Compare Fentanyl Nasal Spray With Intravenous Opioids to Treat Severe Pain: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Compare Fentanyl Nasal Spray With Intravenous Opioids to Treat Severe Pain (NCT02459964), a randomized phase 4 study. The primary outcome is Non-inferiority of Fentanyl Nasal Spray Versus Intravenous Opioids in the Change in the Numeric Rating Scale (NRS) Pain Intensity Score at One Hour, Starting From the Time of Drug Delivery (Treatment Initiation)..

2. Clinical Question

3. Trial Design

The registry describes a completed, randomized, open-label, parallel-group phase 4 trial sponsored by M.D. Anderson Cancer Center. The trial title and brief description identify the comparison as fentanyl nasal spray versus intravenous hydromorphone for cancer-related severe pain among patients in an emergency department. The registry lists enrollment as 84. The reported treatment-arm counts are 42 for intranasal fentanyl and 40 for intravenous hydromorphone; the adverse-event record also has a no-medication group with 2 participants at risk. These registry counts should be read as posted, rather than reconstructed or combined.

Allocation was randomized, while masking was listed as none. In practical terms, the registry does not describe the study as blinded. It lists a parallel model, meaning the treatment arms were compared as groups rather than as a crossover in which participants switch between treatments. Details about the randomization procedure, concealment of allocation, and whether outcome assessment was blinded are not reported in the registry record.

The intranasal arm received fentanyl nasal spray 100 mcg at time 0, defined as the time the spray was administered. The arm description allowed a rescue dose at 0.5 hour. The intravenous arm received hydromorphone hydrochloride 1.5 mg by IV push at time 0, defined as completion of the opioid IV push, and also allowed a rescue dose at 0.5 hour. A study nurse was to call participants 24 hours after participation to ask about side effects. The registry does not report the criteria for rescue dosing, how often rescue medication was used, or whether use differed between arms.

The listed start date is 2015-09-14, the primary completion date is 2020-06-21, and the status is completed. The record does not describe the study’s sample-size justification, a protocol amendment history, or interim analyses. Those design details are not reported in the registry record.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Before assignment: One (1) participant withdrew consent before treatment and 1 participant was found ineligible due to abnormal EKG.

Overall Study

Reasons for not completing this period:

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 Non-inferiority of Fentanyl Nasal Spray Versus Intravenous Opioids in the Change in the Numeric Rating Scale (NRS) Pain Intensity Score at One Hour, Starting From the Time of Drug Delivery (Treatment Initiation)., measured over Baseline, One hour post time of drug delivery/treatment initiation.

The primary endpoint is the median change in Numeric Rating Scale pain intensity from baseline to one hour after treatment initiation. The registry defines the NRS as an 11-point Likert scale, with 0 meaning no pain and 10 meaning worst pain. It gives the assessment timeframe as baseline and one hour post time of drug delivery or treatment initiation. The one-hour outcome therefore concerns change over the stated period, rather than a pain score at an unspecified point during the emergency department visit.

A change score is used to express how a participant’s measured pain differs from the baseline assessment. Comparing change between randomized groups can address whether the groups’ reported pain trajectories differ over the specified treatment period. The registry labels the primary objective as non-inferiority, which is a particular question: whether the intervention is not worse than the comparator by more than a prespecified margin. The non-inferiority margin and the method for testing it are not reported in the registry record, so the objective’s statistical criterion cannot be assessed from the posted description alone.

The results list a median and a standard deviation for each treatment arm. The record does not explain the direction or coding convention for the change score, and it does not post a confidence interval or a between-group estimate for this primary outcome. Accordingly, the posted group summaries describe the reported values but do not by themselves establish the direction of improvement or answer the formal non-inferiority question.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The posted results include a continuous pain-change summary and a secondary count outcome. These require different statistical interpretations. The primary outcome is described using median change and standard deviation within each treatment arm. The secondary outcome counts participants meeting the registry’s description of having a change in NRS pain intensity. The two summaries should not be treated as interchangeable: a group summary of a change score is not the same estimand as the proportion classified as having a change.

