Safety and Efficacy of Pioglitazone as an Anti-inflammatory for the Treatment of Cystic Fibrosis (CF) Lung Disease: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Safety and Efficacy of Pioglitazone as an Anti-inflammatory for the Treatment of Cystic Fibrosis (CF) Lung Disease (NCT00322868), a na na study. The primary outcome is Sputum White Cell Count.

2. Clinical Question

3. Trial Design

This was a completed, single-group pilot study of pioglitazone in cystic fibrosis lung disease. The registry lists 21 participants as enrolled. It identifies University Hospitals Cleveland Medical Center as the sponsor, gives a start of 2006-04 and primary completion of 2006-12, and lists the phase and allocation as NA. Masking is listed as none. No randomization or separate control arm is described.

All subjects in the treatment group were assigned pioglitazone, 30 mg orally once daily, for 28 days. The primary outcome description specifies sputum assessments at Day 0 and Day 29. The planned measures included markers of inflammation—neutrophils, elastase, cytokines and bacteria—assessed in induced sputum before and after treatment. The record lists eight outcome measures and eight analyses, but does not give their complete numerical results in the information available for interpretation.

The primary-outcome description says that one of 20 enrolled subjects did not produce sputum at baseline and that a post-treatment sample for another subject was lost because of a laboratory error. It states that 19 subjects remained for that outcome and that 18 had both baseline and post-treatment samples. The record does not reconcile its enrollment count of 21 with the description’s count of 20.

A within-participant comparison can describe how measurements differed over time among people with data at both time points. Without a concurrent control group, however, a change cannot be separated from other influences over the same period. Details of the protocol, sample-size justification, and prespecified statistical plan 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: Washout: A 4 week washout from chronic use of NSAIDs, corticosteroids and oral meds to treat diabetes will be required. Subjects will be excluded if they take these medication within 72 hours of Visit 1. Subjects using alternate month tobramycin for inhalation (TOBI) will need to be on an off cycle to participate in this study.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Of the 20 enrolled subjects, one did not produce sputum at Baseline and a second subject's Post-Treatment sputum sample was lost due to a lab error. This reduced the number of subjects per group to 19. Of these 19 subjects, 18 subjects had both Baseline and Post-Treatment sputum samples.

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: 21 subjects were screened and enrolled with 1 subject withdrawal due to disease progression.

6. Primary Endpoint

The registry lists the primary outcome as Sputum White Cell Count, measured over Day 0 and Day 29.

The primary endpoint is sputum white cell count, reported on a log 10 (cells/mL) scale. It represents the total number of white cells measured in induced sputum, with the count transformed to a logarithmic scale. White cells are included among the inflammatory markers specified in the study description. The registry identifies the endpoint as a mean with standard deviation and names Day 0 and Day 29 as the measurement time points.

As a general statistical matter, a measure of cell count can help describe the inflammatory material present in a sputum sample. A logarithmic scale can make a wide range of counts easier to summarize and compare. Interpretation must remain on the scale actually reported: changes on a log scale are not the same as simple differences in the original cell counts.

Repeated measurements from the same participant are useful for describing within-person change, because each participant contributes observations at both time points when both samples are available. The record states that 18 subjects had both samples for this endpoint. It does not report the baseline or post-treatment mean, standard deviation, estimated change, confidence interval, or test result. Therefore, the endpoint’s observed magnitude and precision are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry describes the primary endpoint as a mean of a continuous measure, with standard deviation, and specifies measurement at Day 0 and Day 29. These details identify the summary scale and the planned time points, but they do not reveal the statistical test or estimation procedure used. The analysis method is not reported in the registry record.

What the mean and standard deviation describe

As a general statistical explanation, the mean is a measure of the center of a set of observations. The standard deviation describes how dispersed the individual observations are around that center. A small standard deviation indicates that observations are relatively close to the mean, while a larger one indicates greater spread. Standard deviation is not the same as the standard error, which describes the uncertainty in an estimated mean. The registry names standard deviation as the dispersion measure but does not report its value for the primary endpoint.

