Smoking Cessation in Subjects With Mild-to-moderate Chronic Obstructive Pulmonary Disease (COPD).: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Smoking Cessation in Subjects With Mild-to-moderate Chronic Obstructive Pulmonary Disease (COPD). (NCT00285012), a randomized phase 3 study. The primary outcome is Number of Subjects With Four Week Continuous Quit Rate (CQR).

2. Clinical Question

The trial asked whether smoking cessation outcomes differed between varenicline and placebo among smokers with mild-to-moderate COPD. Its primary endpoint was a count of participants meeting a defined continuous quit criterion, rather than a measure of lung function or a comparison of changes in COPD symptoms.

For this endpoint, participants had to report no smoking and no use of other nicotine-containing products since the last study visit on the Nicotine Use Inventory. They also had to have an end-expiratory exhaled carbon monoxide measurement less than or equal to 10 parts per million for the relevant weeks. The endpoint therefore combines participant report with a biochemical measure. It focuses on sustained abstinence over the final four weeks of the treatment period, as specified in the registry record.

The posted results show 105 participants meeting the endpoint in the varenicline group and 22 in the placebo group. The corresponding group sizes in the primary results are 248 and 251 participants. The comparison and its uncertainty are; the counts alone do not establish how the trial was analyzed or how broadly its findings apply. The registry record does not report additional eligibility details beyond the description of smokers with mild-to-moderate COPD and the primary analysis population.

3. Trial Design

This was a Phase 3 randomized, double-blind, placebo-controlled, multicenter, parallel-group trial. The registry lists triple masking. It records an enrollment of 504 and a completed status. The study title describes a 12-week treatment period followed by 40 weeks of follow-up. The recorded start was 2006-05, and primary completion was 2008-07.

The two interventions were placebo and varenicline tartrate. The registry's arm descriptions are blank, so details such as dose, dosing schedule, and any treatment adjustments are not reported in the registry record. Random allocation and masking are stated, but the allocation procedure and the parties included in the masking are not described in the record.

The primary endpoint was assessed during Week 9 through Week 12. Its analysis population is described as all subjects who received at least one dose, including partial doses, of randomized study drug. The posted primary counts are based on 248 participants in the varenicline group and 251 in the placebo group. These group sizes should be distinguished from the enrollment figure: the registry does not explain the flow from enrollment to the populations used in the posted result.

The registry lists multiple other outcome measures, including abstinence measures, lung function, COPD questionnaire scores, cigarette use, inflammatory biomarkers, and body weight. The primary result is a smoking cessation endpoint; it should not be read as a direct estimate of changes in COPD severity. Specific details about site locations, concealment procedures, adherence, reasons for discontinuation, and the trial's original sample-size justification 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: Subjects were current cigarette smokers 35 years of age or older with mild to moderate Chronic Obstructive Pulmonary Disease (COPD) as defined by Global Initiative for Chronic Obstructive Lung Diseases (GOLD) criteria.

Assigned to Study Treatment (Tx)

Reasons for not completing this period:

Received Study Tx

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All subjects population: received at least 1 dose, including partial doses, of randomized study drug.

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 Number of Subjects With Four Week Continuous Quit Rate (CQR), measured over Week 9 through Week 12.

The primary endpoint is the number of subjects with a four-week continuous quit rate during Week 9 through Week 12. A participant had to report no smoking and no use of other nicotine-containing products since the last study visit on the Nicotine Use Inventory, and have an end-expiratory exhaled carbon monoxide measurement less than or equal to 10 parts per million for the specified weeks. The registry defines the analysis population as subjects who received at least one dose, including partial doses, of randomized study drug.

This kind of endpoint is used to represent abstinence sustained across a defined period, rather than abstinence reported at only a single time point. The use of both participant report and a carbon monoxide measurement adds a biochemical component to the classification. These features describe how the endpoint was defined; they do not eliminate possible measurement error or establish that all participants had complete outcome data.

