This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. The registry reports that results have been posted, but the ClinicalTrials.gov record does not contain formal statistical analyses for the registered primary endpoint.
1. Trial at a Glance
CAMP was a randomized, triple-masked, parallel phase 3 trial in pulmonology evaluating three interventions—placebo, nedocromil, and budesonide—in the context of childhood asthma. The registered primary endpoint was pulmonary function measured by normalized FEV1 over a 4-6 year period.
| Feature | CAMP |
|---|---|
| Trial name | Childhood Asthma Management Program (CAMP) Phases I (Trial), II (CAMPCS), III (CAMPCS/2), and IV (CAMPCS/3) |
| Phase | Phase 3 |
| Status | Completed |
| Therapeutic area | Pulmonology |
| Conditions | Asthma; Lung Diseases |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Triple |
| Primary purpose | Treatment |
| Enrollment | 1041 |
| Interventions | Placebo (drug); Nedocromil (drug); Budesonide (drug) |
| Lead sponsor | Johns Hopkins Bloomberg School of Public Health |
| Start | 1991-09 |
| Primary completion | 1999-10 |
| ClinicalTrials.gov | NCT00000575 |
2. Clinical Question
The central statistical question is whether pulmonary function, measured as change in FEV1 % of predicted from baseline to the end of treatment, can be compared across the three randomized intervention groups over the 4-6 year treatment period.
Population
The trial concerns childhood asthma, with the condition listed as asthma and lung diseases in the registry record.
Interventions
The three registered interventions were placebo, nedocromil, and budesonide.
Comparator
Placebo is one of the three randomized trial arms and provides a direct comparison group for the active interventions.
Primary question
How does pulmonary function, expressed as change in FEV1 % of predicted, differ among the randomized intervention groups at the end of treatment?
3. Trial Design
Budesonide
- Registered intervention: Budesonide (drug)
- Serious AEs: 0/311 affected/at risk
Nedocromil
- Registered intervention: Nedocromil (drug)
- Serious AEs: 1/312 affected/at risk
Placebo
- Registered intervention: Placebo (drug)
- Serious AEs: 4/418 affected/at risk
4. Registered Primary Endpoint
| Endpoint | Registry definition | Time frame | Type |
|---|---|---|---|
| Pulmonary Function as Measured by Normalized FEV1 Over a 4-6 Year Period | Change in FEV1 % of predicted, post-bronchodilator use, from baseline to the end of treatment (4-6 years after randomization). Percent predicted determined from three separate published sets of reference equations for white, black, and Hispanic children. | At the end of treatment, 4-6 years from baseline assessment | Binary as listed in the ClinicalTrials.gov record |
The endpoint definition contains an important statistical distinction. The underlying measurement described by the registry is a change in FEV1 % of predicted, which is naturally a quantitative change score. The ClinicalTrials.gov record nevertheless identify the registered primary endpoint type as Binary. That distinction should be preserved rather than silently replacing the registry classification with an inferred analysis type.
5. Statistical Analysis Status
ClinicalTrials.gov indicates that results have been posted for CAMP and that seven outcome measures are posted. However, the registry-reported statistical-analysis section contains no formal statistical analyses for the registered primary endpoint: the ClinicalTrials.gov record has no formal statistical analyses and reports zero analyses omitted for length.
What can be concluded from the ClinicalTrials.gov record
The ClinicalTrials.gov record does not provide a treatment-effect estimate, confidence interval, or p-value for the registered primary endpoint.
6. Planned Analysis
Because the ClinicalTrials.gov record contains no formal statistical analysis for the registered primary endpoint, the appropriate focus is the analysis that would ordinarily be used for an endpoint of this type, while keeping separate what is documented from what is methodological explanation.
What the registry says will be measured
The primary endpoint is pulmonary function as measured by normalized FEV1 over a 4-6 year period. More specifically, the registry defines it as change in FEV1 % of predicted, post-bronchodilator use, from baseline to the end of treatment 4-6 years after randomization.
How a quantitative change endpoint would typically be analyzed
If the endpoint is retained as the quantitative change score described in the definition, a conventional comparison would estimate the difference in mean change among the randomized groups, using an analysis appropriate for a continuous outcome. With three randomized groups, the overall comparison could be framed as a one-way analysis of variance or, when baseline FEV1 is incorporated explicitly, an ANCOVA comparing end-of-treatment pulmonary function while adjusting for baseline pulmonary function.
