Iontophoresis With Dexamethasone and Physical Therapy to Treat Apophysitis of the Knee in Pediatrics: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Iontophoresis With Dexamethasone and Physical Therapy to Treat Apophysitis of the Knee in Pediatrics (NCT03606980), a randomized phase 2 study. The primary outcome is Functional Outcome as Measured in Length of Time to Meet Return To Sport Criteria (Measured in Days).

2. Clinical Question

The registry describes a randomized, triple-masked, parallel-group Phase 2 trial. Its stated objective was to compare time to meet return-to-sport criteria across the three treatment approaches. The dexamethasone and sodium chloride groups both received iontophoresis and physical therapy; the third group received physical therapy alone.

The registry record lists enrollment of 45 and a terminated status. Its brief summary says approximately 147 patients would be enrolled. These entries describe the recorded enrollment and the approximate enrollment stated in the summary; the registry record does not explain the difference. It also does not report why the trial was terminated.

3. Trial Design

This was a randomized, triple-masked, parallel-group Phase 2 trial in pediatric patients with apophysitis of the knee. The three groups were iontophoresis with dexamethasone plus physical therapy, iontophoresis with sodium chloride plus physical therapy, and physical therapy alone. The saline iontophoresis group served as a placebo comparison for the iontophoresis component, while physical therapy alone provided a comparison without iontophoresis.

The registry describes the delivery system as an adhesive patch and an iontophoresis system. Its arm descriptions are truncated in the record, so the complete administration details are not reported in the registry record. The physical-therapy-alone description says that participants received the standard physical therapy protocol, involving up to 20 visits.

Enrollment is listed as 45, and the trial status is terminated. The brief summary states that approximately 147 patients would be enrolled. The reason for termination, the number completing each treatment approach, and the reasons participants may have left the study are not reported in the registry record. The start date and primary completion date are recorded, but the record does not give further information about how those dates relate to the termination.

The primary outcome was time, in days, from Visit 1 until all return-to-sport checklist items could be passed. Assessment was planned at each treatment visit, for a maximum of 12 visits over a maximum of 8 weeks, or until criteria were met, whichever occurred sooner. The reported primary descriptive statistics concern participants who met the criteria during the study. The registry record does not report the number who did not meet them or how such cases were handled in the time-to-event analyses.

Randomization procedures, allocation concealment, details of who was masked, the statistical analysis plan, and the rationale for the planned enrollment are not reported in the registry record. The word “triple” records the masking designation but does not identify the masked roles.

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

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Analysis of return to sport time in patients who met return to sport criteria during 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 Functional Outcome as Measured in Length of Time to Meet Return To Sport Criteria (Measured in Days), measured over Measured at each Study Treatment Visit for a maximum of 12 Study Treatment Visits over a maximum of 8 weeks or until Return to Sport Criteria are met, whichever is sooner.

The primary endpoint measures the elapsed time from Visit 1 until a participant meets every item on the return-to-sport checklist. The checklist items are passed or not passed; the endpoint converts the date on which all are passed into a duration measured in days. Assessments were scheduled at treatment visits, up to 12 visits over a maximum of 8 weeks, or until the criteria were met, whichever came first.

Time-to-event outcomes are useful when both whether an event occurs and how long it takes to occur matter. In this trial, the event is meeting the return-to-sport criteria. A time-to-event approach can account for different observed durations, rather than treating all participants as if they had the same follow-up time. That general advantage depends on appropriate handling of participants whose event is not observed during follow-up.

The registry record identifies the primary analysis population as patients who met return-to-sport criteria during the study. It also reports group means and standard deviations for time among those participants, and sponsor analyses using Cox proportional hazards models. The record does not report the event and censoring counts or fully describe how participants who did not meet criteria were represented in those models. Therefore, the posted group means should be read as summaries among the stated analysis population, not necessarily as summaries of every randomized participant.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a mean and standard deviation describe?

A mean is the arithmetic average of a continuous outcome, such as the number of days to meet return-to-sport criteria. A standard deviation describes how dispersed individual observations are around that average. It is not a confidence interval and does not show how precisely the mean has been estimated. Precision depends in part on the number of observations; the primary descriptive groups are small.

The posted means are calculated for the registry’s stated population of patients who met the criteria during the study. This restriction matters: the averages do not, by themselves, describe outcomes for every randomized participant. If participants who did not meet criteria differ from those who did, an analysis limited to observed achievers can give a partial picture of the randomized groups.

