Crohn Disease Exclusion Diet After Single Medication De-escalation: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Crohn Disease Exclusion Diet After Single Medication De-escalation (NCT02472457), a non-randomized na study. The primary outcome is Sustained Relapse-free Remission.

2. Clinical Question

This question combines a dietary comparison with a clinical course after medication withdrawal. Those components matter when interpreting the result: an observed difference between diet groups could reflect differences in diet, differences between the people who entered each group, or other factors not described in the registry record. Because allocation was non-randomized, the comparison cannot by itself isolate a causal effect of diet.

The primary outcome is a status at a specified follow-up point, whereas time-to-relapse concerns when relapse occurs over follow-up. These are related but not identical questions. The registry describes the primary remission definition using the Pediatric Crohn Disease Activity Index, C-reactive protein, and stool calprotectin. It does not report a full analysis comparing the groups or an estimate of the size and precision of any difference.

3. Trial Design

The registry lists this as a completed, non-randomized, parallel-group study with no masking. It records enrollment of 19 and names the Children's Hospital of Philadelphia as sponsor. The study start is recorded as 2015-08 and primary completion as 2019-05. The phase is listed as NA.

The two named interventions are the Crohn Disease Exclusion Diet and a free diet. The free diet is described as having no dietary restrictions. CDED is described as a palatable diet that excludes foods suspected to have a role in intestinal inflammation. These descriptions identify the diet contrast, but details such as how adherence was assessed, how diet assignment was made, and how medication de-escalation was carried out are not reported in the registry record.

Non-randomized allocation means the registry does not describe assignment by chance. As a result, participants in the diet groups may have differed at the outset or in ways related to their subsequent outcomes. No adjustment method or other approach to account for such differences is reported in the registry record. The lack of masking also means the record does not describe concealment of group assignment from participants or study personnel.

The posted primary outcome is sustained relapse-free remission at 52 weeks following enrollment. The record defines remission using the Pediatric Crohn Disease Activity Index, C-reactive protein, and stool calprotectin. It also lists time-to-relapse and microbial composition of the gastrointestinal tract among secondary outcomes. The microbiome outcome appears twice in the secondary-outcome list, and the record does not explain whether these entries represent different assessments.

The record gives outcome and safety information but reports no analyses. It does not specify a statistical analysis plan, a sample-size justification for the trial, or whether any interim analyses were conducted. Those details should not be inferred from the study being marked completed or from the separate planning illustrations.

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: 1 participant in the free diet group followed the specific carbohydrate diet and was in remission at 52 weeks. 1 participant in the CDED group withdrew prior to 52 weeks.

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 Sustained Relapse-free Remission, measured over At 52 weeks following enrollment.

The primary endpoint is sustained relapse-free remission, recorded as a count of participants and assessed at 52 weeks following enrollment. The registry defines remission using the Pediatric Crohn Disease Activity Index, C-reactive protein, and stool calprotectin. Thus, the endpoint is not described as relying on a single measurement; it combines a disease activity index with blood and stool markers. The registry record does not give thresholds, the precise rule for combining these measures, or details about how assessments were scheduled.

A fixed-time remission endpoint asks whether a participant meets the stated remission definition at a specified follow-up time. It can offer a direct clinical summary of status at that point, but it does not convey when a relapse may have occurred during follow-up or how long remission lasted. A time-to-relapse analysis addresses timing, which is why the registry’s separate secondary outcome is relevant, but no such analysis is reported in the registry record.

The registry’s primary-outcome description notes that one participant in the free diet group followed the specific carbohydrate diet and was in remission at 52 weeks. It also notes that one participant in the CDED group withdrew before 52 weeks. These details affect how the endpoint should be read: the first describes a diet different from the assigned group description, and the second indicates that a 52-week outcome was not available for that participant. The record does not state how either circumstance was handled in an analysis.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a non-randomized comparison imply?

