Cigarette Pack Inserts for Smoking Cessation: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Cigarette Pack Inserts for Smoking Cessation (NCT04075682), a randomized na study. The primary outcome is Self-efficacy to Quit Smoking.

2. Clinical Question

The trial was designed to examine psychological responses and behaviors associated with smoking cessation during exposure to different cigarette-pack messages. The primary endpoint is a measure of confidence in quitting, not a record of whether participants actually quit smoking. The registry describes repeated daily assessments and reports group means and standard deviations, together with mixed-model analyses. The outcome summaries therefore speak to average self-efficacy scores during the study, rather than establishing longer-term cessation outcomes.

3. Trial Design

The registry describes a randomized, single-masked, factorial study of cigarette-pack inserts and pictorial warnings. Its brief description characterizes the design as a 2-by-2 between-subject experiment, with four groups: standard packs, inserts only, pictorial warnings only, and both inserts and pictorial warnings. Enrollment was 387. The study status is completed.

Participants received a 14-day supply of their preferred cigarette brand. In the inserts-only group, the packs included rotating messages about cessation benefits and quitting tips. In the pictorial-warning group, packs carried rotating pictorial warnings. The combined group received both types of messages. Standard packs retained a small textual health warning on the side. The registry describes the pictorial warnings as covering 50% of the front and back of the pack.

The primary measure was assessed approximately 4-5 times a day over the 14 day study period. The registry states that analyses were limited to days 2-14 because day 1 coverage varied with study orientation time. The primary analysis population included repeated observations from daily cigarette surveys. Sample sizes for the reported arm summaries were 101 for standard packs, 86 for inserts only, 90 for pictorial warnings only, and 89 for the combined group.

The study record lists 18 outcome measures and 166 analyses. It names mixed models as the method for the sponsor-reported analyses. The trial’s original statistical plan, including its planned analysis details and sample-size justification, is not reported in the registry record. The registry lists study dates, but those dates do not establish the timing or conduct of any interim analyses; interim analyses 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:

Analysis population for the primary outcome, as defined in the registry: Analyses include repeated observations from daily cigarette surveys. Analytic samples for H6 differ due to some missing data for each moderating variable: education (control=101; insert only=84; pictorial warnings only=87; insert + pictorial warnings=88); health literacy (control=101; insert only=86; pictorial warnings only=90; insert + pictorial warnings=88); and delayed discounting (control=101; insert only=84; pictorial warnings only=88; insert + pictorial warnings=88).

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: The analysis population is those participants who were included in the analytic sample.

6. Primary Endpoint

The registry lists the primary outcome as Self-efficacy to Quit Smoking, measured over Evaluated approximately 4-5 times a day over the 14 day study period, with analyses limited to days 2-14. Day 1 was dropped due to varying coverage of the day depending on study orientation time..

The primary endpoint is self-efficacy to quit smoking: a single continuous-scale question about how confident a participant feels about quitting. The scale runs from 1, “Not at all,” described as the worse outcome, to 7, “Extremely,” described as the better outcome. The registry reports average scores and standard deviations for each group.

Self-efficacy is used as a measure of confidence or perceived ability to carry out a behavior. In this trial, it captures a psychological response to cigarette-pack messages. It is not itself a measure of cessation, abstinence, or smoking frequency. A higher score represents stronger confidence on the stated scale, but a difference in this score alone does not establish a change in quitting behavior.

Participants were asked the question repeatedly, approximately 4-5 times a day, across the 14 day study period. The primary analyses excluded day 1 and covered days 2-14. Repeated assessments can describe how reported confidence varies during the study and can contribute multiple observations per participant to an analysis. Such observations are not independent in the same way as observations from different people, which is why the registry’s use of mixed models is relevant. The registry record does not report all details of the model specification or how missing repeated assessments were handled.

For the posted arm summaries, average scores were 2.61 in the standard-pack group, 2.76 with inserts only, 2.62 with pictorial warnings only, and 2.56 with both inserts and pictorial warnings. These descriptive averages give the scale context for the reported comparisons; they do not, by themselves, indicate whether a group difference is statistically distinguishable from uncertainty.

