The Effect of Per Oral Immunotherapy in Severe Milk, Peanut and Egg Allergy in Adults: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for The Effect of Per Oral Immunotherapy in Severe Milk, Peanut and Egg Allergy in Adults (NCT03361072), a non-randomized na study. The primary outcome is Desensitisation.

2. Clinical Question

This question concerns a change in food tolerance after treatment, rather than a randomized comparison of OIT against an alternative treatment or no treatment. Milk, peanut and egg are recorded in separate allergy groups. Those groups differ in the food being treated and are not described as interchangeable comparison groups. The study’s summary also mentions lung function follow-up before OIT and a year after OIT, and a secondary outcome concerning worries associated with food allergy. The primary outcome record does not specify how an increase in tolerated dose was operationalized beyond the description of no symptoms if accidental exposure occurred. The registry record does not report the outcome results, the analysis population, or a statistical comparison against a control group.

3. Trial Design

The study, NCT03361072, is titled “The Effect of Per Oral Immunotherapy in Severe Milk, Peanut and Egg Allergy in Adults.” It is listed as completed, with an enrollment of 30. The recorded start date is 2017-08-03 and the primary completion date is 2023-03-11. Its phase is listed as NA. Helsinki University Central Hospital is named as the sponsor.

The registry describes a non-randomized, unmasked, parallel design. The arms are labeled Milk allergy, Peanut allergy and Egg allergy, and each is associated with oral immunotherapy for that allergy. The registry cautions that this is not a randomized comparison and that the arms are not a like-for-like control comparison. In particular, the groups concern different foods, so differences between them would not by themselves isolate an effect of treatment. The record does not describe an untreated parallel control arm.

The brief summary states that allergy diagnosis is verified using a positive history, skin prick tests and egg- and milk-specific IgE antibodies. It describes an open-label challenge test for milk allergy and blind challenge tests for peanut and egg allergy. These details describe the diagnosis verification in the summary; further information about challenge procedures is not reported in the registry record. The summary says OIT is carried out according to a detailed plan but does not give that plan’s dosing or escalation details.

The primary outcome is described over “1 year, through study completion.” The summary also says lung function parameters are followed before OIT and a year after OIT, and lists worries associated with food allergy as a secondary outcome. The registry record does not report the statistical analysis plan, allocation procedures beyond the non-randomized designation, masking beyond none, or the handling of participants who did not complete follow-up.

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 Desensitisation, measured over 1 year, through study completion.

The primary endpoint is titled “Desensitisation” and is measured as a count of participants. Its description defines the event as an increase in tolerated food dose compared with the pre-OIT level, with no symptoms if accidental exposure occurs. The listed timeframe is “1 year, through study completion.” Thus, the endpoint is intended to record how many participants meet the stated increase criterion, rather than the average dose tolerated or the size of each participant’s change.

A count can be useful when the clinical question is whether participants reach a specified outcome: it gives a direct tally of those meeting the registry’s definition. However, interpreting a count requires knowing the relevant group size, the exact operational criteria, and the time point used to determine whether the outcome occurred. The registry gives the outcome’s unit and general description, but the observed outcome counts are not reported in the registry record. It also does not report a more detailed threshold for an increase, how symptom assessment was conducted for this endpoint, or how participants without an evaluable one-year assessment were handled.

Because the outcome is defined against pre-OIT tolerance, the reference is within the participant rather than a separate randomized control group. A difference from baseline can describe a change over time, but in a non-randomized study it does not, on its own, establish that treatment caused that change. The endpoint’s result and its uncertainty are not reported in the registry record, so the magnitude and precision of the observed outcome cannot be assessed from the posted results.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry identifies the primary outcome as a participant count and the study design as non-randomized, unmasked and parallel. These labels describe the planned structure and type of outcome; they do not specify which statistical model was used to analyze the results. The number of analyses recorded is 0, and no analysis results are listed. The statistical analysis plan is not reported in the registry record.

What does a participant count represent?

