Integrating Use of Self-Affirmation Content Into a Mobile App to Promote Quit Attempts With Text-Based Smoking Cessation Intervention Messaging: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Integrating Use of Self-Affirmation Content Into a Mobile App to Promote Quit Attempts With Text-Based Smoking Cessation Intervention Messaging (NCT03027466), a randomized na study. The primary outcome is Number of Participants Reporting a Quit Attempt Within the Past Week at 1 Month and 3 Months.

2. Clinical Question

The study addresses whether adding self-affirmation material to a mobile smoking-cessation app may be associated with quit-attempt reporting. The rationale in the registry record is that self-affirmation—focusing on personal strengths and values—could offset a threat to self-identity and support healthier behavior. That rationale motivates the research question; it is not evidence that the content changed behavior.

At issue statistically is not simply whether people received affirmation content, but how outcomes were compared across the different combinations of the baseline quiz and affirmation messages. The registry record describes those combinations but does not report the planned contrasts or an analysis of the primary outcome. Because the primary outcome has two follow-up time points and is not reduced in the record to one comparison, the posted material does not answer a single, clearly specified between-group question with an effect estimate.

3. Trial Design

The registry record describes a randomized, parallel-group study with no masking. Its status is completed. Enrollment is reported as 7899. The registered phase is NA. The National Cancer Institute is listed as the sponsor. The recorded study start is 2018-03-06 and primary completion is 2019-09-01.

The study took place entirely on a mobile device. According to the description, participants entered demographic and smoking information and a desired quit date, then received quitting-related texts. Follow-up surveys were scheduled at 1 month and 3 months. The study therefore combined app-based content with later self-reported outcomes; the registry record does not report the amount of follow-up completed or the number of participants contributing outcome data at either follow-up.

The arm descriptions vary two content features: a baseline affirmation quiz and affirmation text messages during the study. They describe assignments with both features, with the quiz but not the messages, with the messages but not the quiz, and with neither. The last description identifies the app experience without affirmation content. Although this structure could support comparisons of individual components or their combination, the registry record does not report which comparisons were planned or analyzed.

Random allocation is a design feature intended to make groups comparable on average at assignment, including characteristics that were not measured. It does not guarantee that groups are identical, ensure that participants complete follow-up, or establish that outcome reporting is free of bias. With no masking, participants knew their assigned content, and the record does not describe masking of outcome assessment or analysis.

The registry record lists a primary outcome and secondary outcomes concerning smoke-free days at follow-up and tobacco-product type. However, no outcome analysis is reported. The record also does not report the allocation procedure in greater detail, the exact timing or frequency of text delivery, or further implementation information. Those details should not be inferred from the brief study description.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

Reasons for not completing this period:

5. Baseline Characteristics

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

6. Primary Endpoint

The registry lists the primary outcome as Number of Participants Reporting a Quit Attempt Within the Past Week at 1 Month and 3 Months, measured over 1 month and 3 months.

The primary outcome is titled as the number of participants reporting a quit attempt within the past week at 1 month and 3 months. Its timeframe is listed as 1 month and 3 months. The description says participants were asked to report the number of times they attempted to quit smoking at those follow-ups. These phrases do not describe the outcome in exactly the same way: one focuses on whether participants reported a quit attempt, while the other refers to the number of attempts. The registry record does not clarify how that distinction was handled in analysis.

A quit attempt is a behavioral outcome rather than a direct measure of long-term abstinence. It can capture whether participants report initiating an effort to stop smoking during the specified period. Follow-up measurements can also describe behavior at different stages after assignment. However, a count of participants reporting an attempt does not by itself indicate how long an attempt lasted, whether smoking stopped permanently, or whether the intervention changed smoking behavior overall.

Using a self-reported endpoint can make mobile follow-up practical, but responses depend on participants completing the survey and accurately recalling or interpreting the question. The registry record does not describe a verification method, a recall procedure beyond the stated past-week period, or how incomplete responses were handled.