7. Results Posted in the Registry

• Cigarette Packs With Inserts Only: mean 2.76 versus 2.61 on control; difference 0.15 (95% CI -0.28 to 0.58), standardized effect d = 0.10; z-test p = 0.496.
• Cigarette Packs With Pictorial Warnings Only: mean 2.62 versus 2.61 on control; difference 0.01 (95% CI -0.41 to 0.43), standardized effect d = 0.01; z-test p = 0.963.
• Cigarette Packs With Inserts and Pictorial Warnings: mean 2.56 versus 2.61 on control; difference -0.05 (95% CI -0.45 to 0.35), standardized effect d = -0.04; z-test p = 0.808.

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.

Strength of Hopefulness About Quitting

• 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:
  • Cigarette Packs With Inserts Only: mean 4.25 versus 3.80 on control; difference 0.45 (95% CI -0.08 to 0.98), standardized effect d = 0.24; z-test p = 0.099.
  • Cigarette Packs With Pictorial Warnings Only: mean 3.96 versus 3.80 on control; difference 0.16 (95% CI -0.36 to 0.68), standardized effect d = 0.09; z-test p = 0.548.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Cigarette Packs With Inserts Only and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Cigarette Packs With Pictorial Warnings Only and the control. The interval gives the range of effects that remain compatible with the data.

Extent of Cognitive Elaboration of Smoking Benefits

• 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:
  • Cigarette Packs With Inserts Only: mean 3.20 versus 3.19 on control; difference 0.01 (95% CI -0.47 to 0.49), standardized effect d = 0.01; z-test p = 0.968.
  • Cigarette Packs With Pictorial Warnings Only: mean 3.16 versus 3.19 on control; difference -0.03 (95% CI -0.52 to 0.46), standardized effect d = -0.02; z-test p = 0.904.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Cigarette Packs With Inserts Only and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Cigarette Packs With Pictorial Warnings Only and the control. The interval gives the range of effects that remain compatible with the data.

Extent of Cognitive Elaboration of Smoking Harms

• 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:
  • Cigarette Packs With Inserts Only: mean 4.40 versus 3.91 on control; difference 0.49 (95% CI -0.03 to 1.01), standardized effect d = 0.27; z-test p = 0.062.
  • Cigarette Packs With Pictorial Warnings Only: mean 3.95 versus 3.91 on control; difference 0.04 (95% CI -0.49 to 0.57), standardized effect d = 0.02; z-test p = 0.883.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Cigarette Packs With Inserts Only and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Cigarette Packs With Pictorial Warnings Only and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcome summaries cover several psychological responses, including hopefulness about quitting and thoughts about smoking benefits and harms. The overall descriptive pattern is mixed rather than a consistent shift across outcomes and groups. Some insert-only averages are higher than standard-pack averages, while other outcome summaries show little separation among groups. This broad pattern does not alter the primary result: the self-efficacy comparisons are close to zero in the simple recalculations, and the registry’s all-group mixed-model result is imprecise and compatible with no difference.

Secondary measures may help describe how participants responded to the messages, but they are not interchangeable with the primary self-efficacy measure, and they do not establish quitting behavior. Their relevance depends on their measurement properties, the specific contrast tested, and whether the analysis was planned in advance. The registry reports multiple outcomes and a large number of analyses. When many tests are considered, chance alone can produce some apparently notable results even if there is no underlying difference. Unless multiplicity is controlled or a clear hierarchy is prespecified, individual secondary p-values carry less weight than a primary analysis specified for the main question.

The registry record does not report a multiplicity adjustment or a complete prespecified hierarchy for interpreting secondary outcomes. The reported secondary findings should therefore be treated as exploratory context, not as independent confirmation of the primary endpoint. The record also does not provide all model details for these analyses. A coherent secondary pattern would require consistent estimates, clear contrasts, and appropriate consideration of the number of tests; those conditions cannot be fully assessed from the registry entries alone.