For a binary outcome, each participant is classified according to whether the specified event occurred. A count records the number classified as having the event. To interpret that count, readers generally also need the denominator: the number of participants assessed for the outcome in the relevant group. A proportion is often used to relate the event count to that denominator, but no proportion should be inferred when the denominator and outcome count are not both reported for the endpoint. Here, the primary endpoint is described as a count of participants, and its observed counts and endpoint-specific denominators are not reported in the registry record.

The endpoint wording makes an increase in tolerated dose relative to pre-OIT the basis for classifying the outcome. A clear binary endpoint generally needs operational rules for how much increase qualifies, when the assessment takes place, and how symptoms or an unevaluable assessment affect classification. The registry provides the broad endpoint description and timeframe, but further operational definitions are not reported in the registry record. Those details matter because different classification rules can change which participants count as having the event.

Why does the design affect interpretation?

Random allocation is a method for creating comparison groups that, in expectation, differ less systematically in characteristics at baseline. This study is explicitly non-randomized, and its arms correspond to different food allergies rather than a like-for-like control comparison. Therefore, any contrast between those groups would mix the difference in allergy category with other possible differences between participants and treatment contexts. The registry does not report a separate randomized control group. A within-participant comparison with pre-OIT tolerance answers a different question from a randomized comparison, and changes over time cannot alone rule out other explanations.

Unmasked means the study is listed as having no masking. In general, masking can reduce the influence of knowledge about treatment assignment on some assessments or reporting. Its relevance depends on the outcome and how it is measured. The record does not describe masking for the endpoint assessment beyond the overall designation of none, and does not report whether outcome assessment followed additional safeguards. The challenge tests described for diagnosis should not be confused with a reported masking plan for the primary endpoint.

How should uncertainty be assessed?

General statistical interpretation considers both an estimated result and its uncertainty. A confidence interval is a range calculated using a specified method to represent uncertainty around an estimate. It does not guarantee that the true value lies within the interval, nor does it correct for bias in study design or measurement. A narrow interval generally signals greater precision than a wide interval, under the assumptions of the method. No primary-outcome estimate or confidence interval is reported in the registry record, so precision cannot be evaluated.

A p-value, when reported, summarizes how unusual data at least as inconsistent with a specified null hypothesis would be under the assumptions of a statistical test. It does not measure the probability that a treatment works, nor the size or clinical importance of an effect. The registry record does not report a p-value or identify a test for the primary outcome. No inference about statistical significance can be made from the record.

Counts and proportions also have distributional considerations. For a binary outcome, a binomial model is one general approach when observations are independent and share the relevant probability within a group. Over-dispersion means variability is greater than the model expects; if present, a model that fails to account for it can understate uncertainty. These are general concepts, not methods reported for this study. The registry record does not identify a model, assumption checks, or a method for handling any extra variability.

What do the planning assumptions mean?

The record includes assumptions for a planning illustration: a two-sided significance level of 0.05, equal allocation between two arms, a normal approximation, and effects expressed as a stated fraction of an observed effect. These assumptions describe an illustration, not the trial’s own design or an observed analysis. The illustration’s numerical sample-size outputs are not reported in the registry record. A planning sample size is not observed power, and it cannot substitute for an analysis of the actual results. The assumptions also do not establish that this study was planned with those allocation or analysis features; the registry describes the trial itself as non-randomized and with allergy-specific arms.

Overall, the statistical methods that were actually used are not reported in the registry record. The design labels and endpoint definition are available, but they do not disclose an analysis model, covariate adjustment, uncertainty calculation, or treatment of missing observations. Without those details and the primary outcome results, readers cannot reproduce or evaluate a quantitative treatment-effect 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("Milk Allergy", "Peanut Allergy", "Egg Allergy")
n      <- c(12, 14, 4)
events <- c(7, 5, 2)
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  Milk Allergy  12  7  58.3  27.7  84.8  32.0  80.7  31.6  80.8
ARM  Peanut Allergy  14  5  35.7  12.8  64.9  16.3  61.2  16.3  61.6
ARM  Egg Allergy  4  2  50.0  6.8  93.2  15.0  85.0  14.7  85.3