The outcome is listed as a count of participants with an event, but the registry record presents it at two follow-up times and does not report results for either time. It also does not report a combined endpoint or a prespecified way to handle results across the follow-ups. As a result, the endpoint’s purpose can be described, but the posted results cannot show the frequency of reported quit attempts or quantify a difference between assignments.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry record identifies a randomized, parallel design without masking, but it does not report a statistical analysis plan or the methods used to analyze the outcomes. The explanations below are general statistical context, not a description of analyses confirmed to have been performed in this trial.

What does random allocation contribute?

In a randomized study, assignment is determined by a chance-based procedure rather than by participants or investigators choosing a group. If allocation is implemented as intended and outcome data are handled appropriately, randomization can support a causal comparison because measured and unmeasured baseline characteristics are expected to be balanced on average. It does not guarantee exact balance in the realized groups. Nor does it eliminate bias arising after assignment, such as different follow-up rates, deviations from assigned content, or differential reporting.

The registry record says allocation was randomized and describes a parallel model. It does not report the sequence-generation method, concealment details, or any analysis that accounts for assignment. Those implementation and analysis details cannot be inferred from the design label.

Why does the comparison need to be specified?

The arm descriptions distinguish a baseline quiz and affirmation messages. A comparison between assignments that differ in both features estimates the contrast between those combinations. A comparison that differs only in the messages addresses a different contrast. If the question is whether one component works differently depending on whether the other component is present, that interaction also requires a defined analysis. General factorial-design reasoning makes these distinctions important, but the registry record does not state which contrast was planned or used.

For a binary outcome such as whether a participant reports a quit attempt, an analysis might compare the proportions of participants with the event in specified groups. Other approaches may model the event while adjusting for prespecified variables or accounting for repeated measurements. The registry record reports no outcome values and no analysis method, so it is not possible to identify the estimator, model, or assumptions used.

How should repeated follow-up measurements be understood?

Measurements at 1 month and 3 months may be analyzed separately or with a model that accounts for repeated observations from the same participant. Separate analyses address time-specific contrasts but can create multiple statistical tests. A repeated-measures approach can use information across follow-ups while recognizing that responses from one participant are related. It requires assumptions about the relationship among measurements and, depending on the model, how missing observations relate to observed data.

These are general options, not reported methods for this study. The registry record does not state whether outcomes at the two follow-ups were analyzed separately, jointly, or according to a hierarchy. Its note that the primary outcome has several categories or time points and was not reduced to one comparison reinforces that a single contrast cannot be extracted from the posted record.

What would uncertainty measures tell a reader?

A confidence interval expresses uncertainty around an estimated effect under a statistical model. A relatively narrow interval indicates more precision than a wider interval, while both the estimate and interval are needed to understand the data. A p-value measures how unusual data at least as incompatible with a specified null model would be, if that model and its assumptions held. It is not the probability that the null hypothesis is true and does not measure clinical importance. The registry record reports no primary-outcome estimate, confidence interval, or p-value, so none of these interpretations can be applied to the quit-attempt outcome.

For a count outcome, variability may exceed what a simple model assumes. General methods for count data sometimes account for over-dispersion, meaning that observed variation is larger than expected under a basic count model. That issue is relevant only if the outcome is analyzed as a count with an appropriate model. The primary outcome is described both as participants reporting a quit attempt and as the number of attempts, and the registry record does not clarify which data representation or model was used. No conclusion about over-dispersion or its handling is possible.

What do the planning assumptions mean?

The registry record describes a planning illustration based on the observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It says the effect to detect is a stated fraction of the observed effect. Such an illustration is a conditional planning calculation: its output depends on the assumed effect and method. It is not an observed trial result, a measure of the study’s actual power, or proof that the original study was designed around those assumptions. The registry record does not report the trial’s own sample-size justification or design assumptions.

The record also describes how planning results are generally expressed: participants per arm for non-time-to-event outcomes, and total events for time-to-event outcomes. The primary endpoint is not described as a time-to-event outcome. No specific planning outputs for the primary endpoint are reported in the registry record, so no sample-size value or planning conclusion can be stated.

Why is an analysis population important?