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

• Strength of Hopefulness About Quitting
• Extent of Cognitive Elaboration of Smoking Benefits
• Extent of Cognitive Elaboration of Smoking Harms
• Extent of Cognitive Elaboration of Cessation Benefits
• Strength of Beliefs About Benefits of Smoking Cessation (i.e., Response Efficacy)
• Strength of Perceived Susceptibility to Smoking Harms
• Expressed Reactance Against Health Messages
• Satisfaction From Smoking

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is continuous: each participant’s answer is represented by a score on a scale, and the posted arm summaries give means and standard deviations. A mean describes the average score in a group. A standard deviation describes how spread out individual scores are around that group mean. Neither statistic alone captures uncertainty in a comparison between groups.

The simple recalculations compare each intervention arm’s mean with the standard-pack mean. A mean difference is expressed in the original scale’s units. Positive values indicate a higher average in the intervention arm than in the standard-pack arm; negative values indicate a lower average. These comparisons are unadjusted: they are recalculated from the posted group summaries rather than from the sponsor’s mixed model. They do not account for repeated observations or any covariates in the way a fitted model might.

A confidence interval gives a range of estimates compatible with the data and the method used to calculate it. An interval that includes zero does not establish that the true difference is zero. It indicates that the data do not clearly distinguish a zero difference from values on either side of zero at the stated confidence level. A wide interval indicates less precision than a narrow interval. Confidence intervals should be read alongside the estimate and the analysis method, not as a guarantee that the true value lies within the interval.

A p-value describes how unusual data at least as inconsistent with a specified null hypothesis would be if that hypothesis and the analysis assumptions held. A p-value is not the probability that the null hypothesis is true, nor is it a measure of clinical importance. The recalculated p-values are unadjusted and arise from simple comparisons. The registry’s mixed-model p-values refer to sponsor-reported analyses and should not be assumed to test exactly the same contrasts as the simple recalculations unless the record identifies them that way.

The outcome was assessed repeatedly within participants. Repeated measurements from one person tend to be related because they come from the same person. A mixed model can represent that within-person dependence while estimating differences over repeated observations. The registry names “Mixed Models Analysis,” but the model structure, covariance assumptions, fixed effects, handling of time, and other specification details are not reported in the registry record. It is therefore not possible to reproduce or fully evaluate the sponsor’s model from the registry summary alone.

The study is described as factorial because it crossed two labeling factors: inserts and pictorial warnings. Such a design can be used to examine the effect associated with each factor and whether the factors interact. An interaction would mean that the association of one labeling factor with the outcome depends on whether the other factor is present. The registry record does not report enough clearly labeled model detail to determine which reported mixed-model estimates represent main effects, interactions, or other contrasts. Interpretation should not assign a factorial meaning to an estimate unless its contrast is specified.

The posted recalculations also include standardized effect summaries and planning illustrations. A standardized effect expresses a mean difference relative to the pooled spread of the scores, allowing scale-free description of magnitude. Such a value is derived from the posted summaries and is not a separate trial outcome. The planning illustrations use observed differences as hypothetical effects for future-study planning; they do not show the probability that this completed study detected an effect.

Over-dispersion is a general issue when observed variability exceeds what a model’s assumed distribution or variance structure allows. If present and not appropriately addressed, standard errors may be too small and confidence intervals too narrow. The registry record does not report an over-dispersion assessment for this continuous endpoint, so no claim can be made about whether it occurred or how it was handled. More generally, model diagnostics and sensitivity analyses are not reported in the registry record.

Missingness matters because participants may not complete every repeated survey, and some moderator analyses had different analytic sample sizes due to missing moderator data. The registry’s primary population description identifies repeated survey observations and notes that moderator analytic samples differ, but it does not fully describe the amount or pattern of missing primary outcome data, the assumptions used to handle missingness, or whether imputation was used. These omissions limit assessment of how missing data could affect the estimates.

Multiplicity is also relevant. The record lists 18 outcome measures and 166 analyses, while the posted results include several secondary outcomes. When many outcomes or contrasts are tested, the chance of at least one apparently notable result can increase. The registry record does not report a multiplicity adjustment or a prespecified hierarchy for interpreting the many analyses. Consequently, isolated secondary p-values should be treated cautiously and should not be given the same weight as a clearly prespecified primary analysis.

