Comparative Acceptability of Tobacco and Menthol Flavored E-cigarettes: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Comparative Acceptability of Tobacco and Menthol Flavored E-cigarettes (NCT04996940), a randomized na study. The primary outcome is Questionnaire Scale "Intentions for Future Use".

2. Clinical Question

3. Trial Design

This was a randomized, unmasked crossover trial of tobacco- and menthol-flavored e-cigarette products. The registry record lists the study as completed, with enrollment of 52. The listed start date is 2021-06-20, and the primary completion date is 2021-08-15. The sponsor was the University of Kansas Medical Center.

In a crossover design, participants experience more than one study condition, allowing outcomes from different conditions to be measured in the same people. In this trial, participants completed a session with each flavor. One sequence was tobacco followed by menthol; the other was menthol followed by tobacco. Each session was described as lasting 30 minutes. Random assignment to sequence was reported, but masking was not. The registry record does not report further details about how sequence assignment was generated or concealed.

This structure matters when interpreting the outcome. The flavor-specific means are summaries of responses recorded after participants used each flavor. They do not describe two groups in which participants were assigned to use only one flavor. A formal comparison would ordinarily need to account for the crossover structure and, where relevant, the order in which flavors were used. The registry record does not report a statistical analysis of the primary outcome or explain whether order was considered.

The primary outcome was the questionnaire scale titled Intentions for Future Use, measured during the intervention. The registry also lists three secondary questionnaire outcomes: Subjective Vaping Experience, Perceptions of Flavor, and Product Demand. The phase is listed as NA. Additional protocol details, including eligibility criteria beyond the description of current menthol smokers, 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

Analysis population for the primary outcome, as defined in the registry: Participants who participated in the study

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 Questionnaire Scale "Intentions for Future Use", measured over during the intervention.

The primary endpoint records participants’ stated intention to use a flavor in the future. The questionnaire asks how likely the participant would be to try the product again if it were the only flavor available. Responses range from 1, extremely unlikely, through 5, extremely likely, with neutral at 3. This is a self-reported rating of intended behavior, not a measure of subsequent product use.

The outcome is useful for describing participants’ immediate reactions to the products because it asks about willingness to try the flavor again after use. It captures a subjective judgment that may be relevant to product acceptability. It does not, on its own, establish what participants later did, whether a flavor changed tobacco use, or whether any product had a health effect. Those questions require outcomes and follow-up not reported for this primary endpoint.

The registry reports the mean score and the full observed response range for each flavor. The mean was 3.1 for tobacco and 3.7 for menthol; each flavor had a full range of 1–5. The range indicates that responses extended across the scale, but it does not describe how many participants chose each response. The record does not report a standard deviation, confidence interval, or a statistical test for this outcome. It also notes that not all participants completed every study measure, and that the number of respondents was lower than enrollment.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

In general, a descriptive mean is the arithmetic average of recorded scores. For this trial, the posted means are 3.1 for tobacco and 3.7 for menthol on a scale from 1 to 5. A mean condenses multiple ratings into one value, which can help describe the overall response level. It does not show whether most ratings were close to the mean or whether ratings were widely spread. The registry record does not report a standard deviation or another measure of dispersion for this endpoint.

What does the response range tell us?

The full range is the lowest through the highest observed response. The reported range was 1–5 for each flavor, which covers the scale’s endpoints. A range is sensitive to extreme observations and does not describe the frequency of each score. It cannot show whether responses clustered around neutral, leaned toward likely, or were divided across the scale. A response distribution or suitable measure of spread would add information, but those details are not reported in the registry record.

Why does the crossover design matter?

In general, a crossover design allows each participant to experience multiple conditions. A comparison can then use the relationship between a participant’s responses under those conditions, rather than treating all observations as if they came from unrelated people. This can be useful when responses from the same person are correlated. The trial used two flavor orders: tobacco then menthol, and menthol then tobacco. The posted flavor summaries do not state whether they are based on matched responses from the same participants, or whether a paired statistical analysis was performed.

Order can also matter in crossover studies. A participant’s response after the later session could reflect not only the flavor used in that session but also the experience of the earlier session or the sequence itself. Statistical approaches for crossover data may account for such features when the design and data support doing so. The registry record does not report an analysis model, an assessment of order effects, or other details about how the crossover was analyzed. It would therefore be inappropriate to assume that these considerations were addressed.

What does uncertainty mean?

In general, a confidence interval gives a range of values compatible with the data under a stated statistical procedure. Its width communicates precision: a narrower interval indicates more precision than a wider one, under the assumptions of that method. A p-value, when reported, describes how compatible the observed data are with a specified null hypothesis under a statistical model; it is not the probability that the null hypothesis is true. Neither a confidence interval nor a p-value for the primary outcome is reported in the registry record. The means alone cannot establish the precision or statistical significance of a flavor comparison.