An analysis population defines which participants contribute to an estimate. An intention-to-treat analysis generally compares participants according to randomized assignment, whether or not they fully followed the assigned intervention. Other analysis populations may require adherence or available outcome data. These choices can affect interpretation and can introduce bias if inclusion depends on post-randomization events. The registry record does not report which population was used for the primary outcome, the extent of follow-up, or how missing observations were treated.

10. Multiplicity, Missing Data and Analysis Populations

The primary endpoint is recorded at 1 month and 3 months, and the study also lists secondary outcomes concerning smoke-free days and tobacco-product type. These create multiple potential outcome comparisons. In general, testing several outcomes, time points, or group contrasts can increase the chance of at least one apparently unusual result when all tests are interpreted without accounting for multiplicity. Prespecifying a primary comparison, using a hierarchy, or applying an adjustment are possible strategies, but the registry record does not report whether any was used.

The posted record does not identify a primary analysis contrast among the described content assignments. It also does not say whether the two follow-up times were treated as separate endpoints, combined, or analyzed in another way. Without those details, readers cannot determine the scope of testing or judge whether any reported statistical significance—if it were available—would reflect an adjusted or unadjusted analysis. No p-values or outcome estimates are reported in any case.

Missing data are especially relevant in mobile follow-up studies because participants may not complete every survey. The registry record describes surveys at 1 month and 3 months but does not report completion, missingness by assignment, reasons for nonresponse, or the method used to handle missing outcomes. General approaches include analyzing available responses, imputing missing values under stated assumptions, or conducting sensitivity analyses. Each approach relies on assumptions, and no approach is identified for this trial.

The record also does not specify whether the primary analysis included everyone as randomized, only participants with available survey responses, or another population. Thus, the enrolled population cannot be assumed to match the analysis population. These omissions leave uncertainty about both the amount of information contributing to the outcome and the robustness of any potential comparison. They are reporting gaps in the registry record, not evidence that a particular method was or was not used outside that record.

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 record lists serious adverse event and other adverse event counts as zero in each described assignment. It also lists the number at risk for each group: 2027 for Baseline Affirmation and Affirmation Texts, 1984 for Baseline Affirmation and No Affirmation Texts, 1903 for No Baseline Affirmation and Affirmation Texts, and 1985 for No Baseline Affirmation and No Affirmation Texts. Deaths are listed as zero in each group.

These are registry-reported counts, not an estimate of comparative risk. A zero count in the posted record means that no events were recorded in that category for the listed group; it does not establish that no participant experienced an event that went unreported, was not collected, or was outside the registry’s reporting definition. The record does not provide event definitions, collection procedures, observation time, or an explanation of how the at-risk figures were determined.

Adverse-event interpretation also depends on the completeness of surveillance and on whether events were assessed consistently across groups. The record lists no top serious-event terms. That absence does not supply further detail about monitoring or reporting. The counts alone do not establish whether affirmation content caused or prevented an adverse event, and they should not be read as a complete account of safety.

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

The registry record establishes that a randomized mobile-app study was conducted to examine self-affirmation content in a text-based smoking-cessation intervention. It describes content assignments, follow-up at 1 month and 3 months, and a primary outcome concerning reported quit attempts. It also reports enrollment and adverse-event counts.

Those design details do not establish what the trial found. The registry record contains no analyzed values for the primary outcome, no between-group estimate, and no measure of uncertainty. It therefore cannot show whether reported quit attempts were more frequent, less frequent, or similar under any of the described assignments. The absence of an analysis in the record should not be interpreted as evidence that the interventions had no effect.

Even with an outcome estimate, interpretation would depend on the comparison being made. Comparing content that includes both a baseline quiz and messages with an assignment containing neither would address the combined content package. Comparing assignments that differ only in message delivery would address a component contrast. The record does not report which contrast was analyzed, so these questions cannot be answered from the posted information.

A report of a quit attempt is also not the same as sustained smoking cessation. The endpoint’s title refers to participants reporting an attempt within the past week, while its description refers to the number of attempts. The registry record does not clarify that wording difference or report quit-attempt outcomes. The listed secondary outcomes concerning smoke-free days and tobacco-product type likewise have no posted analysis results.

