Clinical Study to Investigate the Effect of the Combination of Psychotropic Drugs and an Opioid on Ventilation: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Clinical Study to Investigate the Effect of the Combination of Psychotropic Drugs and an Opioid on Ventilation (NCT04310579), a randomized phase 1 study. The primary outcome is Part 1 - Minute Ventilation at the 55 mm Hg End Tidal Carbon Dioxide (CO2) Point (VE55) on Day 1 for Oxycodone Combined With Midazolam and Oxycodone Alone..

2. Clinical Question

The study’s stated objective was to assess whether selected drugs, when combined with an opioid, change breathing compared with the opioid alone. Its rebreathing procedure increases oxygen and carbon dioxide in the inhaled gas mixture; the increased carbon dioxide prompts participants to breathe more. The posted primary result is therefore about ventilation under a controlled challenge. It does not by itself describe a participant’s breathing in other settings, nor does it answer the separate Part 2 questions about paroxetine or quetiapine.

3. Trial Design

This was a phase 1 randomized, quadruple-masked, crossover study sponsored by the Food and Drug Administration (FDA). The registry lists 55 participants enrolled and the status as completed. The recorded study start date is 2020-06-15, and the primary completion date is 2021-05-25. The study concerned hypercapnia and ventilatory depression.

The study had a lead-in reproducibility phase and two main parts. The primary endpoint belongs to Part 1, which compared oxycodone combined with midazolam with oxycodone alone. In a crossover design, participants receive treatments in different periods, allowing comparisons to account for the fact that one participant may have a different baseline response from another. The registry describes randomized crossover periods for Part 1. It does not report the sequence details in the complete information available for this analysis, so the specific order of each participant’s treatments cannot be described.

Part 2 involved oxycodone combined with paroxetine or quetiapine, with oxycodone plus placebo as another condition. These treatment combinations are distinct from the primary Part 1 comparison. The registry lists secondary outcomes covering ventilation and oxycodone concentration measures, including maximum observed plasma concentration and area under the plasma concentration-time curve. Eight secondary outcomes are posted, alongside 11 outcome measures and 17 analyses in total.

Quadruple masking indicates that the study was recorded as masked, but the registry information does not identify which groups of people were masked. The exact sample-size justification, detailed allocation procedures, and complete descriptions of all treatment-period schedules are not reported in the registry record. The study design therefore establishes the broad comparison and setting, but does not answer every operational question about conduct.

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.

Only the first 6 of 11 groups are shown.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All participants who finished more than 1 study period and completed rebreathing assessments at the 2-hour time point on day 1 were included in the primary analysis without imputation of missing data. One participant completed the study but did not have any completed rebreathing assessments at the 2-hour time point. Midazolam dose was escalated after the first five subjects, so those subjects who received the lower dose are excluded from summaries and comparison as specified in the protocol

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 Part 1 - Minute Ventilation at the 55 mm Hg End Tidal Carbon Dioxide (CO2) Point (VE55) on Day 1 for Oxycodone Combined With Midazolam and Oxycodone Alone., measured over Part 1: 2 hour timepoint on treatment period Day 1.

The primary endpoint is minute ventilation at the point where end-tidal carbon dioxide pressure is 55 mm Hg, abbreviated VE55. Minute ventilation measures the volume of air breathed over time. In this study, VE55 represents the estimated ventilation corresponding to a specified carbon dioxide level during a rebreathing assessment. The registry gives the unit as L/min and the assessment time as the 2 hour timepoint on treatment period Day 1.

The endpoint was obtained by fitting a linear regression to rebreathing data relating minute ventilation to partial pressure of end-tidal carbon dioxide. This approach produces an estimate of ventilation at the specified carbon dioxide point, rather than relying on a single breath or an unadjusted observation. A fixed point on that relationship makes it possible to compare the response under different treatment conditions.

In general, a controlled carbon-dioxide rebreathing challenge is useful for assessing ventilatory responsiveness because increasing carbon dioxide stimulates breathing. The study description notes that opioids and benzodiazepines can reduce this response, and that other drugs could potentially do so as well. That rationale explains why the endpoint is relevant to the study question; it does not establish what any particular drug combination did.

