A Study Designed to Evaluate ODSH in Subjects With Exacerbations of COPD: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study Designed to Evaluate ODSH in Subjects With Exacerbations of COPD (NCT00457951), a randomized phase 2 study. The primary outcome is Incidence of Treatment Failure.

2. Clinical Question

3. Trial Design

This was a phase 2, randomized, double-blind, parallel-group study with a placebo comparator. It included an open-label phase followed by a randomized phase, as reflected in the official study title. The registry record describes an initial open-label safety stage: the first six subjects were to receive ODSH, followed by six additional patients after a safety review. The registry does not report the outcome of that review or explain how it affected the subsequent conduct of the trial.

In the randomized phase, the ODSH arm received standard-of-care treatment with ODSH, and the comparator arm received standard-of-care treatment with 0.9% sodium chloride placebo. The arm description specifies randomization 1:1, ODSH 8mg/kg, and a continuous infusion dose of 0.375 mg/kg/hr over 96 hours. The placebo description also gives an infusion dose of 0.375mg/kg/hr over 96 hours. Both descriptions mention bolus dosing. The precise bolus schedule is not reported in the registry record.

The study concerned subjects with COPD exacerbations and was designed to compare ODSH added to conventional treatment with conventional treatment alone. The registry lists an enrollment of 158 and reports 138 subjects randomized. For the primary outcome, 132 subjects were analyzed in the intent-to-treat population: 64 in the placebo arm and 68 in the ODSH arm. The record states that six randomized subjects were excluded from that population, with reasons described in the missing-data discussion.

The study began in 2007-04, had a primary completion date of 2009-08, and is listed as terminated. The reason for termination is not reported in the registry record. The registry does not report the planned duration of recruitment, the rationale for the enrollment target, or further details of the randomization and allocation-concealment procedures.

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

Analysis population for the primary outcome, as defined in the registry: Of the 138 subjects randomized, 132 were analyzed in the intent-to-treat population. Of the 6 excluded from the intent-to-treat population, 4 did not receive study drug and 2 lacked information for assessment.

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.

Population: Open-Label followed by multi-center, randomized, double-blind, placebo-controlled trial in hospitalized subjects with exacerbations of COPD. In the double-blind phase, 145 subjects consented and 138 met criteria and were randomized.

6. Primary Endpoint

The registry lists the primary outcome as Incidence of Treatment Failure, measured over Time to hospital discharge and 21 days post-treatment, up to 31 days.

The primary endpoint was incidence of treatment failure. The registry defines failure as either failure to discharge from hospital based on GOLD criteria or relapse after discharge. It is therefore a composite outcome: a participant could meet the endpoint through either of the listed events. The reported timeframe covers the period to hospital discharge and 21 days post-treatment, up to 31 days.

Using a treatment-failure endpoint makes the outcome relevant to the acute course of a COPD exacerbation: it captures continued inability to leave hospital under the stated criteria as well as relapse after discharge. Combining these events can provide a single outcome for the study question, but it also means the overall percentage does not show how often each component occurred separately. Component-specific results are not reported in the registry record.

The endpoint is expressed as a percentage of failures, and the registry reports a value and 95% confidence interval for each randomized arm. The posted figures are 24.6% for 0.9% sodium chloride, with a 95% confidence interval of 14.0% to 43.0%, and 32.4% for ODSH, with a 95% confidence interval of 21.0% to 42.0%. These are arm-specific estimates. A formal estimate of the between-arm difference, and its uncertainty, is not reported in the registry record.

The endpoint description does not provide additional adjudication details or explain how discharge eligibility was assessed. Such details are not reported in the registry record, so the definition should be interpreted as written rather than supplemented with assumptions about clinical review or outcome verification.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is reported as a percentage of treatment failures in each arm, with a 95% confidence interval. The result is a proportion-type summary: it describes the share of analyzed subjects in an arm who met the study’s treatment-failure definition during the stated follow-up period. A confidence interval communicates the uncertainty around that estimate. The registry lists 24.6% with an interval from 14.0% to 43.0% for 0.9% sodium chloride, and 32.4% with an interval from 21.0% to 42.0% for ODSH.

