Melatonin in Smoke-induced Vascular Injury: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Melatonin in Smoke-induced Vascular Injury (NCT02591238), a randomized phase 4 study. The primary outcome is Smoke-induced Vascular Injury and Melatonin's Effect in This Process Assessed by the Concentration of Fbg..

2. Clinical Question

3. Trial Design

This was a randomized, double-blind, parallel-group Phase 4 study. The registry record identifies Peking Union Medical College Hospital as the sponsor and gives enrollment as 68. The listed study status is completed. The study start is recorded as 2015-06, with primary completion in 2015-07.

The four outcome groups combine smoking status with assigned treatment: non-smoker plus placebo, non-smoker plus melatonin, smoker plus placebo, and smoker plus melatonin. The group sizes posted for the primary outcome are 15, 16, 17, and 15, respectively. These outcome-group counts are not identical in total to the registry's enrollment figure; the reason is not reported in the registry record.

Randomization and double masking are recorded, but the method used to generate or conceal the allocation, who was masked, and whether masking was maintained are not reported in the registry record. The record describes the study as population-based, randomized, double-blind, and placebo-controlled. It does not explain how smoking status was determined or whether smoking status was itself assigned; the listed groups distinguish smokers from non-smokers while randomization concerns treatment assignment.

The registry lists six outcome measures and no posted secondary outcome results. It lists zero analyses. The detailed statistical analysis plan, original sample-size justification, analysis populations, handling of missing observations, and any interim analyses are not reported in the registry record. The primary outcome timeframe is listed as Three months, while its description says the concentration data followed 2 weeks of oral melatonin. The record does not reconcile those descriptions.

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: First in included 96 participants of both genders, 28 participants excluded (refused to randomly assigned n=9, not meeting inclusion criteria n=9, time improper n=8, unknown n=2). 5 participants withdraw during the trial (pregnant n=1, influenza n=1, personal circumstance n=2, unknown reason n=1).

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Bellow concentration data of Fbg was after treatment with oral melatonin 2 weeks.

5. Baseline Characteristics

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

Population: According to the requirements of the randomized controlled trials and statistical methods to determine a total of 63 participants, and signed the informed consent.

6. Primary Endpoint

The registry lists the primary outcome as Smoke-induced Vascular Injury and Melatonin's Effect in This Process Assessed by the Concentration of Fbg., measured over Three months.

The primary outcome is fibrinogen concentration, identified as Fbg in the registry title and reported in g/L. The record describes the measure as a continuous outcome summarized by its mean and standard deviation. Fibrinogen is used in this study as a laboratory measure relevant to the stated question of smoke-induced vascular injury and the possible role of melatonin in that process. The registry does not provide a further explanation of the biological rationale, measurement procedure, or threshold for interpreting the result.

The posted group means are 2.76 g/L for non-smokers receiving placebo, 2.38 g/L for non-smokers receiving melatonin, 2.89 g/L for smokers receiving placebo, and 2.49 g/L for smokers receiving melatonin. The corresponding standard deviations are 0.34, 0.43, 0.58, and 0.54. These summaries describe the observed concentrations and their variability within each group; they do not, on their own, establish why groups may differ.

The outcome timeframe is listed as Three months, but the population description says the Fbg concentration data were after treatment with oral melatonin 2 weeks. The exact measurement schedule and which statement defines the primary measurement time are not reported in the registry record. That ambiguity matters when interpreting the endpoint because the timing of a measurement determines what period of treatment or follow-up it represents.

7. Results Posted in the Registry

• III Non-smoker + Melatonin: mean 2.38 versus 2.76 on control; difference -0.38 (95% CI -0.65 to -0.11), standardized effect d = -0.98; z-test p = 0.006.
• II Smoker + Placebo: mean 2.89 versus 2.76 on control; difference 0.13 (95% CI -0.19 to 0.45), standardized effect d = 0.27; z-test p = 0.433.
• I Smoker + Melatonin: mean 2.49 versus 2.76 on control; difference -0.27 (95% CI -0.59 to 0.05), standardized effect d = -0.60; z-test p = 0.101.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is continuous: each participant’s fibrinogen concentration is measured on a numerical scale in g/L. For a continuous outcome, a group mean describes the central value, while a standard deviation describes how spread out individual measurements are around that mean. General: a mean difference compares the average outcome in one group with the average in another. A negative difference means the first group’s mean is lower when the comparison is calculated as the listed group minus the non-smoker placebo mean.

What does a mean difference describe?

The registry posts group means and standard deviations, but it does not post sponsor-reported analysis results. The comparisons are recalculated from the posted summary statistics and are unadjusted. For example, the recalculated mean difference for non-smokers assigned melatonin versus non-smokers assigned placebo is -0.38 g/L. The comparison for smokers assigned melatonin is against non-smokers receiving placebo, not against smokers receiving placebo. That distinction is important: a contrast against a shared reference does not directly answer whether melatonin and placebo differ among smokers.