In general, random allocation is used to make groups comparable in expectation, including for factors that may not be measured. A non-randomized study does not have that design feature. Differences in outcomes between groups may reflect the intervention, but may also reflect differences in the participants or circumstances associated with group assignment. Statistical adjustment can sometimes address measured differences, but it cannot automatically remove bias from unmeasured factors. The registry record does not report a method of adjustment, how group assignment occurred, or the characteristics needed to judge comparability.

The parallel-group label indicates that the study has distinct diet groups rather than describing a crossover comparison. It does not by itself establish how participants entered those groups or whether the groups were comparable. The record lists no masking. In general, masking can limit the influence of knowledge of group assignment on assessment or behavior, but it does not correct all sources of bias. No specific masking procedure is reported in the registry record.

How should a binary count outcome be understood?

In general, a binary outcome classifies each participant into one of two states, such as meeting or not meeting a defined outcome. A count outcome then reports how many participants meet the specified condition. The primary outcome is described as a count of participants who sustain relapse-free remission at 52 weeks. To compare such outcomes, an analysis ordinarily needs a clear group denominator, a consistent outcome definition, and a stated approach to participants whose outcome is unavailable or whose care differs from the intended group.

The registry gives the endpoint definition and two relevant participant details, but it does not report a complete primary-outcome analysis. It does not state the group denominators used for that endpoint, how the withdrawal was treated, or how the participant following a different diet was classified for analysis. Without those details and a comparative estimate, readers cannot determine the magnitude or precision of a group contrast from the primary outcome description alone.

What would a confidence interval communicate?

In general, a confidence interval expresses the uncertainty around an estimate under the assumptions of the statistical method used. A narrow interval indicates greater precision than a wide interval, while an interval spanning a broad range of possible effects signals substantial uncertainty. A confidence interval is not itself a measure of clinical importance, and it does not resolve design limitations such as non-random allocation or incomplete outcome information.

No comparative confidence interval is reported in the registry record. Therefore, the precision of any difference in remission or relapse between the diet groups cannot be assessed from the posted results. A count alone, without a complete comparison and its uncertainty, is not enough to establish whether an apparent difference is stable or could be compatible with a wide range of underlying effects.

What does time-to-event analysis add?

In general, time-to-event methods consider both whether an event occurs and when it occurs. They can also account for participants whose event has not been observed by the end of their follow-up, provided that follow-up and censoring are appropriately defined. Such methods require information about event times, follow-up, and the analysis procedure. The registry lists time-to-relapse as a secondary outcome, but reports no analysis or details about follow-up and censoring. It is therefore not possible to evaluate a time-to-event result from this record.

Time-to-relapse also differs from the primary fixed-time status. Two groups could have similar status at 52 weeks while differing in the timing of relapses, or the reverse. Those are general possibilities, not findings from this study. The record supplies no event-time estimates with which to examine that distinction.

How should planning illustrations be read?

In general, sample-size planning calculations describe the number of participants or events that would be needed under stated assumptions to detect a specified effect with a chosen statistical procedure. They are not observed outcomes, and they do not show that a study had a particular probability of detecting its true effect. That latter interpretation would incorrectly treat a planning illustration as observed power.

The planning assumptions described for this study use a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The target effect is expressed as a stated fraction of the observed effect. For time-to-event planning, the reported quantity is the total number of events; for other planning calculations, it is participants per arm. These calculations are illustrations based on the observed effect, not the trial’s own design justification. The registry record does not report the trial’s sample-size justification or a protocol-based target. No specific planning count is stated in the registry record text.

In general, an observed effect used to plan a future study can be unstable, especially when the underlying study is small or its comparison is non-randomized. A planning result depends on the assumed target effect and analysis conditions; it is not a forecast of the result of a future study. It should not be confused with the actual enrollment or the number of participants with observed outcomes in this study.

Why can over-dispersion matter?

