Usefulness of Selegiline as an Aid to Quit Smoking - 1: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Usefulness of Selegiline as an Aid to Quit Smoking - 1 (NCT00129311), a randomized phase 2 study. The primary outcome is 7 Day Point Prevalence of Cigarette Abstinence.

2. Clinical Question

The registry describes the study as a phase 2 trial intended to assess the safety and effectiveness of selegiline as an aid to smoking cessation. The reported primary endpoint focuses on recent abstinence at follow-up: it concerns cigarette abstinence during the 7 days associated with the assessment, rather than continuous abstinence throughout the period since treatment began. The record reports 6 participants with this outcome in the selegiline arm and 8 in the placebo arm. The comparison is therefore about the observed outcome counts and their uncertainty, not a definitive statement about the effect of selegiline in other populations or settings.

The broader clinical context in the registry summary is that some smokers are unable to quit despite widespread use of nicotine replacement therapies. The record does not report whether participants used nicotine replacement, received other cessation support, or had restrictions on other treatments. It also does not give enough detail to characterize the participants beyond the study’s smoking-cessation focus. These gaps matter when considering how closely the enrolled group and study procedures correspond to other clinical contexts.

3. Trial Design

The registry identifies this as a randomized, triple-masked, parallel-group phase 2 study. The two arms were selegiline and placebo. Enrollment was 101. The study is recorded as completed, with a start date of 2004-07 and primary completion in 2012-12. The registry lists Yale University as sponsor and tobacco use cessation as the condition.

Randomized allocation means that assignment to the study arms was intended to be made by a random process. In general, randomization is used to reduce systematic differences between groups at baseline, although it does not guarantee that the groups will be identical. The allocation sequence method and details of how assignments were concealed are not reported in the registry record. The record describes masking as triple, but does not identify which groups of people were masked or explain how masking was maintained.

The parallel model means participants were assigned to one arm rather than receiving both interventions in sequence. The recorded arm counts for the primary outcome are 51 in the selegiline arm and 50 in the placebo arm. The study’s original target enrollment, its recruitment history, and any changes to its planned enrollment are not reported in the registry record.

The primary outcome was a count of participants with 7-day point-prevalence cigarette abstinence at 6-month follow-up. The record lists two outcome measures but does not post any secondary outcomes. It also lists zero posted analysis entries. The reported arm results and comparison statistics permit a reading of the primary result, but they do not provide a full account of the analysis procedures, the handling of participants without follow-up data, or the protocol’s planned statistical methods.

The registry describes the study as examining safety and effectiveness, but that description alone does not specify the detailed protocol or analysis plan. The treatment dose, administration schedule, duration, eligibility criteria, and accompanying cessation support 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.

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 7 Day Point Prevalence of Cigarette Abstinence, measured over 6-month follow up.

The primary endpoint is titled “7 Day Point Prevalence of Cigarette Abstinence,” with the time frame listed as 6-month follow-up. It is recorded as a number of participants, so the reported outcome is a count of participants meeting the endpoint in each arm. The posted count is 6 for selegiline and 8 for placebo.

Point-prevalence abstinence captures whether a participant meets an abstinence definition at a specified assessment window. It differs from continuous abstinence, which would require abstinence over an extended interval. A 7-day point-prevalence measure can describe recent smoking status around follow-up, but it does not by itself establish whether abstinence was maintained throughout the time between treatment and assessment. The record does not describe how abstinence was assessed or whether it was verified by a biochemical measure.

Assessing abstinence at 6-month follow-up gives the outcome a clinically relevant time context, while the brief 7-day window defines what the endpoint counts. Interpretation depends on the endpoint definition and the completeness and reliability of follow-up measurements. The registry record does not report further endpoint details, such as the rules for classifying missing assessments or the precise assessment procedures.

7. Results Posted in the Registry

• Selegiline: 11.8% (6/51) versus 16.0% (8/50) on control; risk difference -4.2 percentage points (95% CI -17.7 to 9.2), risk ratio 0.74 (0.27–1.97); unadjusted two-proportion z-test p = 0.538.