What does a Cox proportional hazards analysis estimate?

A Cox proportional hazards model compares the instantaneous rate of an event between groups over time, conditional on the model. Here the event is described as meeting the return-to-sport criteria. The hazard ratio is a relative measure, not a difference in days and not a probability that an individual participant will meet criteria. A value of 1 corresponds to equal modeled event rates; values on either side represent relative rates whose practical direction depends on the coding and reference group.

The proportional hazards model assumes that the relative hazard between groups is reasonably constant over follow-up. Whether that assumption was checked, how event times and censoring were handled, and what covariates were included are not reported in the registry record. The record also does not provide the number of events used in each pairwise model. Those details are important for assessing whether a model is appropriate and how stable its estimates may be.

How should a confidence interval be read?

A confidence interval expresses uncertainty around an estimated quantity under the statistical model. A narrower interval indicates greater precision; a wider interval indicates less precision. For a hazard ratio, an interval containing 1 is compatible with equal relative hazards under the model, while an interval that excludes 1 is inconsistent with that value at the corresponding confidence level, subject to the model and analysis assumptions. This is not proof that the model is correct or that a clinically important difference exists.

The reported primary intervals are broad. Two exclude 1 and one includes 1, but the intervals do not pin down the magnitude of any relative association precisely. A result that does not exclude 1 is not proof of no difference. Likewise, an interval excluding 1 does not establish a clinically important effect, particularly when the analysis details and event coding are not fully described.

What is the distinction between the sponsor analysis and recalculation?

The registry reports sponsor analyses using Cox proportional hazards models for the primary endpoint. A recalculation from posted group summaries would be an unadjusted comparison based on those summary statistics; it is not interchangeable with the sponsor’s time-to-event model. The record does not provide the event-level data needed to reproduce the Cox analysis from first principles. The sponsor’s own analysis is therefore the relevant reported inferential analysis, while any simple summary-based calculation would be a separate, limited description.

For continuous secondary outcomes, unadjusted calculations based on posted means and standard deviations can describe contrasts between groups. Such calculations do not necessarily match the trial’s original analysis, and they do not account for adjustment or modeling choices that are not reported. They should be treated as exploratory summaries rather than substitutes for the sponsor’s analysis.

Why does over-dispersion matter?

Over-dispersion arises when the observed variability is greater than a model assumes. In general, failing to account for over-dispersion can make standard errors too small and confidence intervals too narrow, overstating precision. The posted continuous outcomes are summarized with means and standard deviations, but the registry record does not report a model for these outcomes that would allow an assessment of over-dispersion. This general issue should not be mistaken for a documented problem in this trial.

What do planning illustrations mean?

Sample-size planning illustrations based on an observed effect are hypothetical design calculations, not evidence of the power the completed trial achieved. They depend on assumptions about the effect worth detecting, significance level, allocation, and statistical model. An effect estimated from a small study may be unstable, so using it as a planning target can produce optimistic or otherwise unreliable projections. The registry record does not report that these illustrations were the trial’s original design rationale.

Why is “equivalence” not enough to establish equivalence?

Equivalence testing requires a prespecified range of differences considered acceptably small and an analysis designed to show that the uncertainty lies within that range. A nonsignificant conventional test does not establish equivalence. Although the registry labels the primary pairwise analyses as equivalence, it does not report an equivalence margin or the relevant testing details. The label alone cannot support a conclusion that two approaches are equivalent.

10. Confidence Intervals for Each Arm

No two-arm comparison is shown for this trial, but each arm's mean can be given with a t-based 95% confidence interval computed from the posted mean, spread and number of participants.

R code and output

arm <- c("Iontophoresis With Dexamethasone", "Iontophoresis With Sodium Chloride", "Physical Therapy Alone")
n   <- c(7, 9, 4)
m   <- c(38.1, 30.2, 37.25)
sp  <- c(9.48, 6.8, 3.86)
sd  <- sp      # posted standard deviations
for (i in seq_along(n)) {
  se <- sd[i] / sqrt(n[i]); h <- qt(0.975, n[i] - 1) * se
  cat(sprintf("CARM|%s|%d|%.3f|%.3f|%.3f|%.3f\n", arm[i], n[i], m[i], sd[i], m[i] - h, m[i] + h))
}

CARM  Iontophoresis With Dexamethasone  7  38.100  9.480  29.332  46.868
CARM  Iontophoresis With Sodium Chloride  9  30.200  6.800  24.973  35.427
CARM  Physical Therapy Alone  4  37.250  3.860  31.108  43.392

11. Analyses Reported by the Sponsor

The registry’s primary inferential results are the sponsor-reported Cox proportional hazards analyses. They compare time to meet return-to-sport criteria using a time-to-event model. The posted descriptive means and standard deviations summarize elapsed time among participants who met the criteria, but they are not a replacement for the Cox analysis.

