Alternate Day Fasting, Exercise, and NAFLD: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Alternate Day Fasting, Exercise, and NAFLD (NCT04004403), a randomized na study. The primary outcome is Change in Hepatic Steatosis.

2. Clinical Question

3. Trial Design

The registry record describes a randomized, single-masked, parallel trial sponsored by the University of Illinois Chicago. It lists an enrollment of 80 and a completed status. The study start is recorded as 2019-09-01 and primary completion as 2024-05-01. The trial is listed as phase NA.

Four groups are described: alternate day fasting, exercise, the combination of alternate day fasting plus exercise, and control. The fasting regimen involves consuming 600 kcal on a fast day and eating ad libitum at home on alternating feed days. The exercise program is supervised aerobic exercise 5 times per week, 40-60 min per session, at 60-85% HRmax. The combination group follows both regimens. Controls are instructed to maintain weight and avoid changes to eating and physical activity habits.

Random allocation can help balance measured and unmeasured characteristics across groups, but the registry record does not report the allocation procedure or whether allocation was concealed. It identifies masking as single, without specifying who was masked. Further details about implementation, adherence, or treatment fidelity are not reported in the registry record.

The primary outcome is change in hepatic steatosis from week 1 to week 12. The posted result gives a mean and 95% confidence interval for each arm, with 20 participants listed for each arm's primary outcome. Although the trial has four randomized groups, the record does not provide numerical two-arm comparisons for this continuous endpoint. As a result, the arm-specific changes should not be read as estimates of the randomized treatment contrasts.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

Analysis population for the primary outcome, as defined in the registry: Adults with obesity and NAFLD

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

Population: Adults with obesity and non-alcoholic fatty liver disease

6. Primary Endpoint

The registry lists the primary outcome as Change in Hepatic Steatosis, measured over Change from week 1 to week 12.

Hepatic steatosis means fat accumulation in the liver. The primary endpoint is the change in hepatic steatosis from week 1 to week 12, expressed as liver fat percentage and measured by magnetic resonance imaging using MRI-PDFF. A change endpoint describes how the measured value shifted between the stated assessment times, rather than reporting only the value at the later assessment.

In general, an imaging measure that quantifies liver fat can address the study question more directly than a measure of a related metabolic feature. The registry record identifies MRI-PDFF as the measurement method but does not report technical details about image acquisition, reading, or quality control.

Negative mean changes indicate a decrease in liver fat percentage over the measured period; positive changes indicate an increase. The reported means are accompanied by 95% confidence intervals, which describe uncertainty around the estimated mean under the analysis assumptions. The registry record does not give a between-group estimate for the primary endpoint, so the arm-specific changes do not by themselves quantify how much the interventions differ from control.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a mean change represent?

The primary outcome is continuous, and the registry record describes the summary measure as a mean. A mean change represents the average difference between the recorded value at week 1 and the value at week 12, within an arm. It is a summary of the group, not a description of every participant's change. Individuals may have changes that differ in direction and magnitude from the reported mean.

The sign conveys direction on the recorded scale. A negative value is a decrease in liver fat percentage over the assessment period, whereas a positive value is an increase. The mean alone does not describe the spread of participant-level changes. No standard deviation or other dispersion measure is posted; the listed uncertainty measure is a 95% confidence interval.

How should a 95% confidence interval be read?

In general, a confidence interval expresses the uncertainty around an estimated quantity under the statistical procedure and its assumptions. A narrower interval indicates greater precision than a wider interval on the same scale, while a wide interval leaves more uncertainty about the possible mean changes. The interval is not a range that contains every participant's result, and it should not be interpreted as a guarantee about the true value.

For the arm-specific results, the intervals are presented around mean changes. An interval that includes zero is compatible with no mean change as well as with changes in either direction at the level represented by that interval. An interval entirely below zero supports a negative mean change under the interval's assumptions. Neither reading determines whether a treatment differs from control; that requires a between-arm contrast and its uncertainty.

Why are direct comparisons important?

In a randomized trial, the main comparative question is generally addressed by contrasting outcomes between assigned groups. A contrast can be based on the difference in mean change or another prespecified measure. It uses the randomized groups to estimate how outcomes differ between treatment strategies. The posted results give arm-specific means and intervals, but the registry record does not report numerical two-arm estimates for the primary continuous endpoint. It notes that values are missing or not estimable for a two-arm comparison.

