Efficacy of Varenicline in Ambivalent Smokers: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Efficacy of Varenicline in Ambivalent Smokers (NCT00595868), a randomized phase 2 study. The primary outcome is Number of Participants With a Quit Attempt.

2. Clinical Question

3. Trial Design

The registry describes a phase 2, randomized, quadruple-masked, parallel-group study comparing varenicline with placebo. It lists enrollment as 220. The posted primary-outcome groups contain 107 participants in the varenicline arm and 111 in the placebo arm; these arm counts should be read as reported, rather than reconciled with the enrollment field. The sponsor was University of Vermont Medical Center. The record lists the study start as 2008-03, primary completion as 2010-01, and status as completed.

The trial’s stated aim was to test whether varenicline could help people who were interested in quitting at some point but did not currently plan to quit reduce smoking and eventually quit. The primary endpoint was participants with a quit attempt, assessed over 6 months. Quit attempts were self-reported: the record describes daily information collected since the previous contact, including whether a quit attempt occurred that day. Contacts comprised visits during the first 2 months and monthly phone calls during months 3–6.

The posted secondary outcome was 7-day point-prevalent abstinence verified by breath carbon monoxide of less than 10 parts per million, also assessed at 6 months. An attempt to quit and verified abstinence are related but different outcomes: an attempt can occur without meeting the abstinence definition, while the abstinence measure concerns smoking status over a specified period and uses a breath measurement. Further details about eligibility criteria, the treatment regimen, allocation procedures, masking procedures, and the trial’s original design assumptions 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: Enrolled sample size of 220; 2 were excluded because found to be ineligible after enrollment

Overall Study

Reasons for not completing this period:

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

6. Primary Endpoint

The registry lists the primary outcome as Number of Participants With a Quit Attempt, measured over 6 months.

The primary endpoint counts participants with a quit attempt over 6 months. The registry defines an attempt as a self-reported attempt to quit smoking on a given day, recorded through a Time Line Follow Back. The record says this information was collected at each visit during the first 2 months and through monthly phone calls during months 3–6. For each day since the prior contact, the measure collected cigarette use, whether study medication was used, and whether a quit attempt occurred.

This endpoint focuses on an attempt to change smoking behavior rather than sustained abstinence. Counting participants with an attempt makes the outcome interpretable as the number of people who experienced the event, rather than the number of separate attempts. It can therefore address a behavioral step toward quitting among people who were described as having no current plans to quit. It does not, by itself, establish that a participant stopped smoking or maintained abstinence.

The secondary endpoint measured 7-day point-prevalent abstinence verified by breath carbon monoxide of less than 10 parts per million. That outcome supplies a distinct, more directly abstinence-focused measure. The registry does not report further endpoint definitions, such as additional rules for classifying missing or conflicting self-reports.

7. Results Posted in the Registry

• Varenicline: 51.4% (55/107) versus 32.4% (36/111) on control; risk difference 19.0 percentage points (95% CI 6.1 to 31.8), risk ratio 1.58 (1.14–2.19); unadjusted two-proportion z-test p = 0.005.

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.

7 Day Point Prevalent Abstinence Verified by Breath Carbon Monoxide of Less Than 10 Parts Per Million

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Varenicline: 14.0% (15/107) versus 7.2% (8/111) on control; risk difference 6.8 percentage points (95% CI -1.3 to 15.0), risk ratio 1.95 (0.86–4.40); unadjusted two-proportion z-test p = 0.102.
• 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 posted secondary endpoint is 7-day point-prevalent abstinence verified by breath carbon monoxide of less than 10 parts per million, assessed at 6 months. Its reported comparison is directionally consistent with the primary endpoint in that the estimates favor the varenicline arm. However, the secondary outcome measures abstinence rather than a quit attempt, so it is not a direct replication of the primary outcome. The secondary estimates are also uncertain: the sponsor-reported confidence interval includes the odds-ratio null value, as does the interval for the unadjusted risk difference on its own scale.