The posted results count 105 participants in the varenicline group and 22 in the placebo group as meeting the endpoint. The group denominators are 248 and 251. The endpoint is specifically about continuous quitting during the stated weeks. It is not itself a measure of long-term abstinence beyond that period, lung function, COPD symptoms, or safety. The registry lists other outcomes addressing some of those topics, but their results are not all detailed in the information.

7. Results Posted in the Registry

• Varenicline: 42.3% (105/248) versus 8.8% (22/251) on control; risk difference 33.6 percentage points (95% CI 26.5 to 40.6), risk ratio 4.83 (3.16–7.39); unadjusted two-proportion z-test p = <0.001.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Change From Baseline in Body Weight

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Varenicline: mean 2.40 versus 1.50 on control; difference 0.90 (95% CI -0.10 to 1.90), standardized effect d = 0.17; z-test p = 0.078.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Varenicline and the control. The interval gives the range of effects that remain compatible with the data.

The registry lists secondary outcomes covering several domains: other definitions of abstinence, lung function, COPD questionnaire scores, daily cigarette use, inflammatory biomarkers, and body weight. These outcomes can add context to a trial whose primary question concerns continuous quitting, but they are not interchangeable with the primary endpoint. Abstinence measures address related but distinct definitions or periods; lung function and questionnaire measures address different aspects of COPD; biomarkers and weight capture other outcomes.

The detailed secondary result is change in body weight from baseline to Week 52. The posted means are 2.4 kg for varenicline and 1.5 kg for placebo, with a recalculated mean difference of 0.9 kg and a confidence interval from -0.101 to 1.901. The unadjusted p-value is 0.078. This interval crosses zero, so the result does not provide a clear unadjusted separation between the groups on that measure. It is a secondary outcome and should not be used to reinterpret the primary cessation comparison.

The registry lists nine posted secondary outcomes, but the detailed results for the other outcomes are not reported in the registry record. Consequently, an overall direction or consistency across all secondary endpoints cannot be established from these details. The absence of those results should not be mistaken for evidence that the outcomes were unchanged.

Secondary analyses generally carry less weight than the prespecified primary question when many outcomes are considered. Repeated testing creates more opportunities for apparently notable findings by chance. The registry record does not report a multiplicity adjustment or a hierarchy for interpreting secondary outcomes. Therefore, secondary results are best treated as supportive or exploratory unless a prespecified analysis plan and its multiplicity approach are available. They can inform the broader picture, but they do not independently confirm the primary result or establish conclusions about every outcome domain.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Number of Subjects With Continuous Abstinence (CA)
• Number of Subjects With Long Term Quit Rate (LTQR)
• Number of Subjects With 7-Day Point Prevalence of Abstinence
• Number of Subjects With 4-Week Point Prevalence of Abstinence
• Change From Baseline in Pre-bronchodilator and Post-bronchodilator Forced Expiratory Volume in First Second (FEV1)
• Change From Baseline in Clinical COPD Questionnaire (CCQ)
• Number of Cigarettes Smoked Daily During First 3 Weeks of the 12-Week Treatment Period
• Change From Baseline in Inflammatory Biomarkers: C-Reactive Protein (CRP) and Fibrinogen Antigen

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is binary at the participant level: each person either met the registry's four-week continuous quit definition or did not. For a binary endpoint, the group count is interpreted alongside the number of participants assessed. The posted results give the number meeting the endpoint and the group denominator, allowing an unadjusted comparison of the observed groups. A risk difference expresses the difference between the groups' event probabilities on the probability scale. Here, the recalculated risk difference is 0.336.

The confidence interval for the risk difference, 0.265 to 0.406, summarizes the uncertainty associated with that unadjusted estimate under the calculation used for the posted comparison. A confidence interval is a range generated by a statistical procedure; it does not mean that there is a 95% probability that this particular interval contains the fixed true effect. Its width reflects both the estimated effect and the amount of information in the data. The interval does not account automatically for every possible source of bias, such as missing outcome data or limitations in the registry description.

