Cognitive Behavioral Therapy for Adolescent Binge Eating and Loss of Control Eating: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Cognitive Behavioral Therapy for Adolescent Binge Eating and Loss of Control Eating (NCT04088097), a randomized na study. The primary outcome is Binge (Loss-of-control) Eating Episode Frequency.

2. Clinical Question

The clinical question is about a between-group comparison at a specified assessment point, not simply whether participants changed over time. The primary outcome is the frequency of eating episodes, expressed as a mean and standard deviation in each group. The record lists 50 enrolled participants, but it does not report a complete account of who contributed data to each outcome analysis or why outcome counts differ from enrollment.

The study was marked completed. Its stated purpose was development and initial efficacy testing, so the registry record describes an early evaluation rather than a definitive estimate. The reported results can indicate the direction and uncertainty of the recorded group comparison, but they do not by themselves establish how well the findings would generalize to other settings or populations. Detailed eligibility criteria, recruitment setting, and participant characteristics are not reported in the registry record.

3. Trial Design

The registry describes a randomized, single-masked, parallel-group study sponsored by Yale University. Participants were assigned to cognitive-behavioral therapy or a control condition involving educational materials related to adolescent health and nutrition. The trial status is completed. The record lists enrollment of 50 and a start date of 2019-12-30, with primary completion on 2023-10-17.

The two assigned conditions are described at a broad level. The intervention is adapted cognitive-behavioral therapy for adolescents with binge eating or loss-of-control eating; the control condition is educational materials. The exact treatment schedule, session content, delivery format, and details of the masking procedure are not reported in the registry record. The record labels masking as single but does not identify which participants, staff, or assessors were masked.

The primary outcome is mean binge or loss-of-control eating episode frequency at post-treatment, with the timeframe given as 4 months. The record also lists follow-up eating episode frequency and weight as secondary outcomes. The eating measure is described as assessed by questionnaire, specifically the Eating Disorder Examination questionnaire. The follow-up outcome timeframe is described as 3 months after the end of treatment.

The primary outcome results list 21.0 participants in the cognitive-behavioral therapy group and 19.0 in the control group. The follow-up results list different group counts, and the weight outcome has its own group counts. These are outcome-specific counts; reasons for the varying numbers are not reported in the registry record. The protocol's planned sample size, original sample-size justification, allocation procedure details, and analysis populations are also not reported in the registry record.

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

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.

6. Primary Endpoint

The registry lists the primary outcome as Binge (Loss-of-control) Eating Episode Frequency, measured over Post (4 months).

The primary endpoint is binge or loss-of-control eating episode frequency at post-treatment, with the registry specifying a 4-month timeframe. It is measured as a count-like behavioral frequency but is summarized in the record as a continuous outcome: the mean and standard deviation of episodes in each group. The outcome description says episodes are assessed by questionnaire using the Eating Disorder Examination questionnaire.

Frequency is relevant to the trial's stated focus because the intervention was developed for adolescents with binge or loss-of-control eating. A group mean gives an overall summary of reported episode frequency, while the standard deviation describes how much individual observations vary around that mean. Neither summary conveys the full distribution of participants' responses, and the record does not report individual-level data or the distributional shape.

The registry reports a mean of 1.52 episodes in the cognitive-behavioral therapy group and 6.05 in the control group at the primary timepoint. The posted unadjusted comparison is a mean difference of -4.53 episodes, with the negative direction indicating a lower recorded mean in the therapy group under the comparison's stated ordering. The sponsor's own analysis is reported separately and takes precedence over a simple recalculation from summary statistics.

The questionnaire-based endpoint captures the measure defined by the registry, not every aspect of eating behavior or clinical status. Details such as the recall period and handling of incomplete questionnaires are not reported in the registry record.

7. Results Posted in the Registry

• Cognitive-Behavioral Therapy: mean 1.52 versus 6.05 on control; difference -4.53 (95% CI -9.27 to 0.21), standardized effect d = -0.62; z-test p = 0.061.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Binge (Loss-of-control) Eating Episode Frequency

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Cognitive-Behavioral Therapy: mean 2.04 versus 5.94 on control; difference -3.90 (95% CI -8.61 to 0.81), standardized effect d = -0.59; z-test p = 0.105.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Cognitive-Behavioral Therapy and the control. The interval gives the range of effects that remain compatible with the data.