The pattern therefore offers related but not identical evidence. A participant may make a quit attempt without meeting the abstinence definition, and the registry does not report individual-level links between these outcomes. The results should not be described as though the primary finding automatically establishes the secondary outcome, or vice versa. The posted data support reading the endpoints separately while noting their broadly similar direction.

In general, secondary outcomes are often interpreted more cautiously than a prespecified primary endpoint. If many outcomes or comparisons are examined, the chance of obtaining an apparently notable result by chance can increase unless multiplicity is addressed. The registry does not report whether a multiplicity adjustment was planned or applied, nor does it provide a comprehensive list of all outcomes considered beyond the posted measures. The secondary result should therefore be treated as supportive and exploratory rather than as an independent confirmatory conclusion. Its interpretation also depends on the outcome definition, verification procedure, missing-data handling, and analysis population, details that are not fully reported in the registry record.

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

• 7 Day Point Prevalent Abstinence Verified by Breath Carbon Monoxide of Less Than 10 Parts Per Million

9. Statistical Interpretation

10. Statistical Methods Explained

The registry lists the sponsor’s primary analysis as a chi-squared comparison reported on the odds-ratio scale. It also contains enough posted counts to calculate simple, unadjusted comparisons. These answer related questions but are not interchangeable: the sponsor’s reported analysis is the registry’s own analysis, while the re-calculations are derived directly from the displayed event counts and group sizes. The record flags a discrepancy, so the sponsor’s analysis takes precedence when describing the trial’s own inferential result.

What does a risk difference describe?

In general, a risk difference subtracts the event proportion in one group from the event proportion in the other. It expresses the contrast on an absolute scale: how far apart the observed event proportions are. The re-calculated primary risk difference is 0.190, with a 95% confidence interval from 0.061 to 0.318. For the secondary endpoint, the re-calculated risk difference is 0.068, with a 95% confidence interval from -0.013 to 0.150. These are unadjusted estimates based on posted summary counts; they do not account for any covariates or analysis choices that may have been used in the sponsor’s analysis.

What does an odds ratio describe?

In general, an odds ratio compares the odds of an event between groups. An odds ratio above 1 indicates higher observed odds in the group named in the comparison, while an odds ratio below 1 indicates lower observed odds. The odds ratio is not the same quantity as a risk difference, and the two measures need not have matching numerical values. The sponsor reported odds ratios for both outcomes: 1.7 for the primary endpoint and 2.1 for the secondary endpoint. The record identifies chi-squared as the method for those sponsor-reported comparisons.

How should confidence intervals be read?

In general, a confidence interval describes statistical uncertainty around an estimate under the assumptions of the method used. A narrower interval indicates greater precision on that scale; a wider interval indicates less precision. An interval that includes the null value—no difference for a risk difference, or 1 for an odds ratio—does not rule out that null value at the corresponding confidence level. It does not prove that the null is true. The primary sponsor-reported odds-ratio interval begins at 1.0, while the secondary sponsor-reported interval spans 1. The intervals should be considered alongside the estimates, outcome definitions, and the discrepancy between sponsor results and unadjusted re-calculations.

What does a p-value contribute?

In general, a p-value describes how unusual data at least as incompatible with a null hypothesis would be if that null hypothesis and the statistical model were correct. It is not the probability that the null hypothesis is true, and it does not measure the size or practical importance of an effect. The sponsor-reported primary p-value is 0.06, while the re-calculated unadjusted primary p-value is 0.005. Given the recorded discrepancy, these values should not be selectively combined or treated as equivalent evidence. The sponsor’s analysis is the registry’s reported trial analysis. For the secondary outcome, the sponsor-reported p-value is 0.10 and the unadjusted re-calculation is 0.102.

Why does the analysis scale matter?

Different effect scales can emphasize different features of the same comparison. Absolute differences can be more directly connected to the difference in observed event proportions, whereas odds ratios compare odds. A simple unadjusted calculation also may not reproduce an analysis that uses other methods or population rules. The registry does not report the full statistical plan, model details beyond the listed chi-squared method, or analysis-population definitions. It is therefore not possible to establish from the record why the sponsor and re-calculated primary results differ.