A risk ratio and an odds ratio are alternative ways to compare binary outcomes. A risk ratio compares the probability of the event between groups. An odds ratio compares event odds, where odds are the event probability divided by the probability of no event. Odds and probabilities are not the same quantity, so an odds ratio should not be translated into a risk difference without further information. The recalculation also reports a risk ratio of 4.83, with a confidence interval from 3.159 to 7.387, and an odds ratio of 7.643, with a confidence interval from 4.613 to 12.66. These are unadjusted calculations from the posted summary counts.

The sponsor's reported analysis uses logistic regression and gives an odds ratio of 8.40, with a 95% confidence interval from 4.99 to 14.14. Logistic regression is a model for the odds of a binary outcome. It can incorporate explanatory variables, but the covariates and model specification are not reported in the registry record. The reported regression result is therefore not necessarily identical to an unadjusted odds ratio calculated directly from the group counts. The sponsor's reported analysis is the trial's own analysis; the recalculated values are a separate, simple summary-based comparison.

The reported p-values are measures of how incompatible the data are with a specified null comparison under the assumptions of the relevant analysis. They are not measures of the size or importance of an effect, and they are not the probability that the null hypothesis is true. For the primary recalculation, p <0.001; for the sponsor's reported logistic-regression result, p < 0.001. These values come from different analysis descriptions and should be read in that context.

Randomization supports causal comparisons when the trial is conducted and analyzed appropriately, but the registry record does not provide all details needed to assess that condition. In particular, the statistical analysis plan, the precise handling of missing data, any sensitivity analyses, and details of the regression model are not reported in the registry record. The primary population is described as participants who received at least one dose, but the record does not explain how that population relates to all randomized participants or how participants without endpoint assessments were treated.

Why distinguish odds from risks?

Risk is the proportion of participants experiencing an outcome over a defined period. Odds compare the number experiencing the outcome with the number not experiencing it. An odds ratio can be useful in logistic regression and is the scale reported by the sponsor, but it can sound more dramatic than a risk comparison when read as though it were a probability ratio. Keeping the measure's scale explicit prevents that interpretive error.

What does an unadjusted recalculation add?

A simple recalculation makes the comparison transparent by using the posted counts and denominators without modeling covariates. It provides a way to inspect the scale of the group contrast and its sampling uncertainty. It does not replace the sponsor's analysis or reconstruct unreported aspects of the trial analysis. If the registry record does not state model details, the recalculation should not be described as adjusted or as reproducing the trial's full statistical plan.

What can and cannot be inferred from a confidence interval?

The interval conveys a range of estimates compatible with the data under the calculation's assumptions. A narrower interval generally indicates greater statistical precision, while a wider interval indicates more uncertainty. Neither width nor statistical significance resolves questions about endpoint validity, missingness, generalizability, or whether an effect matters to patients. Those require information beyond the interval itself.

11. Confidence Intervals and Precision

The primary risk-difference estimate is 0.336, with a 95% confidence interval from 0.265 to 0.406. The interval lies above zero, and the recalculated p-value is p <0.001. This indicates that, under the unadjusted calculation, the data are inconsistent with a zero risk difference at conventional significance levels. The estimate and interval describe the magnitude and precision of this particular group comparison; they do not address every limitation in the trial record.

The posted regression result is an odds ratio of 8.40, with a 95% confidence interval from 4.99 to 14.14. Its interval also lies above the no-association value for an odds ratio. The estimate is on the odds scale, and the record does not describe the regression covariates. It should therefore be kept distinct from the risk-difference interval rather than read as a second expression of the same quantity.

Confidence intervals summarize sampling uncertainty under the methods used. They do not incorporate unreported design features, possible errors in outcome classification, or uncertainty caused by unreported missing-data handling. The primary counts and intervals give useful information about the observed comparison, but the record does not include enough detail to evaluate every source of uncertainty.