Weight (e.g., Percent Loss)

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Cognitive-Behavioral Therapy: mean 0.02 versus 0.01 on control; difference 0.01 (95% CI -0.05 to 0.07), standardized effect d = 0.09; z-test p = 0.758.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Cognitive-Behavioral Therapy and the control. The interval gives the range of effects that remain compatible with the data.

The secondary results extend the assessment beyond the primary post-treatment timepoint. The registry reports follow-up binge or loss-of-control eating episode frequency and weight as a percentage of baseline BMI, both at the stated short-term follow-up. Taken together, the posted summaries show a lower mean episode frequency in the therapy group than in the control group at follow-up, while the weight means are close. The unadjusted recalculated intervals for these comparisons cross zero.

The eating episode follow-up is directionally consistent with the primary group means, but it is a separate timepoint and has its own outcome counts and uncertainty. It should not be treated as a confirmation independent of the primary result. The weight comparison addresses a different outcome and does not directly answer whether episode frequency differed. The registry reports a sponsor ANOVA result for the follow-up eating episode outcome, but it does not provide a sponsor-reported analysis for weight.

Secondary outcomes generally carry less inferential weight than a designated primary outcome. Testing multiple outcomes or timepoints increases the chance that at least one result may appear notable by chance, especially if no adjustment or prespecified hierarchy governs interpretation. The registry record does not report whether multiplicity was addressed, how outcomes were prioritized, or whether the analyses were prespecified. Therefore, the secondary results are best read as supporting descriptive evidence rather than as a collection of independent confirmatory tests.

Outcome-specific participant counts differ across the posted results, and the reasons are not reported in the registry record. That limits direct comparisons across timepoints and outcomes. No additional causal or clinical conclusion should be drawn from the secondary pattern beyond the particular measures and timeframes recorded.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Binge (Loss-of-control) Eating Episode Frequency
• Weight (e.g., Percent Loss)

9. Statistical Interpretation

10. Statistical Methods Explained

The registry reports an ANOVA for the sponsor's primary comparison and gives its result as a superiority analysis. The record does not describe the full model specification, the particular null hypothesis parameterization, covariate adjustment, or the analysis population. The sponsor's reported p-value is therefore the available account of that analysis, rather than a complete statistical methods description.

What does a mean difference describe?

For a continuous outcome, a mean difference compares the average outcome in one group with the average in another. The posted recalculation uses the control mean as the reference and subtracts it from the cognitive-behavioral therapy mean. A negative estimate therefore indicates a lower mean in the therapy group for the eating episode outcomes. This is a group-level contrast; it does not imply that every participant had the same direction or size of change.

The recalculation is explicitly unadjusted and based on group summary statistics. It is useful as a straightforward descriptive comparison, but it may not match an analysis that uses participant-level information, adjusts for baseline measurements, or applies another model. The registry does not report whether or how those features were used in the sponsor's ANOVA. Since the reported sponsor result differs from the recalculation, the sponsor's analysis takes precedence and the discrepancy should remain visible rather than being resolved by speculation.

What does a confidence interval convey?

A confidence interval describes uncertainty around an estimated effect under the assumptions of the analysis. A relatively wide interval indicates that a broad range of effects is compatible with the data and method; an interval that crosses zero includes both directions for a mean difference. It is not a range containing a known proportion of individual participants' outcomes, and it does not express the probability that the true effect lies within the particular interval. The primary recalculated interval crosses zero, so the simple estimate is not highly precise relative to the range of values it allows.

Why do standard deviations matter?

A standard deviation describes the spread of observations around a group mean. For the primary outcome, the reported standard deviations are 2.09 in the therapy group and 10.35 in the control group. The difference in these spreads indicates that variability was not similar across the groups in the posted summaries. General statistical context: when variability differs substantially, an analysis that assumes equal variance may not describe uncertainty well; methods that allow different group variances or model the outcome distribution can be relevant. The record does not report whether the sponsor's ANOVA addressed unequal variances.

General statistical context: episode counts can be concentrated at low values, may include many zero observations, and can have a long right tail. In such settings, a mean and standard deviation remain interpretable summaries, but their relationship to a normal-distribution model may be imperfect. Over-dispersion means that the observed variability is greater than a model's assumed variability; if a count model fails to account for it, standard errors and resulting inference can be unreliable. The registry does not report the distribution of individual episode counts or a diagnostic assessment, so these possibilities should not be asserted as features of this trial's data.

What is the role of the p-value?

