Study analysis · Nutrition, metabolism, and cardiovascular diseases : NMCD · 2025
Cutting junk food only helped people lose 3.4 kg more in a year — and it might not matter at all.
People who ate less junk food lost a tiny bit more weight than others, but the difference was so small it probably doesn’t help your health.
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 gave two groups of people different diets and saw who lost more weight. Because they randomly picked who got which diet, we can guess that the diet caused the difference in weight loss — but the difference was tiny. So it’s not a huge win, and we can’t say it works for everyone.
What’s the bottom line?
Scientists tested if telling people to eat less junk food helps them lose more weight when they're already eating fewer calories.
How strong is this study?
The study did a good job by randomly assigning people to diets, which helps make it fair. But we don’t know if people knew which diet they were on, and we don’t have all the details — so we can’t be totally sure the results are perfect. That’s why we should be careful not to overhype it.
0 / 100
- COI disclosureconflicts of interest not disclosed
- Data availabilitydata not shared
- Code availabilitycode not shared
69 / 100
- Randomization+20/20
- Blindingblinding unclear
- Control group+15/15
- Sample size (n=148)+10.5/20
- Follow-up+10/10
100 / 100
46 / 100
- P-values+15/15
- Effect sizeno effect size reported
- Confidence intervals+15/15
- Pre-registrationnot pre-registered
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 556 / 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 was explicitly stated, allowing causal inference, but lack of blinding information and abstract-only access limit confidence in effect size and control of confounders.
No Conflicts
No conflicts of interest identified
No conflicts of interest or funding information were disclosed in the provided text.
The study lacks any declared funding sources or conflict of interest disclosures. While the methodology appears rigorous, the absence of transparency about funding or author affiliations limits the ability to fully assess potential bias.
Key takeaways
- 01
People who tried to eat less junk food ate it 14% of the time (down from 21%) and lost a little more weight (82.9 kg vs.
- 02
86.3 kg) than those who just ate fewer calories.
- 03
The extra weight loss was very small — less than 4 kg — and likely not meaningful for most people's health.
Surprising findings
- UPF intake was already low at baseline (~22%) and barely changed in the control group.Most public health claims assume people eat 50%+ UPF — but this study’s participants were already eating less than expected.
Practical takeaways
If you're already eating under 25% ultra-processed foods, cutting further may not give you meaningful weight loss.
This study only looked at obese adults already on calorie restriction — results may not apply to those eating 50%+ UPF.
low confidenceWhy this study matters
Junk Food Cut, But Still High
Participants started eating about 22% ultra-processed foods (UPF) and ended at 14% in the restricted group — still above 20%. The control group stayed around 20%.
Even after trying to cut junk food, most people were still eating it daily — suggesting it’s hard to avoid even with guidance.
Statistically Significant, Clinically Useless
The UPF-restricted group lost 82.9 kg vs. 86.3 kg in the control group — a 3.4 kg difference that was statistically significant (p=0.01) but labeled 'non-clinically significant'.
A number that looks impressive on paper might not mean anything in real life — like losing a bag of sugar over a year.
No Other Health Improvements
Despite reduced UPF intake and slight weight loss, no changes occurred in metabolic, biochemical, or body composition markers.
You might lose a little weight, but your blood sugar, cholesterol, or energy levels didn’t improve — challenging the idea that UPF is uniquely toxic.
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
Scientists tested if telling people to eat less junk food helps them lose more weight when they're already eating fewer calories.
Research results
People who tried to eat less junk food ate it 14% of the time (down from 21%) and lost a little more weight (82.9 kg vs. 86.3 kg) than those who just ate fewer calories.
What this means - more context
The extra weight loss was very small — less than 4 kg — and likely not meaningful for most people's health.
To evaluate the effectiveness and metabolic effects of restricting ultra-processed food (UPF) consumption in obese individuals undergoing energy restriction.
In a 12-month randomized trial, participants assigned to UPF restriction reduced their UPF intake and NOVA-UPF score more than controls, and lost slightly more weight (82.9 kg vs. 86.3 kg, p=0.01), but the difference was statistically significant yet non-clinically significant. Baseline UPF intake was low (~22%) and remained above 20% in both groups. No other metabolic outcomes changed.
Methods Used
Randomized, parallel clinical trial with 148 obese adults; two groups: generic energy restriction (ER-G) and energy restriction with UPF restriction (ER-UPF). Energy requirements were determined by calorimetry and accelerometry. Anthropometric, dietary, body composition, metabolic, and biochemical data were collected over 12 months.
Main Finding
The ER-UPF group reduced UPF intake from 21.2% to 13.9% and the NOVA-UPF score from 2.74 to 1.86 (p=0.03), while the ER-G group showed minimal change (23.7% to 20.0%, p=0.08). Weight loss was greater in ER-UPF (final mean 82.9 kg) vs. ER-G (86.3 kg; p=0.01), but described as statistically significant yet non-clinically significant.
Confidence Level
Limited - based on abstract only, full methodology not available
Study Flags
Red Flags
- •Full text not available - methodology details cannot be verified
- •Baseline UPF intake was low and remained high, limiting generalizability
- •Effect size for weight loss was statistically significant but described as non-clinically significant
No biological mechanisms were identified in this study. This may be an epidemiological, observational, or survey-based study that reports associations rather than proposing causal biological pathways.
