Effect of Alternate-Day Fasting on Weight Loss, Weight Maintenance, and Cardioprotection Among Metabolically Healthy Obese Adults

This was a head-to-head comparison of two popular weight loss strategies — alternate-day fasting (ADF) versus daily calorie restriction (CR) — against a no-intervention control group. The trial had two phases: a 6-month weight-loss phase followed by a 6-month weight-maintenance phase.

The three groups:

• Alternate-day fasting (ADF): Participants alternated between "fast days" where they consumed 25% of their daily energy needs (roughly 500–600 calories for most people) and "feast days" where they consumed 125% of their daily energy needs. This created an average daily deficit of about 25% of energy needs across the week.

• Daily calorie restriction (CR): Participants consumed 75% of their daily energy needs every day — no fasting days, no feast days. This also created an average daily deficit of about 25% of energy needs.

• Control: Participants received no dietary intervention and were asked to maintain their usual eating habits.

Primary outcome: Change in body weight (percentage of initial body weight lost).

Secondary outcomes: Adherence to the prescribed diet (measured via self-reported food diaries and urinary ketones), plus a panel of cardiovascular risk indicators including blood pressure, heart rate, triglycerides, fasting glucose, fasting insulin, insulin resistance (HOMA-IR), C-reactive protein (CRP, a marker of inflammation), homocysteine, HDL cholesterol ("good" cholesterol), and LDL cholesterol ("bad" cholesterol).

• Sample size: 100 participants (86 women, 14 men)
• Age: Mean 44 years (standard deviation 11 years; range 18–64)
• Body mass index (BMI): Mean 34 (obese range; 30+ is considered obese)
• Health status: "Metabolically healthy obese" — meaning they were obese but did not have diagnosed diabetes, uncontrolled hypertension, or other major metabolic diseases
• Setting: Single academic medical center in Chicago, Illinois
• Recruitment period: October 2011 to January 2015
• Exclusions: People with diabetes, cardiovascular disease, eating disorders, or those taking medications affecting weight or metabolism

Important limitation for generalisation: 86% of participants were women, and all were from one urban area. Results may not apply equally to men or to people with existing metabolic conditions.

• Body weight: Measured on a calibrated scale at baseline, then monthly for 12 months. Percentage change was calculated relative to starting weight.
• Dietary adherence: Participants completed 7-day food diaries at baseline, month 3, month 6, month 9, and month 12. These were analysed by a registered dietitian using nutrition software (Nutrition Data System for Research). Adherence was defined as how closely actual intake matched prescribed intake.
• Urinary ketones: Measured at each visit using dipsticks (a positive result indicates the body is in ketosis, which can occur during fasting).
• Cardiovascular risk markers: Blood samples were taken after a 12-hour fast at baseline, month 6, and month 12. These were analysed for lipids (HDL, LDL, triglycerides), glucose, insulin, CRP, and homocysteine. Blood pressure and heart rate were measured using an automated monitor after 5 minutes of seated rest.
• Insulin resistance: Calculated using the HOMA-IR formula (fasting insulin × fasting glucose / 405). Higher values indicate greater insulin resistance.

Study design: This was a single-center, parallel-group, randomized clinical trial (RCT). Participants were randomly assigned to one of three groups using a computer-generated randomisation sequence with a 1:1:1 allocation ratio.

Randomisation: Yes, participants were randomly assigned. The randomisation sequence was generated by a statistician not involved in the study, and allocation was concealed using sealed envelopes. This is important because it prevents selection bias — the researchers could not influence which group a participant ended up in.

Blinding: This was an open-label trial — neither participants nor researchers were blinded to group assignment. This is a major methodological limitation. Participants knew whether they were fasting or restricting calories daily, and researchers knew which group each participant was in. This can introduce bias in several ways:

• Participants in the ADF group may have had different expectations about results, affecting their motivation and self-reported adherence.
• Researchers may have unconsciously treated groups differently during counseling sessions.
• The control group knew they were not receiving an active intervention, which could affect their behavior.

Duration: 12 months total — 6 months of active weight loss followed by 6 months of weight maintenance. This is a relatively long duration for a diet trial, which is a strength. Many fasting studies last only 8–12 weeks.

Statistical approach: The primary analysis used intention-to-treat (ITT) — meaning all participants were analysed in the group they were assigned to, regardless of whether they completed the study. Missing data were handled using multiple imputation. This is the gold standard for RCTs because it preserves the benefits of randomisation and avoids bias from dropout.

