Study analysis · Annals of internal medicine · 2025
This one weird diet trick lost people 3kg more than daily calorie cutting — and they stuck with it longer.
People who ate very little on 3 days a week and ate normally the other 4 days lost nearly 3kg more over a year than people who ate a little less every day.
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 where you eat less 3 days a week, and one where you eat less every day. It found that the 3-days-a-week diet helped people lose a little more weight, but only because everyone got lots of help and coaching. It doesn't prove that one diet is better for everyone, just for people in this special program.
What’s the bottom line?
People who ate very little on 3 random days a week and ate normally the other 4 days lost more weight than people who ate a little less every single day — even though both groups were supposed to cut the same total calories per week.
How strong is this study?
This study did a good job by randomly assigning people to diets and measuring weight accurately, which makes the results trustworthy. But it didn't hide which diet people were on, and most participants were women — so we can't be sure it would work the same for men or people without a coach. That's why we should be careful not to say it works for everyone.
0 / 100
- COI disclosureconflicts of interest not disclosed
- Data availabilitydata not shared
- Code availabilitycode not shared
70 / 100
- Randomization+20/20
- Blindingnot blinded
- Control group+15/15
- Sample size (n=165)+11.2/20
- 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 570 / 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 and intention-to-treat analysis support causal inference, but lack of blinding of participants and staff introduces potential performance bias, and the study was conducted within a highly structured behavioral program, limiting generalizability to real-world settings.
No Conflicts
No conflicts of interest identified
No conflicts of interest identified; study was funded by the National Institutes of Health with no involvement in design, analysis, or publication.
Funders
Conflict Details
National Institutes of Health: Funding source for the study (grants R01 DK111622, R01 DK111622-02S1, P30 DK048520, UL1 TR002535, F32 DK122652)
Independent Analysis Safeguards
- Core laboratory staff analyzing blood samples, doubly labeled water, and diet diaries were blinded to randomized assignment
- Statistical analysis performed by an author (Z.P.) using SAS, with no indication of funder influence
- Independent safety officer reviewed adverse events and trial progress annually
- Study protocol approved by institutional review board (COMIRB)
The study was publicly registered, followed rigorous methodological controls, and explicitly stated funder non-involvement in all aspects of the research. No industry ties or financial conflicts of interest were disclosed among authors.
Key takeaways
- 01
The 3-days-a-week group lost 2.89 kg more than the daily-restriction group; 58% lost at least 5% of their body weight vs.
- 02
47% in the daily group; fewer people quit the 3-day plan (19% vs.
- 03
29.6%).
- 04
Losing nearly 3 kg more over a year is meaningful — it’s like losing a small suitcase of weight, and more people stuck with the plan, which matters for long-term success.
Surprising findings
- Both groups fell short of their targeted 34.3% weekly energy deficit, yet the 4:3 group still lost significantly more weight.Common belief: if you don’t hit your calorie target, you won’t lose weight. But here, even under-targeted, the 4:3 group outperformed — suggesting behavioral adherence matters more than perfect math.
- The 4:3 group showed greater improvement in binge eating scores than the daily restriction group.You’d expect daily restriction to reduce binge eating by controlling intake — but the opposite happened. Periodic fasting may reduce the psychological pressure that triggers binges.
- The 4:3 group had higher caloric restriction than the daily group — even though they only counted calories on 3 days.People assume daily tracking leads to better control. But the 4:3 group, tracking only 3 days, ended up eating less overall — suggesting psychological relief on non-fast days improved overall compliance.
Practical takeaways
Try 4:3 intermittent fasting: restrict calories to 20% of normal on 3 nonconsecutive days per week, and eat normally (but mindfully) on the other 4 days.
This worked best with behavioral support and calorie tracking on fast days — doing it alone without structure may not yield the same results.
high confidenceIf you’ve quit diets before because they felt too restrictive, try a 4:3 approach — the freedom on 4 days may make it easier to stick with long-term.
This study used a medically supervised program; if you have diabetes, eating disorders, or other health conditions, consult a doctor first.
medium confidenceTrack calories only on your fast days — this reduces mental fatigue and may improve long-term adherence compared to daily logging.
Without any tracking at all, you may overeat on non-fast days. Start with tracking fast days, then gradually reduce reliance.
high confidenceWhy this study matters
3kg More Weight Loss — With Less Effort
The 4:3 intermittent fasting group lost 2.89kg more than the daily calorie restriction group over 12 months, despite both groups being targeted for the same weekly calorie deficit. The fasting group also achieved a 9.75% average caloric reduction versus just 5.64% in the daily group.
You don’t need to count calories every single day to lose more weight — just 3 days of strict restriction per week can outperform daily moderation, making it easier to stick with long-term.
