 / 100
- Randomization+20/20
- Blindingnot blinded
- Control group+15/15
- Sample size (n=347)+16.5/20
- Follow-up+10/10
100 / 100
100 / 100
- P-values+15/15
- Effect size+20/20
- Confidence intervals+15/15
- Pre-registration+15/15
Each component is scored out of 100 and then capped by the study design — a case series cannot reach the ceiling a randomised trial can, however well it is reported.
Where it sits
RCT reviewsReviews of RCTs (Meta-analyses)
Max 100Randomized TrialsRandomized Trials
Max 90Reviews of Cohort StudiesReviews of Cohort Studies
Max 85Cohort StudiesCohort Studies
Max 72Reviews of Case-Control StudiesReviews of Case-Control Studies
Max 63Case-Control StudiesCase-Control Studies
Max 58Cross-Sectional & Case SeriesCross-Sectional & Case Series
Max 50Expert OpinionExpert Opinion
Max 580 / 100
Probability of being correct
Participants are randomly assigned to treatment or control groups, minimizing bias. The gold standard for testing whether an intervention causes an effect.
This design can establish causation. Randomization allows causal inference between the personalized nutrition program and observed outcomes, but lack of blinding and unequal intervention intensity may introduce performance bias, limiting the strength of causation.
Major COI
Major conflicts that significantly reduce study credibility
The study was conducted by ZOE, a commercial company offering personalized nutrition services, and the personalized program tested is their proprietary product, creating a clear financial conflict of interest.
Funders
Conflict Details
ZOE: Authors are affiliated with ZOE, the company developing and commercializing the personalized nutrition program being tested.
The study was conducted and likely analyzed by the company whose product is being evaluated. No independent oversight or third-party analysis is mentioned. The trial is registered under ClinicalTrials.gov, but no funding source or conflict of interest statement is disclosed in the provided text. The results strongly favor the company's product, raising concerns about bias.