8. Statistical Interpretation

9. Statistical Methods Explained

The registry posts counts for a binary outcome: each participant either met the abstinence endpoint or did not, according to the endpoint definition. For each arm, the count of participants with the outcome is reported alongside the arm size. The comparison statistics summarize how the observed endpoint counts differ between the randomized arms. The record does not post the individual participant data, so the comparison cannot be independently reconstructed beyond the summary statistics recorded.

What does the risk difference describe?

The risk difference is an absolute measure. In general, it is the outcome probability in one group minus the outcome probability in the comparison group. A value of zero corresponds to equal observed probabilities. The posted estimate for selegiline relative to placebo is -0.042. Its negative sign indicates that the observed endpoint probability was lower in the selegiline arm. This direction is descriptive; the uncertainty interval includes zero and extends to either side of it. The interval therefore does not isolate a single direction of difference.

An absolute measure can be useful because it expresses a contrast on the outcome-probability scale. It should be interpreted in light of the outcome definition, the comparison groups, and uncertainty. The record also posts arm-specific outcome probabilities as p0 and p1, but the analysis should not be reduced to those values alone: the contrast and its uncertainty are central to the comparison.

How do relative measures differ?

A risk ratio compares outcome probabilities by division, while an odds ratio compares the odds of the outcome between groups. Both are relative measures. A ratio of one represents no relative difference. The registry reports a risk ratio of 0.735 and an odds ratio of 0.7. Their confidence intervals include one, so the estimates do not rule out no relative difference. Relative measures can tell a different numerical story from absolute measures because they use a different scale; each should be understood according to its definition rather than treated as interchangeable with the risk difference.

For an outcome that is not common, odds and probabilities may be close in some settings, but they are not the same quantity. General statistical interpretation should retain the distinction. The registry’s comparison presents both measures, which helps readers see the result on absolute and relative scales. Neither measure resolves uncertainty that is evident in the intervals.

What does the p-value add?

The posted unadjusted p-value is 0.538. In general, a p-value measures how compatible the observed data are with a specified null hypothesis under the assumptions of a statistical test. It is not the probability that the null hypothesis is true, and it does not give the probability that the result occurred by chance in an everyday sense. A p-value also does not quantify the size of the observed difference or establish whether that difference matters clinically.

The registry labels the value unadjusted. No multiplicity adjustment is reported for this comparison, and the record does not describe the test or its assumptions. A p-value is most informative when read with the estimated effect and confidence interval. Here, the interval conveys that the data leave substantial uncertainty about the direction and magnitude of the difference.

Why does randomization matter?

In general, randomized allocation is intended to make treatment groups comparable, on average, with respect to measured and unmeasured characteristics at the start of a trial. This supports a causal interpretation when allocation is implemented appropriately and outcomes are collected and analyzed as planned. Randomization cannot prevent chance imbalances, missing outcome data, deviations from assigned treatment, or measurement problems. The registry identifies the study as randomized but does not report the sequence-generation or concealment methods, adherence information, or detailed follow-up procedures.

The parallel design means the arms were compared as separate groups. The endpoint is a participant count at follow-up, not a time-to-event outcome. No time-to-event analysis is reported. The registry record also does not state the analysis population, such as whether the primary comparison followed an intention-to-treat principle or included only participants with observed follow-up data.

What is over-dispersion, and does it apply?

General statistical note: over-dispersion occurs when observed variability is greater than a model assumes. It can matter for outcomes analyzed with count models, because underestimated variability can lead to overly narrow uncertainty intervals. The primary endpoint in this record is a binary participant outcome reported as counts by arm, and no count-model analysis or over-dispersion assessment is reported. It would therefore be inappropriate to infer that such a model was used or that over-dispersion was assessed in this trial.

What can be checked from the registry record?