The outcome is on a log 10 (cells/mL) scale. In general, transforming a measure can make its distribution or scale more suitable for summarization. A difference expressed on a logarithmic scale represents a multiplicative relationship on the original scale, rather than a direct additive difference in cell counts. This is why the scale should be kept in view when interpreting any estimate. The numerical values needed to interpret the observed change are not reported in the registry record.

Repeated measurements and within-person change

Because the primary endpoint was assessed before and after treatment in the same study group, participants with both samples can contribute paired measurements. A paired analysis would use the relationship between each participant’s two observations when estimating within-person change. This can differ from treating the observations at the two time points as unrelated groups. The record states that 18 participants had both samples, but it does not say whether the statistical analysis used a paired method. It also does not report the estimated paired change or its uncertainty.

For any estimate, a confidence interval is a way of expressing statistical uncertainty under specified assumptions. It helps readers assess both the range of effects compatible with the data and the precision of the estimate. A p-value, if reported, would address how compatible the data are with a specified null hypothesis under the test’s assumptions; it would not indicate the size or importance of a change. Neither a confidence interval nor a p-value for the primary endpoint is reported in the registry record. No claim about statistical significance can be made from the endpoint description alone.

Design and causal interpretation

General statistical interpretation depends on the design as well as on the calculation. A single-group before-and-after analysis can describe observed change over time among participants with measurements. It cannot isolate the effect of treatment from other changes that occur between assessments. A concurrent comparison group would provide a reference for changes over the same period, but this record describes no such group. The registry lists allocation as NA and masking as none. It does not report additional safeguards or design features that would allow a causal comparison.

Variation in sputum measurements can arise from differences among participants and from measurement or sample-handling processes. A standard deviation can describe observed spread, but it cannot explain its sources. The record notes that one post-treatment specimen was lost because of a laboratory error. It does not report any further quality-control information or the extent to which that loss affected the analysis.

Planning illustrations and observed power

The planning assumptions described in the registry use a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. They describe an effect to detect as a stated fraction of the observed effect. These assumptions are not a description of this trial’s actual design: the trial is listed as single-group, with no randomized comparison. Sample-size calculations based on two arms are hypothetical planning illustrations, not evidence that the study had a particular probability of detecting an effect. They are not observed power.

The registry record does not report the trial’s own sample-size justification or explain whether the planning assumptions were used in designing this study. No planning sample-size results are included in the information available for interpretation. Since the primary observed effect is also not reported, the calculations cannot be independently assessed from this record. A planning illustration should not be confused with the number enrolled or with the number whose sputum samples were available.

Other distributional considerations

As a general statistical concept, over-dispersion occurs when variability is greater than a chosen model assumes. It can make standard errors too small if the model does not account for the extra variation. The registry information does not identify a model for the primary endpoint or report a dispersion assessment, so it is not possible to determine whether over-dispersion was evaluated or relevant to the analysis. This is a general consideration, not a statement about the study’s findings.

The record lists eight outcome measures and eight analyses. It does not state whether the analyses were prespecified, whether any were exploratory, or whether a multiplicity adjustment was used. These omissions matter when interpreting a collection of statistical tests: repeated testing can increase the chance of at least one apparently notable result by chance. The record does not permit an assessment of that issue or of the specific methods used for the other outcomes.

10. Confidence Intervals for Each Arm

No two-arm comparison is shown for this trial, but each arm's mean can be given with a t-based 95% confidence interval computed from the posted mean, spread and number of participants.