General: an unadjusted comparison uses the group summaries without accounting for other variables or a prespecified model. A randomized design can support comparisons between assigned treatments, but interpretation still depends on the analysis population, missing data, adherence, measurement timing, and statistical plan. Those details are not reported in the registry record. No adjustment for multiple comparisons is reported either. The recalculated p-values should therefore be understood as unadjusted results for the listed contrasts, not as a complete analysis of all treatment and smoking-status questions.

How should the confidence intervals be read?

A confidence interval gives a range of values reasonably compatible with the observed data under the method used to calculate it. General: narrower intervals indicate greater precision, while wider intervals indicate greater uncertainty. If an interval for a mean difference includes zero, the data are compatible with no difference as well as with values on either side of zero. If the interval does not include zero, it indicates a difference under that calculation, but it does not establish clinical importance or remove concerns about bias, multiplicity, or the choice of comparison.

The non-smoker melatonin comparison has a 95% confidence interval from -0.652 to -0.108 g/L. The intervals for the smoker placebo and smoker melatonin comparisons against non-smoker placebo include zero. These intervals are based on the recalculated comparisons and should not be mistaken for intervals from a sponsor’s adjusted primary analysis. The registry lists no analysis results, and the record does not specify the inferential method used in an original analysis.

Why do sample-size illustrations depend on the assumed effect?

General: power is the probability that a planned analysis will detect a specified effect under stated assumptions, if that effect is present. Planning requires choosing a target effect, a significance level, an allocation pattern, and a statistical method. The registry’s planning illustrations use the observed effect or a stated proportion of it, with a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. These are planning assumptions for the illustrations, not reported assumptions from the trial’s original design.

For example, the illustration for the non-smoker melatonin comparison lists 17 participants per arm for 80% power when planning for 100% of the observed effect, and 30 participants per arm when planning for 75% of the observed effect. Planning for 50% of that observed effect lists 66 participants per arm for 80% power. For 90% power, the corresponding listed values are 23, 40, and 89 participants per arm. The other comparison-specific illustrations differ because the observed differences and variability differ. These values are not estimates of how many participants this completed study needed under its own design; the original justification is not reported in the registry record.

General: smaller effects are harder to distinguish from random variation and therefore require more information, often through a larger sample. Greater variability also makes estimates less precise. These principles explain why the planning values change across comparisons and target effects. The group summaries are small, so the observed effects used for planning may themselves be uncertain. A planning illustration based on an observed effect should not be treated as a reliable guarantee of the effect a future study will observe.

What limits the analysis that can be reconstructed?

The registry posts means, standard deviations, sample sizes, recalculated mean differences, confidence intervals, and unadjusted p-values. It does not post participant-level data, so the calculations cannot account for individual covariates, unusual observations, outcome distributions, or departures from the assumptions underlying the summary-statistic method. The record also does not report the original statistical plan, model specification, missing-data approach, analysis population, or whether the posted group counts reflect all randomized participants.

General: a randomized trial’s principal treatment comparison is normally defined by the assigned groups, but an interpretation depends on knowing which participants were included and how post-randomization events were handled. Without those details, the posted summaries support a descriptive reading and straightforward unadjusted comparisons, but not a full reconstruction of the planned analysis. No discrepancy is flagged in the record, but that does not supply missing protocol details or establish that every analytical decision can be evaluated.

10. Confidence Intervals and Precision

The recalculated confidence intervals show how uncertain the mean differences are, beyond the point estimates alone. For non-smokers receiving melatonin compared with non-smokers receiving placebo, the estimated mean difference is -0.38 g/L, with a 95% confidence interval from -0.652 to -0.108. That interval lies below zero, but it still spans a range of possible differences rather than identifying a single exact value.

For smokers receiving placebo compared with non-smokers receiving placebo, the estimated difference is 0.13 g/L, and the 95% confidence interval ranges from -0.195 to 0.455. For smokers receiving melatonin compared with non-smokers receiving placebo, the estimate is -0.27 g/L, with a 95% confidence interval from -0.593 to 0.053. Both intervals cross zero, indicating that the estimates are not precise enough to rule out no difference under these calculations. They also include values on either side of zero, so the direction is uncertain.

General: a confidence interval describes uncertainty under a specified statistical method; it is not a range containing a fixed proportion of individual participant values, nor does it account for unmeasured bias or all possible analytic choices. Here, the intervals come from recalculated unadjusted comparisons based on group means, standard deviations, and sample sizes. The registry does not report an adjusted sponsor analysis or its confidence intervals. Small outcome-group sizes and observed variability contribute to uncertainty, but the registry does not report a precision target or original sample-size rationale.