The record posts two outcome measures, zero analysis entries, and no posted secondary outcomes. It also provides a primary comparison estimate, confidence intervals, and an unadjusted p-value. The absence of posted analysis entries does not mean that no analysis was conducted; it means that the registry record does not provide analysis entries as part of its posted results. The registry does not report the protocol’s statistical plan, the approach to missing outcomes, subgroup analyses, or sensitivity analyses.

The planning calculations associated with the observed effect are separate from the analysis of the trial’s observed data. They illustrate participant counts under stated assumptions and are not evidence about the study’s actual statistical power. The original design assumptions and sample-size justification are not reported in the registry record. Statistical interpretation should consequently distinguish the posted study result from hypothetical planning scenarios.

10. Confidence Intervals and Precision

The confidence interval for the risk difference runs from -0.177 to 0.092 around a point estimate of -0.042. Its span indicates substantial uncertainty in the estimated absolute difference. Because the interval includes zero, the data are compatible with no difference as well as with differences in either direction. The point estimate alone would obscure this uncertainty.

The relative estimates are also imprecise. The confidence interval for the risk ratio is 0.275 to 1.967, and the interval for the odds ratio is 0.224 to 2.186. Both ranges include one, the value representing no relative difference. The intervals extend across a broad range of possible relative effects, so the registry result does not tightly determine the magnitude or direction of the comparison.

A confidence interval is a statistical procedure that, under its assumptions, yields intervals that contain the true parameter at the stated confidence level in a specified proportion of repeated studies. It is not a probability statement about the fixed parameter after observing a particular interval. The confidence level for the posted intervals is not reported in the registry record, so it should not be inferred from these results.

Precision depends on factors including the amount of information in the data, the outcome frequency, and the analysis method. The posted endpoint counts are small, and the registry’s interval widths show that the comparison is uncertain. The record does not report enough methodological detail to assess all assumptions used to construct the intervals. The appropriate reading is that the observed estimates are accompanied by considerable uncertainty, rather than that the point estimates establish a stable effect.

11. 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 estimates are participants per arm, not total events. They are illustrations based on the observed effect rather than the trial’s own design. The stated assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation; each scenario uses the stated fraction of the observed effect as the effect to detect. The original design assumptions and sample-size justification are not reported in the registry record.

The 80% power column gives the number of participants per arm used in the planning calculation to have an 80% chance of detecting the specified effect, if the assumptions and effect are correct. The 90% power column gives the corresponding per-arm planning number for a 90% chance. These are target planning probabilities for a hypothetical study, not probabilities that the observed trial result is correct. This is not observed power.

The scenarios consider planning for 100% of the observed effect, 75% of the observed effect, and 50% of the observed effect. The listed participant counts rise as the effect used for planning becomes smaller: detecting a more modest difference generally requires more information, often through a larger planned enrollment. For example, the registry lists 1,046 participants per arm for 80% power when planning for 100% of the observed effect, and 4,448 participants per arm when planning for 50% of that effect. For 90% power, the corresponding listed figures are 1,399 and 5,954 participants per arm.

These calculations are planning illustrations computed from the observed result. They should not be read as the original sample-size target, a retrospective assessment of the study’s design quality, or an estimate of the chance that the trial detected an effect. The registry does not report the trial’s original assumptions.

12. 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 = 0.160; treatment proportion p₁ = 0.118; average p̄ = 0.139.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.4890 = 0.9584.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.4881 = 0.4108.
4. n = (0.9584 + 0.4108)² / (-0.0424)² = 1,045.0, rounded up to 1,046 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 <- 0.16                          # control proportion
p1 <- 0.117647058823529                          # 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(129311)
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 = 1,046 (80% power), 1,399 (90% power)
 75% of observed effect: n per arm = 1,918 (80% power), 2,567 (90% power)
 50% of observed effect: n per arm = 4,448 (80% power), 5,954 (90% power)
power.prop.test: 1045 
Simulated power at n = 1,046 per arm: 0.799 (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.