What do the confidence intervals describe?

In general, a confidence interval is a range produced by a statistical procedure intended to reflect uncertainty in an estimate. It is not the range of individual participant outcomes, and it does not mean that every value inside it is equally plausible. Wider intervals indicate less precision than narrower intervals, all else being equal. Here, both intervals cover a broad range of possible arm-specific failure percentages. The registry does not name the interval-construction method, so a more specific account of how these intervals were calculated is not reported in the registry record.

The two reported intervals belong to separate arm estimates. In general, comparing two separate confidence intervals is not the same as estimating the difference between arms. A direct comparison would ordinarily report an effect estimate, such as a difference or ratio, together with uncertainty or a test result. The registry does not report such a comparison for this endpoint. The fact that the arm-specific intervals overlap cannot, on its own, be used to declare the arms equivalent or to establish whether a difference is statistically significant.

Why does the endpoint format matter?

The outcome is described as incidence of treatment failure, measured as a percentage, and its parameter is listed as a number. The registry classifies the endpoint as count/other and notes that values are missing or not estimable for a two-arm comparison. This means the posted record supplies arm-level summaries but not a usable inferential result for the comparison. It does not report the underlying event counts alongside the percentages in the outcome entry, so those counts should not be reconstructed from the percentages.

The endpoint combines failure to discharge under GOLD criteria with relapse after discharge. In general, a composite endpoint can capture more than one clinically relevant type of failure in a single summary. It can also obscure whether one component accounts for more of the overall result than another. Component-specific event counts and analyses are not reported in the registry record.

What analysis details are absent?

The registry lists zero posted analyses. It does not report the statistical model for the primary outcome, whether any covariates were included, how the confidence intervals were calculated, or whether an inferential comparison was planned. It also does not report an analysis plan, a protocol-defined primary analysis population beyond the stated intent-to-treat population, or the handling of missing outcome data. These details cannot be inferred from the arm-specific percentages.

General statistical practice distinguishes a prespecified analysis from an exploratory or post hoc analysis, but the registry record does not report enough information to make that distinction for any comparative analysis. The arm-level estimates can be described as posted results; no further adjustment or model should be assumed.

How should planning illustrations be understood?

The study’s planning illustration is described as based on the observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It describes the effect to detect as a stated fraction of the observed effect. For outcomes other than time-to-event outcomes, such illustrations are expressed as participants per arm. The registry record does not report the trial’s own sample-size justification or design assumptions, and no numeric planning illustration is available in the registry record summarized for this analysis.

In general, a sample-size calculation is a planning tool that estimates how many participants may be needed under specified assumptions. A calculation that uses an observed effect to illustrate planning is not an observed power calculation and does not show the probability that the trial’s result is true. It is also not evidence that the study was originally designed using those assumptions. Because the registry does not report the original justification, the illustration should not be treated as a description of the trial’s prospective design.

What does over-dispersion mean generally?

For count outcomes, over-dispersion describes variability greater than a simple statistical model expects. In general, ignoring over-dispersion can make uncertainty appear smaller than it is. The primary result in this study is presented as a percentage of failures, and the registry does not report a count model or an over-dispersion assessment. This concept therefore provides general context about count-data methods, not a claim about the analysis used in this trial.

Likewise, no claim about statistical significance, equivalence, or non-inferiority can be made from the posted arm-specific values alone. Each requires an appropriate comparison and a defined analytical framework. Neither is reported in the registry record.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists one outcome measure and no posted analyses. It does not report secondary outcomes in the summarized record. The presence of one listed outcome measure does not establish whether other outcomes were assessed during the trial or whether multiple analyses were planned; those details are not reported in the registry record. The record also does not report a multiplicity adjustment or explain whether any adjustment was considered. Without information on the analysis plan, it is not possible to determine how many comparisons were intended or how results were selected for reporting.