11. Confidence Intervals and Precision

The recalculated primary comparisons have confidence intervals that extend to both sides of zero. This means the posted summary data do not pin down a precise direction or magnitude for the difference between any intervention arm and the standard-pack group. For inserts only, the estimated difference is 0.15; for pictorial warnings only, it is 0.01; and for the combined group, it is -0.05. Their unadjusted p-values are 0.496, 0.963, and 0.808, respectively. These estimates are close to zero, but their uncertainty means that a small estimate should not be interpreted as proof of equivalence.

The sponsor’s reported all-group mixed-model estimate is 0.26 with a 95% confidence interval from -0.73 to 1.25 and p = 0.60. This interval also spans zero and includes a range of possible differences in both directions. Its endpoints are materially farther from zero than the point estimate, illustrating why the point estimate alone is an incomplete description of precision. The record reports other mixed-model results for different group sets, which should be interpreted according to their stated contrasts.

Confidence intervals describe uncertainty under a particular analysis and its assumptions. The simple recalculated intervals and sponsor-reported model interval are based on different approaches, so they should not be treated as interchangeable. The sponsor’s analysis takes precedence where the recalculated results differ. The record does not report enough model detail to assess every assumption behind the sponsor’s intervals, and it does not report an equivalence margin. Therefore, neither a null-compatible interval nor a small observed difference establishes that the groups are equivalent.

Precision is also affected by the repeated-measures structure and the available observations. The registry reports arm-level participant counts, but full information about the number and completeness of repeated outcome observations is not reported in the registry record. This limits a fuller assessment of the information contributing to the mixed-model estimates.

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 illustrations translate an observed effect into hypothetical planning requirements. For these continuous-outcome comparisons, the figures are participants per arm, not total participants or total events. The planning assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The assumed effect to detect is the stated fraction of the observed effect. The trial’s own 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 estimated to provide an 80% chance of detecting the specified effect under those planning assumptions, if that effect and the assumptions are correct. The 90% power column uses a higher target chance of detection and therefore generally requires more participants per arm. These are planning illustrations, not observed power.

Rows for 100% of the observed effect use the observed difference as the hypothetical target. Rows for 75% and 50% of the observed effect plan around progressively smaller differences. Smaller effects are harder to distinguish from random variation, so they require more participants per arm to reach the same target power. This is why the sample-size figures increase when the assumed fraction of the observed effect decreases.

The illustrations should not be read as evidence that the observed effects are reliable or reproducible. They use effect estimates from the posted results, which may be imprecise, and the resulting requirements can change substantially if the true effect differs. They also do not reproduce the trial’s repeated-measures model or necessarily match the study’s original design. Their role is to show how planning requirements depend on assumptions about effect size and analysis, not to establish whether the completed trial was adequately powered.

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

Worked calculation for this trial

1. Observed difference Δ = 0.150; pooled standard deviation σ = 1.492.
2. n = 2 × (1.96 + 0.842)² × 1.492² / 0.150² = 1,552.1, rounded up to 1,553 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 <- 0.15                      # observed difference in means
sigma <- 1.49154826144386                     # 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(4075682)
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 = 1,553 (80% power), 2,078 (90% power)
 75% of observed effect: n per arm = 2,760 (80% power), 3,694 (90% power)
 50% of observed effect: n per arm = 6,209 (80% power), 8,312 (90% power)
power.t.test: 1553.1 
Simulated power at n = 1,553 per arm: 0.800 (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.

14. Analyses Reported by the Sponsor

The registry reports sponsor analyses using mixed models, while the recalculated comparisons are simple, unadjusted contrasts based on the posted arm means and standard deviations. The approaches need not produce the same estimates or uncertainty: the mixed models may use the repeated observations and model structure, whereas the recalculations do not reproduce those features.

For the primary outcome, one mixed-model entry listing all four groups reports a mean difference (net) of 0.26, a 95% confidence interval from -0.73 to 1.25, and p = 0.60. Other entries list different combinations of groups and report different estimates. The record does not clearly identify every entry’s contrast in a way that supports treating all of them as direct comparisons of an intervention arm with the standard-pack group.