What this design can and cannot prove:

• Can prove: That the two dietary strategies produce different (or similar) average weight loss and changes in cardiovascular markers over 12 months in this specific population. The randomisation means that differences between groups at the end of the study can be attributed to the interventions (assuming no major confounds).

• Cannot prove: Why the interventions worked or didn't work (e.g., whether ADF has unique metabolic effects beyond calorie restriction). The design cannot separate the effects of calorie restriction from the effects of the fasting pattern itself, because the ADF group was also in calorie deficit. The design also cannot prove long-term effects beyond 12 months, or effects in people with existing metabolic disease.

• Cannot prove individual-level effects: RCTs tell us about group averages. Your personal response to ADF might be very different from the average.

Major methodological weaknesses:

1. No blinding — as noted above, this is a significant source of potential bias.
2. High dropout rate — 38% in the ADF group, 29% in CR, 26% in control. If the people who dropped out had different outcomes than those who stayed, the results could be misleading.
3. Self-reported dietary intake — food diaries are notoriously inaccurate. People underreport calories by 20–50% on average, and this may differ between groups.
4. Single center, mostly women — limits generalisability.
5. No measure of physical activity — changes in activity could have confounded results.

Primary outcome — Weight loss:

• At 6 months (end of weight-loss phase):

  • ADF group lost 6.8% of initial body weight (95% CI: -9.1% to -4.5%)
  • CR group lost 6.8% of initial body weight (95% CI: -9.1% to -4.6%)
  • Control group lost 0.2% (not significant)
  • Difference between ADF and CR: 0.0% (not significant)

• At 12 months (end of weight-maintenance phase):

  • ADF group lost 6.0% of initial body weight (95% CI: -8.5% to -3.6%)
  • CR group lost 5.3% of initial body weight (95% CI: -7.6% to -3.0%)
  • Control group gained 0.3% (not significant)
  • Difference between ADF and CR: 0.7% (not significant)

Key interpretation: Both diets produced clinically meaningful weight loss (5% or more is considered significant for health benefits), but ADF was not superior to daily calorie restriction at either time point.

Secondary outcomes — Adherence:

• ADF participants ate more than prescribed on fast days (mean intake ~35% of energy needs instead of the prescribed 25%) and less than prescribed on feast days (mean intake ~110% instead of 125%).
• CR participants generally met their prescribed energy goals (mean intake ~78% of energy needs vs prescribed 75%).
• Urinary ketones were positive in the ADF group on fast days (confirming they were in a fasted state) but not on feast days.
• Dropout rate was highest in ADF (38% vs 29% for CR vs 26% for control), suggesting ADF was harder to sustain.

Secondary outcomes — Cardiovascular risk markers (at 6 and 12 months):

• Blood pressure, heart rate, triglycerides, fasting glucose, fasting insulin, HOMA-IR, CRP, and homocysteine: No significant differences between ADF and CR at either 6 or 12 months.
• HDL cholesterol ("good" cholesterol):
  • At 6 months: ADF group had a 6.2 mg/dL increase compared to CR (95% CI: 0.1 to 12.4 mg/dL; p < 0.05). This was statistically significant.
  • At 12 months: ADF group had only a 1.0 mg/dL increase compared to CR (95% CI: -5.9 to 7.8 mg/dL; not significant). The early benefit disappeared.
• LDL cholesterol ("bad" cholesterol):
  • At 6 months: No significant difference between groups.
  • At 12 months: ADF group had an 11.5 mg/dL increase compared to CR (95% CI: 1.9 to 21.1 mg/dL; p < 0.05). This was statistically significant and clinically concerning — a rise in LDL of this magnitude is associated with increased cardiovascular risk.

Key interpretation: ADF did not provide any cardiovascular benefits beyond daily calorie restriction. The early HDL improvement was not sustained, and LDL actually worsened at 12 months.

• Weight loss: Both diets produced about a 6–7% reduction in body weight. For a 200-pound person, that's 12–14 pounds. This is a moderate effect — enough to improve health markers in most people, but not dramatic weight loss.

• HDL increase at 6 months: 6.2 mg/dL is a modest increase. For context, a typical HDL range is 40–60 mg/dL, so this represents about a 10–15% improvement. However, it was not sustained.