In general, over-dispersion refers to variation in count data that exceeds what a chosen statistical model expects. If a model does not allow for enough variability, its uncertainty estimates can be too small, making results appear more precise than they are. Whether over-dispersion is relevant depends on the outcome and model. The primary endpoint is described as a participant count, but no model or inferential analysis is reported in the registry record, so it is not possible to assess whether over-dispersion was considered or was relevant to a particular analysis.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("Free Diet", "Crohn Disease Exclusion Diet")
n      <- c(8, 9)
events <- c(6, 5)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  Free Diet  8  6  75.0  34.9  96.8  40.9  92.9  40.0  92.5
ARM  Crohn Disease Exclusion Diet  9  5  55.6  21.2  86.3  26.7  81.1  26.2  81.3

11. Multiplicity, Missing Data and Analysis Populations

The registry lists five outcome measures and reports zero analyses. Its secondary outcomes include time-to-relapse and microbial composition of the gastrointestinal tract, with the microbiome outcome appearing twice in the list. The record does not explain whether the repeated entry represents separate measures, time points, or a duplicate listing. It does not report which outcomes were analyzed or how results were selected for reporting.

When a study examines multiple outcomes or makes multiple comparisons, the chance of finding an apparently notable result by chance can increase. In general, a prespecified hierarchy or a method for adjusting comparisons can help address this issue. Neither a multiplicity strategy nor a statistical analysis plan is reported in the registry record. Since no analyses are posted, the effects of any multiplicity approach cannot be assessed.

Missing outcome information is relevant to the primary endpoint: the registry states that one participant in the CDED group withdrew before 52 weeks. It also says that one participant in the free diet group followed the specific carbohydrate diet and was in remission at 52 weeks. The record does not specify the analysis population, whether the withdrawal was excluded or otherwise handled, or how the dietary deviation was classified. It does not report imputation, sensitivity analyses, or other missing-data procedures.

In general, missingness can bias a comparison when the availability of an outcome is related to prognosis or to the intervention experience. A stated withdrawal alone does not reveal whether such bias occurred. Without the analysis population and missing-data method, readers cannot assess how the primary comparison would change under alternative assumptions about unavailable outcomes. The registry record also does not report whether all enrolled participants were included in each outcome analysis.

12. 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 reports zero serious adverse events among the 8 participants listed at risk in the free diet group and zero among the 10 listed at risk in the Crohn Disease Exclusion Diet group. It also reports zero other adverse events and zero deaths for each group’s listed at-risk count. No serious adverse event terms are listed.

These figures describe events recorded in this study. They do not establish that the diets or the surrounding medication changes carry no risk. In general, zero observed events can occur even when an event is possible, particularly when the amount of observation is limited. The registry record does not report exposure time, detailed event definitions, monitoring procedures, or a statistical comparison of safety between groups. The counts also do not establish whether all participants had the same duration of safety follow-up.

The at-risk counts are safety denominators as reported in the registry, and should not be treated as proof that every outcome has the same denominator. The primary outcome description separately notes a withdrawal before 52 weeks. Safety counts and efficacy outcome availability answer different questions and need not be interpreted as interchangeable. No estimate of comparative safety or precision is reported in the registry record.

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

The registry record identifies a clinically focused question: whether pediatric Crohn disease patients in remission remain relapse-free longer after medication de-escalation when following CDED rather than a free diet. It reports a 52-week sustained remission endpoint and describes its clinical and biomarker components. It also records one participant in the free diet group who followed the specific carbohydrate diet and was in remission at 52 weeks, and one CDED participant who withdrew before that follow-up.

Those details do not constitute a complete comparison. No group-level primary outcome analysis, effect estimate, confidence interval, or p-value is reported. The record therefore does not establish how the groups differed, how precisely any difference was estimated, or whether a difference could be attributed to diet. The trial’s non-randomized allocation adds uncertainty because differences between groups may reflect factors beyond the diet contrast.

The study is marked completed, but that status is not a substitute for a reported statistical analysis. The record does not describe outcome handling for the withdrawal, the analysis population, adjustment for differences between groups, or a multiplicity strategy. Time-to-relapse and microbiome outcomes are listed, but their analyses are not reported. The safety counts record no events in the listed categories, but they do not show that risk is absent or provide a precise estimate of rare harms.

In general, a registry result is most informative when it presents clear definitions, group denominators, comparative estimates, uncertainty intervals, and methods for missing data and analysis. Those elements are not reported for the primary comparison in this record. The posted information should therefore be read as a limited description of the study and its outcome reporting, not as evidence that one diet does or does not produce a particular clinical effect.

