Online Survey in Studying Smoking Attitudes and Behaviors Among Current Smokers: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Online Survey in Studying Smoking Attitudes and Behaviors Among Current Smokers (NCT03863795), a registered phase not stated study. The primary outcome is Quit Intentions.

2. Clinical Question

3. Trial Design

This is a completed online survey study, not a reported clinical treatment trial. The registry title describes a survey of smoking attitudes and behaviors among current smokers, and the official title places the study in the context of self-affirmation and stigma-induced identity threat. The listed intervention is Survey Administration. The record says participants were recruited and prescreened through MTurk and responded to a one-time research survey using SurveyGizmo. The survey was described as taking over 20 minutes; the primary outcome timeframe is stated as the time of survey response within approximately 20 minutes.

The record lists enrollment of 1100 and gives a start date of 2019-02-04, a primary completion date of 2019-03-23, and a status of COMPLETED. It identifies four groups with labels combining smoking stigma or non-stigma control and self-affirmation or no-affirmation control. However, it does not state how participants were allocated to those groups. It explicitly cautions that the groups do not form a randomized, like-for-like control comparison. Group labels therefore should not be treated as proof of random assignment or as evidence that differences between groups are attributable to an assigned condition.

The record lists six outcome measures, but the primary outcome is the quit-intentions item. It is a continuous measure summarized by a mean and standard deviation. The registry record does not report the survey instrument beyond the described single item, the exact group sizes, the demographic composition, or the procedures used to implement the stigma and affirmation conditions. It also does not report a statistical analysis plan or analysis results. The listed intervention and group descriptions give a broad account of the survey setting, but they are not enough to reconstruct a full protocol or to evaluate whether all participants experienced the same procedures.

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: Respondents were directed to complete a 15-20 min online survey on smoking attitudes and behaviors. Participants provided electronic informed consent and then answered two questions to confirm their current smoking status. Current established smokers (smoked at least 100 cigarettes in lifetime and now smoke every day or some days) were eligible. Eligible participants completed survey items assessing smoking characteristics and smoking self-concept.

Overall Study

Analysis population for the primary outcome, as defined in the registry: Listwise deletion: 1 participant did not respond to the survey item on quit intentions. Required summary level data only. Interpretation of these unadjusted results is not recommended. Please refer to manuscript for adjusted linear regression results.

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 Quit Intentions, measured over At the time of survey response within approximately 20 minutes.

The primary endpoint is “Quit Intentions,” measured in units on a scale. It is based on one question: “Are you considering quitting smoking during the next 6 months?” The item is described as originating from the 2014-2015 Tobacco Use Supplement to the Current Population Survey. Its response options distinguish plans to stop within the next 30 days, plans to stop within the next 6 months but not within 30 days, and not thinking of quitting in the next 6 months. The options were reverse scored so that higher scores represent greater intentions to quit.

Intentions are used as a self-reported measure of a participant’s stated orientation toward quitting at the time of the survey. They are not the same as a later quit attempt, sustained abstinence, or another observed behavior. The endpoint therefore describes reported intention, not whether a participant subsequently stopped smoking. The record specifies the mean and standard deviation as the summary format for this continuous outcome, but it does not report the numerical mean, standard deviation, or a group contrast.

The record notes that one participant did not respond to the quit-intentions item and that listwise deletion was used. It also cautions that interpretation of the unadjusted results is not recommended and directs readers to manuscript adjusted linear regression results. Those adjusted results are not reported in the registry record. Consequently, the registry description identifies what the endpoint measures and how it is coded, but does not provide an estimate from which to judge the observed level or precision of quit intentions.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a mean and standard deviation describe?

The primary endpoint is continuous, and the registry identifies the mean and standard deviation as its summary statistics. In general, a mean describes the center of a set of numerical observations. A standard deviation describes how spread out observations are around that center. Neither statistic alone gives the full distribution, and both depend on the underlying observations and the way the scale is coded. Here, the record explains the direction of scoring: higher scores indicate greater intention to quit. It does not report the numerical mean or standard deviation, so the outcome’s observed level and variability cannot be characterized from the registry information.

A mean is not a percentage of participants and should not be read as the proportion who plan to quit. The endpoint combines ordered response categories into a numerical score, and interpretation depends on that coding. The record explains the reverse scoring but does not report a group-specific distribution. The question asks about intentions during the next six months, not verified quitting behavior during that period.

What would an uncertainty interval mean?

In general, a confidence interval expresses the statistical uncertainty around an estimate under the assumptions of the analysis. A narrower interval indicates greater precision for that estimate, while a wider interval indicates less precision. It does not describe the range of individual participants’ scores; the standard deviation is more directly related to variation among observations. The registry record does not report a confidence interval for the primary endpoint or a group contrast, so the precision of an estimated association cannot be assessed from this record.