13. Multiplicity, Missing Data and Analysis Populations

The registry posts one primary outcome and lists two outcome measures in total, but it does not post any secondary outcomes. It also lists zero analysis entries. The record does not explain whether additional outcomes were planned, collected, or analyzed but not posted, so those possibilities cannot be resolved from the registry record.

Multiplicity arises when researchers conduct multiple statistical tests or examine multiple outcomes, subgroups, or time points. In general, the more comparisons made, the greater the chance that at least one result appears unusual under a null hypothesis. Methods such as prespecifying a primary comparison or adjusting significance thresholds can address multiplicity, but the statistical plan and any multiplicity procedure are not reported in the registry record. The posted p-value is identified as unadjusted.

Missing data are also important for a follow-up outcome. If participants do not complete the abstinence assessment, the way those missing outcomes are handled can affect the estimated arm results. The registry record does not report follow-up completeness, the number of missing endpoint assessments, the missing-data assumptions, or any imputation or sensitivity analysis. It is therefore not possible to determine from the posted record how missing outcomes influenced the comparison.

The analysis population is not reported in the registry record. In general, an intention-to-treat analysis retains participants according to randomized assignment, while other analysis populations may apply eligibility, treatment receipt, or outcome-observation criteria. These choices can affect interpretation. The available summary does not identify which approach was used or how deviations from assigned treatment were handled.

The record identifies randomized allocation and triple masking, but does not provide enough detail to assess the operational conduct of allocation, masking, outcome assessment, or follow-up. These missing details do not erase the posted outcome counts, but they limit assessment of how robustly the estimates reflect the intended randomized comparison.

14. 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 0% reporting threshold.

Most frequent non-serious adverse event terms

The serious adverse event figures report zero affected participants in the selegiline arm among 51 at risk and zero in the placebo arm among 50 at risk. The registry lists no serious adverse event terms. These are counts recorded for the study; they do not establish that serious adverse events are impossible or that the interventions have equivalent safety.

For other adverse events, the registry reports 47 affected participants among 51 at risk in the selegiline arm and 45 among 50 at risk in the placebo arm. These are descriptive arm-level counts. The record does not list the specific other adverse events, their severity, timing, duration, or whether investigators judged them related to study treatment. It also does not report a comparative statistical analysis for these safety counts.

Deaths are not reported in the registry record. They should not be interpreted as zero. The safety information is limited to the categories and figures posted, and it does not support conclusions about the frequency or cause of individual event types.

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

The posted primary outcome shows 6 participants meeting the 7-day point-prevalence abstinence endpoint in the selegiline arm and 8 in the placebo arm at 6-month follow-up. The risk-difference estimate is -0.042, with an interval from -0.177 to 0.092; the unadjusted p-value is 0.538. The interval crosses zero, while the relative-effect intervals also include their respective no-difference values. Together, these results show an uncertain comparison rather than a precisely determined difference.

The result does not establish that selegiline improves smoking cessation, nor does it establish that selegiline has no effect. A non-significant p-value is not proof of equivalence, and the point estimate is not a definitive estimate of the effect that would be expected in other studies. The interval allows a range of possibilities, including a difference in either direction.

The endpoint is a 7-day point-prevalence measure at 6-month follow-up. It should not be interpreted as continuous abstinence or as evidence about abstinence at other times. The registry does not report how abstinence was verified, how missing follow-up assessments were handled, or which analysis population was used. Those details are relevant to judging the result’s validity and generalizability.

The trial was randomized and parallel-group, but the registry provides limited detail about eligibility, treatment delivery, masking procedures, and participant follow-up. The findings apply most directly to the study as recorded, not automatically to all people who smoke or to other cessation programs. The record describes the study as assessing safety and effectiveness, but the available outcome information does not support a broad conclusion about overall benefit or harm.

The safety counts are likewise limited. No serious adverse events were recorded in either arm, and other adverse events were recorded in both. The record does not provide enough detail to evaluate event types, severity, attribution, or deaths. These data should be described as observed registry counts, not as a complete safety profile.