For binary outcomes, the posted values are counts of participants with an event. The primary outcome is not a count of repeated attempts; it is the number of participants classified as having a quit attempt. Similarly, the secondary outcome counts participants meeting the stated abstinence endpoint. The distinction matters because the interpretation of an event count depends on the outcome definition and its observation process. The record describes self-reported daily quit-attempt information gathered at scheduled contacts, while the secondary outcome includes breath carbon monoxide verification.

11. Confidence Intervals and Precision

The primary re-calculated risk difference is 0.190, and its 95% confidence interval runs from 0.061 to 0.318. This interval is entirely above zero, indicating that the unadjusted estimate is compatible with a positive difference on the risk-difference scale under the calculation used. It does not resolve the conflict with the sponsor-reported primary analysis, whose odds ratio is 1.7 with a 95% confidence interval from 1.0 to 2.9 and p = 0.06. The interval’s lower endpoint reaches the odds-ratio null value.

For the secondary endpoint, the re-calculated risk difference is 0.068, with a 95% confidence interval from -0.013 to 0.150. This interval includes zero and allows both a small negative difference and a positive difference. The sponsor-reported odds ratio is 2.1, with a 95% confidence interval from 0.9 to 5.2, which includes 1. These intervals are broad relative to their point estimates, so the registry’s results do not pin down the size of the secondary association precisely.

In general, a confidence interval is an uncertainty range for an estimate under specified statistical assumptions; it is not a guarantee that the true effect lies within the particular interval. Width reflects precision, while the location of the interval reflects the estimated direction and scale. Comparisons across risk-difference and odds-ratio intervals require care because the measures use different scales. The record does not report the full calculation details needed to explain the discrepancy between the primary re-calculation and sponsor analysis. That discrepancy is an important limitation on precision claims, and the sponsor’s own analysis takes precedence.

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 planning figures are participants per arm, not total events. The registry’s planning illustration assumes a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. It uses the observed effect as its starting point and considers planning for 100%, 75%, or 50% of that observed effect. These are planning assumptions for an illustration, not the trial’s original design assumptions; those original assumptions are not reported in the registry record.

For the primary outcome, planning for 100% of the observed effect corresponds to 106 participants per arm for 80% power and 141 participants per arm for 90% power. Planning for 75% corresponds to 185 participants per arm for 80% power and 247 participants per arm for 90% power. Planning for 50% corresponds to 407 participants per arm for 80% power and 544 participants per arm for 90% power. The secondary outcome has its own planning figures, based on its observed effect: for 100%, 320 participants per arm for 80% power and 428 participants per arm for 90% power; for 75%, 529 participants per arm for 80% power and 708 participants per arm for 90% power; and for 50%, 1,098 participants per arm for 80% power and 1,469 participants per arm for 90% power.

In general, the 80% and 90% power columns describe the planned chance of detecting the specified effect under the stated assumptions if that effect is real. The higher target requires more participants in these illustrations. Smaller target effects also require more participants, because they are harder to distinguish from random variation. These figures are planning illustrations, not observed power.

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 = 0.324; treatment proportion p₁ = 0.514; average p̄ = 0.419.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6978 = 1.3677.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6848 = 0.5763.
4. n = (1.3677 + 0.5763)² / (0.1897)² = 105.0, rounded up to 106 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 <- 0.324324324324324                          # control proportion
p1 <- 0.514018691588785                          # 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(595868)
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 = 106 (80% power), 141 (90% power)
 75% of observed effect: n per arm = 185 (80% power), 247 (90% power)
 50% of observed effect: n per arm = 407 (80% power), 544 (90% power)
power.prop.test: 105 
Simulated power at n = 106 per arm: 0.800 (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 registry’s own primary analysis is a chi-squared analysis reported as an odds ratio of 1.7, with a 95% confidence interval from 1.0 to 2.9 and p = 0.06. A separate simple, unadjusted re-calculation from the posted event counts gives a risk difference of 0.190, with a 95% confidence interval from 0.061 to 0.318 and p = 0.005. The measures differ, and the p-values do not agree. The registry flags this as a discrepancy.