The endpoint is a laboratory response under the study procedure. It should not be treated as a direct measure of ventilation during ordinary activity, or as a complete description of a participant’s breathing across circumstances. The registry also reports quantitative pupillometry as part of the broader study, but pupillary measurements are not the primary endpoint analyzed in the posted comparison.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry describes VE55 as a continuous outcome: a measured quantity that can take a range of values rather than only categories. It reports least-squares means and 95% confidence intervals for the treatment groups. A least-squares mean is a model-based estimate of a group mean. Unlike a raw average, it can reflect adjustment for other terms included in the model. The primary analysis description says that baseline VE55 was included as a continuous variable, while treatment, sequence, and period were categorical variables and participant was a random effect.

Why use a mixed-effects model?

General statistical explanation: a mixed-effects model can represent both systematic differences associated with study design terms and variation among participants. In a crossover study, the same participant may contribute observations under more than one treatment condition. Those observations are related because they come from the same person. Treating participant as a random effect allows the model to represent this within-participant structure rather than treating every observation as unrelated. Including period and sequence terms can account for differences associated with when a treatment was given and with the order of treatment periods. The registry names these model terms but does not give further detail about their exact coding.

Baseline VE55 was included as a continuous covariate. In general, covariate adjustment can account for variation in the outcome associated with a baseline measurement, potentially improving the comparison’s precision. The posted result is therefore not simply the difference between the two displayed least-squares means. Those means describe model-based group estimates, while the reported analysis is specifically labeled a mixed-model comparison of final values. The registry does not report enough information to reconstruct the fitted model or its full estimate from the group summaries alone.

What does the confidence interval express?

General statistical explanation: a confidence interval describes the uncertainty around an estimated quantity under the statistical model and its assumptions. A narrower interval indicates greater precision than a wider interval on the same scale. For this comparison, the registry reports a 95% confidence level and gives the upper endpoint as 2.5, but leaves the lower endpoint absent. The interval is therefore incomplete in the posted result. It is not possible to state its full width, whether it crosses zero, or the complete set of values it supports. The group-specific intervals are available, but they are not a substitute for the missing endpoint of the interval for the modeled difference.

How should the p-value be read?

General statistical explanation: a p-value describes how incompatible the observed data are with a specified null hypothesis, given the model and its assumptions. It is not the probability that the null hypothesis is true, and it does not measure the magnitude or clinical importance of an effect. The reported p-value is 0.22. That value does not by itself establish equivalence, prove that there is no difference, or show that the treatments have the same effect. A claim of equivalence or non-inferiority would require a suitable design, a prespecified margin, and an analysis addressing that question; such a margin or analysis is not reported in the registry record.

The primary endpoint description says two independent assessors reviewed the rebreathing data while blinded to study treatment and assessment time, evaluating whether data were complete for study outcomes. This review concerns the assessment data, and the registry does not report further details about agreement between assessors or procedures for resolving differing judgments. The primary analysis used no imputation of missing data. In general, no-imputation analysis uses available qualifying observations rather than filling in missing values with modeled or substituted values. This can make the analyzed population smaller than enrollment and means that results pertain to the participants with eligible completed assessments.

The registry states that R software was used for all analyses. Software identification alone does not specify every analytic choice: for example, the complete model formula, handling of any unusual observations, or diagnostic checks are not reported in the registry record. The result should be interpreted with the method actually described, without assuming unreported procedures.

How does crossover structure affect interpretation?

General statistical explanation: crossover comparisons can use each participant as a source of information under multiple conditions, which can help separate treatment-related differences from stable differences between people. They also require attention to period and sequence, as named in this analysis. The registry’s model includes those terms, but the information available does not describe washout intervals, carryover assessments, or how any carryover concern was handled. Those details should not be inferred from the fact that the design was crossover.

Planning calculations are distinct from observed analysis results. The sample-size assumptions described for planning illustrations use the observed effect as a basis, a two-sided significance level of 0.05, equal allocation, and a normal approximation; the illustrations are not the trial’s original design justification and are not observed power. No numerical planning estimates are reported in the registry information for this analysis. The trial’s own sample-size justification and any prespecified power calculation are not reported in the registry record. A planning illustration cannot replace those missing design details or establish what the study was intended to detect.

10. Analyses Reported by the Sponsor

The registry reports the sponsor’s primary comparison as a mixed-model analysis, with a mean difference (final values) of -2.2 and p = 0.22. It gives the upper endpoint of the 95% confidence interval as 2.5, while the lower endpoint is not reported in the registry record. The group-specific least-squares means and their confidence intervals are also posted, but those summaries do not independently supply the missing endpoint of the confidence interval for the difference.