11. Multiplicity, Missing Data and Analysis Populations

The registry lists a secondary outcome concerning worries associated with food allergy, as well as the primary endpoint of desensitisation. Its summary also mentions lung function parameters followed before OIT and a year after OIT. The record does not report results or statistical analyses for these measures, and it does not describe a prespecified hierarchy for testing multiple outcomes. Multiplicity methods, such as an adjustment to account for testing several hypotheses, are not reported in the registry record. It is therefore not possible to tell whether any reported statistical conclusions would account for multiple outcomes or analyses.

The number of analyses recorded is 0. This does not establish that no analyses were conducted; it means the registry record does not list analyses. The statistical analysis plan, including any interim analyses or stopping rules, is not reported in the registry record. No assumptions about such procedures can be made from the design description alone.

Missing data handling is also not reported. The record does not state how many participants completed the primary endpoint assessment, how missing one-year measurements were classified, or whether any imputation or sensitivity analysis was used. These details are important because the endpoint is based on whether participants meet a criterion at follow-up. If participants without follow-up differ systematically from those assessed, the observed outcomes among assessed participants may not represent the enrolled group. This is a general concern, not a claim that missingness occurred in this study.

The analysis population is not reported in the registry record. There is no description of whether analyses included all enrolled participants, only participants who began OIT, or another defined group. The registry’s enrollment figure should not be assumed to be the denominator for every outcome. Without these details, the scope and robustness of any endpoint result cannot be judged.

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.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: allergic reaction (33.3%).

The serious adverse-event entries are reported separately for Milk Allergy, Peanut Allergy and Egg Allergy. For milk, the record lists 12 at risk and 4 affected for serious events; for peanut, it lists 14 at risk and 2 affected; for egg, it lists 4 at risk and 0 affected. The record lists 0 deaths in each group. These are registry entries for participants at risk and participants affected, not a randomized comparison of treatment against a control group.

The record lists “Allergic reaction” as the top serious term at 33.3%. The denominator or calculation behind that percentage is not reported in the registry record, so its precise basis cannot be established from the entry. It should not be combined with the group-level serious-event counts or treated as a group-specific rate without further information.

Other adverse-event entries list 12 at risk and 8 affected for milk, 14 at risk and 12 affected for peanut, and 4 at risk and 4 affected for egg. These are separate from the serious-event figures. Affected-participant counts do not necessarily indicate the number of event episodes; the registry record does not report episode counts or further detail about severity, timing, relatedness to treatment, or how events were collected.

These figures are descriptive safety information. Because the study is non-randomized and the allergy groups are not like-for-like controls, the entries do not establish comparative safety between treatments or groups. Further context, including exposure time and event definitions, is not reported in the registry record.

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

The record establishes that the study examined per oral immunotherapy in adults with severe milk, peanut or egg allergy and defined a primary outcome based on an increase in tolerated dose relative to pre-OIT. It also records the study as completed and reports safety counts by allergy group. These registry entries describe the study question and the information entered in the registry.

The record does not report the primary outcome results. It therefore cannot show how many participants met the desensitisation definition, the size of any change, or how precise an estimate might be. No confidence interval, p-value or analysis result is reported. Missing results should not be interpreted as evidence of either benefit or lack of benefit.

The study’s non-randomized design also limits causal interpretation. The pre-OIT reference allows a before-and-after description if those measurements are analyzed, but a change over time is not by itself proof that OIT caused the change. The three arms concern different food allergies and are explicitly not like-for-like controls. Comparisons across them cannot be read as randomized treatment comparisons.

The safety entries show that serious and other adverse events were recorded, with affected-participant counts varying across the allergy groups. Without event definitions, exposure time, relatedness assessments and a suitable control comparison, those counts cannot establish comparative risk. The listed allergic-reaction percentage also lacks a reported denominator in the registry record.