Precision is not synonymous with clinical importance. An interval can be statistically separated from a null value while leaving other questions unanswered, including the durability of abstinence beyond the measured weeks and the relevance of the result to people outside the trial population. The registry's endpoint and follow-up structure define what this comparison can address.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The sample-size display is a planning illustration based on the observed effect, not a record of the trial's original design calculation. For the binary primary endpoint, the figures are participants per arm, rather than total events. The planning assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The effect to detect is the stated fraction of the observed effect.

The table compares planning for 100% of the observed effect with planning for 75% and 50% of the observed effect. For each effect size, it gives participants per arm for 80% power and participants per arm for 90% power. In plain language, a power target describes the chance, under the planning assumptions and a specified true effect, of detecting that effect using the chosen statistical criterion. The 90% column therefore represents a higher target chance of detection than the 80% column, and it calls for more participants under the same assumptions.

The displayed curve helps show how the planning requirement changes as the assumed effect changes. Smaller effects need more participants to detect with the same target power. The values are conditional on the stated assumptions; changing allocation, significance level, analysis method, or the effect being targeted would change the planning illustration. For continuous outcomes, the same record notes that the illustration is based on the observed effect and gives participants per arm; time-to-event displays, if present, use total events.

These figures are not observed power. They do not measure the probability that the completed trial's result is correct, nor do they explain how the actual trial enrollment was chosen. The original design assumptions and sample-size justification are not reported in the registry record.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.088; treatment proportion p₁ = 0.423; average p̄ = 0.256.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6168 = 1.2089.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.5693 = 0.4791.
4. n = (1.2089 + 0.4791)² / (0.3357)² = 25.3, rounded up to 26 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
p0 <- 0.0876494023904383                          # control proportion
p1 <- 0.423387096774194                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(285012)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 26 (80% power), 34 (90% power)
 75% of observed effect: n per arm = 41 (80% power), 54 (90% power)
 50% of observed effect: n per arm = 78 (80% power), 104 (90% power)
power.prop.test: 25.3 
Simulated power at n = 26 per arm: 0.842 (20000 replicates)

The simulation draws many pairs of event counts at the planned sample size and counts how often the two-proportion z-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

14. Analyses Reported by the Sponsor

The sponsor's own primary analysis is reported as logistic regression, with an odds ratio of 8.40, a 95% confidence interval from 4.99 to 14.14, and p < 0.001. The posted counts also permit an unadjusted recalculation. That calculation gives an odds ratio of 7.643, with a confidence interval from 4.613 to 12.66, and a risk difference of 0.336, with a confidence interval from 0.265 to 0.406. The recalculated risk ratio is 4.83, with a confidence interval from 3.159 to 7.387.

The sponsor's estimate and the recalculated estimate are not identical because the sponsor reports a regression analysis while the recalculation is an unadjusted comparison from summary counts. The registry does not report the model covariates or enough detail to reconstruct the sponsor's model. The difference between these values should not be treated as evidence of an error. The record does not flag a discrepancy, and the sponsor's own analysis is the trial's reported analysis.

The measures also use different scales. The sponsor's odds ratio is not the same as the risk difference, and neither should be substituted for the other. Read the sponsor result as the reported regression comparison and the recalculation as a transparent, unadjusted summary-based comparison. Both accompany the same posted group counts, but their methods and interpretations differ.

15. Multiplicity, Missing Data and Analysis Populations

The registry names a primary endpoint and multiple secondary outcomes, but it does not report a statistical analysis plan or a strategy for controlling multiplicity. Multiplicity arises when investigators assess several outcomes, time points, or comparisons: the more tests considered, the greater the possibility of observing a small p-value by chance somewhere among them. Whether and how to adjust depends on the planned hierarchy and the intended interpretation of the outcomes. The record does not state whether any adjustment was used.

It is also not reported in the registry record whether interim analyses were planned or conducted. No interim analysis schedule, stopping rule, or related adjustment is described. These details matter because repeated looks at accumulating data can affect the properties of conventional significance tests if not accounted for in the analysis plan. Their absence from the record does not establish that such procedures did or did not occur.