The treatment effect would quantify the difference in average change between two groups. With three arms, the overall analysis would first address whether the groups differ before focusing on specific pairwise comparisons.
Why baseline adjustment can matter
When a trial measures a continuous outcome both before and after treatment, baseline adjustment can improve precision when baseline values are associated with follow-up values. An ANCOVA-style analysis would use the baseline measurement as a covariate rather than relying only on raw post-treatment differences.
Why the registry classification should not be overwritten
The ClinicalTrials.gov record explicitly labels the primary endpoint type as binary even though its accompanying definition describes a continuous change in FEV1 % of predicted. These are not interchangeable descriptions. A production statistical analysis should follow the prespecified statistical analysis plan or protocol to determine whether the endpoint was transformed, categorized, or otherwise analyzed as binary. The ClinicalTrials.gov record does not provide that information.
7. Statistical Methodology
Randomization
Randomization is the foundation of the comparative design. In a randomized trial, assignment is intended to prevent systematic selection of participants into particular intervention groups. This supports causal interpretation of differences between randomized groups, subject to the quality of implementation, adherence, follow-up, and analysis.
Parallel-group analysis
The registry identifies CAMP as a parallel-design trial. In a parallel trial, the treatment groups are compared across participants assigned to different interventions. The statistical comparison therefore concerns between-group differences rather than within-person treatment-period contrasts.
Continuous change from baseline
The endpoint definition describes a change score: the end-of-treatment FEV1 % of predicted minus the baseline FEV1 % of predicted. Change scores are useful because they directly describe how the pulmonary-function measurement moved during treatment.
For each participant, the conceptual change score compares pulmonary function at the end of treatment with the baseline assessment. The registry specifies that the endpoint is evaluated 4-6 years after randomization.
FEV1 % of predicted
Using percent predicted standardizes measured FEV1 relative to reference values. The registry specifically states that percent predicted was determined from three separate published sets of reference equations for white, black, and Hispanic children. This means the definition of the endpoint incorporates reference-equation methodology rather than relying only on the raw pulmonary-function measurement.
Three-arm comparisons
With placebo, nedocromil, and budesonide, the trial has more than one active-versus-control comparison available. A statistically careful analysis must distinguish the overall question of whether the three groups differ from individual pairwise comparisons. If multiple pairwise hypotheses are formally tested, multiplicity becomes relevant.
Confidence intervals
A confidence interval would quantify uncertainty around an estimated treatment difference or other effect measure. It is more informative than a p-value alone because it communicates both the estimated direction and the range of values compatible with the specified statistical model and confidence level.
P-values
A p-value addresses the compatibility of the observed data with a specified null hypothesis under the statistical model. It does not measure the size of an effect, the clinical importance of an effect, or the probability that a treatment is effective.
8. Statistical Methods Explained
Why was randomization important?
Randomization makes the intervention groups comparable in expectation before treatment is administered. That is particularly important when the endpoint is measured years after baseline, because many characteristics can influence pulmonary function. Randomization does not guarantee numerically identical groups, but it provides the central design mechanism for reducing systematic allocation bias.
Why measure change from baseline?
A change score directly asks how pulmonary function moved during the treatment period. This can be more informative than examining only the final measurement because participants may begin the study with different levels of pulmonary function.
Why express FEV1 as percent predicted?
FEV1 % of predicted places an observed pulmonary-function value on a standardized scale relative to reference equations. The CAMP registry definition specifically identifies separate reference-equation sets for white, black, and Hispanic children, making the reference standard part of the endpoint definition.
Why does the 4-6 year time frame matter?
The primary endpoint is not an immediate post-treatment measurement. It is defined at the end of treatment, 4-6 years after baseline assessment. A statistical comparison therefore summarizes a relatively long treatment period rather than a single short-term change.
Why would ANCOVA be considered for a continuous pulmonary-function endpoint?
ANCOVA can compare treatment groups while accounting for baseline pulmonary function. Because baseline and follow-up measurements of the same physiological variable are often related, baseline adjustment can increase precision. However, the registry-reported CAMP data do not state that ANCOVA was actually used, so this is a methodological explanation rather than a claim about the original analysis.
Why does a three-arm trial raise a multiplicity question?
Three groups create several possible pairwise comparisons. If each comparison is tested independently at the same nominal significance level, the chance of at least one false-positive result across the family of tests can exceed the nominal level. The appropriate multiplicity strategy depends on the prespecified hypotheses and analysis plan, which are not reported here.