The re-calculation is useful as a transparent description of the unadjusted contrast implied by the posted counts, but it does not override the sponsor’s analysis. The registry’s sponsor-reported result takes precedence for describing the trial’s own inferential analysis. The record does not explain the source of the discrepancy, so it is not possible to determine whether it reflects a difference in analysis method, population, data handling, or another factor. No specific explanation should be inferred.

The same distinction applies to the secondary endpoint: the sponsor-reported odds ratio is 2.1, with a 95% confidence interval from 0.9 to 5.2 and p = 0.10, while the unadjusted re-calculation is on the risk-difference scale. These are complementary descriptions, not interchangeable outputs. Readers should keep the registry’s analysis separate from the re-calculated comparison and avoid selecting whichever result appears more persuasive.

15. Multiplicity, Missing Data and Analysis Populations

The registry record does not report whether the trial used a prespecified plan to control multiplicity across endpoints, time points, or other comparisons. This matters because repeated testing can make nominal p-values less informative: across multiple tests, some results may appear statistically notable by chance even when there is no underlying effect. In general, a multiplicity strategy defines which comparisons are confirmatory and how inference is adjusted or ordered. Without that information, the secondary result should not be treated as though its nominal significance level necessarily reflects a family-wide error rate.

Missing-data procedures are not reported in the registry record. The posted group sizes and outcome counts do not establish how missing observations were handled, whether every randomized participant was included in each analysis, or whether any imputation or sensitivity analysis was performed. These details can affect interpretation, particularly when outcomes are assessed over time and the primary endpoint relies on self-reported information gathered at visits and calls. No assumptions about missing outcomes should be inferred from the posted counts alone.

The analysis populations are also not reported in sufficient detail. The registry does not state whether the comparisons followed an intention-to-treat approach, a per-protocol approach, or another population definition. It does not report participant flow, treatment discontinuation, adherence, or exclusions from each analysis. The record’s listed enrollment and posted arm denominators should therefore be described as recorded rather than used to infer unreported analysis rules.

The primary discrepancy adds another interpretive limitation. The sponsor-reported odds-ratio analysis and the unadjusted risk-difference re-calculation differ, and the registry does not explain why. The sponsor’s own analysis takes precedence, but the absence of a detailed statistical plan prevents a fuller assessment of the analysis choices. Together, unreported multiplicity procedures, missing-data methods, and analysis populations constrain how confidently the posted p-values can be interpreted.

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 5% reporting threshold.

The serious adverse event record lists 0 serious adverse events among 107 participants at risk in the varenicline group and 0 among 111 participants at risk in the placebo group. No serious adverse event terms are listed. These are the registry’s posted serious-event figures and should be read as such; they do not establish that no adverse experiences of any kind occurred.

The fields for other adverse events show 0 participants at risk in both groups, so those fields do not provide a usable comparison of other adverse events. Death counts are not reported in the registry record. The record also does not provide further detail about adverse-event definitions, collection, observation periods, or whether all participants contributed safety follow-up. The serious-event figures alone cannot characterize the full safety experience of the trial.

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

The primary endpoint concerns whether participants made a quit attempt during the 6-month assessment period. The posted counts and estimates describe a comparison between randomized groups on that outcome, and the unadjusted re-calculation places the event proportion higher in the varenicline arm. The sponsor’s own primary analysis reports an odds ratio of 1.7 with p = 0.06; because the registry flags a discrepancy, that analysis takes precedence over the re-calculation for describing the trial’s reported result.

The result does not establish that a quit attempt led to abstinence. The secondary endpoint measures verified 7-day point-prevalent abstinence, a separate outcome, and its sponsor-reported interval includes the null value. The trial record also does not report enough methodological detail to resolve the primary-analysis discrepancy or assess missing data, multiplicity, and analysis-population choices.

In general, an estimate is not a guarantee of an individual outcome or proof of a causal mechanism. A group-level comparison does not explain why participants differed, and the listed endpoints do not measure every aspect of smoking behavior or longer-term abstinence. The registry’s summary describes a phase 2 study, but the record does not provide a basis for extrapolating these findings beyond the enrolled study population or its specified follow-up.