Surprising Findings
UPF intake was already low at baseline (~22%) and barely changed in the control group.
Most public health claims assume people eat 50%+ UPF — but this study’s participants were already eating less than expected.
Practical Takeaways
If you're already eating under 25% ultra-processed foods, cutting further may not give you meaningful weight loss.
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 556 / 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
Moderate probability
on the GRADE evidence scale
This study gave two groups of people different diets and saw who lost more weight. Because they randomly picked who got which diet, we can guess that the diet caused the difference in weight loss — but the difference was tiny. So it’s not a huge win, and we can’t say it works for everyone.
No conflicts of interest were detected in this study. No score impact.
Strengths
- Explicit randomization
- Control group used
- Objective measures of energy intake and body weight
Weaknesses
- Blinding status unknown
- Full methodology not available - based on abstract only
- No reporting of attrition or adherence rates
Methodology
Evidence Keywords
Statistical Reporting
Not medical advice. For informational purposes only. Always consult a healthcare professional. Terms
Scientists tested if telling people to eat less junk food helps them lose more weight when they're already eating fewer calories.
Research results
People who tried to eat less junk food ate it 14% of the time (down from 21%) and lost a little more weight (82.9 kg vs. 86.3 kg) than those who just ate fewer calories.
What this means - more context
The extra weight loss was very small — less than 4 kg — and likely not meaningful for most people's health.
To evaluate the effectiveness and metabolic effects of restricting ultra-processed food (UPF) consumption in obese individuals undergoing energy restriction.
In a 12-month randomized trial, participants assigned to UPF restriction reduced their UPF intake and NOVA-UPF score more than controls, and lost slightly more weight (82.9 kg vs. 86.3 kg, p=0.01), but the difference was statistically significant yet non-clinically significant. Baseline UPF intake was low (~22%) and remained above 20% in both groups. No other metabolic outcomes changed.
Methods Used
Randomized, parallel clinical trial with 148 obese adults; two groups: generic energy restriction (ER-G) and energy restriction with UPF restriction (ER-UPF). Energy requirements were determined by calorimetry and accelerometry. Anthropometric, dietary, body composition, metabolic, and biochemical data were collected over 12 months.
Main Finding
The ER-UPF group reduced UPF intake from 21.2% to 13.9% and the NOVA-UPF score from 2.74 to 1.86 (p=0.03), while the ER-G group showed minimal change (23.7% to 20.0%, p=0.08). Weight loss was greater in ER-UPF (final mean 82.9 kg) vs. ER-G (86.3 kg; p=0.01), but described as statistically significant yet non-clinically significant.
Confidence Level
Limited - based on abstract only, full methodology not available
Study Flags
Red Flags
- •Full text not available - methodology details cannot be verified
- •Baseline UPF intake was low and remained high, limiting generalizability
- •Effect size for weight loss was statistically significant but described as non-clinically significant
No biological mechanisms were identified in this study. This may be an epidemiological, observational, or survey-based study that reports associations rather than proposing causal biological pathways.
Surprising Findings
UPF intake was already low at baseline (~22%) and barely changed in the control group.
Most public health claims assume people eat 50%+ UPF — but this study’s participants were already eating less than expected.
Practical Takeaways
If you're already eating under 25% ultra-processed foods, cutting further may not give you meaningful weight loss.
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 556 / 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
Moderate probability
on the GRADE evidence scale
This study gave two groups of people different diets and saw who lost more weight. Because they randomly picked who got which diet, we can guess that the diet caused the difference in weight loss — but the difference was tiny. So it’s not a huge win, and we can’t say it works for everyone.
No conflicts of interest were detected in this study. No score impact.
Strengths
- Explicit randomization
- Control group used
- Objective measures of energy intake and body weight
Weaknesses
- Blinding status unknown
- Full methodology not available - based on abstract only
- No reporting of attrition or adherence rates
Methodology
Evidence Keywords
Statistical Reporting
Scoring
How strong is this study?
The study did a good job by randomly assigning people to diets, which helps make it fair. But we don’t know if people knew which diet they were on, and we don’t have all the details — so we can’t be totally sure the results are perfect. That’s why we should be careful not to overhype it.
0 / 100
- COI disclosureconflicts of interest not disclosed
- Data availabilitydata not shared
- Code availabilitycode not shared
69 / 100
- Randomization+20/20
- Blindingblinding unclear
- Control group+15/15
- Sample size (n=148)+10.5/20
- Follow-up+10/10
100 / 100
46 / 100
- P-values+15/15
- Effect sizeno effect size reported
- Confidence intervals+15/15
- Pre-registrationnot pre-registered
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 556 / 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 was explicitly stated, allowing causal inference, but lack of blinding information and abstract-only access limit confidence in effect size and control of confounders.
No Conflicts
No conflicts of interest identified
No conflicts of interest or funding information were disclosed in the provided text.
The study lacks any declared funding sources or conflict of interest disclosures. While the methodology appears rigorous, the absence of transparency about funding or author affiliations limits the ability to fully assess potential bias.