Separate within-arm changes can reflect more than the intervention itself, including the passage of time, changes common to participants, or variation in measurement. Randomization helps support a comparison between groups, but it does not make a within-group change an estimate of the treatment effect. Statistical significance or comparative benefit should not be inferred from the direction of the arm means alone.

What can be said about the statistical plan?

The registry record identifies the primary outcome, time frame, summary measure, and confidence interval, but it does not report the detailed model, covariance assumptions, adjustment variables, or test procedure used for the posted results. The analysis population and handling of missing observations are also not reported in the registry record. These omissions limit assessment of how the estimates were produced and whether the analysis followed a prespecified plan.

The record includes a sample-size planning illustration based on a fraction of the observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. Such illustrations describe hypothetical design requirements under stated assumptions; they are not observed power and do not show that the trial was designed using those assumptions. No trial-specific sample-size justification is reported in the registry record.

In general, planning estimates based on observed effects are sensitive to the estimated effect and its uncertainty. Using a fraction of that effect changes the planning question, but does not create evidence that a future or current study has a particular power. The record does not provide a protocol-based justification, interim-analysis plan, or further details of the statistical plan, so those aspects cannot be reconstructed from the posted outcomes.

Why does dispersion matter?

In general, the variation in participant responses affects how precisely a mean is estimated. When data are more variable, an estimate can be less precise unless other aspects of the study compensate. If outcomes show over-dispersion relative to a model's assumptions, a model that understates the variation can produce confidence intervals that are too narrow. Over-dispersion is a general statistical concern; the registry record does not report that it was assessed or present participant-level data to evaluate it.

The stated intervals provide some information about precision, but they do not reveal all analytic choices. Without the model description, standard errors, or participant-level data, the precision estimates cannot be independently evaluated from the registry record. The appropriate interpretation is therefore descriptive and limited to the reported arm-level means and intervals.

10. Multiplicity, Missing Data and Analysis Populations

The registry record lists seven secondary outcomes: change in body weight, alanine aminotransferase (ALT), aspartate aminotransferase (AST), fasting glucose, fasting insulin, insulin resistance, and HbA1c. Together with the primary outcome, the record identifies eight outcome measures. No analyses are posted for the secondary outcomes, and the registry record does not report a multiplicity procedure or explain how multiple outcomes or comparisons were handled.

In general, testing many outcomes or contrasts can increase the chance of finding an apparently notable result by chance. A multiplicity strategy may define which comparisons are confirmatory or adjust interpretation across tests. Whether such a strategy was planned or used is not reported in the registry record. The posted arm-specific confidence intervals for the primary endpoint do not answer this question.

Information about missing primary outcome data, reasons for missingness, imputation, and sensitivity analyses is not reported in the registry record. The record also does not state whether the displayed results use an intention-to-treat population, a per-protocol population, or another analysis set. Although 20 participants are listed for each arm's primary result, those counts alone do not establish how missing data or deviations from assigned treatment were handled.

These details matter because exclusions or missing observations can affect the comparability created by randomization. Without information on analysis populations and missing-data handling, readers cannot determine whether the posted estimates include all assigned participants or how robust they are to alternative assumptions. No conclusions about the unreported secondary outcomes or the multiplicity approach can be drawn.

11. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

The registry record reports zero serious adverse events and zero other adverse events for each of the four groups. It also reports zero deaths in each group. For each group, 20 participants are listed as at risk for serious adverse events and 20 as at risk for other adverse events. No serious adverse event terms are listed.

These figures describe what was reported in the registry record; they do not establish that adverse events could not occur or that the interventions are free of risk. A zero count can reflect the events observed and recorded under the trial's procedures, and it does not provide a precise estimate of uncommon harms. The record does not report the adverse-event collection period, event definitions, monitoring procedures, or details about severity grading and attribution.

The safety counts should be read separately from the primary outcome findings. They do not resolve the uncertainty in the liver fat estimates, nor do they substitute for longer-term safety information. The registry record gives no additional safety narrative or event terms that would clarify the circumstances behind the zero counts.

12. What the Result Does, and Does Not, Mean

The posted results describe mean change in MRI-PDFF-measured hepatic steatosis within the four trial groups over the interval from week 1 to week 12. The most negative arm-level mean is reported for the combination group, while the reported control mean is close to zero. The uncertainty intervals show that these are estimates rather than exact values, with differing degrees of precision and different relationships to zero.