For a paired crossover analysis, the relevant uncertainty would generally be based on within-participant comparisons, not merely on the two flavor-specific means viewed independently. The registry does not report the paired differences or their variability. No new comparative estimate can be recovered from the posted means and ranges alone, because those summaries do not encode how each participant’s two responses relate to one another.

How should missing responses be considered?

The registry states that not all participants completed all study measures and identifies 41 respondents for each flavor on the primary outcome, against enrollment of 52. In general, missing outcomes can affect both the number of observations available for analysis and the representativeness of the analyzed responses. The impact depends on why values are missing and how missingness relates to participant characteristics or outcomes. The reasons for missing primary outcome responses and the handling of those responses are not reported in the registry record.

Equal respondent counts by flavor do not establish that the same participants contributed to both summaries. Nor do they show whether participants with missing data differed from those with observed scores. In a crossover design, the count of complete pairs is important to interpreting a paired comparison; that count is not reported. No assumptions about missingness or imputation should be attributed to the trial without a reported analysis plan or results.

What can a planning illustration show?

In general, a sample-size calculation is a planning exercise based on assumptions about an effect to detect, variability, significance level, allocation, and analysis method. A calculation using an effect observed in the trial is a retrospective planning illustration, not evidence that the completed study had a particular probability of detecting an effect. Such illustrations do not replace an analysis of the observed outcomes. The registry record describes planning assumptions based on a two-sided significance level of 0.05, equal allocation, and a normal approximation, with the target effect expressed as a fraction of the observed effect. It does not report the trial’s own design justification in the information available for this outcome. Any calculated sample-size illustration should therefore be interpreted as conditional on its stated assumptions, not as observed power or as a trial result.

For the primary outcome, the registry lists no analysis results. There is consequently no posted inferential procedure to evaluate for its assumptions, no reported adjusted or unadjusted comparative estimate, and no basis to characterize the analysis as confirming or rejecting a hypothesis. The descriptive mean scores remain informative about the recorded responses, but they answer a narrower question than a comparative statistical analysis would.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists the primary Intentions for Future Use scale and three secondary questionnaire outcomes: Subjective Vaping Experience, Perceptions of Flavor, and Product Demand. Considering multiple outcomes can raise the general issue of multiplicity: the more comparisons explored, the greater the chance that at least one result may appear noteworthy by chance. Methods such as prespecifying a primary outcome or adjusting inference can address this issue, depending on the analysis goals. The registry record does not report a multiplicity strategy, a hierarchy for testing outcomes, or inferential analyses for the listed measures.

The record lists zero analyses. It therefore does not provide a reported statistical analysis plan for the posted outcomes, adjustment method, model specification, or inferential results. Values are also described as missing or not estimable for a two-arm comparison for another endpoint. That note does not identify which specific analysis was attempted or explain the reason the values were unavailable.

For the primary outcome, the registry says that not all participants completed all measures and reports 41 respondents for each flavor, compared with enrollment of 52. The record does not report the number with complete paired responses, the reasons for missingness, the analysis population, or any method used to handle incomplete questionnaires. It also does not say whether participants were analyzed according to randomized sequence or according to the flavor sessions completed. These details matter for interpreting a crossover comparison and cannot be inferred from the respondent counts alone.

In general, missing-data methods rely on assumptions about the relationship between observed and missing outcomes. The assumptions and any sensitivity analyses used for this trial are not reported in the registry record. As a result, the summaries should be interpreted as describing the recorded responses, without assuming they represent every enrolled participant or that missingness was addressed in a particular way.

11. 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.

The registry reports serious adverse event counts separately for Tobacco Pod and Menthol Pod. For each group, 52 participants were listed as at risk, and zero serious adverse events were reported. The same counts are listed for other adverse events: 52 at risk and zero affected in each group. The death count was zero for each group. No serious adverse event terms are listed.

These figures are counts within the registry’s recorded safety categories. They indicate that no events were posted in those categories for the listed groups during the reported study record. They do not establish that no symptoms or other experiences occurred, because the registry presents only these event summaries and lists no serious event terms. The record does not provide additional detail about event collection, definitions, monitoring procedures, or the period over which events were assessed.

The safety groups are labeled by pod, while the trial design describes participants using both flavors in separate sessions. These group labels should not be read as evidence that the participants were independent parallel groups. The registry does not report an inferential comparison of safety outcomes. With zero recorded events in the listed categories, the counts are descriptive; they do not by themselves quantify the likelihood of an event outside the recorded study experience.

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

The registry record describes a randomized crossover study in which menthol cigarette smokers used both tobacco- and menthol-flavored e-cigarette products. For the primary questionnaire outcome, the recorded mean intention score was 3.1 after tobacco flavor and 3.7 after menthol flavor, on a scale from 1 to 5. The full observed response range was 1–5 for each flavor. These findings describe reported ratings among respondents, rather than later behavior.