14. Why This Trial Matters Statistically

This study is statistically instructive because the clinical question is understandable, but its design and reporting constrain what can be concluded. The registry places dietary choices alongside medication de-escalation in pediatric Crohn disease and defines remission with a disease activity index and inflammatory markers. At the same time, the comparison is non-randomized, so group differences cannot be interpreted as though allocation had been determined by chance.

The record also illustrates why an endpoint definition and an endpoint result are different things. It specifies sustained relapse-free remission at 52 weeks and describes the measures used to define remission, but does not report a complete group comparison. A clearly named endpoint does not by itself provide an effect estimate, uncertainty, or a method for handling participants whose outcome is unavailable or whose diet differs from the group description.

Another important distinction is between a completed study and a fully reported analysis. The record lists outcomes and safety counts, yet reports no analyses. It does not describe missing-data procedures, multiplicity handling, or an analysis population. The distinction matters because the same observed counts can be interpreted differently depending on the denominator, the treatment of withdrawals, and the planned comparison.

Finally, planning illustrations should be separated from evidence about the study’s results. They use stated assumptions and an observed effect to illustrate future planning; they are not observed power, a trial design justification, or a substitute for an uncertainty interval. Together, these features make the record useful for discussing design, endpoint interpretation, and reporting limits while leaving the comparative clinical effect unresolved in the registry record.

15. 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 trial-specific limitation is non-randomized allocation. The registry does not report how participants entered the diet groups or whether measured baseline differences were addressed. Without random allocation or a described adjustment strategy, an observed contrast could be influenced by differences among participants rather than diet alone. The absence of masking is also recorded, but the registry does not explain how outcome assessment was conducted or whether any assessments were protected from knowledge of group assignment.

The primary outcome description is incomplete for comparative interpretation. It gives the 52-week endpoint definition and two participant-level details, but not a complete group-specific outcome summary, comparative estimate, or precision measure. One CDED participant withdrew before 52 weeks, and the record does not say how that outcome was handled. One participant assigned to the free diet group followed the specific carbohydrate diet and was in remission at 52 weeks; the record does not state how this dietary deviation was treated analytically.

Additional design and analysis details are not reported in the registry record, including eligibility criteria beyond the remission description, medication-specific details, adherence assessment, the method of allocation, a sample-size justification, the analysis plan, and any interim analyses. The secondary outcome list includes a repeated microbiome entry, which is not explained. The registry reports zero analyses, so there are no posted comparative estimates for the primary or secondary outcomes.

Safety reporting is also limited to event counts and at-risk counts. Detailed event definitions, duration of exposure, monitoring methods, and uncertainty estimates are not reported. These limitations do not show that the study findings are invalid; they define what can and cannot be inferred from the registry record.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Crohn Disease Exclusion Diet After Single Medication De-escalation — NCT02472457.
• Registry record last updated 2020-10-20; status last verified 2020-09.
• Results first posted 2020-05-27.
• 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.

19. Record Summary

The registry describes a completed, non-randomized, unmasked, parallel-group study sponsored by the Children's Hospital of Philadelphia. Enrollment is listed as 19. The study start is recorded as 2015-08 and primary completion as 2019-05; phase is listed as NA.

The comparison is between the Crohn Disease Exclusion Diet and a free diet. The primary outcome is sustained relapse-free remission at 52 weeks following enrollment, with remission defined using the Pediatric Crohn Disease Activity Index, C-reactive protein, and stool calprotectin. The record notes one free diet participant followed the specific carbohydrate diet and was in remission at 52 weeks, and one CDED participant withdrew before 52 weeks.

Time-to-relapse and microbial composition are listed among secondary outcomes. The registry reports zero analyses, no comparative effect estimate, and no statistical plan or missing-data method. It lists zero serious adverse events, zero other adverse events, and zero deaths in both groups’ reported safety counts. These results describe the registry entries but do not resolve the comparative effect of the diets.