A p-value, in general, measures how compatible the observed data are with a specified null hypothesis under the model assumptions. It does not measure the size or practical importance of an association, and it does not establish causality. No primary-outcome p-value is reported in the registry record. Since no analysis estimate is posted, statistical significance cannot be inferred from the group names, enrollment, or the study’s completed status.

Why does the group structure matter?

Four groups are named, but allocation is not stated. The record specifically warns that the comparison is not randomized and the arms are not a like-for-like control comparison. In general, random assignment can help balance measured and unmeasured characteristics across groups, while a nonrandom comparison may reflect pre-existing differences as well as any condition-related association. Even if the groups were described as controls, their labels do not demonstrate comparability. The registry does not report group sizes, baseline comparisons, or a causal identification strategy.

The record also points away from unadjusted interpretation and toward adjusted linear regression results in a manuscript. In general, linear regression can estimate the association between a continuous outcome and one or more explanatory variables while accounting for included covariates. Adjustment does not automatically remove bias: interpretation depends on the variables selected, their measurement, model assumptions, and the study design. The covariates, model form, estimates, and diagnostics are not reported in the registry record. No adjusted conclusion can therefore be drawn from the registry entry.

What do the planning calculations represent?

The record describes sample-size planning illustrations based on an observed effect, using a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The effect to detect is defined as a stated fraction of the observed effect. These are planning assumptions, not a record of the study’s original design justification. The registry does not report the trial’s own sample-size justification, and the calculations should not be read as evidence that the enrolled sample had a particular probability of detecting an effect. Sample sizes from such calculations are planning illustrations, never observed power.

In general, a planning calculation asks what sample size would be needed under specified assumptions about an effect, variability, allocation, and error rates. If an assumed effect is taken from observed data, the resulting illustration inherits uncertainty in that observed estimate. It does not establish that an effect is real or that an analysis was adequately powered. The record does not provide the numerical planning results in the registry record for this explanation, so no sample-size value or achieved power can be stated.

What is not available for evaluating the analysis?

The registry record lists zero analyses. It does not report an analysis plan, model details, outcome estimates, confidence intervals, or p-values. It also does not report interim analyses or multiplicity adjustments. These absences limit what can be said about the inferential methods actually used. General explanations of means, intervals, regression, and planning calculations describe statistical concepts; they do not fill in unreported details of this study.

10. Confidence Intervals for Each Arm

No two-arm comparison is shown for this trial, but each arm's mean can be given with a t-based 95% confidence interval computed from the posted mean, spread and number of participants.

R code and output

arm <- c("Smoking Stigma, Self-Affirmation", "Smoking Stigma, No-Affirmation Control", "Non-Stigma Control, Self-Affirmation", "Non-Stigma Control, No-Affirmation Control")
n   <- c(306, 269, 244, 280)
m   <- c(1.9, 1.72, 1.89, 1.79)
sp  <- c(0.74, 0.71, 0.71, 0.7)
sd  <- sp      # posted standard deviations
for (i in seq_along(n)) {
  se <- sd[i] / sqrt(n[i]); h <- qt(0.975, n[i] - 1) * se
  cat(sprintf("CARM|%s|%d|%.3f|%.3f|%.3f|%.3f\n", arm[i], n[i], m[i], sd[i], m[i] - h, m[i] + h))
}

CARM  Smoking Stigma, Self-Affirmation  306  1.900  0.740  1.817  1.983
CARM  Smoking Stigma, No-Affirmation Control  269  1.720  0.710  1.635  1.805
CARM  Non-Stigma Control, Self-Affirmation  244  1.890  0.710  1.800  1.980
CARM  Non-Stigma Control, No-Affirmation Control  280  1.790  0.700  1.708  1.872

11. Multiplicity, Missing Data and Analysis Populations

The registry lists six outcome measures and identifies quit intentions as the primary outcome. It reports zero analyses. The registry record does not describe a multiplicity strategy, such as whether secondary outcomes were treated as exploratory or whether any adjustment was planned for evaluating several outcomes. When multiple outcomes or comparisons are examined, the chance of finding at least one apparently notable result can increase. Whether that concern applies to the reported analyses cannot be judged from the registry because the analyses and their plan are not reported.

For the primary endpoint, the record states that one participant did not respond to the quit-intentions item and that listwise deletion was used. Listwise deletion excludes a participant without an observed value for the outcome from an analysis that requires that value. The record does not report the numerical analysis sample size, reasons for the missing response, or whether the omitted response was related to the participant’s likely score. It also does not describe any sensitivity analysis for missing data. The enrollment count should therefore not be treated as the exact number analyzed for the primary endpoint.

The record cautions that unadjusted results should not be interpreted and refers readers to adjusted linear regression results in a manuscript. The adjusted results, model covariates, and analysis population are not reported in the registry record. It also does not report whether any interim analyses occurred or give a statistical analysis plan. These missing details prevent a fuller assessment of how outcome multiplicity, missing observations, and the choice of analysis population were handled.