9. Safety Results
The ClinicalTrials.gov record provides serious adverse events by randomized arm as affected participants divided by the number at risk. These figures are descriptive safety information; the ClinicalTrials.gov record does not provide a formal comparative statistical analysis of serious adverse events.
| Arm | Serious adverse events | Interpretation |
|---|---|---|
| Budesonide | 0/311 | No affected participants among 311 at risk in the ClinicalTrials.gov record. |
| Nedocromil | 1/312 | One affected participant among 312 at risk in the ClinicalTrials.gov record. |
| Placebo | 4/418 | Four affected participants among 418 at risk in the ClinicalTrials.gov record. |
The ClinicalTrials.gov record is counts, not treatment-effect estimates. The figures show the number affected and the corresponding number at risk in each arm, but they do not provide a confidence interval, hypothesis test, or adjusted comparison. In particular, the counts alone should not be converted into a claim that one intervention has a statistically different serious-adverse-event rate from another.
The denominators also matter. A count of four events cannot be interpreted independently of the 418 participants at risk in the placebo arm. Similarly, the absence of an event in the 311-person budesonide safety denominator is a descriptive observation rather than proof that the true event probability is zero.
10. Clinical Biostats Interpretation of the Primary Endpoint
The registered primary endpoint asks about change in FEV1 % of predicted from baseline to the end of treatment. Thus, the relevant quantity is change over time, not simply the final FEV1 value.
No mean change, treatment difference, standard error, confidence interval, test statistic, or p-value is posted on ClinicalTrials.gov for the registered primary endpoint. Therefore, the magnitude and statistical uncertainty of any intervention effect cannot be reconstructed from the provided record.
Even if a p-value were available, it would not describe how large the pulmonary-function difference was. An effect estimate and its confidence interval would be needed to understand magnitude and precision. The registry-reported CAMP data contain neither.
The registry's endpoint-type field says Binary, while the detailed endpoint definition describes a continuous change in FEV1 % of predicted. That discrepancy should be resolved using the original protocol or statistical analysis plan before asserting a specific inferential model.
11. Design Features That Affect Statistical Interpretation
Blinding
The trial was triple-masked. Masking can reduce the influence of treatment knowledge on trial behavior, assessment, and reporting.
Randomization
The allocation was randomized, providing the core framework for comparing the three intervention groups.
Parallel design
The parallel model means the statistical comparison is fundamentally between randomized groups rather than sequential treatments within the same participants.
Long follow-up
The primary endpoint was assessed 4-6 years after baseline, making follow-up duration central to the interpretation of the pulmonary-function outcome.
Multiplicity
The presence of three randomized arms creates multiple possible comparisons. The ClinicalTrials.gov record does not specify a multiplicity-control procedure, hierarchy of hypotheses, or familywise error strategy. Consequently, this page does not assign a confirmatory interpretation to any particular pairwise comparison.
Interim analysis
The ClinicalTrials.gov record does not report an interim analysis, alpha-spending procedure, stopping boundary, or other group-sequential method. No such procedure is inferred here.
Missing data and imputation
The ClinicalTrials.gov record does not report a missing-data strategy or imputation method for the primary endpoint. This is particularly relevant for a 4-6 year endpoint because participants may not all contribute an end-of-treatment measurement. The appropriate handling of incomplete follow-up should therefore be taken from the trial's prespecified statistical documents.
Stratification
The ClinicalTrials.gov record identifies randomization but do not specify stratification factors. No stratified analysis is therefore described.
Bayesian methods
The ClinicalTrials.gov record does not identify a Bayesian analysis. No posterior probability, prior distribution, Bayesian model, or credible interval is reported.
12. Limitations
- No formal primary analysis reported: no formal statistical analyses were posted, so the effect estimate and inferential results for the registered primary endpoint cannot be reported from the ClinicalTrials.gov record.
- Endpoint-type discrepancy: the endpoint is classified as binary in the ClinicalTrials.gov record even though the definition describes change in a continuous pulmonary-function measure.
- No baseline characteristics: the ClinicalTrials.gov record does not provide a baseline demographic or pulmonary-function table.
- No subgroup results: the ClinicalTrials.gov record contains no subgroup estimates or interaction analyses.
- No missing-data method: the ClinicalTrials.gov record does not specify how incomplete 4-6 year pulmonary-function assessments were handled.
- No multiplicity procedure: the ClinicalTrials.gov record does not state how multiple intervention comparisons were controlled statistically.
- No formal safety comparison: serious-adverse-event counts are available by arm, but no comparative statistical analysis is reported.