16. Why This Trial Matters Statistically

This study illustrates why a clinical-trial result should be read through both its point estimate and its uncertainty. The observed primary-outcome counts lean in one direction, but the confidence interval for the risk difference crosses zero and spans values on either side. Looking only at which arm has more observed participants meeting the endpoint would miss the uncertainty in the comparison.

It also demonstrates the importance of distinguishing absolute and relative measures. The registry reports a risk difference, risk ratio, and odds ratio. These statistics describe the same comparison on different scales, and each has its own interpretation. None should be treated as conclusive in isolation, particularly when the confidence intervals are broad and include values corresponding to no difference.

The trial’s endpoint makes clear that outcome definitions matter. A 7-day point-prevalence measure at 6-month follow-up captures recent abstinence at a specified assessment. It does not answer whether participants remained abstinent continuously, and the registry does not report verification procedures. Readers need the endpoint’s time window and measurement details to interpret what the counts represent.

The registry record also highlights the value of complete reporting. Details about missing outcomes, analysis populations, statistical methods, and the protocol’s sample-size assumptions are not reported. Their absence limits assessment of how the comparison was conducted and whether the posted estimate reflects the planned analysis. Such limitations are distinct from the numerical uncertainty captured by the confidence interval.

Finally, the planning estimates are useful as illustrations of how the required enrollment changes with the effect a future study aims to detect, but they are not observed power or the original design target. Keeping observed results, uncertainty, and hypothetical planning scenarios separate is essential to a sound statistical reading.

17. 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 gives limited information about participant eligibility and characteristics. It does not report the specific inclusion or exclusion criteria, baseline smoking patterns, or other participant features needed to judge how representative the enrolled group may be. The study’s size and setting should not be generalized to populations that are not characterized in the record.

Details of the intervention are also limited. The record names selegiline and placebo but does not report dose, route, schedule, treatment duration, adherence, or concurrent cessation support. The precise implementation of the triple masking is not described, and the people masked are not identified. These omissions constrain assessment of treatment delivery and potential sources of bias.

The outcome is a 7-day point-prevalence measure at 6-month follow-up. The registry does not describe the assessment method or report whether abstinence was biochemically verified. It also does not state how missing follow-up assessments were handled, the analysis population, or whether sensitivity analyses were conducted. The effect estimates therefore cannot be evaluated against those methodological details.

The original sample-size rationale, statistical plan, and design assumptions are not reported in the registry record. The listed planning calculations are based on the observed effect and are not a substitute for the original plan. The record also does not post secondary outcomes or detailed analysis entries, and it does not explain whether other analyses were performed.

Safety reporting is limited to serious-event and other-event counts. Specific event terms for other adverse events, event severity, relatedness, timing, and deaths are not reported. Consequently, the posted safety figures do not provide a complete account of the study’s safety experience.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Usefulness of Selegiline as an Aid to Quit Smoking - 1 — NCT00129311.
• Registry record last updated 2015-03-31; status last verified 2014-07.
• Results first posted 2014-07-16.
• 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.

21. Record Summary

The registry record describes a completed, randomized, triple-masked, parallel-group phase 2 study sponsored by Yale University. It began in 2004-07, reached primary completion in 2012-12, and recorded enrollment of 101. The two study arms were selegiline and placebo.

The primary endpoint was 7-day point-prevalence cigarette abstinence at 6-month follow-up. The posted outcome counts were 6 participants for selegiline and 8 for placebo. The reported risk difference was -0.042, with a confidence interval from -0.177 to 0.092, and the unadjusted p-value was 0.538. The comparison is imprecise and does not establish a clear difference.

The registry lists two outcome measures, no posted secondary outcomes, and zero analysis entries. It reports zero participants with serious adverse events in each arm and other adverse events in both arms. Further details about methods, missing outcomes, participant characteristics, and event types are not reported in the registry record.