The primary population is described as subjects who received at least one dose, including partial doses, of randomized study drug. The registry does not fully describe the analysis population beyond that definition, nor does it report how missing primary outcomes were handled. The disposition of participants, reasons for missing outcome assessments, and any imputation or sensitivity analyses are not reported in the registry record. This leaves uncertainty about how incomplete data may have affected the primary comparison.

The registry also does not report whether the outcome assessment was analyzed according to an intention-to-treat principle, whether participants were retained in their assigned groups after discontinuation, or how protocol deviations were handled. The randomized design is relevant to interpreting group comparisons, but the analysis details are needed to evaluate whether the posted estimate preserves that design's intended comparison.

These reporting gaps do not negate the posted estimates. They define limits on what can be independently assessed from the registry record. A fuller evaluation would require the analysis plan, participant flow, missing-data methods, and details of any multiplicity or interim-analysis procedures. Without those details, p-values and confidence intervals should be understood as attached to the reported analyses rather than as a complete account of all analytic uncertainty.

16. 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 2% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: lung neoplasm malignant (0.8%), acute myocardial infarction (0.4%), angina pectoris (0.4%), cellulitis (0.4%), back pain (0.4%).

The registry reports serious adverse events in 6 of 248 participants in the varenicline group and 11 of 251 participants in the placebo group. It also reports other, non-serious adverse events in 157 of 248 and 121 of 251 participants, respectively. These are counts of participants with events as presented in the record. They are not a full description of event timing, severity gradation, duration, recurrence, or the statistical comparison between groups.

Several terms are listed among the top serious adverse-event terms, with the percentages shown in the record: lung neoplasm malignant at 0.8%, acute myocardial infarction at 0.4%, angina pectoris at 0.4%, cellulitis at 0.4%, back pain at 0.4%, and cerebrovascular accident at 0.4%. The list does not assign each term to a particular group in the details, so group-specific attribution should not be inferred. Death counts are not reported in the registry record.

Adverse-event counts do not by themselves establish that an event was caused by a study drug. They also do not determine comparative safety without considering event definitions, observation time, participant characteristics, and the handling of incomplete follow-up. These figures are descriptive registry results, not medical advice.

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

The primary posted data show that more participants in the varenicline group than in the placebo group met the registry's definition of four-week continuous quitting during Week 9 through Week 12. The unadjusted risk difference is 0.336, with a 95% confidence interval from 0.265 to 0.406. The sponsor's logistic-regression analysis also reports a group difference on the odds scale. Together, these results describe the trial's observed comparison under its endpoint definition.

The endpoint requires reported abstinence from smoking and other nicotine-containing products, together with an exhaled carbon monoxide measurement less than or equal to 10 parts per million. It is a measure of abstinence for a defined period, not a measure of permanent cessation. The trial title describes follow-up beyond the treatment period, but the primary endpoint is specifically defined for Week 9 through Week 12. Results for other abstinence measures are not detailed.

The result does not establish the effect in every person with COPD, in people with different disease severity, or in populations unlike those enrolled. The registry describes the intended population as cigarette smokers with mild-to-moderate COPD, but it does not provide enough participant characteristics to judge representativeness. The result also does not establish changes in lung function, COPD symptoms, or inflammatory biomarkers; those are distinct outcomes.

A statistical result is not a treatment recommendation. The group comparison does not predict an individual participant's outcome, and the record's safety counts should not be interpreted as a complete benefit-risk assessment. The registry summary also omits important analytic details, including missing-data handling and the sponsor model specification. Those limitations mean the estimates should be read as evidence about this trial's posted data, not as a complete account of all relevant clinical questions.

Finally, a small p-value does not mean there is a corresponding probability that the conclusion is true. The confidence interval describes statistical precision under the analysis assumptions, while questions of generalizability, measurement, missingness, and clinical importance require separate judgment and information. The trial supports a comparison of the two randomized groups on its defined endpoint; it does not settle every broader question about smoking cessation or COPD.