Any recalculated comparison based only on group summary statistics would be unadjusted and would not reproduce the time-to-event model. It would not use participant-level event timing or clarify how observations without a recorded event were treated. The registry record does not report enough detail to independently reconstruct the sponsor’s model. The sponsor’s own analysis therefore takes precedence over a simple recalculation for interpretation of the reported primary inferential result.

The registry’s “equivalence” label should be read cautiously. No equivalence margin or full equivalence-testing method is reported in the registry record. In particular, the dexamethasone-versus-physical-therapy comparison has a confidence interval that includes 1 and is broad; its p-value is 0.64. That is not evidence that the approaches are equivalent. More generally, the reported estimates remain uncertain, and the record does not provide all the information needed to assess the model assumptions, event handling, or clinical importance of the estimates.

12. Multiplicity, Missing Data and Analysis Populations

The trial has a primary outcome and several secondary outcomes, with pairwise comparisons among treatment approaches and overall analyses for some secondary measures. Testing multiple outcomes and comparisons raises the chance of at least one apparently unusual result arising by chance. The registry record does not report a multiplicity-control procedure, a prespecified testing hierarchy, or how confirmatory and exploratory analyses were distinguished. The reported secondary p-values should therefore be interpreted as unadjusted evidence rather than as results controlling the overall false-positive risk.

Information about missing data is limited. The primary analysis population is described as patients who met return-to-sport criteria during the study, while the registry does not state how participants who did not meet criteria were handled in the Cox analyses. For the secondary outcomes, the number of observations differs by measure. Reasons for missing assessments, imputation methods, and sensitivity analyses are not reported in the registry record.

The registry record does not explain whether analyses followed randomized assignment regardless of treatment received, whether there was a per-protocol analysis, or how departures from the assigned treatment were handled. It does not report whether the analysis populations were defined before examining outcomes. These details matter because excluding participants or handling missing observations differently can change the groups represented by an analysis and the interpretation of its results.

Randomization is a design strength in principle, but it does not by itself ensure that the posted analysis preserves the protections of random assignment. Clear reporting of analysis populations, participant flow, missingness, and prespecified multiplicity handling would help readers assess the risk of bias and the reliability of the estimates. Those details are not reported in the registry record.

13. 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.

Most frequent non-serious adverse event terms

The registry reports no serious adverse events in any of the three groups and no deaths. For other adverse events, it reports 14 participants affected among 16 at risk in the dexamethasone group, 10 among 15 in the sodium chloride group, and 9 among 14 in the physical-therapy-alone group. These are the registry’s posted counts, not adjusted comparisons.

The counts indicate that non-serious adverse events were recorded, but the record does not provide the event terms, severity, timing, or attribution needed to describe their nature. No serious-event terms are listed. The number at risk for safety reporting differs from the number contributing to the primary outcome summaries, so these denominators should not be conflated.

Zero reported serious events in these groups does not demonstrate that serious events cannot occur. Safety interpretation depends on the size and duration of exposure, completeness of reporting, and the types of events assessed. The terminated trial and the limited posted detail mean that the safety figures should be read as a description of recorded events, not as a comprehensive comparison of safety between approaches.

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

The posted results describe a randomized comparison of three treatment approaches and report time-to-event analyses for meeting return-to-sport criteria. The sponsor’s Cox analyses show different point estimates across pairwise comparisons, with wide confidence intervals. The dexamethasone-versus-physical-therapy interval includes 1, while the other two reported intervals exclude 1. These findings describe statistical estimates under the reported analyses; they do not establish the size, clinical importance, or cause of any difference.

The descriptive means apply to participants who met the criteria during the study. They should not be read as average times for every randomized participant. Nor can the hazard ratios be translated into a number of days saved or added. The registry does not report the model’s event coding and reference convention in sufficient detail for a definitive directional translation, and a hazard ratio is not itself a difference in days.

The result does not show that treatments are equivalent. A nonsignificant comparison is not evidence of equivalence, and the registry record does not report an equivalence margin. Similarly, intervals that exclude 1 do not by themselves establish a meaningful advantage. The small primary analysis groups and broad intervals leave substantial uncertainty.