What does a median summarize?

In general, the median is the central value when observations are ordered. It can describe the middle of a distribution without requiring the distribution to be symmetric. The registry reports median values for the primary outcome, but does not provide the underlying distribution, the individual observations, a confidence interval for the median, or a direct contrast between treatment arms. Consequently, the posted medians alone do not quantify uncertainty about the treatment-group comparison. The listed standard deviations describe spread, but do not supply the missing between-group inferential result.

The record’s use of a median alongside a standard deviation does not establish which test or model was used. The statistical method for the primary comparison is not reported in the registry record. It would be inappropriate to assume that the test used a particular distributional model, adjusted for baseline characteristics, or accounted for other variables. The record likewise does not report a non-inferiority margin or a confidence interval for the primary contrast.

How should the secondary risk difference be read?

For the secondary count outcome, the registry reports a risk difference of 0.131 for intravenous hydromorphone relative to intranasal fentanyl. In general, a risk difference is the contrast between the proportions in two groups for a binary outcome. The posted 95% confidence interval runs from -0.035 to 0.297. A confidence interval represents uncertainty in the estimated contrast under the calculation’s assumptions; it is not a range containing a known true value with certainty. Since this interval includes zero, the reported result does not clearly distinguish a nonzero difference from no difference at the conventional level indicated by the accompanying unadjusted p-value of 0.125.

The registry also posts a risk ratio of 2.1 with a confidence interval from 0.786 to 5.608, and an odds ratio of 2.467 with a confidence interval from 0.761 to 8.0. These are alternative scales for expressing the same binary comparison, not independent findings. Their intervals are wide and include the null value for each scale. That pattern is consistent with substantial uncertainty in the estimated association. It does not resolve the ambiguity in the outcome definition: the record does not state what amount or type of change was required for a participant to be counted.

The reported p-value is unadjusted. In general, a p-value measures how compatible the observed data are with a specified null model under the analysis assumptions; it is not the probability that the null hypothesis is true, nor a measure of clinical importance. The record does not describe adjustments for multiplicity or other covariates. The secondary result should therefore be treated as exploratory or descriptive rather than as a stand-alone confirmatory conclusion.

What do the planning illustrations mean?

The registry includes planning illustrations based on the observed secondary effect, not the trial’s own design. The stated assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. For planning at 100% of the observed effect, the reported figures are 137 participants per arm for 80% power and 183 participants per arm for 90% power. At 75% of the observed effect, the figures are 227 participants per arm for 80% power and 303 participants per arm for 90% power. At 50% of the observed effect, they are 471 participants per arm for 80% power and 630 participants per arm for 90% power.

These figures are hypothetical planning calculations under their stated assumptions. They are not the trial’s planned sample size, do not describe what was enrolled, and are not observed power. Using the observed effect as a planning target can be unstable, especially when that estimate is uncertain. The registry does not report a sample-size justification for the trial itself, so the planning illustrations should not be used to infer the original design rationale.

What is not reported about the analyses?

The registry does not report the analysis method for the primary endpoint, a primary treatment-effect estimate, its confidence interval, a non-inferiority margin, the analysis population, or a missing-data strategy. The record lists no reported secondary analyses, although it posts the secondary counts and comparison statistics. Details such as model covariates, handling of rescue medication, assumptions for the comparison, and sensitivity analyses are not reported in the registry record. These omissions matter because statistical conclusions depend not only on the observed summaries but also on how endpoints, participants, and uncertainty are handled.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists one primary endpoint and one secondary endpoint. It does not report a multiplicity adjustment, a testing hierarchy, or a plan specifying how conclusions across outcomes were to be controlled. In general, when several outcomes or comparisons are tested, the chance of at least one apparently unusual result can increase unless the analysis plan accounts for the number and structure of tests. The secondary comparison is explicitly unadjusted, so it should not be interpreted as if it had been protected by a reported multiplicity procedure.