R code and output

arm <- c("Baseline", "Post-Treatment")
n   <- c(19, 19)
m   <- c(6.93, 6.81)
sp  <- c(0.4, 0.41)
sd  <- sp      # posted standard deviations
for (i in seq_along(n)) {
  se <- sd[i] / sqrt(n[i]); h <- qt(0.975, n[i] - 1) * se
  cat(sprintf("CARM|%s|%d|%.3f|%.3f|%.3f|%.3f\n", arm[i], n[i], m[i], sd[i], m[i] - h, m[i] + h))
}

CARM  Baseline  19  6.930  0.400  6.737  7.123
CARM  Post-Treatment  19  6.810  0.410  6.612  7.008

11. Multiplicity, Missing Data and Analysis Populations

The registry lists eight outcome measures and eight analyses. It does not report whether the outcomes were organized into primary and secondary families beyond the named primary endpoint, whether the analyses were prespecified, or whether any adjustment for multiple comparisons was planned or applied. As a general statistical matter, evaluating several outcomes or tests can increase the chance of a chance finding. Adjustment methods address that issue in different ways, but the registry record does not identify a method. No inference about multiplicity control can be made.

For the primary endpoint, the record describes two missing-sample circumstances. One of 20 enrolled subjects did not produce sputum at baseline, and another subject’s post-treatment sample was lost because of a laboratory error. The endpoint description says this left 19 subjects in the endpoint group and that 18 had both baseline and post-treatment samples. Separately, the overall enrollment entry lists 21. The record does not explain the difference between the enrollment counts or state how the subject with only one available measurement was included in the analysis.

The reason for one missing post-treatment sample is identified as a laboratory error, but no broader missing-data plan is reported. The registry does not describe imputation, sensitivity analyses, or whether the analysis was restricted to participants with complete pairs. It also does not report missingness details for each of the other listed outcomes. Consequently, the potential impact of missing data on the reported estimates cannot be evaluated from this record.

The analysis population is not defined beyond the endpoint description of available sputum samples. The record does not report an intention-to-treat analysis or another formal analysis-population definition. These omissions leave uncertainty about which participants contributed to each analysis and how incomplete measurements were handled.

12. 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 non-serious adverse event terms

The adverse-event entry lists 20 participants at risk for serious adverse events and 0 affected. It lists 20 at risk for other adverse events and 1 affected. The record does not name the other adverse event in the information. The death count is not reported in the registry record.

These figures are counts for the listed group, not estimates of comparative risk. Because the study had a single treatment group and no separate control group, the adverse-event data do not provide a direct comparison with an untreated or alternative-treatment group. The registry record also does not report details such as event timing, severity beyond the serious/non-serious grouping, or the relationship of the other event to treatment.

As a general statistical point, observing no serious events in a study group does not establish that serious events cannot occur. The amount of safety information depends on the number observed, the number at risk, follow-up, and how events were collected and reported. The record’s 0 serious events should therefore be read as the reported count, not as proof of absence of risk. The event figures are separate from the sputum endpoint and do not resolve the missing primary-outcome results.

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

This record establishes that a pilot study assessed pioglitazone in clinically stable patients with cystic fibrosis, using a single-group before-and-after design. It documents a 28-day treatment regimen and identifies sputum white cell count as the primary endpoint, with measurements planned at Day 0 and Day 29. It also describes missing sputum samples and reports adverse-event counts for the treatment group.

The record does not report the numerical primary endpoint results. It therefore does not show the average baseline or post-treatment count, the direction or size of any change, or the uncertainty around an estimated change. It cannot support a conclusion about whether the measured endpoint changed during treatment. The absence of these values is different from a finding of no change.

Even if before-and-after values were available, the design would provide a description of change within the treated group rather than a controlled estimate of treatment effect. Without a concurrent comparator, changes over time could have explanations other than the intervention. No causal conclusion can be drawn from the design description alone.

The adverse-event entry reports 0 serious events among 20 at risk and 1 other event among 20 at risk. Those counts do not establish comparative safety, and the death count is not reported in the registry record. The study’s pilot status and available sample information also do not substitute for outcome estimates or a precision assessment.