Fewer People Quit — The Real Secret to Weight Loss
Attrition was 19% in the 4:3 group versus 29.6% in the daily restriction group — meaning nearly 1 in 5 people stuck with intermittent fasting for a full year, while nearly 1 in 3 dropped out of daily dieting.
Weight loss isn’t about how much you lose — it’s about who stays in the game. This study proves adherence is the hidden key to success.
You Can Eat Normally 4 Days a Week — And Still Lose Weight
On non-fast days, the 4:3 group ate ad libitum — no calorie counting, no restrictions — yet still lost more weight than the group eating less every day. Their average intake on non-fast days was still lower than baseline due to behavioral coaching.
This shatters the myth that you must restrict every day to lose weight. Freedom on 4 days makes the 3 days of restriction feel sustainable, not punishing.
It Didn’t Kill Your Workout — And You Might Even Move More
Both groups increased moderate-to-vigorous physical activity by the same amount (300 mins/week target), proving that 4:3 fasting didn’t sap energy or motivation to exercise.
People fear fasting will make them tired or weak — but this study shows you can still hit your gym goals and even improve adherence.
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
People who ate very little on 3 random days a week and ate normally the other 4 days lost more weight than people who ate a little less every single day — even though both groups were supposed to cut the same total calories per week.
Research results
The 3-days-a-week group lost 2.89 kg more than the daily-restriction group; 58% lost at least 5% of their body weight vs. 47% in the daily group; fewer people quit the 3-day plan (19% vs. 29.6%).
What this means - more context
Losing nearly 3 kg more over a year is meaningful — it’s like losing a small suitcase of weight, and more people stuck with the plan, which matters for long-term success.
To compare 4:3 intermittent fasting (80% energy restriction on 3 nonconsecutive days/week) with daily caloric restriction (34% daily deficit) for weight loss over 12 months in adults with overweight or obesity, both within a high-intensity behavioral program.
In a 12-month randomized trial, 4:3 intermittent fasting led to 2.89 kg greater weight loss than daily caloric restriction, with higher adherence (lower attrition: 19% vs. 29.6%), greater caloric restriction (−9.75% vs. −5.64%), and improved binge eating scores, without reducing physical activity levels.
Methods Used
165 adults (BMI 27–46 kg/m², mean age 42, 73.9% female) were randomized 1:1 to 4:3 intermittent fasting or daily caloric restriction, both receiving identical high-intensity behavioral support, group counseling, and physical activity guidance; primary outcome was weight change at 12 months using intention-to-treat analysis.
Main Finding
4:3 intermittent fasting resulted in 2.89 kg greater weight loss at 12 months compared to daily caloric restriction (95% CI, 5.65 to 0.14 kg; P = 0.040), with 58% achieving ≥5% weight loss vs. 47% in the daily restriction group.
Confidence Level
High confidence due to randomized controlled trial design, intention-to-treat analysis, pre-registered protocol, reported effect sizes and confidence intervals, and objective dietary and weight measurements.
Study Flags
Red Flags
- •Limited generalizability due to recruitment from one metropolitan area
- •No blinding of participants or staff due to intervention nature
- •Both groups fell short of targeted 34.3% weekly energy deficit
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
Both groups fell short of their targeted 34.3% weekly energy deficit, yet the 4:3 group still lost significantly more weight.
Common belief: if you don’t hit your calorie target, you won’t lose weight. But here, even under-targeted, the 4:3 group outperformed — suggesting behavioral adherence matters more than perfect math.
Practical Takeaways
Try 4:3 intermittent fasting: restrict calories to 20% of normal on 3 nonconsecutive days per week, and eat normally (but mindfully) on the other 4 days.
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 570 / 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 is like a fair race between two diets — one where you eat less 3 days a week, and one where you eat less every day. It found that the 3-days-a-week diet helped people lose a little more weight, but only because everyone got lots of help and coaching. It doesn't prove that one diet is better for everyone, just for people in this special program.
No conflicts of interest were detected in this study. No score impact.
Strengths
- Randomized controlled trial with block randomization and stratification by sex
- Intention-to-treat analysis performed
- Objective outcome measures (e.g., dual-energy X-ray absorptiometry, doubly labeled water)
Weaknesses
- No blinding of participants or intervention staff (high risk of performance bias)
- 24% attrition rate with differential dropout between groups
- Limited generalizability due to homogenous sample (mostly female, low racial diversity)
Methodology
Evidence Keywords
Statistical Reporting
Not medical advice. For informational purposes only. Always consult a healthcare professional. Terms
People who ate very little on 3 random days a week and ate normally the other 4 days lost more weight than people who ate a little less every single day — even though both groups were supposed to cut the same total calories per week.