No separate recalculated estimate is reported in the registry record for this comparison. The sponsor’s reported analysis is therefore the available inferential result to interpret, rather than a basis for contrasting sponsor and recalculated estimates. The discrepancy indicator is false, so there is no identified disagreement between the sponsor result and a recalculation. This does not remove the limitation that the interval is incomplete: the estimate and p-value can be quoted as reported, but the full uncertainty interval cannot be assessed from the posted result.

The registry’s analysis description identifies adjustment for baseline VE55, treatment, sequence, and period, with participant as a random effect. Any simple comparison of the displayed group means would not reproduce that model-based analysis. The sponsor’s model-based result should not be replaced by an unadjusted calculation from summary values, especially in a crossover design where participant and period structure matter.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 11 outcome measures and 17 analyses, including eight posted secondary outcomes. Those secondary outcomes cover VE55 in Parts 1 and 2 and pharmacokinetic measures of oxycodone, including maximum observed plasma concentration and area under the plasma concentration-time curve. Multiple outcomes and analyses create a multiplicity issue: when many statistical comparisons are considered, the chance of at least one apparently unusual result can increase. This is a general statistical concern, not a claim that any particular posted finding is spurious.

The registry record does not report a multiplicity adjustment, a testing hierarchy, or a complete prespecified plan for deciding which outcomes would be treated as confirmatory. It is therefore not possible to determine from the record whether the reported p-value was intended to be interpreted as one test within a controlled family of comparisons or as an exploratory result. The number of posted outcomes alone does not establish which analyses were planned in advance or which were primary for decision-making.

The primary population was defined as participants who completed more than one study period and had completed rebreathing assessments at the 2-hour time point on Day 1. Missing data were not imputed. The registry notes that one participant completed the study but had no completed rebreathing assessments at the primary time point. The primary analysis counts were 14 for the combination group and 19 for oxycodone alone. The record does not fully explain the difference between the enrolled total and the analysis counts, or describe all missing-data patterns and reasons.

In general, analyzing only available qualifying observations can be appropriate for some estimands, but interpretation depends on why observations are missing and whether missingness is related to treatment or outcome. Those reasons and any sensitivity analyses are not reported in the registry record. The protocol-level analysis population, handling of missing secondary outcomes, interim analyses, and any additional missing-data methods are also not reported. These gaps limit assessment of how robust the primary result is to analysis choices and how the multiple outcomes were managed.

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 non-serious adverse event terms

The registry reports zero serious adverse events in every listed group and zero deaths in every listed group. For the lead-in group, 10 participants were at risk for serious adverse events and 0 were affected. In each of the listed Part 1 groups—placebo, midazolam, oxycodone, and oxycodone plus midazolam—20 were at risk and 0 were affected. In each listed Part 2 group—oxycodone plus placebo, oxycodone plus paroxetine, and oxycodone plus quetiapine—25 were at risk and 0 were affected.

These are group-specific registry counts, not a single count of unique participants across the whole study. Because participants could take part in multiple treatment periods in a crossover study, the group counts should not be added together to estimate the number of distinct people. The registry also reports other adverse events separately: 6 in the lead-in group; 6, 5, 8, and 5 in the listed Part 1 groups; and 16, 17, and 20 in the listed Part 2 groups. These are not the serious-event counts and should not be described as serious events.

Zero reported serious events in these groups means none were recorded as affected in the registry’s serious-event summary. It does not establish that no adverse events of any kind occurred, or that risk is absent. The record does not report the duration of safety follow-up or enough detail to assess event ascertainment and comparison precision. The figures are descriptive and should be read with the group sizes and study structure in mind.

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

The posted primary result describes a model-based comparison of VE55 at a specific time during a controlled rebreathing assessment in Part 1. The group estimates were 21.9 L/min for oxycodone combined with midazolam and 24.1 L/min for oxycodone alone. The mixed-model estimate was -2.2, with p = 0.22. The registry reports only the upper endpoint, 2.5, of the stated 95% confidence interval for the difference, so the full uncertainty around that estimate is unavailable.

The result does not establish that the treatments are equivalent. A non-significant p-value is not evidence that any difference is exactly zero, and the incomplete confidence interval prevents a full assessment of the range of differences compatible with the model. Conversely, the point estimate alone does not establish a reliable difference. The result should be understood as an estimate with incomplete precision information, not as a definitive yes-or-no conclusion.

The outcome measures ventilatory response under a specific carbon-dioxide challenge, at the stated time point and in participants included in the primary analysis. It is not an assessment of breathing during ordinary activity, and it does not by itself describe other outcomes such as symptoms, long-term effects, or effects in populations not enrolled in this study. The study’s Part 2 drug combinations are separate from the primary Part 1 comparison.