The recalculated intervention-versus-standard mean differences are 0.15 for inserts only, 0.01 for pictorial warnings only, and -0.05 for the combined group. Their unadjusted p-values are 0.496, 0.963, and 0.808. Because the registry-reported analyses and the simple recalculations differ, the sponsor’s own analysis takes precedence. The recalculations are useful as transparent summaries of the posted group statistics, but they do not override the sponsor’s mixed-model results or establish the exact reason for any discrepancy. Model specifications and further details needed to reconcile the analyses are not reported in the registry record.

15. Multiplicity, Missing Data and Analysis Populations

The registry lists 18 outcome measures and 166 analyses. That breadth creates a multiplicity concern: as the number of statistical tests increases, the possibility of at least one low p-value arising by chance also increases. The record does not report a multiplicity adjustment, such as a procedure to control the overall false-positive rate, nor does it describe a prespecified hierarchy that distinguishes confirmatory analyses from exploratory ones. The statistical analysis plan is not reported in the registry record.

Missing data are relevant because the primary endpoint was assessed repeatedly and the registry notes that some moderator analyses had different analytic samples due to missing moderator data. For the education moderator, the listed samples were control=101, insert only=84, pictorial warnings only=87, and insert plus pictorial warnings=88. For the health literacy moderator, the record begins to list arm counts but the description is incomplete. These moderator-specific numbers should not be confused with the primary arm-summary sample sizes.

The registry does not fully report the amount or pattern of missing primary outcome observations, whether missingness differed by group, or the method used to handle missing repeated responses. It also does not report whether imputation or sensitivity analyses were used. As a result, the assumptions required to assess potential bias from missing data cannot be evaluated from the record.

The record identifies repeated survey observations and mixed-model analyses, but gives limited information about the analysis population, model specification, or handling of incomplete records. It does not establish whether all randomized participants were included in each analysis or whether any exclusions occurred. These omissions make it harder to assess how the analyses align with the randomized design and how robust the estimates are to missingness and multiple testing.

16. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

Most frequent serious adverse event terms

Serious events with the highest arm-level rate: heart attack (1.1%), mini stroke (1.1%).

The registry’s serious adverse event entries list 0 affected participants in the standard-pack group, 1 in the inserts-only group, 1 in the pictorial-warning-only group, and 0 in the combined group. The corresponding numbers at risk were 108, 94, 94, and 91. The record lists 0 other adverse events in each group and 0 deaths in each group.

The serious-event terms listed are “Heart Attack” and “Mini stroke,” each shown as 1.1%. These are registry-listed serious-event terms; the record does not provide enough detail to infer timing, causality, or whether the events were considered related to the study materials. The group counts are small, and these entries are descriptive safety reporting rather than a basis for comparing risk between groups. In particular, a count of zero in a group does not establish that an event cannot occur.

Safety results should be read separately from the self-efficacy endpoint. The registry’s event reporting does not establish whether any event was caused by an intervention, and no causal safety conclusion can be drawn from these figures alone.

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

The posted results describe average self-efficacy-to-quit scores during repeated surveys in a four-group randomized study of cigarette-pack messages. The group means were 2.61 for standard packs, 2.76 for inserts only, 2.62 for pictorial warnings only, and 2.56 for both inserts and pictorial warnings. Simple unadjusted comparisons with standard packs produced small mean differences, and their confidence intervals included zero.

The registry’s mixed-model analyses also report uncertainty around the estimates. The all-group entry gives a mean difference (net) of 0.26, with a 95% confidence interval from -0.73 to 1.25 and p = 0.60. Because the sponsor’s analyses differ from the simple recalculations, the sponsor’s own analysis takes precedence. The available registry results do not establish a clear difference in the primary score among the labeling groups.

This does not mean the groups have been shown to be exactly equivalent. The estimates are uncertain, no equivalence margin is reported, and the confidence intervals allow differences in either direction. Nor do these results establish whether participants quit smoking: the primary endpoint measures reported confidence, not cessation itself. The study’s repeated measures capture responses over the study period, not necessarily lasting changes after it.

The registry record does not report the complete statistical plan, the full mixed-model specification, or all details of missing-data handling and multiplicity control. Those omissions limit how fully the analysis can be evaluated. The secondary outcomes provide additional psychological measures, but their interpretation is tempered by the many outcomes and analyses listed and the absence of reported multiplicity procedures.