• LDL increase at 12 months: 11.5 mg/dL is a moderate increase. For context, a typical LDL range is 100–130 mg/dL, so this represents about a 9–11% increase. This is enough to offset some of the cardiovascular benefits of weight loss.

• Dropout rate: 38% in ADF means that nearly 4 out of 10 people assigned to alternate-day fasting dropped out before completing the study. This is a high rate and suggests the regimen is difficult to maintain.

Acknowledged by authors:

• Open-label design (no blinding)
• Single center, mostly female sample
• Self-reported dietary intake (known to be inaccurate)
• High dropout rate, especially in ADF group
• No measure of physical activity or energy expenditure
• Short follow-up (12 months) for a chronic condition like obesity

Additional critical observations:

• No true "fasting" condition: The ADF group ate 25% of calories on fast days, which is not true fasting (zero calories). This means the study tested "modified alternate-day fasting" rather than water-only fasting. Results may not apply to stricter fasting protocols.
• No measure of hunger or satisfaction: The study did not report whether participants in the ADF group experienced more hunger or food cravings, which would be important for understanding why dropout was higher.
• No adjustment for multiple comparisons: The authors tested many secondary outcomes (blood pressure, heart rate, 7 different blood markers, at 2 time points each). Without statistical correction, some "significant" findings may be due to chance. The HDL finding at 6 months could be a false positive.
• Industry funding: The study was funded by the National Institutes of Health (NIH), so no direct industry conflict. However, the broader field of intermittent fasting has commercial interests that may influence public perception.
• Population limits: "Metabolically healthy obese" adults — results may not apply to people with diabetes, prediabetes, or other metabolic conditions. The study also excluded people over 64, so results may not apply to older adults.

For someone running their own n=1 experiment:

What to test

• Intervention: Alternate-day fasting with a "modified" approach — consume ~25% of normal calories (about 500–600 kcal) on fast days, and eat normally (or up to 125% of normal) on feast days. Alternatively, test true zero-calorie fasting on fast days if you can tolerate it.
• Comparator: Daily calorie restriction at ~75% of normal intake (a 25% deficit every day). Or simply compare ADF to your usual eating pattern.
• Dose: The study used a 25% average daily deficit. For most people, this means:
  • Fast days: ~500–600 calories
  • Feast days: ~2,000–2,500 calories (depending on your size and activity)
  • Daily restriction: ~1,500–1,800 calories every day

Minimum meaningful duration

• At least 3 months to see initial weight loss effects
• 6 months is better to see if the pattern is sustainable
• 12 months is ideal to assess long-term adherence and cardiovascular effects (especially LDL changes)

The study showed that early benefits (like HDL increase) may disappear by 12 months, so short experiments may give misleading results.

What to measure (specific metrics)

• Body weight: Weigh yourself at the same time each morning, after using the bathroom, before eating or drinking. Use a consistent scale. Track percentage change from baseline.
• Waist circumference: Measure at the narrowest point (or at the navel if you don't have a narrow waist) weekly. This captures fat loss better than weight alone.
• Blood lipids: Get a fasting lipid panel (HDL, LDL, triglycerides) at baseline, 3 months, and 6 months. The study showed LDL can rise at 12 months, so longer follow-up is valuable.
• Fasting glucose and insulin: If you have access, these are more sensitive markers of metabolic health than weight alone.
• Adherence tracking: Keep a daily log of what you ate on fast days and feast days. The study found that people on ADF ate more than prescribed on fast days — this is a common pitfall.
• Subjective experience: Rate your hunger, energy, mood, and sleep quality daily on a 1–10 scale. This helps you understand whether the intervention is sustainable.

Key confounds to control for

• Calorie intake: The study showed that ADF works only if you actually restrict calories on fast days. If you overeat on feast days, you won't lose weight. Track your intake for at least the first 2 weeks to calibrate.
• Physical activity: Keep your exercise routine constant during the experiment. Changes in activity can dramatically affect weight loss and confound your results.
• Sleep: Poor sleep increases hunger hormones and can cause weight gain. Track your sleep duration and quality.
• Menstrual cycle (for women): Weight fluctuates 2–5 pounds during the menstrual cycle due to water retention. Track your cycle and compare weights at the same phase.
• Stress: Chronic stress raises cortisol, which promotes fat storage. Note major stressors in your log.
• Hydration: Dehydration can cause weight fluctuations. Drink consistent amounts of water each day.
• Timing of measurements: Always measure