These findings do not provide a direct estimate of how much the fasting, exercise, or combination strategies differ from control. The record does not report numerical two-arm comparisons for the primary continuous endpoint, and it notes that comparison values are missing or not estimable. It would therefore be inappropriate to treat the numerical difference between two arm means as a reported treatment effect or to claim a comparative effect based on the arm-specific intervals alone.

The results also do not establish a change in clinical outcomes beyond the measured liver fat percentage. The registry lists secondary outcomes related to body weight, liver enzymes, glucose, insulin, insulin resistance, and HbA1c, but no analyses for those outcomes are posted. No conclusions about those measures can be drawn from the primary endpoint results.

In general, a clinical trial estimate is interpreted in light of design, measurement, missingness, analytic methods, precision, and the comparison being addressed. Here, the available information supports a description of the posted within-arm changes and confidence intervals. It does not support broader claims about effectiveness, comparative benefit, durability beyond the measured period, or safety beyond the adverse-event counts recorded.

13. Why This Trial Matters Statistically

This trial raises a useful statistical distinction between describing change within groups and estimating differences between randomized groups. The primary results report mean change for each arm, but the central comparative question requires a between-arm estimate. That distinction is important whenever baseline-to-follow-up change is presented: a group's movement over time is not, by itself, the effect attributable to its assigned intervention.

The four-arm design includes fasting, exercise, their combination, and control. This structure allows multiple potentially relevant comparisons, but the registry record does not provide numerical estimates for those comparisons. The absence of comparison results also means that the relative contribution of fasting and exercise, and whether their combination differs from either component, cannot be evaluated from the posted primary results.

Precision is another important feature. The confidence intervals communicate uncertainty around the arm means and make clear that the estimates should not be treated as exact. The result summaries do not include participant-level variability measures or a reported model, limiting independent assessment of the uncertainty calculations.

The registry record also lists several secondary outcomes without posted analyses. Prespecifying and interpreting multiple outcomes requires clarity about which questions are primary, how comparisons are prioritized, and whether multiplicity is addressed. The record identifies a primary endpoint but does not report those additional analysis details. Taken together, this trial is statistically informative as an example of why outcome summaries, direct contrasts, and transparent analysis reporting serve different purposes.

14. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The main limitation for interpreting the primary findings is that the posted results are arm-specific changes, while numerical two-arm comparisons are not reported in the registry record. This prevents direct estimation of how each intervention compares with control. The record specifically notes that values are missing or not estimable for a two-arm comparison of the continuous endpoint.

The registry record provides limited detail about the statistical methods. It does not report the model used, the analysis population, the handling of missing observations, adjustment choices, or whether sensitivity analyses were performed. The masking designation is single, but who was masked is not specified. Allocation concealment and implementation details are not reported in the registry record.

The outcome is measured from week 1 to week 12, so the posted primary result addresses change over that stated period. Longer-term outcomes are not included in the primary result. Technical details about MRI-PDFF measurement and quality control are not reported. The listed secondary outcomes have no posted analyses, and no multiplicity strategy is described.

The adverse-event record reports zero serious events, other events, and deaths in each group, but does not give collection periods or event definitions. Those omissions limit interpretation of the safety counts. Finally, the registry record's completed status and listed enrollment do not resolve questions about adherence, outcome completion, or participant flow; those details are not reported in the registry record.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Alternate Day Fasting, Exercise, and NAFLD — NCT04004403.
• Registry record last updated 2025-09-25; status last verified 2025-09.
• Results first posted 2025-09-25.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

18. Record Summary

NCT04004403, titled “Alternate Day Fasting, Exercise, and NAFLD,” is a completed, randomized, single-masked, parallel trial sponsored by the University of Illinois Chicago. It lists adults with obesity and NAFLD as the population and an enrollment of 80. The study start is recorded as 2019-09-01 and primary completion as 2024-05-01.

The primary outcome is mean change in hepatic steatosis, measured as liver fat percentage by MRI-PDFF from week 1 to week 12. The record describes four groups: alternate day fasting, exercise, their combination, and control. Primary outcome means and 95% confidence intervals are posted for each arm, with 20 participants listed per arm. Numerical two-arm comparisons are not reported in the registry record. Seven secondary outcomes are listed without posted analyses. Safety reporting lists zero serious adverse events, other adverse events, and deaths in each group.