For the primary endpoint, the registry states that 138 subjects were randomized and 132 were included in the intent-to-treat population. Six randomized subjects were excluded: four did not receive study drug, and two lacked information for assessment. The analyzed arm counts are 64 for 0.9% sodium chloride and 68 for ODSH. The record does not explain whether the two subjects without assessment information had missing outcomes for other reasons or how their outcomes were handled beyond exclusion from this population.

Although the population is labeled intent-to-treat, the stated analysis excludes some randomized subjects. The label should therefore be read alongside the explicit population description rather than as proof that all randomized subjects contributed to the endpoint analysis. The registry does not report an analysis including all randomized subjects, a sensitivity analysis for missing data, an imputation method, or a comparison of included and excluded subjects.

The record does not report interim analyses, stopping rules, or whether the termination affected the planned analysis. The reason for termination is also not reported in the registry record. These omissions limit assessment of how missingness, multiple testing, and the final analysis may have influenced the posted results.

11. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: respiratory failure (7.7%), guillain-barre syndrome (1.5%), ischaemic stroke (1.5%), acute respiratory failure (1.5%), chronic obstructive pulmonary disease (1.5%).

The registry reports safety counts in separate groups: an open-label group, a placebo comparator group, and an ODSH treatment group. Serious adverse events were reported for 1 of 13 subjects in the open-label group, 4 of 65 in the placebo group, and 2 of 69 in the ODSH group. Other adverse events were reported for 0 of 13, 13 of 65, and 48 of 69, respectively. These are descriptive counts in the registry’s safety groups, not estimates from a formal comparative safety analysis.

The serious-event terms listed include respiratory failure at 7.7%, and Guillain-barre syndrome, ischaemic stroke, acute respiratory failure, chronic obstructive pulmonary disease, and cardiopulmonary failure at 1.5% each. The record does not provide the context needed to attribute these terms to a particular randomized arm in this summary, so they should not be read as arm-specific comparisons. The registry does not report death counts in the safety information.

Safety counts can be affected by group size, follow-up, event definitions, and how events were collected. The registry record does not report enough detail to determine whether exposure time differed across groups or whether events were independently adjudicated. A difference in counts alone does not establish that treatment caused an event or that one group had greater underlying risk. These data describe what was posted, without supporting a causal conclusion.

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

The registry record addresses a focused question: whether ODSH added to conventional treatment was more effective for COPD exacerbations than conventional treatment alone, using treatment failure as the primary outcome. It reports arm-specific failure percentages and confidence intervals for the randomized groups. Those figures show the observed summaries for the analyzed population, but the registry does not provide a direct statistical comparison between arms.

The precision of the estimates is limited by the breadth of the reported confidence intervals. In general, a broad interval means the data do not pin down an arm’s underlying event percentage closely. The record does not report a confidence interval for the difference between ODSH and placebo, a p-value, or an adjusted model. Therefore, the posted outcome values do not establish whether an apparent difference between the arms is statistically distinguishable from chance.

The endpoint combines failure to discharge from hospital based on GOLD criteria and relapse after discharge. The overall percentage does not reveal the contribution of each component, and component-specific results are not reported. Nor does the record provide enough methodological detail to assess outcome adjudication or the statistical method used to construct the arm-level intervals.

The analyzed population excluded six of the 138 randomized subjects. Four did not receive study drug, and two lacked information for assessment. The registry does not report sensitivity analyses or further missing-data methods. This limits assessment of how exclusion and incomplete assessment may affect the posted estimates.

The record’s terminated status is relevant context, but the reason for termination is not reported. It should not be treated as evidence for or against a treatment effect. The registry also does not report the trial’s own sample-size justification, interim analyses, or full statistical analysis plan. These omissions prevent a fuller assessment of the intended power, decision rules, and analysis choices.