A p-value is a measure of how compatible the observed data are with a specified null model, assuming the analysis assumptions hold. It is not the probability that the null hypothesis is true, nor does it measure clinical importance. The sponsor's p-value is reported as .01 for the ANOVA. The simple unadjusted recalculation gives 0.061. These values come from different reported analyses, and the registry does not give enough methods detail to identify the source of the difference. The sponsor result takes precedence; the recalculation offers a separate, limited check from the posted summaries.

The record also labels the sponsor analysis as superiority. That label does not supply the missing effect estimate, confidence interval, or detailed analysis specification. Without those details, the sponsor's result cannot be used to quantify the magnitude or precision of the group contrast. The general statistical distinction is that a test result and an effect estimate answer different questions: the test summarizes evidence under a null model, while the estimate gives the contrast on the outcome's scale.

How should follow-up outcomes be read?

The follow-up eating episode outcome and weight outcome are separate comparisons at a later timeframe. Their recalculated estimates are unadjusted and their sponsor-reported results are available only for the eating episode outcome. These secondary analyses should be interpreted cautiously, particularly because multiple outcomes and timepoints create more opportunities for apparently notable results. The registry does not report whether a multiplicity adjustment or a prespecified hierarchy was used.

The statistical plan, original sample-size assumptions, handling of missing data, and analysis populations are not reported in the registry record. Those omissions limit the ability to assess whether the presented analyses match the trial's planned methods. General methodological interpretation can clarify what the posted summaries mean, but it cannot fill in these missing design and analysis details.

11. Confidence Intervals and Precision

The primary recalculated mean difference is -4.53 episodes. Its 95% confidence interval ranges from -9.269 to just above zero, and the unadjusted p-value is 0.061. This interval crosses zero, so the summary-statistic recalculation is compatible with a range that includes no mean difference as well as a lower mean in the therapy group. The interval is broad relative to the point estimate, which limits precision.

A confidence interval is an uncertainty summary conditional on the analysis and its assumptions. It does not show the range of individual participants' episode frequencies, and it does not establish that the true effect is equally likely to be any value inside the interval. The primary outcome's observed standard deviations differ considerably between groups, and the record does not report the full model or whether the sponsor's analysis accounted for that feature.

The sponsor reported an ANOVA result of p = .01 but no associated effect estimate or confidence interval. That result therefore does not supply a directly reported estimate of the sponsor analysis's magnitude or precision. The sponsor analysis takes precedence over the recalculation because the two results differ, but the missing interval means the precision of the sponsor's estimate cannot be assessed from the registry record.

The follow-up eating episode comparison also has an interval crossing zero in the unadjusted recalculation. For weight, the recalculated interval likewise crosses zero. These are secondary outcomes and do not remove the uncertainty surrounding the primary comparison. All intervals and p-values should be interpreted in light of outcome-specific sample counts, the unreported handling of missing data, and the absence of detailed analysis methods in the registry record.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The sample-size figures are planning illustrations based on the observed effect, not the trial's own original design assumptions. They use a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The assumed effect to detect is the stated fraction of the observed effect. The figures are participants per arm for these continuous outcomes; the convention of reporting total events applies to time-to-event planning displays, not these outcomes.

The 80% power column gives the planned number of participants per arm under the stated assumptions for a design with an 80% chance of detecting the specified effect if that effect is the true one. The 90% power column describes the corresponding planning target for a 90% chance. These are design probabilities under a model, not guarantees for a future study. The displayed numbers do not mean that the completed trial had a particular probability of obtaining its observed result.

The rows consider planning for 100%, 75%, and 50% of the observed effect. A smaller target effect generally requires more participants to distinguish it from random variation at the same significance level and power target. Consequently, the participant requirements increase as the planning effect is made smaller. The estimates are illustrations derived from the observed result and should not be read as the original rationale for enrolling participants in this trial.

This is not observed power. The original sample-size justification and the assumptions used to plan the trial are not reported in the registry record. The planning illustrations also rely on simplified assumptions, including equal allocation and a normal approximation; they do not account for any unreported design features or missing-data process.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = -4.530; pooled standard deviation σ = 7.283.
2. n = 2 × (1.96 + 0.842)² × 7.283² / -4.530² = 40.6, rounded up to 41 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
delta <- -4.53                      # observed difference in means
sigma <- 7.28294149654556                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(4088097)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 41 (80% power), 55 (90% power)
 75% of observed effect: n per arm = 73 (80% power), 97 (90% power)
 50% of observed effect: n per arm = 163 (80% power), 218 (90% power)
power.t.test: 41.6 
Simulated power at n = 41 per arm: 0.792 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

14. Analyses Reported by the Sponsor

The sponsor's primary analysis is reported as an ANOVA superiority analysis with p = .01. The registry does not report an effect estimate or confidence interval for this analysis, and it does not provide enough detail to reconstruct its model. The result is therefore the sponsor's stated inferential finding, but its magnitude and precision cannot be read directly from the reported p-value.