Missing-data handling is not reported in the registry record. The posted arm counts identify participants contributing to the reported summaries, but the record does not describe whether any outcome assessments were missing, how missing values were treated, or whether the analysis used imputation, complete cases, or another approach. The outcome-specific analysis population is also not reported. No assumptions about missingness or exclusions should be inferred from the displayed counts.

The trial is described as randomized and open-label. The registry does not report whether the analysis followed an intention-to-treat principle, a per-protocol approach, or another analysis population. This is particularly relevant to the stated non-inferiority objective, because different analysis populations can affect the interpretation of a non-inferiority comparison. The record also does not report how participants who received rescue medication were handled in the primary analysis.

Details of allocation concealment, blinded outcome assessment, protocol deviations, and sensitivity analyses are not reported in the registry record. Nor is a statistical analysis plan described. These absences do not demonstrate that such procedures were not used; they mean the posted record does not document them. Without the analysis population, missing-data approach, and multiplicity strategy, readers cannot fully assess how robust or confirmatory the posted comparisons are.

11. 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 zero serious adverse events among the 42 participants at risk in the intranasal fentanyl group and zero among the 40 participants at risk in the intravenous hydromorphone group. The no-medication group lists 2 participants at risk and zero serious adverse events. No deaths are reported in any of these groups. These are event counts in the registry, not evidence that serious events are impossible or that the treatments have equivalent safety.

The record separately lists 1 other, non-serious adverse event in the intranasal fentanyl group and 3 in the intravenous hydromorphone group, with none in the no-medication group. The registry does not provide serious adverse-event terms, and no top serious-event terms are listed. It also does not report a formal comparative safety analysis, the event definitions, severity details for other events, or the follow-up completeness for safety assessment. The planned nurse call 24 hours after participation was intended to ask about side effects, but the record does not describe the completeness of those calls or the method used to classify events.

Rare-event counts are inherently limited in how precisely they describe risk, and zero observed events should not be read as zero possible risk. These figures are descriptive registry results, not a basis for a treatment recommendation.

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

The registry records a randomized comparison of intranasal fentanyl and intravenous hydromorphone for cancer-related pain in the emergency department. Its primary objective is described as non-inferiority in one-hour change in NRS pain intensity. The posted primary arm summaries are 5.14 with a standard deviation of 2.16 for intranasal fentanyl and 4.9 with a standard deviation of 2.31 for intravenous hydromorphone. Those values do not, on their own, report the treatment contrast or its uncertainty.

The title’s non-inferiority wording is not enough to conclude that non-inferiority was demonstrated. To interpret such a claim, readers need the prespecified margin and the relevant estimate and confidence interval or test result. Those details are not reported in the registry record. The record also does not define the direction of the change score, so the arm values cannot safely be translated into a statement about improvement based only on their numerical size.

The secondary count-based outcome gives a separate, unadjusted comparison. Its confidence interval includes no difference, and the registry does not explain the rule for classifying a participant as having a change. It should therefore be regarded as an uncertain secondary observation, not as a replacement for the primary analysis or as confirmation of the trial’s stated objective.

The adverse-event record contains zero serious adverse events and zero deaths in the listed groups, with a small number of other adverse events in the treatment arms. These observations are descriptive; they do not establish the absence of risk or a comparative safety advantage. The study record also lacks details needed to judge missing data, analysis populations, multiplicity, and the handling of rescue medication.

In general, a trial result is strongest when the outcome definition, estimand, analysis method, effect estimate, and uncertainty are reported together. In this record, several of those pieces are absent. The posted data can be described accurately, but they do not support a definitive statistical conclusion about whether intranasal fentanyl met the non-inferiority objective.