In general, statistical results describe evidence under specified definitions, comparisons, and assumptions. They do not automatically establish how a result would apply to people outside the study or whether a behavioral change would persist. Here, the immediate limitation is more basic: the registry record does not supply outcome analyses from which an effect or its precision could be interpreted. No treatment conclusion follows from the design information alone.

13. Why This Trial Matters Statistically

This trial is statistically notable for the structure of its intervention question. The registry record distinguishes a baseline affirmation quiz from affirmation text messages. Evaluating these components separately is not the same as evaluating them as a combined package. A well-defined contrast is therefore central: it specifies which assignments are being compared and what effect that comparison is intended to represent.

The outcome is also measured at two follow-up times. Multiple follow-ups can describe how an outcome varies over time, but they raise questions about whether each time point is a separate primary test or whether a joint analysis is intended. Those choices affect interpretation and the handling of multiplicity. The registry record does not report a chosen analysis or results, making the absence of a clearly reported comparison important to understanding what can be learned from the entry.

The mobile format brings practical considerations for outcome measurement. Surveys conducted through an app can collect self-reported behavior at follow-up, while incomplete responses may affect the amount and representativeness of available data. The registry record does not report follow-up completion or missing-data methods. These details matter because randomization at assignment does not by itself protect an analysis from bias caused by differential loss of outcome information.

The study also illustrates the difference between a registered research question and an interpretable result. A completed status, enrollment figure, and described outcome do not substitute for outcome counts, an effect estimate, and its uncertainty. Without those elements, the trial’s statistical contribution to the question about quit attempts cannot be quantified from the registry record. This is a reporting limitation, not evidence for or against the intervention.

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 principal limitation for interpreting this trial is that the registry record does not report analyzed primary-outcome results. It gives no arm-level quit-attempt counts, effect estimate, confidence interval, or p-value. Although enrollment is reported as 7899, the record does not state how many participants posted data at each follow-up or contributed to an analysis.

The primary outcome’s wording is not fully consistent. Its title concerns the number of participants reporting a quit attempt within the past week, while its description asks about the number of times participants attempted to quit. Those represent different ways to characterize behavior. The record does not say how responses were coded or how the discrepancy was resolved.

The intended comparison is also not specified. The arm descriptions distinguish the baseline quiz and affirmation messages, but the record does not report which assignments were contrasted for the primary outcome or whether component effects were examined. With measurements at 1 month and 3 months, the absence of a stated analysis plan also leaves unclear how time points and multiple potential outcomes were handled.

The registry record does not report the method of randomization, the analysis population, missing-data handling, survey completion, adherence to assigned content, or detailed measurement procedures. It does not report a sample-size justification for the trial; the described planning illustration is not the trial’s own design rationale. These details could affect the interpretation of any outcome analysis, but their absence does not show what methods were used outside the record.

Finally, the safety information is limited to reported event counts and at-risk figures, without event definitions or monitoring detail. The study’s mobile self-report approach and the behavioral nature of the endpoint also mean that the registry record alone cannot establish the accuracy or persistence of reported quit attempts. These limitations constrain conclusions from the entry and should not be replaced by assumptions.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Integrating Use of Self-Affirmation Content Into a Mobile App to Promote Quit Attempts With Text-Based Smoking Cessation Intervention Messaging — NCT03027466.
• Registry record last updated 2021-07-23; status last verified 2021-07.
• Results first posted 2021-07-23.
• 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

The registry record describes NCT03027466, a completed randomized, parallel-group mobile study with no masking, sponsored by the National Cancer Institute. It enrolled 7899 and evaluated combinations of a baseline affirmation quiz and affirmation text messages within a smoking-cessation app. Follow-up surveys were scheduled at 1 month and 3 months.

The primary outcome concerns participants reporting a quit attempt, and secondary outcomes concern smoke-free days and tobacco-product type. The record lists zero serious and other adverse events in each assignment. It reports no outcome analyses, primary-outcome values, comparative estimates, confidence intervals, or p-values. The intended primary comparison, missing-data methods, and analysis population are not reported in the registry record. The entry therefore describes the trial question and design but does not permit an assessment of the effect on reported quit attempts.