Interpretation also depends on defining the outcome time point. The primary timeframe is listed as Three months, while the outcome description refers to Fbg concentrations after 2 weeks of oral melatonin. Until that timing is clarified, the precision of the numerical estimates does not resolve what period of treatment or follow-up they represent.

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 figures are participants per arm, not total events. The registry’s 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 or a stated proportion of that observed effect as the effect to detect. The trial’s own original design assumptions and sample-size justification are not reported in the registry record.

The 80% power column gives the planned number of participants per arm under those assumptions for an 80% chance of detecting the chosen effect if that effect is present. The 90% power column gives the per-arm number for a higher planned chance of detection under the same stated framework. For the non-smoker melatonin comparison, planning for 100% of the observed effect lists 17 participants per arm for 80% power and 23 for 90% power. Planning for 75% of the observed effect lists 30 and 40 participants per arm; planning for 50% lists 66 and 89.

The values depend on the comparison because each has a different estimated effect and variability. Smaller effects need more participants to distinguish them from random variation, which is why planning for 75% or 50% of the observed effect calls for larger group sizes than planning for the full observed effect. The curve expresses that same relationship visually: required group size rises as the target effect becomes smaller. These are planning illustrations based on observed summaries, not observed power for the completed trial.

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, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = -0.380; pooled standard deviation σ = 0.389.
2. n = 2 × (1.96 + 0.842)² × 0.389² / -0.380² = 16.5, rounded up to 17 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
delta <- -0.38                      # observed difference in means
sigma <- 0.389159128874304                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^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 (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(2591238)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 17 (80% power), 23 (90% power)
 75% of observed effect: n per arm = 30 (80% power), 40 (90% power)
 50% of observed effect: n per arm = 66 (80% power), 89 (90% power)
power.t.test: 17.5 
Simulated power at n = 17 per arm: 0.791 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-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 lists six outcome measures, but no secondary outcome results are posted. It lists zero analyses. For the primary measure, the posted comparisons include three contrasts against the non-smoker placebo group, with unadjusted p-values. The record does not report whether these contrasts were prespecified, whether one was designated as the main statistical comparison, or whether any correction for multiple testing was planned or applied.

General: when several outcomes or comparisons are examined, the chance of at least one apparently unusual result can increase if each is assessed separately without accounting for the number of tests. A multiplicity procedure can control that risk, but the choice depends on the trial’s planned questions and analysis strategy. Since that strategy is not reported in the registry record, the unadjusted p-values should be read as nominal results for their individual comparisons rather than as a family-wise conclusion.

The registry does not report participant-level missingness, reasons for missing outcome data, imputation, exclusions, or the analysis population. The posted primary outcome group sizes are 15, 16, 17, and 15, while total enrollment is 68; the record does not explain the difference between those figures. It is therefore not possible to determine from the record whether all randomized participants contributed to the outcome summaries, or how any missing measurements were handled.

The record also does not report a treatment-by-smoking interaction analysis. Separate comparisons against non-smoker placebo cannot establish whether the melatonin-placebo contrast differs between smokers and non-smokers. A formal interaction question requires a suitable direct model or contrast, and no such result is posted. The absence of those details limits the strength and scope of statistical conclusions that can be drawn from the available summaries.

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.

The safety record lists zero serious adverse events among 15 non-smokers receiving placebo, 16 non-smokers receiving melatonin, 17 smokers receiving placebo, and 15 smokers receiving melatonin. It also lists zero other adverse events among participants at risk in each group. Death counts are not reported in the registry record.

These figures describe events recorded during the safety reporting period, but the period and the definitions used for serious and other adverse events are not reported in the registry record. Zero recorded events does not establish that an event is impossible or that a treatment has no safety risks. Small group sizes also mean that the absence of a recorded event cannot precisely characterize uncommon harms. The record does not provide event narratives, severity details beyond the serious-versus-other classification, or information about withdrawals related to adverse events.

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

The posted summaries describe fibrinogen concentrations across four groups defined by smoking status and treatment assignment. The recalculated comparison among non-smokers shows a mean difference of -0.38 g/L for melatonin versus placebo, with a 95% confidence interval that lies below zero and an unadjusted p-value of 0.006. The two listed comparisons against non-smoker placebo involving smokers have confidence intervals that include zero. Together, these figures describe the specific group contrasts available in the registry record; they do not answer every possible treatment question.

The results do not establish that melatonin caused a change in fibrinogen, nor do they establish a treatment effect specifically among smokers. The smoker melatonin group is compared in the posted calculations with non-smoker placebo, rather than with smoker placebo. No direct treatment contrast within smokers or interaction analysis is posted. The record also does not provide an adjusted analysis, participant-level data, or a complete statistical plan.