The summaries do not establish a statistically reliable difference between flavors. The registry record reports no primary outcome analysis, confidence interval, p-value, or measure of variability. It also does not report the matched participant-level comparisons that would clarify how each person rated the two flavors. Without those elements, the degree of uncertainty around a comparative effect is not available from the posted results.

The scores measure stated intentions in response to a questionnaire during the intervention. They do not establish whether participants subsequently tried or used either flavor, whether their tobacco use changed, or whether either product caused a health outcome. They also do not establish that the ratings would generalize beyond the participants and circumstances recorded in this study.

The trial’s crossover structure is an important part of the interpretation. Participants were assigned to one of two flavor orders and had a session with each flavor. A flavor comparison should account for the relationship between responses from the same participants and may need to consider order. The registry record does not report how these design features were incorporated into an analysis.

Thus, the supported conclusion is descriptive and limited: the posted mean rating was higher for menthol than for tobacco, while both flavor conditions included responses across the full scale. The registry does not provide the inferential analysis needed to quantify uncertainty in that contrast or to establish what it means for actual future use.

13. Why This Trial Matters Statistically

This study illustrates why outcome summaries need to be read in the context of trial design. A crossover trial exposes participants to more than one condition, so observations under different conditions may be linked within a person. Flavor-specific means are useful descriptions, but they do not reveal the within-person pattern needed to understand a paired comparison. Recognizing that distinction prevents a reader from treating the study as if it were a conventional parallel-group trial.

The primary outcome also illustrates the difference between an immediate self-reported intention and observed behavior. A questionnaire rating can describe participants’ stated likelihood of trying a flavor again. It is not interchangeable with later product use, nor does it directly answer questions about changes in smoking or health. Matching the interpretation to the outcome actually measured is central to sound statistical reading.

Missing outcome data are another key feature. Enrollment was 52, while 41 respondents were recorded for each flavor on the primary measure. The record notes that not all participants completed all measures but does not describe missingness or the analysis population. In a crossover design, the availability of responses under both conditions affects how many participant-level comparisons can be made. A count for each condition alone does not reveal the number of complete pairs.

Finally, the absence of reported inferential analyses limits what can be concluded from the descriptive means. Statistical significance, precision, and the influence of order cannot be assessed from those means and ranges. This is not a reason to disregard the observations; it is a reason to distinguish what the record documents from what would require a comparative analysis. The study is a useful example of how design, outcome definition, missingness, and reporting jointly shape the interpretation of clinical research.

14. 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 offers limited detail about the statistical analysis of the primary endpoint. It reports flavor-specific means and full response ranges, but no standard deviations, confidence intervals, p-values, participant-level comparisons, or model results. As a result, the precision and inferential strength of the flavor comparison cannot be assessed from the record.

Although the study is described as randomized and crossover, the record does not state how randomization was generated or concealed. Masking is listed as none. The two flavor orders are described, but there is no reported assessment of sequence, period, or carryover effects, and no explanation of how order was handled analytically. These details are relevant to interpreting outcomes after participants use both flavors.

The primary outcome was recorded for 41 respondents in each flavor condition, while enrollment was 52. The record says that not all participants completed all measures but does not report reasons, completeness of paired responses, missing-data methods, or characteristics of respondents with missing outcomes. It is therefore unclear how well the recorded scores represent the enrolled participants.

The endpoint measures stated future-use intentions during the intervention, not subsequent product use. The registry does not report follow-up results for actual use in the primary outcome description. The primary question is therefore limited to the questionnaire response recorded after product use. The record also lists three secondary outcomes but no analyses, so their results cannot be incorporated into an overall assessment of the study.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Comparative Acceptability of Tobacco and Menthol Flavored E-cigarettes — NCT04996940.
• Registry record last updated 2023-09-13; status last verified 2023-08.
• Results first posted 2023-09-13.
• 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.

18. Record Summary

NCT04996940, titled Comparative Acceptability of Tobacco and Menthol Flavored E-cigarettes, is a completed randomized, unmasked crossover trial sponsored by the University of Kansas Medical Center. Enrollment was 52. The listed study start date was 2021-06-20, and primary completion was 2021-08-15. Participants completed sessions with tobacco and menthol flavors in one of two orders.

The primary outcome was the Intentions for Future Use questionnaire scale, with responses from 1 to 5. The reported mean was 3.1 for tobacco and 3.7 for menthol, with a full observed range of 1–5 for each and 41 respondents per flavor. The registry reports zero analyses and does not provide inferential results for the primary outcome. It lists three secondary questionnaire outcomes. Safety summaries report zero serious adverse events, zero other adverse events, and zero deaths in each pod group, with 52 at risk in each.