The secondary outcomes provide additional descriptive information but do not settle the primary question. Their analyses involve multiple outcomes, and no multiplicity adjustment is reported. Safety figures describe recorded adverse events, but do not establish comparative safety or rule out uncommon harms.

Finally, the trial is recorded as terminated, with enrollment of 45 and an approximate planned enrollment of 147 stated in the brief summary. The registry record does not explain the termination or the difference in enrollment figures. Together with the missing details on event handling, missing observations, analysis populations, and model checks, this limits how broadly the findings can be interpreted.

15. Why This Trial Matters Statistically

This trial illustrates why the outcome definition and analysis population matter as much as the treatment-group averages. The primary endpoint is a time-to-event measure, while the posted means are restricted to participants who met the return-to-sport criteria. Those quantities answer related but different questions. A mean among observed achievers does not incorporate the experience of participants who did not reach the event during follow-up.

The record also illustrates the difference between statistical labels and statistical evidence. The analyses are tagged as equivalence, but the record does not state an equivalence margin. Equivalence requires a design and analysis that can rule out differences beyond a prespecified acceptable range; a p-value that does not indicate a difference is not sufficient. Making this distinction helps prevent overinterpretation of conventional hypothesis tests.

Precision is especially important in small studies. Wide confidence intervals communicate that a point estimate is not known precisely, even when an interval excludes the null value. Conversely, an interval that includes the null does not demonstrate that no difference exists. Readers need to consider the full interval and the analysis assumptions, rather than treating a single p-value as a complete summary.

Finally, termination, incomplete reporting of participant flow, and the contrast between recorded enrollment and the approximate planned enrollment underscore the importance of transparent reporting. Details about missing outcomes, event counts, analysis populations, model assumptions, and multiplicity are essential for evaluating what a trial can support. Their absence does not invalidate the recorded results, but it narrows the conclusions readers can draw from them.

16. 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 limitations specific to this trial affect interpretation. The registry records a terminated status and an enrollment of 45, while the brief summary says approximately 147 patients would be enrolled. The record does not explain the termination or the difference between these entries. Without that context, readers cannot determine whether the termination was related to recruitment, operations, safety, efficacy, or another reason.

The primary descriptive groups are small and include only patients who met the return-to-sport criteria during the study. The registry does not report how many randomized participants failed to meet criteria or how they were handled in the time-to-event analyses. The posted means may therefore represent a selected subset rather than all randomized participants. The record also does not provide event counts, censoring details, or tests of the proportional hazards assumption.

The registry’s equivalence designation is not accompanied by an equivalence margin or enough methodological detail to establish that an equivalence analysis was conducted. Randomization and masking procedures are not described beyond the overall design labels. Allocation concealment and the identities of the masked roles are not reported in the registry record.

For secondary outcomes, measurement counts vary and reasons for missing measurements are not stated. Multiplicity handling, analysis populations, protocol deviations, and missing-data sensitivity analyses are also not reported. Safety reporting includes counts, but not event terms or detailed timing. These limitations do not change the reported values, but they constrain assessment of precision, bias, and generalizability.

17. Related Tutorials

18. Related Statistical Calculators

19. Sources

• ClinicalTrials.gov: Iontophoresis With Dexamethasone and Physical Therapy to Treat Apophysitis of the Knee in Pediatrics — NCT03606980.
• Registry record last updated 2024-03-06; status last verified 2024-02.
• Results first posted 2024-03-06.
• 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.

20. Record Summary

The registry describes a randomized, triple-masked, parallel-group Phase 2 trial in pediatric patients with apophysitis of the knee. It compared dexamethasone iontophoresis plus physical therapy, sodium chloride iontophoresis plus physical therapy, and physical therapy alone. The trial status is terminated, and recorded enrollment is 45; the brief summary states that approximately 147 patients would be enrolled.

The primary outcome was time in days to meet return-to-sport criteria. The sponsor reported pairwise Cox proportional hazards analyses, while descriptive means and standard deviations are posted for participants who met the criteria. Secondary outcomes included activity, pain, function, and adverse events. The registry reports no serious adverse events or deaths, alongside other adverse-event counts.

Wide primary confidence intervals, small analysis groups, and unreported details about termination, event handling, missing data, analysis populations, and multiplicity limit interpretation. The registry’s equivalence label is not accompanied by a reported equivalence margin.