13. Why This Trial Matters Statistically

This trial is statistically notable because it poses a non-inferiority question rather than merely asking whether two groups differ. Non-inferiority is not established by observing similar-looking group summaries or by failing to find a statistically significant difference. The assessment depends on a prespecified margin and an analysis that estimates the treatment contrast with enough precision to rule out an unacceptable loss. The registry’s primary endpoint is framed in that way, but the margin and primary comparison results are not reported.

The study also illustrates why outcome summaries need to match the question being asked. The primary endpoint is described as median change in pain intensity, while the secondary endpoint counts participants with a change. A continuous or ordinal change summary and a binary classification can convey different aspects of the data. Without a definition for the binary threshold, the secondary outcome is particularly difficult to interpret. Clear endpoint definitions support reproducibility and help readers distinguish a clinically meaningful change from any recorded change.

The randomized, open-label parallel design provides the basic framework for comparing treatment groups, but the registry record does not report enough analysis detail to evaluate all potential sources of bias or uncertainty. The absence of reported information about missing data, analysis populations, rescue medication handling, and multiplicity limits the strength of conclusions that can be drawn from the posted outcomes.

Finally, the sample-size illustrations emphasize that planning depends on the effect size assumed and the statistical assumptions used. Their figures are hypothetical calculations based on the observed secondary effect, not evidence of the study’s original design or an assessment of observed power. The statistical value of the record is therefore not a definitive answer about the treatment comparison, but a useful example of why non-inferiority claims require transparent margins, analyses, and uncertainty estimates.

14. 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 registry record leaves important trial-specific information unreported. It does not provide detailed eligibility criteria, baseline pain distributions, cancer characteristics, or participant-level outcome data. This makes it difficult to assess how closely the enrolled population represents other emergency department patients with cancer-related pain. The record describes the setting and condition broadly, but those labels do not define the clinical range of the participants.

The primary endpoint is reported as a median change with a standard deviation, but no between-arm estimate, confidence interval, p-value, or testing method is posted. The direction convention for change is not stated. Most importantly for the stated non-inferiority question, the margin is not reported. This prevents evaluation of whether the primary objective was met. The secondary endpoint’s classification rule is also unclear, which limits the clinical interpretation of its participant counts.

The study was open-label, and the registry does not report whether outcome assessment was blinded. Pain intensity is a participant-reported measure, so knowledge of treatment assignment may be relevant to interpretation; the record does not provide information to evaluate that possibility. The rules for rescue dosing and its use by arm are not reported, even though rescue doses were allowed. The registry also does not describe how rescue treatment was incorporated into the primary outcome analysis.

Missing-data handling, analysis populations, protocol deviations, multiplicity procedures, and sensitivity analyses are not reported in the registry record. Safety reporting is limited to event counts, without detailed event descriptions or a formal comparison. These limitations concern what is documented in the record; they do not establish that unreported procedures or information were absent from the trial itself.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Compare Fentanyl Nasal Spray With Intravenous Opioids to Treat Severe Pain — NCT02459964.
• Registry record last updated 2021-06-16; status last verified 2021-06.
• Results first posted 2021-06-16.
• 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.

18. Record Summary

NCT02459964 is a completed, randomized, open-label, parallel-group phase 4 trial sponsored by M.D. Anderson Cancer Center. It compared intranasal fentanyl with intravenous hydromorphone for severe cancer-related pain in an emergency department setting. The registry lists enrollment as 84, with reported treatment-arm counts of 42 and 40; the adverse-event record also includes a no-medication group.

The primary endpoint was median change in NRS pain intensity from baseline to one hour after treatment initiation. The posted arm summaries are 5.14 with a standard deviation of 2.16 for intranasal fentanyl and 4.9 with a standard deviation of 2.31 for intravenous hydromorphone. The record does not report the primary non-inferiority margin or a direct inferential comparison.

A secondary participant-count outcome and unadjusted comparison are posted, along with descriptive adverse-event counts. Important analytic details, including missing-data handling, analysis populations, and multiplicity procedures, are not reported in the registry record.