The registry also reports no serious adverse events among the participants at risk in the listed treatment groups. That is a descriptive safety summary for this study, not proof that an event could not occur or a complete assessment of safety across broader use. Taken together, the posted information characterizes one early-phase laboratory comparison and its recorded event counts. The study does not provide a complete interval for its primary treatment contrast, and the record does not report several design and analysis details needed to extend interpretation beyond that comparison.

14. Why This Trial Matters Statistically

This study illustrates why the statistical structure should match the clinical design. A crossover trial produces observations linked to individual participants across treatment periods. The reported mixed model reflects that structure by including participant as a random effect and by accounting for treatment, sequence, and period. A simple comparison of group summaries would not represent all of those features, so the model description matters when interpreting the estimate.

It also illustrates the importance of reporting complete uncertainty measures. The registry provides a point estimate and p-value, along with an upper endpoint of the confidence interval, but not the lower endpoint. Without both endpoints, readers cannot assess the full interval around the treatment contrast. Group-level confidence intervals do not repair this gap because they describe uncertainty in each estimated mean, not the interval for the modeled difference.

The endpoint further demonstrates how a physiological measure depends on the conditions under which it was collected. VE55 is estimated from rebreathing data at a specified carbon dioxide pressure and time point. It answers a focused question about ventilatory response under that procedure; it should not be read as interchangeable with breathing under ordinary conditions.

Finally, the registry’s listed secondary outcomes and analyses raise a general question about multiplicity. Without a reported hierarchy or adjustment plan, readers cannot determine how the primary and secondary tests were intended to work together. Clear reporting of the analysis population, missing observations, full model results, and multiplicity strategy helps distinguish a focused estimate from a broader set of exploratory comparisons. These are statistical reporting issues, not conclusions about the clinical value of any treatment.

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 most direct limitation in the posted primary result is the missing lower endpoint of its 95% confidence interval. That omission restricts assessment of the full range of model-compatible differences and means the interval cannot be checked for its position relative to zero. The group-specific confidence intervals are complete, but they answer a different question from the interval for the treatment contrast.

The primary analysis applies to a subset defined by completed study periods and rebreathing assessments, rather than necessarily to every participant enrolled. The analysis group counts are 14 and 19, while enrollment was 55. The registry states that one participant completed the study without completed rebreathing assessments at the primary time point, but does not fully account for all exclusions or missing observations. No imputation was used, and sensitivity analyses addressing missing data are not reported in the registry record.

The study includes more than one part and multiple outcomes, but the available result concerns only the Part 1 primary comparison. The complete multiplicity strategy and status of each analysis as confirmatory or exploratory are not reported. The exact sample-size justification, complete protocol-level analysis plan, and any interim analyses are not reported in the registry record.

Some operational details are also absent from the result information: the specific masking roles, complete treatment sequence information, and further handling of possible carryover effects are not reported. The endpoint itself is tied to a laboratory rebreathing procedure and one time point; it does not directly characterize ventilation in other settings. The study’s early-phase designation and the limited primary analysis counts should be kept in view when interpreting how broadly the posted estimate can be applied.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Clinical Study to Investigate the Effect of the Combination of Psychotropic Drugs and an Opioid on Ventilation — NCT04310579.
• Registry record last updated 2024-05-21; status last verified 2023-11.
• Results first posted 2024-05-21.
• 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

NCT04310579 is a completed phase 1 randomized, quadruple-masked crossover study sponsored by the Food and Drug Administration (FDA). It enrolled 55 participants and examined combinations of psychotropic drugs with an opioid in relation to ventilation. The primary comparison was oxycodone combined with midazolam versus oxycodone alone in Part 1.

The primary endpoint was VE55, minute ventilation at an end-tidal carbon dioxide pressure of 55 mm Hg, measured at the 2 hour timepoint on treatment-period Day 1. The reported least-squares means were 21.9 L/min and 24.1 L/min, based on 14 and 19 participants. The mixed-model analysis reported a mean difference of -2.2 and p = 0.22; the lower endpoint of the 95% confidence interval is not reported in the registry record.

The study also included a lead-in reproducibility phase and Part 2 comparisons involving oxycodone with paroxetine or quetiapine. The registry reports zero serious adverse events and zero deaths in each listed group. The primary result is specific to the rebreathing assessment and analyzed population; incomplete interval reporting and unreported analysis details limit further interpretation.