The endpoint is a laboratory concentration used in the study’s question about vascular injury. The registry does not report a clinical endpoint or establish how a difference in this measure relates to a clinical outcome. The numerical results should therefore be understood as laboratory outcome summaries, not as evidence of a clinical benefit or harm.

The confidence intervals show substantial uncertainty for two of the three listed comparisons, while the non-smoker contrast is more clearly separated from zero under its unadjusted calculation. The number and type of comparisons, missing analysis details, and unclear measurement timing all matter to interpretation. In particular, the record lists a Three months timeframe but also describes the measurement as occurring after 2 weeks of oral melatonin. That mismatch is unresolved.

The safety entries report no serious or other adverse events in the listed groups, with deaths not reported in the registry record. Given the group sizes and absent reporting details, these counts describe the registry entry rather than providing a complete assessment of safety.

16. Why This Trial Matters Statistically

This study raises several statistical questions that commonly arise when a trial includes both treatment assignment and a participant characteristic such as smoking status. The group structure can describe outcome patterns across smokers and non-smokers, but the comparisons must be chosen carefully. Comparing each group with non-smoker placebo does not directly estimate the melatonin-versus-placebo difference among smokers. Nor does it test whether a treatment contrast changes according to smoking status.

The distinction is useful because a set of group means may appear to suggest patterns that are not formally tested by the posted comparisons. A treatment-by-smoking interaction requires a direct statistical assessment, and no such analysis is reported. This is a reminder that descriptive group differences and treatment effects answer different questions.

The record also illustrates why estimates should be read with their uncertainty. The recalculated mean differences are accompanied by confidence intervals, which show that the smoker comparisons allow a range of values including zero. Reporting a point estimate alone could obscure that uncertainty. The planning illustrations likewise make clear that sample-size needs depend on the target effect: planning for a smaller effect requires more participants per arm under the stated assumptions.

Finally, registry descriptions need internal clarity for reproducible interpretation. Here, the primary outcome timeframe is listed as Three months while the outcome description refers to data after 2 weeks of oral melatonin. The original analysis plan, missing-data methods, multiplicity strategy, and sample-size rationale are not reported. Those omissions do not negate the posted descriptive results, but they limit the degree to which the trial’s intended statistical analysis can be reconstructed.

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 outcome-group counts are 15, 16, 17, and 15, while enrollment is listed as 68. The registry does not explain why the primary outcome summaries include those group counts or whether any randomized participants lacked measurements. There is no reported accounting of missing data, exclusions, adherence, withdrawals, or analysis populations.

Timing is ambiguous. The outcome timeframe is listed as Three months, but its description refers to Fbg concentration after oral melatonin for 2 weeks. The record does not clarify whether these statements refer to different stages of the study, nor does it provide a precise measurement schedule. The treatment dose, route details beyond oral administration, and other protocol procedures are not reported in the registry record.

The posted comparisons are recalculated and unadjusted. The registry does not post sponsor-reported analysis results, a model specification, a direct melatonin-versus-placebo contrast among smokers, or an interaction test by smoking status. It also does not state whether the multiple comparisons were planned or whether any multiplicity adjustment was used. Participant-level data are not available in the registry entry, limiting assessment of distributions, influential observations, covariate balance, or alternative analyses.

The original sample-size calculation, target effect, and planned statistical assumptions are not reported. The illustrations based on observed effects are not a substitute for the trial’s original design rationale. Safety reporting gives zero serious and other adverse events in each group, but event definitions, reporting duration, and death counts are not described. These limitations make the registry most informative as a record of group summaries and basic recalculated contrasts, rather than a complete report of the trial’s methods and analysis.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Melatonin in Smoke-induced Vascular Injury — NCT02591238.
• Registry record last updated 2016-03-03; status last verified 2015-06.
• Results first posted 2016-03-03.
• 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.

21. Record Summary

NCT02591238 is a completed, randomized, double-blind, parallel-group Phase 4 study titled “Melatonin in Smoke-induced Vascular Injury.” The sponsor is Peking Union Medical College Hospital. Enrollment is listed as 68, with study start in 2015-06 and primary completion in 2015-07.

The study included smokers and non-smokers assigned to oral melatonin or oral placebo. Its primary outcome was mean fibrinogen concentration in g/L, with standard deviation reported for each group. The registry lists a Three months timeframe, while the outcome description refers to measurement after 2 weeks of oral melatonin.

The posted summaries support unadjusted recalculated comparisons, but no sponsor analysis results or secondary outcome results are posted. The record does not report the original sample-size justification, statistical plan, missing-data handling, or analysis populations. Safety entries list zero serious and other adverse events in each group; deaths are not reported in the registry record.