Study analysis · JMIR mHealth and uHealth · 2024
This fasting app made people lose weight—but also lost them muscle, hunger, and energy.
A weekly fasting app helped people lose a little weight, but they lost muscle, felt hungrier, and didn’t burn fat.
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 test where half the people got a special app and free meals to fast one day a week, and the other half just got regular health tips. The group with the app lost a little weight, so we can say this specific program probably helped. But we can't say all fasting diets work the same way.
What’s the bottom line?
People used an app that gave them a low-calorie meal once a week to fast, while others just got health tips. After 12 weeks, the fasting group lost a little weight.
How strong is this study?
The study did a good job by randomly assigning people and comparing groups, which makes it trustworthy. But since people knew which group they were in and the group was small, we can't be 100% sure the weight loss was only because of the app — maybe people just tried harder because they knew they were being watched.
40 / 100
- COI disclosure+40/40
- Data availabilitydata not shared
- Code availabilitycode not shared
62 / 100
- Randomization+20/20
- Blindingnot blinded
- Control group+15/15
- Sample size (n=57)+5.0/20
- 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 575 / 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 control group allow causal inference for weight loss, but lack of blinding and small sample size introduce potential bias. The effect is modest and may be influenced by unmeasured factors like diet on non-fasting days.
Moderate COI
Moderate conflicts that may influence study outcomes
The study was administered by researchers affiliated with institutions collaborating with private companies, and a commercial app and meal products were used, raising concerns about industry influence despite no explicit funding disclosure.
Conflict Details
Healthtech Laboratory [HTL], Inc: The Kenko-Nikki app used in the intervention is developed by Healthtech Laboratory, Inc, but no author affiliation with this company is disclosed.
Cold Storage Japan Inc: One recruitment site is Cold Storage Japan Inc, a company that may benefit from health promotion initiatives among employees, but no financial ties to authors are disclosed.
Buddy Training Co, Ltd: One recruitment site is Buddy Training Co, Ltd, a fitness company that may have interest in weight loss outcomes, but no author affiliation or funding link is disclosed.
Zoom Video Communications, Inc: Zoom is used for online guidance, but no financial relationship between authors and Zoom is disclosed.
Independent Analysis Safeguards
- Outcome assessors and statistician analyst were blinded
- Randomization was conducted by a separate researcher (KK) via computer-generated sequences
No conflict of interest statement or funding disclosure is present. The use of commercial products (meal replacements, app) and recruitment from corporate sites introduces potential industry influence. However, no author is disclosed as employed by or financially tied to these entities. The lack of funding transparency and industry-linked tools raises moderate concern about potential bias, though direct funder control cannot be confirmed.
Key takeaways
- 01
The fasting group lost 1.6 kg more than the other group, but also lost 1.3 kg of muscle, felt hungrier, more tired, and had stronger food cravings.
- 02
Losing 1.6 kg is modest — about the weight of a large bag of sugar — but losing muscle while not losing fat may not be healthy long-term.
Surprising findings
- Participants lost 1.3 kg of muscle without losing fat.Intermittent fasting is often marketed as ‘preserving muscle’—but here, muscle vanished while fat stayed the same, contradicting popular claims.
- Control group gained weight while intervention group lost it—yet both groups reported similar numbers of hunger events overall.The control group reported more total hunger events (43 vs 27) and cravings (70 vs 42), suggesting the intervention group’s hunger was more intense but less frequent.
Practical takeaways
If you try a weekly fasting app, pair it with 2–3 strength training sessions per week to protect muscle mass.
This study didn’t measure exercise, so muscle loss may have been avoidable with movement—but the app didn’t include it.
medium confidenceTrack body composition, not just weight—use a scale that measures muscle and fat, not just total pounds.
Most consumer scales aren’t perfectly accurate, but trends over time matter more than exact numbers.
high confidenceAvoid fasting programs that don’t track or recommend protein intake or resistance training.
This study used meal replacements with only 13% protein—likely too low to preserve muscle.
medium confidenceWhy this study matters
Weight Loss Without Fat Loss
The intervention group lost 1.6 kg more than the control group, but fat mass didn’t change significantly (–0.1 kg, 95% CI –1.3 to 1.4 kg). Instead, they lost 1.3 kg of muscle mass.