In general, a study’s design and outcome definitions help frame what it could answer, while results and analysis details are needed to assess what it did show. For this record, the intended endpoint is identifiable, but the posted information is not sufficient to quantify the primary outcome, assess its statistical uncertainty, or attribute an observed change to treatment.

14. Why This Trial Matters Statistically

This study illustrates why the statistical question must be matched to the design. The primary endpoint asks whether participants’ tolerated dose increased relative to their pre-OIT level. That is a within-participant change question. It is not the same as asking whether treatment outperforms an untreated or alternative-treatment group. Because the registry identifies the study as non-randomized and says the arms are not like-for-like controls, the design does not support treating the allergy groups as randomized comparators.

The outcome is expressed as a participant count, which makes the endpoint definition and denominator especially important. Readers need to know what qualifies as an increase, which participants were assessed at follow-up, and how missing assessments were handled. The registry describes the general criterion and timeframe, but the observed endpoint counts, analysis population and missing-data approach are not reported. Without those elements, neither an event frequency nor its precision can be evaluated.

The record also demonstrates the difference between a planned question and an available result. A completed status and an enrollment figure do not reveal the primary outcome. Likewise, safety counts provide useful descriptive context but do not replace the efficacy endpoint or establish a treatment comparison. Statistical interpretation depends on both numerical results and information about how they were obtained.

For clinical researchers, these distinctions matter when designing future studies and when reading registry records. A clear comparator, explicit outcome rules, prespecified analysis methods, and transparent reporting of denominators and uncertainty help readers distinguish descriptive change from a causal treatment effect. The available record identifies important parts of the research question, while leaving the primary quantitative findings and analytic approach unreported.

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 central limitation for interpreting this record is that the primary outcome results are not reported in the registry record. The record gives the endpoint definition and timeframe, but not the number of participants meeting the definition, endpoint-specific denominators, or measures of uncertainty. As a result, the magnitude and precision of the primary outcome cannot be evaluated.

The study is non-randomized, unmasked and organized into allergy-specific arms. The registry cautions that these arms are not like-for-like controls. The record does not describe a separate randomized or untreated comparator, so it does not provide a basis for a randomized causal contrast. Pre-OIT tolerance is the reference described for the endpoint, but the registry does not report the analysis of change from that reference.

Important details of outcome measurement are not reported, including a more precise threshold for an increase in tolerated dose, the endpoint assessment procedures, and how symptom reports were used to classify the outcome. The record also does not report an analysis population, completion counts for the primary endpoint, missing-data methods, or sensitivity analyses. Without these details, it is difficult to assess how the endpoint was constructed or how representative the observed results would be.

The statistical analysis plan and methods are not reported. The record does not identify the model, any adjustment for participant characteristics, the approach to uncertainty, or how multiple outcomes were handled. Although the study summary mentions lung function follow-up and a secondary outcome concerning worries associated with food allergy, their results are not reported. Further limitations may exist, but they cannot be assessed from the registry information.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: The Effect of Per Oral Immunotherapy in Severe Milk, Peanut and Egg Allergy in Adults — NCT03361072.
• Registry record last updated 2025-05-09; status last verified 2025-05.
• Results first posted 2025-05-09.
• 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

NCT03361072 is a completed, non-randomized, unmasked, parallel study of oral immunotherapy for severe milk, peanut or egg allergy in adults. Enrollment is listed as 30, and Helsinki University Central Hospital is named as sponsor. The registry records a start date of 2017-08-03 and a primary completion date of 2023-03-11.

The primary endpoint, “Desensitisation,” is a count of participants with an increase in tolerated food dose compared with pre-OIT, assessed over “1 year, through study completion.” The record describes milk, peanut and egg allergy groups and notes that they are not like-for-like control arms. It also lists adverse-event entries by group and a secondary outcome concerning worries associated with food allergy.

The primary outcome counts, statistical analysis methods, analysis population and missing-data procedures are not reported in the registry record. The record therefore describes the study’s question and design but does not permit a quantitative assessment of the primary result or its precision.