18. Why This Trial Matters Statistically

This trial illustrates why the outcome definition matters in clinical research. “Quitting” can refer to abstinence at a single assessment, abstinence across a sustained interval, or a longer-term outcome. Here the primary measure requires no reported smoking or other nicotine use across the specified weeks, plus an exhaled carbon monoxide criterion. The precise definition lets readers understand what event the statistical comparison represents.

It also shows why counts, effect measures, and uncertainty should be read together. The group counts are directly visible, while the risk difference, risk ratio, and odds ratio express the comparison on different scales. The confidence intervals show the precision of the summary estimates. Keeping those elements distinct helps prevent an odds ratio from being mistaken for a probability difference, or a p-value from being treated as a measure of importance.

The registry presents both the sponsor's logistic-regression result and a simple unadjusted recalculation. Seeing both emphasizes that analysis choices can affect the reported estimate and that a registry summary may not include enough information to reproduce a model. The unadjusted comparison is transparent from the group counts, while the sponsor's result is the reported trial analysis. Neither makes the missing model details available.

Finally, the record demonstrates the importance of reporting analysis populations, missing-data methods, multiplicity plans, and participant flow. These details affect how confidently readers can assess what the posted estimates mean. A clear primary result is valuable, but interpreting a clinical trial also requires knowing how outcomes were selected, measured, and analyzed.

19. 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 several trial-specific questions unanswered. It does not describe the randomization sequence or allocation concealment, identify which parties were masked, or provide detailed arm descriptions. The trial is listed as randomized, double-blind, and triple-masked, but the practical procedures supporting those design features are not reported in the registry record.

The primary analysis population is defined as subjects who received at least one dose, including partial doses. The record does not show participant flow from enrollment into the analyzed groups, explain how missing primary outcome assessments were treated, or state whether the analysis followed an intention-to-treat approach. Reasons for withdrawal, adherence, and protocol deviations are also not reported in the registry record. These gaps make it difficult to assess potential effects of incomplete follow-up on the comparison.

The sponsor reports a logistic-regression odds ratio, but the regression covariates, model specification, and handling of missing data are not described. The statistical analysis plan, multiplicity approach, interim-analysis procedures, and original sample-size justification are not reported in the registry record. The sample-size display is a planning illustration based on the observed effect and should not be mistaken for the trial's design rationale.

The primary endpoint covers a defined four-week period. It does not by itself establish longer-term abstinence, and the detailed results for many listed secondary outcomes are not reported. The record also offers limited information for evaluating how representative the trial participants were of other people with COPD. Finally, the safety counts are descriptive and do not establish causation or provide a complete safety profile.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Smoking Cessation in Subjects With Mild-to-moderate Chronic Obstructive Pulmonary Disease (COPD). — NCT00285012.
• Registry record last updated 2010-04-20; status last verified 2010-04.
• Results first posted 2009-12-03.
• 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.

23. Record Summary

This completed Phase 3 trial compared varenicline tartrate with placebo in cigarette smokers with mild-to-moderate COPD. The registry describes randomization, a parallel-group design, double-blind and triple masking, a 12-week treatment period, and 40 weeks of follow-up.

The primary endpoint was the number of subjects meeting a four-week continuous quit definition during Week 9 through Week 12. The posted counts were 105 of 248 in the varenicline group and 22 of 251 in the placebo group. The unadjusted recalculated risk difference was 0.336, with a 95% confidence interval from 0.265 to 0.406; the sponsor reported a logistic-regression odds ratio of 8.40, with a 95% confidence interval from 4.99 to 14.14.

The registry lists additional abstinence, COPD, lung-function, biomarker, cigarette-use, and body-weight outcomes. Detailed results for most listed secondary outcomes, along with important analysis and missing-data details, are not reported in the registry record. Serious adverse events were recorded in 6 of 248 participants in the varenicline group and 11 of 251 in the placebo group.