Most people think weight loss equals fat loss—but this shows you can lose weight by losing muscle, which is unhealthy long-term and slows metabolism.
79% Adherence, But at What Cost?
79% of participants stuck with the weekly fasting plan (22 out of 28), yet reported significantly more hunger (89% vs 11%), fatigue (64% vs 22%), and cravings (60% vs 11%).
High adherence sounds great—but if people feel miserable, it’s not sustainable. This challenges the idea that ‘easy’ diets are better.
No Metabolic Gains—Just Weight Shift
Despite weight loss, there were no improvements in HbA1c, cholesterol, triglycerides, or quality of life. Muscle loss even lowered basal metabolic rate by 37.8 kcal/day.
You can lose weight without getting healthier. This flips the script on ‘fasting = detox’ claims popular on social media.
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 used an app that gave them a low-calorie meal once a week to fast, while others just got health tips. After 12 weeks, the fasting group lost a little weight.
Research results
The fasting group lost 1.6 kg more than the other group, but also lost 1.3 kg of muscle, felt hungrier, more tired, and had stronger food cravings.
What this means - more context
Losing 1.6 kg is modest — about the weight of a large bag of sugar — but losing muscle while not losing fat may not be healthy long-term.
This pilot randomized controlled trial assessed whether a 12-week web-based intermittent fasting program with weekly meal replacements could promote weight loss in Japanese adults with overweight or obesity.
The intervention group lost 1.6 kg more than the control group on average, with 79% adherence and no serious adverse events, but also experienced greater hunger, fatigue, and cravings, and lost muscle mass without significant fat loss or metabolic improvements.
Methods Used
12-week, 1:1 randomized controlled trial with 57 Japanese adults (BMI 23–<35); intervention group received weekly 407-kcal meal replacements, Zoom guidance, and app messaging; control group received only health-focused app messages; outcomes measured via intention-to-treat analysis.
Main Finding
The web-based intermittent fasting program resulted in a statistically significant adjusted mean weight loss of 1.6 kg (95% CI –2.5 to –0.8) compared to minimal care, but also led to a mean muscle mass reduction of 1.3 kg (95% CI –2.5 to –0.2) without significant fat mass loss or changes in biomarkers or quality of life.
Confidence Level
Moderate — RCT design with intention-to-treat analysis and pre-registered protocol support reliability, but small sample size, nonblinded intervention, and potential bias from missing data imputation limit certainty.
Study Flags
Red Flags
- •Small sample size (n=57)
- •Nonblinded intervention and outcome assessment
- •Muscle loss without fat loss and no dietary intake data
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
Participants lost 1.3 kg of muscle without losing fat.
Intermittent fasting is often marketed as ‘preserving muscle’—but here, muscle vanished while fat stayed the same, contradicting popular claims.
Practical Takeaways
If you try a weekly fasting app, pair it with 2–3 strength training sessions per week to protect muscle mass.
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 575 / 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 test where half the people got a special app and free meals to fast one day a week, and the other half just got regular health tips. The group with the app lost a little weight, so we can say this specific program probably helped. But we can't say all fasting diets work the same way.
Moderate conflicts detected — such as industry funding with partial involvement. A meaningful penalty has been applied.
Strengths
- Randomized controlled trial design with intention-to-treat analysis
- Clear primary outcome (weight change) with statistical adjustment for confounders
- Low dropout rate (11%) for a web-based intervention
Weaknesses
- No blinding of participants or intervention provider
- Small sample size (n=57) limits precision and power
- Missing data imputed using BOCF, which may bias results
Methodology
Evidence Keywords
Statistical Reporting
Not medical advice. For informational purposes only. Always consult a healthcare professional. Terms
People used an app that gave them a low-calorie meal once a week to fast, while others just got health tips. After 12 weeks, the fasting group lost a little weight.
Research results
The fasting group lost 1.6 kg more than the other group, but also lost 1.3 kg of muscle, felt hungrier, more tired, and had stronger food cravings.
What this means - more context
Losing 1.6 kg is modest — about the weight of a large bag of sugar — but losing muscle while not losing fat may not be healthy long-term.
This pilot randomized controlled trial assessed whether a 12-week web-based intermittent fasting program with weekly meal replacements could promote weight loss in Japanese adults with overweight or obesity.