- Reference-equation dependence: FEV1 % of predicted depends on the reference equations specified in the endpoint definition, so interpretation is tied to the chosen reference standards.
13. Why This Trial Matters Statistically
CAMP is a useful teaching example because it combines a randomized three-arm design with a long-term continuous physiological endpoint and a reference-standardized outcome. It also illustrates an important practical issue in clinical-trial analysis: the registry can tell us what was measured without necessarily supplying the complete inferential analysis needed to quantify treatment effects.
| Concept | How it appears in CAMP |
|---|---|
| Randomization | Participants were assigned using a randomized allocation. |
| Three-arm design | The interventions were placebo, nedocromil, and budesonide. |
| Parallel groups | The registry identifies the design model as parallel. |
| Triple masking | The registry identifies the masking as triple. |
| Continuous physiological measurement | The primary endpoint is defined using change in FEV1 % of predicted. |
| Baseline change | The primary endpoint compares baseline with the end-of-treatment measurement. |
| Long-term endpoint | The primary endpoint is assessed 4-6 years after baseline assessment. |
| Reference equations | Percent predicted uses three separate published reference-equation sets identified in the registry definition. |
| Multiplicity | Three arms create multiple possible comparisons, although the ClinicalTrials.gov record does not specify an adjustment procedure. |
| Safety denominators | Serious adverse events are reported as affected participants divided by participants at risk in each arm. |
| Incomplete statistical record | Results are posted, but no formal statistical analyses were posted. |
14. What a Complete Statistical Report Would Ideally Provide
The registry-reported CAMP record establishes the primary endpoint and trial architecture, but a complete statistical report would need additional information to quantify the treatment comparison. For this endpoint, the most useful additional outputs would include the number of participants contributing evaluable baseline and end-of-treatment measurements, the distribution of FEV1 change in each arm, the prespecified inferential model, the estimated treatment differences, confidence intervals, and the associated multiplicity strategy.
Because the registry describes the endpoint as a change in FEV1 % of predicted, the analysis should also make clear whether the prespecified model used the change score directly, adjusted the end-of-treatment value for baseline FEV1, or converted the outcome into a binary category. The ClinicalTrials.gov record does not answer that question.
A treatment comparison becomes much easier to interpret when the effect estimate, its uncertainty, the exact hypothesis being tested, and the population included in the analysis are all reported together.
15. Planned Analysis vs Reported Results
| Item | What the ClinicalTrials.gov record supports |
|---|---|
| Trial status | Completed |
| Results posted | Yes |
| Outcome measures posted | 7 |
| Registered primary endpoint | Pulmonary Function as Measured by Normalized FEV1 Over a 4-6 Year Period |
| Primary endpoint time frame | At the end of treatment, 4-6 years from baseline assessment |
| Formal statistical analyses in the ClinicalTrials.gov record | None |
| Primary treatment-effect estimate | Not reported |
| Primary confidence interval | Not reported |
| Primary p-value | Not reported |
| Serious adverse events | Reported descriptively by arm |
This distinction prevents a common registry-analysis error: treating the presence of posted outcomes as evidence that the complete statistical comparison is available in the extracted dataset. Here, the ClinicalTrials.gov record supports a detailed description of the design and endpoint, but not a numerical efficacy conclusion.
16. Related Tutorials
Learn more about the methods used in this trial:
17. Related Statistical Calculators
18. Sources
- ClinicalTrials.gov: CAMP — NCT00000575.
- Linked PubMed record: PMID 42282026.
- Linked PubMed record: PMID 42010358.
- Linked PubMed record: PMID 38056635.
- Linked PubMed record: PMID 32118022.
- Linked PubMed record: PMID 31187518.
Continue through the Clinical Biostats knowledge graph
Connect randomized-trial design and continuous-outcome methodology to deeper statistical tutorials and practical calculators.
19. Record Summary
CAMP is a randomized, triple-masked, parallel phase 3 trial with 1041 participants and three intervention arms: placebo, nedocromil, and budesonide. Its registered primary endpoint is pulmonary function measured by normalized FEV1 over a 4-6 year period, defined as change in FEV1 % of predicted, post-bronchodilator use, from baseline to the end of treatment.
The statistical lesson is as important as the clinical endpoint itself. The ClinicalTrials.gov record provides enough information to reconstruct the trial's randomized structure, masking, intervention groups, endpoint definition, time frame, and descriptive serious-adverse-event counts. It does not, however, provide the formal primary statistical analysis.