Overall, the registry record is informative about the study question, intervention, outcome measurement, and some missing-data and safety details. It does not provide the outcome statistics needed to judge the observed primary endpoint. Any stronger interpretation would require results and analysis details that are not reported in the registry record.

14. Why This Trial Matters Statistically

This study raises useful statistical issues because it combines repeated measurements with a single-group design and a continuous endpoint on a logarithmic scale. Readers need to distinguish a within-participant description of change from a treatment effect estimated against a concurrent comparator. The design determines what a statistical estimate can support; a more elaborate calculation cannot replace the missing comparison group.

The endpoint description also shows why reporting data availability matters. The record distinguishes participants with an endpoint-group measurement from those with paired samples, and identifies a laboratory loss for one post-treatment specimen. Those details help readers understand that the analysis may not include the same observations at each time point. However, without an account of the analysis population or handling of incomplete pairs, the consequences remain uncertain.

The record lists eight outcome measures and eight analyses but does not describe multiplicity handling. This highlights the importance of explaining which analyses were planned and how results across outcomes should be interpreted. It also lists planning assumptions based on a two-arm comparison, while the described trial is single-group. Separating planning illustrations from the actual design prevents a sample-size calculation from being mistaken for evidence of study power.

Most importantly, the registry entry demonstrates that a clear endpoint description is not a substitute for numerical results. Means, measures of variability, estimates of change, and uncertainty intervals are needed to assess magnitude and precision. When these are not reported, the appropriate conclusion is limited: the measurement plan and study structure can be described, but the primary outcome cannot be evaluated from the registry record.

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 principal limitation for interpreting this trial is that the registry record does not report numerical results for the primary endpoint. It gives the measure, scale, summary type, and time points, but not the observed values or an estimate of change. Confidence intervals and test results are also not reported in the registry record. The primary outcome’s direction, magnitude, and precision therefore cannot be assessed.

The single-group design has no separate comparator. A before-and-after comparison can describe observations over time in participants who contributed both samples, but cannot distinguish treatment-associated change from other changes over the same period. The registry lists masking as none, and does not report other design features that would provide a controlled comparison.

There is an unresolved difference between the overall enrollment entry of 21 and the primary-outcome description, which refers to 20 enrolled subjects. That description states that 19 had endpoint-group data and 18 had paired samples. The record does not explain the enrollment discrepancy, define the analysis population, or state how the participant with only one available sample was handled.

The entry lists eight outcomes and eight analyses but does not report the analysis methods, multiplicity approach, or whether the analyses were prespecified. The study’s own sample-size justification and statistical plan are also not reported in the registry record. Planning assumptions based on a two-arm comparison do not match the described single-group design and cannot establish the study’s observed power.

Safety interpretation is limited as well: the registry lists one other adverse event, does not identify it in the available information, and does not report a death count. These limitations constrain what can be concluded from this record, independent of the study’s clinical motivation.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Safety and Efficacy of Pioglitazone as an Anti-inflammatory for the Treatment of Cystic Fibrosis (CF) Lung Disease — NCT00322868.
• Registry record last updated 2018-02-23; status last verified 2018-02.
• Results first posted 2018-02-23.
• 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

NCT00322868 is a completed pilot study of pioglitazone for cystic fibrosis lung disease. The registry lists 21 participants enrolled and describes a single-group design with no masking. Participants received pioglitazone, 30 mg orally once daily, for 28 days. The primary endpoint was sputum white cell count on a log 10 (cells/mL) scale, assessed at Day 0 and Day 29.

The primary-outcome description refers to 20 enrolled subjects, 19 with endpoint-group data, and 18 with both sputum samples. It does not reconcile those counts with the overall enrollment entry. Numerical primary-outcome results and the statistical analysis methods are not reported in the registry record. The adverse-event entry lists 0 serious events and 1 other event among 20 at risk; the death count is not reported. Thus the record describes the study design and data collection but does not permit evaluation of the primary outcome’s magnitude or precision.