Research results
The 3-days-a-week group lost 2.89 kg more than the daily-restriction group; 58% lost at least 5% of their body weight vs. 47% in the daily group; fewer people quit the 3-day plan (19% vs. 29.6%).
What this means - more context
Losing nearly 3 kg more over a year is meaningful — it’s like losing a small suitcase of weight, and more people stuck with the plan, which matters for long-term success.
To compare 4:3 intermittent fasting (80% energy restriction on 3 nonconsecutive days/week) with daily caloric restriction (34% daily deficit) for weight loss over 12 months in adults with overweight or obesity, both within a high-intensity behavioral program.
In a 12-month randomized trial, 4:3 intermittent fasting led to 2.89 kg greater weight loss than daily caloric restriction, with higher adherence (lower attrition: 19% vs. 29.6%), greater caloric restriction (−9.75% vs. −5.64%), and improved binge eating scores, without reducing physical activity levels.
Methods Used
165 adults (BMI 27–46 kg/m², mean age 42, 73.9% female) were randomized 1:1 to 4:3 intermittent fasting or daily caloric restriction, both receiving identical high-intensity behavioral support, group counseling, and physical activity guidance; primary outcome was weight change at 12 months using intention-to-treat analysis.
Main Finding
4:3 intermittent fasting resulted in 2.89 kg greater weight loss at 12 months compared to daily caloric restriction (95% CI, 5.65 to 0.14 kg; P = 0.040), with 58% achieving ≥5% weight loss vs. 47% in the daily restriction group.
Confidence Level
High confidence due to randomized controlled trial design, intention-to-treat analysis, pre-registered protocol, reported effect sizes and confidence intervals, and objective dietary and weight measurements.
Study Flags
Red Flags
- •Limited generalizability due to recruitment from one metropolitan area
- •No blinding of participants or staff due to intervention nature
- •Both groups fell short of targeted 34.3% weekly energy deficit
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
Both groups fell short of their targeted 34.3% weekly energy deficit, yet the 4:3 group still lost significantly more weight.
Common belief: if you don’t hit your calorie target, you won’t lose weight. But here, even under-targeted, the 4:3 group outperformed — suggesting behavioral adherence matters more than perfect math.
Practical Takeaways
Try 4:3 intermittent fasting: restrict calories to 20% of normal on 3 nonconsecutive days per week, and eat normally (but mindfully) on the other 4 days.
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 570 / 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 is like a fair race between two diets — one where you eat less 3 days a week, and one where you eat less every day. It found that the 3-days-a-week diet helped people lose a little more weight, but only because everyone got lots of help and coaching. It doesn't prove that one diet is better for everyone, just for people in this special program.
No conflicts of interest were detected in this study. No score impact.
Strengths
- Randomized controlled trial with block randomization and stratification by sex
- Intention-to-treat analysis performed
- Objective outcome measures (e.g., dual-energy X-ray absorptiometry, doubly labeled water)
Weaknesses
- No blinding of participants or intervention staff (high risk of performance bias)
- 24% attrition rate with differential dropout between groups
- Limited generalizability due to homogenous sample (mostly female, low racial diversity)
Methodology
Evidence Keywords
Statistical Reporting
Scoring
How strong is this study?
This study did a good job by randomly assigning people to diets and measuring weight accurately, which makes the results trustworthy. But it didn't hide which diet people were on, and most participants were women — so we can't be sure it would work the same for men or people without a coach. That's why we should be careful not to say it works for everyone.
0 / 100
- COI disclosureconflicts of interest not disclosed
- Data availabilitydata not shared
- Code availabilitycode not shared
70 / 100
- Randomization+20/20
- Blindingnot blinded
- Control group+15/15
- Sample size (n=165)+11.2/20
- 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 570 / 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 and intention-to-treat analysis support causal inference, but lack of blinding of participants and staff introduces potential performance bias, and the study was conducted within a highly structured behavioral program, limiting generalizability to real-world settings.
No Conflicts
No conflicts of interest identified
No conflicts of interest identified; study was funded by the National Institutes of Health with no involvement in design, analysis, or publication.
Funders
Conflict Details
National Institutes of Health: Funding source for the study (grants R01 DK111622, R01 DK111622-02S1, P30 DK048520, UL1 TR002535, F32 DK122652)
Independent Analysis Safeguards
- Core laboratory staff analyzing blood samples, doubly labeled water, and diet diaries were blinded to randomized assignment
- Statistical analysis performed by an author (Z.P.) using SAS, with no indication of funder influence
- Independent safety officer reviewed adverse events and trial progress annually
- Study protocol approved by institutional review board (COMIRB)
The study was publicly registered, followed rigorous methodological controls, and explicitly stated funder non-involvement in all aspects of the research. No industry ties or financial conflicts of interest were disclosed among authors.