Accordingly, the results can be described as arm-specific estimates with uncertainty, in a partially analyzed randomized population. They do not, on their own, establish that ODSH reduced treatment failure, increased it, or had no comparative effect. Such conclusions would require a reported and interpretable between-arm analysis, together with adequate information about its methods and assumptions.

13. Why This Trial Matters Statistically

This trial is statistically instructive because it illustrates the difference between reporting an outcome by arm and reporting a comparison between arms. The registry gives a treatment-failure percentage and confidence interval for each randomized group, but no formal comparative estimate or test. For a randomized study, the scientific question is usually comparative; arm-specific summaries alone do not directly answer it. Readers should distinguish the descriptive results from the evidence needed to quantify a treatment contrast.

The confidence intervals also show why precision matters. Each arm estimate has uncertainty, and the intervals are broad. A reader cannot reliably judge the comparative uncertainty by looking only at the separate arm intervals. A direct contrast would account for uncertainty in both arms and report an estimate on a defined scale. That result is not reported in the registry record.

The population description is another important statistical detail. The registry reports 138 randomized subjects but an intent-to-treat population of 132, with six exclusions and stated reasons. The analysis label therefore needs to be read together with the population count and exclusions. Transparent reporting of who contributed data and how unavailable assessments were handled is important because missingness can affect the interpretation of an estimate.

Finally, a trial’s registry entry may contain useful design and outcome information while leaving key analytical details unreported. Here, the registry does not report the original sample-size justification, analysis plan, multiplicity procedures, or reason for termination. A planning illustration based on an observed effect is not a substitute for the prospective design rationale. These distinctions help keep interpretation within the evidence actually documented.

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 primary limitation is the absence of a reported inferential comparison for the primary endpoint. The registry provides a percentage and 95% confidence interval for each arm, but no between-arm estimate, p-value, or adjusted analysis. As a result, comparative uncertainty cannot be assessed from the registry record.

Only 132 of 138 randomized subjects were included in the stated intent-to-treat population. The record identifies four who did not receive study drug and two who lacked information for assessment, but does not report additional detail about missing outcomes, the analysis of excluded subjects, imputation, or sensitivity analyses. This leaves uncertainty about the effect of exclusions on the outcome estimates.

The endpoint is a composite of failure to discharge based on GOLD criteria or relapse after discharge. Component-specific results and further outcome-assessment procedures are not reported. This limits interpretation of what drove the overall treatment-failure percentages and how consistently the criteria were applied.

The study is listed as terminated, but the reason is not reported in the registry record. The record also does not report the original sample-size justification, an analysis plan, interim analyses, stopping rules, or multiplicity procedures. The total enrollment of 158 and randomized count of 138 are both reported, but the difference is not explained.

The safety information is descriptive and does not include a formal comparative analysis in the registry record. Details about event ascertainment, exposure time, and attribution are not reported. These limitations make the posted results useful as a record of reported arm-level outcomes, but constrain causal and comparative conclusions.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: A Study Designed to Evaluate ODSH in Subjects With Exacerbations of COPD — NCT00457951.
• Registry record last updated 2021-12-02; status last verified 2016-10.
• Results first posted 2016-12-16.
• 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

NCT00457951 is a phase 2 randomized, double-blind, parallel-group study of intravenous ODSH added to standard-of-care treatment for COPD exacerbations, compared with 0.9% sodium chloride placebo and standard-of-care treatment. An open-label phase preceded the randomized phase. The registry lists enrollment of 158, reports 138 randomized subjects, and states that 132 were analyzed for the primary outcome.

The primary endpoint was treatment failure, defined as failure to discharge from hospital based on GOLD criteria or relapse after discharge. The reported percentages were 24.6% for placebo and 32.4% for ODSH, with respective 95% confidence intervals of 14.0% to 43.0% and 21.0% to 42.0%. No formal between-arm analysis is reported in the registry record. The study began in 2007-04, had primary completion in 2009-08, and is listed as terminated; the reason for termination is not reported.