The intervention group lost 1.6 kg more than the control group on average, with 79% adherence and no serious adverse events, but also experienced greater hunger, fatigue, and cravings, and lost muscle mass without significant fat loss or metabolic improvements.
Methods Used
12-week, 1:1 randomized controlled trial with 57 Japanese adults (BMI 23–<35); intervention group received weekly 407-kcal meal replacements, Zoom guidance, and app messaging; control group received only health-focused app messages; outcomes measured via intention-to-treat analysis.
Main Finding
The web-based intermittent fasting program resulted in a statistically significant adjusted mean weight loss of 1.6 kg (95% CI –2.5 to –0.8) compared to minimal care, but also led to a mean muscle mass reduction of 1.3 kg (95% CI –2.5 to –0.2) without significant fat mass loss or changes in biomarkers or quality of life.
Confidence Level
Moderate — RCT design with intention-to-treat analysis and pre-registered protocol support reliability, but small sample size, nonblinded intervention, and potential bias from missing data imputation limit certainty.
Study Flags
Red Flags
- •Small sample size (n=57)
- •Nonblinded intervention and outcome assessment
- •Muscle loss without fat loss and no dietary intake data
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
Participants lost 1.3 kg of muscle without losing fat.
Intermittent fasting is often marketed as ‘preserving muscle’—but here, muscle vanished while fat stayed the same, contradicting popular claims.
Practical Takeaways
If you try a weekly fasting app, pair it with 2–3 strength training sessions per week to protect muscle mass.
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 575 / 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 test where half the people got a special app and free meals to fast one day a week, and the other half just got regular health tips. The group with the app lost a little weight, so we can say this specific program probably helped. But we can't say all fasting diets work the same way.
Moderate conflicts detected — such as industry funding with partial involvement. A meaningful penalty has been applied.
Strengths
- Randomized controlled trial design with intention-to-treat analysis
- Clear primary outcome (weight change) with statistical adjustment for confounders
- Low dropout rate (11%) for a web-based intervention
Weaknesses
- No blinding of participants or intervention provider
- Small sample size (n=57) limits precision and power
- Missing data imputed using BOCF, which may bias results
Methodology
Evidence Keywords
Statistical Reporting
Scoring
How strong is this study?
The study did a good job by randomly assigning people and comparing groups, which makes it trustworthy. But since people knew which group they were in and the group was small, we can't be 100% sure the weight loss was only because of the app — maybe people just tried harder because they knew they were being watched.
40 / 100
- COI disclosure+40/40
- Data availabilitydata not shared
- Code availabilitycode not shared
62 / 100
- Randomization+20/20
- Blindingnot blinded
- Control group+15/15
- Sample size (n=57)+5.0/20
- 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 575 / 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 control group allow causal inference for weight loss, but lack of blinding and small sample size introduce potential bias. The effect is modest and may be influenced by unmeasured factors like diet on non-fasting days.
Moderate COI
Moderate conflicts that may influence study outcomes
The study was administered by researchers affiliated with institutions collaborating with private companies, and a commercial app and meal products were used, raising concerns about industry influence despite no explicit funding disclosure.
Conflict Details
Healthtech Laboratory [HTL], Inc: The Kenko-Nikki app used in the intervention is developed by Healthtech Laboratory, Inc, but no author affiliation with this company is disclosed.
Cold Storage Japan Inc: One recruitment site is Cold Storage Japan Inc, a company that may benefit from health promotion initiatives among employees, but no financial ties to authors are disclosed.
Buddy Training Co, Ltd: One recruitment site is Buddy Training Co, Ltd, a fitness company that may have interest in weight loss outcomes, but no author affiliation or funding link is disclosed.
Zoom Video Communications, Inc: Zoom is used for online guidance, but no financial relationship between authors and Zoom is disclosed.
Independent Analysis Safeguards
- Outcome assessors and statistician analyst were blinded
- Randomization was conducted by a separate researcher (KK) via computer-generated sequences
No conflict of interest statement or funding disclosure is present. The use of commercial products (meal replacements, app) and recruitment from corporate sites introduces potential industry influence. However, no author is disclosed as employed by or financially tied to these entities. The lack of funding transparency and industry-linked tools raises moderate concern about potential bias, though direct funder control cannot be confirmed.