These results should therefore be read as registry-reported comparisons with uncertainty, not as medical advice or a definitive statement about treatment. The primary outcome, the abstinence outcome, the sponsor analysis, and the unadjusted re-calculation each answer somewhat different questions. Keeping those distinctions visible is essential to avoid overstating what the posted data establish.

18. Why This Trial Matters Statistically

This trial raises a useful statistical distinction between attempting to quit and achieving a defined abstinence outcome. The primary endpoint counts participants reporting a quit attempt, while the secondary endpoint uses a specific period of abstinence verified by breath carbon monoxide. Treating these as separate outcomes makes clear that behavioral change and verified abstinence are not interchangeable measures.

The record also illustrates why the reported analysis and a simple calculation from event counts should be kept distinct. The re-calculated primary risk difference and p-value differ from the sponsor-reported odds ratio and p-value. A transparent re-calculation can help readers understand the posted counts, but it cannot replace the trial’s own analysis when the registry identifies a discrepancy. Reporting both, with the sponsor analysis taking precedence, avoids concealing that uncertainty.

Precision is another central issue. The primary sponsor-reported interval reaches the null value, and the secondary outcome has intervals that include their null values. These estimates do not settle the magnitude of any association. The sample-size illustrations further show, under their stated assumptions, that planning around smaller effects calls for more participants per arm; those calculations should not be confused with the trial’s original design rationale or with observed power.

Finally, the record’s omissions matter statistically. The absence of reported multiplicity procedures, missing-data methods, analysis populations, and original design assumptions limits what can be concluded from the posted inferential results. These are not merely technical details: they determine how uncertainty, selection, and repeated comparisons should be understood.

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.

Several limitations are specific to the information available for this trial. The main one is the discrepancy between the sponsor-reported primary analysis and the unadjusted re-calculation from posted counts. The registry does not explain its source, so the re-calculated result cannot be used to resolve the sponsor’s reported conclusion. The sponsor analysis takes precedence, but its p-value and interval still indicate uncertainty around the estimated comparison.

The primary endpoint depends on self-reported quit attempts collected at visits and by phone. The registry describes the collection schedule and daily information requested, but does not report validation of the quit-attempt reports or the rules used when information was incomplete. The secondary endpoint includes breath carbon monoxide verification, but additional details of its assessment and missing-data handling are not reported.

The record does not give the trial’s original sample-size assumptions, a full statistical analysis plan, multiplicity procedures, or analysis-population definitions. It also does not describe participant flow, adherence, treatment discontinuation, or reasons for missing observations. These omissions constrain assessment of the reported comparisons and prevent explanation of how the planned analysis relates to the re-calculation.

The record lists enrollment as 220, while the posted primary outcome groups have denominators of 107 and 111. No reconciliation is reported, so these figures should remain as stated rather than being treated as evidence about exclusions or analysis populations. The registry also provides limited safety information: serious-event counts are posted, other-event risk fields are zero, and deaths are not reported. These data do not constitute a complete account of safety outcomes.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Efficacy of Varenicline in Ambivalent Smokers — NCT00595868.
• Registry record last updated 2011-12-02; status last verified 2011-11.
• Results first posted 2011-11-22.
• 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

The registry describes NCT00595868, “Efficacy of Varenicline in Ambivalent Smokers,” as a completed phase 2 randomized, quadruple-masked, parallel-group study sponsored by University of Vermont Medical Center. It lists enrollment as 220, with a start date of 2008-03 and primary completion in 2010-01. The study compared varenicline with placebo in ambivalent smokers.

The primary endpoint was the number of participants with a quit attempt over 6 months, assessed through self-report. The secondary outcome was 7-day point-prevalent abstinence verified by breath carbon monoxide of less than 10 parts per million. The registry reports a sponsor analysis for each outcome and also supports unadjusted re-calculations from posted counts. Because the record flags a primary-analysis discrepancy, the sponsor’s own analysis takes precedence. Important details about multiplicity, missing data, analysis populations, and original design assumptions are not reported in the registry record.