The appropriate reading is therefore bounded: the posted primary scores and reported analyses do not show a precise, clearly established difference in self-efficacy between the groups. They do not prove no effect, establish cessation benefit or harm, or settle the consequences of cigarette-pack messaging beyond the measured outcomes and study period.

18. Why This Trial Matters Statistically

This trial is statistically informative because it separates two communication features—pack inserts and pictorial warnings—in a factorial randomized design. That structure can, when fully reported, help distinguish responses associated with each feature from responses associated with their combination. It also illustrates why the interpretation depends on clearly specified contrasts: an estimate comparing selected groups is not automatically a main effect or an interaction.

The primary endpoint is measured repeatedly, rather than once at the end of the study. Repeated measures can capture changes in reported confidence over time, but they also require methods that account for observations being clustered within participants. The registry names mixed models, making the distinction between sponsor-model results and simple recalculations especially important. The latter provide transparent summaries of group means but do not reproduce the repeated-measures analysis.

The record also demonstrates the importance of precision and multiplicity. Point estimates close to zero may look reassuring at first glance, but the confidence intervals and analysis details are needed to understand what differences remain compatible with the data. At the same time, the registry lists many outcomes and analyses; individual secondary findings should not be treated as definitive without knowing how testing was planned and whether multiplicity was addressed.

Finally, self-efficacy is a psychological outcome, not a direct measure of quitting. Statistical interpretation must preserve that distinction. The trial contributes evidence about reported confidence during exposure to pack messages, while the registry record does not establish the effect of those messages on sustained cessation. Clear reporting of model contrasts, missing-data methods, and analysis plans would make the results easier to assess and reproduce.

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 primary endpoint is self-reported confidence to quit, not observed cessation or a longer-term smoking outcome. It therefore addresses a psychological response to pack messages rather than whether participants stopped smoking. The registry describes measurement during the 14 day study period, with analyses limited to days 2-14, so the posted outcome does not establish persistence after that period.

The primary outcome was measured repeatedly, but the registry record does not report the complete number of observations contributing to each estimate or the extent and pattern of missing primary surveys. The record notes differing analytic samples for moderator analyses, but does not fully describe missing-data methods. This limits assessment of possible bias from incomplete responses.

The sponsor’s results are reported as mixed-model analyses, but the full model specification and clear labeling of every contrast are not available in the registry record. The simple recalculations are unadjusted and do not account for the repeated-measures structure. Because the posted results differ, the sponsor’s own analysis takes precedence, while the available detail does not allow a complete reconciliation.

The record lists many outcome measures and analyses, but does not report multiplicity control or a full hierarchy of confirmatory and exploratory tests. Secondary findings may therefore be vulnerable to chance patterns across multiple tests. The registry also does not report the trial’s original sample-size justification or planned analysis assumptions, making it difficult to assess the relationship between the completed study and its original design goals.

Finally, the registry’s safety figures are sparse and descriptive. The record does not give enough detail to assess causality or compare event risks reliably. These limitations constrain what can be concluded from the reported scores and event counts.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Cigarette Pack Inserts for Smoking Cessation — NCT04075682.
• Registry record last updated 2024-05-13; status last verified 2023-11.
• Results first posted 2024-05-13.
• 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

NCT04075682, “Cigarette Pack Inserts for Smoking Cessation,” is a completed randomized, single-masked factorial study sponsored by the University of South Carolina. It enrolled 387 participants and compared standard cigarette packs, inserts only, pictorial warnings only, and both inserts and pictorial warnings.

The primary outcome was repeated self-efficacy-to-quit scores on a 1-to-7 scale, analyzed for days 2-14. The posted arm means were 2.61, 2.76, 2.62, and 2.56, respectively. Simple unadjusted comparisons were close to zero and their confidence intervals included zero. The registry also reports mixed-model analyses; where those results differ from recalculations, the sponsor’s analysis takes precedence.

The record lists secondary outcomes and safety events, but does not report the full statistical plan, original sample-size justification, complete model details, or multiplicity and missing-data procedures. The results concern reported confidence and other psychological measures, not demonstrated smoking cessation.