A simple unadjusted recalculation from the posted group means and standard deviations gives a mean difference of -4.53 episodes, a 95% confidence interval crossing zero, and an unadjusted p-value of 0.061. This does not reproduce the sponsor's reported result. The record does not explain the discrepancy; it would be inappropriate to guess whether it reflects adjustment, a different analysis population, model choices, or another reason.

When the sponsor's result and a summary-statistic recalculation disagree, the sponsor's own analysis takes precedence. The recalculation remains a useful check that describes what follows from the posted summaries under a simple unadjusted approach, but it should not replace the registry's reported analysis. The missing model details and effect estimate limit independent evaluation of the sponsor's result.

15. Multiplicity, Missing Data and Analysis Populations

The registry identifies binge or loss-of-control eating episode frequency at post-treatment as the primary outcome and lists additional outcomes, including follow-up eating episode frequency and weight. It reports three analyses in total, but does not provide a complete statistical analysis plan. The planned hierarchy among outcomes, any prespecified contrasts, and any adjustment for multiple testing are not reported in the registry record.

Multiplicity matters because interpreting several outcomes and timepoints creates more than one opportunity for a result to appear noteworthy. If tests are treated as independent confirmatory evidence without accounting for that multiplicity, the overall chance of a false positive can be greater than the chance associated with an individual test. General statistical context: methods such as family-wise error control or a prespecified testing sequence can address particular multiplicity goals, but the registry record does not say whether any such approach was used. It is not appropriate to infer an adjustment from the results alone.

Missing-data methods are also not reported. The primary result lists 21.0 participants in the therapy group and 19.0 in the control group, while the trial enrollment is 50. The follow-up eating outcome and weight outcome list different counts. The record does not explain whether these differences reflect missing questionnaires, eligibility for a particular analysis, exclusions, or another cause. It also does not state whether analyses followed intention-to-treat, per-protocol, or another population definition.

The registry record does not describe imputation, sensitivity analyses for missingness, or any method for handling incomplete outcome questionnaires. Without these details, it is not possible to assess whether missing outcomes could affect the group comparisons. Baseline adjustment, model diagnostics, and the treatment of unequal variability are also not described. These omissions do not prove that such methods were absent; they mean that they are not reported in the registry record and cannot be evaluated from the posted information.

16. 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 lists 25 participants at risk for serious adverse events in each group. It reports zero serious adverse events affecting participants in the cognitive-behavioral therapy group and zero in the control group, with zero deaths in both groups. The record lists no serious adverse event terms.

These figures describe the safety entries posted for the registry record. They do not establish that no other adverse experiences occurred, because the figures concern serious adverse events and the record does not provide a full account of non-serious events, event collection procedures, or follow-up completeness. The safety denominators also differ from the outcome-specific participant counts, and the record does not explain the difference.

With no serious events recorded in either group, this summary does not provide a comparative estimate of relative risk. The registry record does not report additional detail about event definitions, timing, relatedness to the intervention, or how adverse events were monitored. These results should therefore be read narrowly as the posted serious adverse event counts, not as a comprehensive assessment of safety.

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

The primary outcome summaries show lower mean questionnaire-reported binge or loss-of-control eating episode frequency in the cognitive-behavioral therapy group than in the control group at post-treatment. The simple unadjusted recalculation gives a negative mean difference, but its confidence interval crosses zero and its unadjusted p-value is 0.061. The sponsor's own ANOVA reports p = .01 and takes precedence over that recalculation; the registry does not report the sponsor analysis's effect estimate or confidence interval.

This combination of reported results means that the sponsor's analysis indicates a statistically notable result under its procedure, while the independent summary-statistic check does not provide a precise estimate. The missing methods and the unexplained discrepancy prevent a fuller reconciliation. Statistical evidence does not by itself establish the size, importance, or durability of a difference. The clinical meaning of a particular difference in episode frequency is not defined in the registry record.