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

The record establishes that a completed online survey study enrolled 1100 current smokers and measured quit intentions using a single self-report item. It also describes four groups related to stigma and self-affirmation. It does not report numerical primary-outcome results, so no value for average intention, variability, group difference, or uncertainty can be stated. In particular, the record does not support a conclusion about the direction or size of an association between the named groups and quit intentions.

The endpoint concerns what participants said about considering quitting, not whether they later attempted to quit or stopped smoking. Stated intention and subsequent behavior are distinct outcomes. The registry record does not report follow-up for actual quitting, and the survey is described as a one-time response. Consequently, this outcome cannot be interpreted as a measure of later smoking behavior.

The four groups do not establish a causal comparison. Allocation is not stated, and the record explicitly cautions that the arms are not a randomized, like-for-like control comparison. Even if numerical group summaries were available, they would need to be interpreted in light of possible differences between participants and the analysis approach. The record itself advises against interpreting unadjusted results and refers to adjusted regression results that are not reported in the registry entry.

The registry’s zero event counts are limited to the listed adverse-event categories and groups. They do not establish that every possible event was assessed or that the survey conditions were compared for safety. Likewise, a completed status indicates the study’s registry status, not that its main findings are statistically conclusive. The appropriate reading is narrow: the record describes a survey and an intended measure, but does not report enough outcome analysis to determine the observed association or its precision.

13. Why This Trial Matters Statistically

This study raises a useful statistical distinction between measuring an attitude and evaluating an intervention effect. The primary outcome is an intention score, while the study topic concerns stigma and self-affirmation. A well-defined measure helps identify the construct under study, but a self-reported intention is not interchangeable with observed quitting behavior. Keeping the outcome’s meaning clear prevents an interpretation from extending beyond what was measured.

The group labels also illustrate why design information matters. A set of named conditions does not establish randomization, and this record says allocation is not stated and the groups are not a like-for-like control comparison. Statistical adjustment can address measured covariates under model assumptions, but it cannot, by itself, guarantee that groups are comparable or resolve every source of bias. The registry’s warning about unadjusted interpretation makes the unavailable adjusted analysis especially relevant to readers assessing the evidence.

Finally, reporting the estimate and its uncertainty is essential for evaluating a continuous outcome. A stated outcome definition and planned summary statistics do not reveal the observed result. With no primary-outcome estimate or interval reported in the registry record, the magnitude and precision of the findings cannot be assessed. This distinction is important in survey research, where a completed study and a specified endpoint can coexist with limited posted information about the statistical results.

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 description gives limited detail for evaluating the study procedures. It says participants were recruited and prescreened through MTurk and completed a one-time survey in SurveyGizmo, but it does not report recruitment dates beyond the overall study dates, eligibility criteria, participant characteristics, or group sizes. The survey is described as taking over 20 minutes, while the outcome timeframe is stated as within approximately 20 minutes. The record does not provide further information about survey completion or data-quality procedures.

Although the official title refers to self-affirmation and stigma-induced identity threat, the registry summary does not describe the exact prompts or experimental materials. It lists four group labels but does not state allocation. The record cautions that the arms are not a randomized, like-for-like control comparison, which limits causal interpretation. The adjusted linear regression results referenced by the primary outcome description are not reported in the registry record, and the model covariates and assumptions are not specified there.

The primary measure is a single item about intentions to quit, not a behavioral outcome. One participant did not respond to that item; listwise deletion was used, but the record does not give the reason for nonresponse or a missing-data sensitivity analysis. It also does not report numerical outcome summaries, analysis estimates, or uncertainty measures. The record lists six outcome measures but does not describe how multiplicity was handled. These limitations concern what can be assessed from the registry entry; they do not establish that unreported procedures or analyses were absent from other study materials.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Online Survey in Studying Smoking Attitudes and Behaviors Among Current Smokers — NCT03863795.
• Registry record last updated 2021-09-08; status last verified 2021-08.
• Results first posted 2021-09-08.
• 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

NCT03863795 is a completed online survey study titled “Online Survey in Studying Smoking Attitudes and Behaviors Among Current Smokers,” with the official title “Self-Affirmation in the Context of Stigma-Induced Identity Threat Among Smokers.” The sponsor is the National Cancer Institute (NCI). The record lists enrollment of 1100, a start date of 2019-02-04, and a primary completion date of 2019-03-23. Participants were recruited and prescreened through MTurk and completed a one-time survey using SurveyGizmo.

The primary outcome is a single-item quit-intentions score, with higher scores indicating greater intentions to quit. The record specifies a mean and standard deviation summary and notes one missing response handled by listwise deletion. It cautions against interpreting unadjusted results and refers to adjusted linear regression results in a manuscript. Numerical outcome results and those adjusted results are not reported in the registry record. Four groups are listed, but allocation is not stated and the record warns that they are not a randomized, like-for-like comparison.