The results concern the enrolled study population, the questionnaire outcome, and the stated assessment times. They do not establish effects in other age groups or settings, because detailed eligibility and setting information are not reported. Nor do the primary results establish changes in all aspects of eating behavior, weight, or broader health outcomes. Follow-up outcomes are secondary and should be interpreted with added caution in light of the unreported multiplicity plan.

The safety summary records no serious adverse events or deaths in either group among the listed at-risk participants. It does not provide a complete account of all possible harms. The registry record also does not explain outcome-specific participant counts or describe missing-data handling. Accordingly, the posted results can be described, but the record does not support claims beyond those particular comparisons and safety entries.

18. Why This Trial Matters Statistically

This trial is statistically informative because it places a sponsor-reported analysis alongside results that can be recalculated from group summaries. Those approaches do not agree: the sponsor reports an ANOVA p-value of .01, whereas the unadjusted recalculation yields a confidence interval crossing zero and p = 0.061. Keeping the discrepancy visible helps readers distinguish the sponsor's inferential analysis from a simpler calculation rather than treating them as interchangeable.

The example also illustrates why an effect estimate, interval, and p-value answer different questions. The recalculated mean difference expresses the contrast in episodes, while its interval conveys uncertainty under the recalculation's assumptions. The sponsor's p-value alone does not give the size or precision of its analysis result. A fuller registry report would make the statistical result easier to assess by including the sponsor estimate, interval, model details, and analysis population.

Unequal reported standard deviations and varying outcome-specific counts further emphasize that group means alone do not tell the full analytical story. Distributional details, missing-data methods, and multiplicity procedures can matter to interpretation, particularly when outcomes are measured at more than one timepoint. These design and analysis details are not reported in the registry record.

Finally, planning illustrations based on an observed effect are not a substitute for a trial's original sample-size justification. They help show how planning targets change when the assumed effect changes, but they do not reveal the study's original design choices or its probability of success. This distinction is important in early evaluations, where estimates may be imprecise and useful primarily for informing better specified future research.

19. 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.

Several trial-specific details needed to interpret the results are not reported in the registry record. These include detailed eligibility criteria and age range, recruitment setting, intervention schedule and delivery, the identity of those masked, allocation procedure details, and the original sample-size justification. The registry describes a randomized, single-masked, parallel-group design but does not elaborate on these features.

The sponsor's primary ANOVA result is not accompanied by an effect estimate or confidence interval, and the analysis model is not specified in detail. The simple unadjusted recalculation differs from the sponsor's result, but the reason is not given. This limits the ability to understand how the sponsor result was produced or to assess its precision. The large difference between the reported group standard deviations for the primary outcome also makes the unspecified model assumptions relevant.

The enrollment count and outcome-specific counts do not match, and the reasons for the varying counts are not reported. Missing-data handling, analysis population definitions, baseline adjustment, and sensitivity analyses are not described. The registry also does not report whether multiplicity was addressed across outcomes and timepoints.

Eating episode frequency is assessed by questionnaire, but the recall period and handling of incomplete questionnaires are not described. Follow-up is identified as 3 months after treatment, but further information about completion and measurement procedures is not reported. The adverse event entries are limited to serious events and deaths; broader safety monitoring details are not reported. These limitations restrict interpretation of both the precision and scope of the findings.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Cognitive Behavioral Therapy for Adolescent Binge Eating and Loss of Control Eating — NCT04088097.
• Registry record last updated 2024-10-10; status last verified 2024-09.
• Results first posted 2024-10-10.
• 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.

23. Record Summary

NCT04088097 is a completed randomized, single-masked, parallel-group trial sponsored by Yale University. It enrolled 50 participants and compared adapted cognitive-behavioral therapy for adolescents with binge or loss-of-control eating against educational materials related to adolescent health and nutrition.

The primary outcome was questionnaire-assessed binge or loss-of-control eating episode frequency at post-treatment, identified as 4 months. The sponsor reported an ANOVA superiority result with p = .01. A simple unadjusted recalculation from the posted summaries gives a mean difference of -4.53 episodes, with a confidence interval crossing zero and p = 0.061; the sponsor's analysis takes precedence.

Secondary results include follow-up eating episode frequency and weight. The record lists zero serious adverse events and zero deaths in both groups. The registry does not report the original sample-size rationale, detailed analysis plan, missing-data methods, or the reason for the difference between the sponsor result and the recalculation.