Examining Brain Responses Linked to Emotion in Individuals Who Smoke Cigarettes: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Examining Brain Responses Linked to Emotion in Individuals Who Smoke Cigarettes (NCT04310735), a randomized na study. The primary outcome is Brain Activation During Smoking-related and Neutral Cues.

2. Clinical Question

The study also describes facial coding as a way to assess subjective affective responses, and lists self-reported urge to smoke and affect among secondary outcomes. Its brief summary says the expectancy manipulation was examined under conditions designed to induce an urge to smoke. Those details establish the broader research question, but they do not substitute for reported outcome estimates. The registry record does not report a numerical finding that would establish the direction, size, or precision of the primary brain-imaging comparison.

3. Trial Design

The registry identifies the study as randomized, unmasked, and parallel-group. The listed intervention is a behavioral verbal smoking-expectancy manipulation. Expect-Yes instructions were intended to create the perception that smoking would be possible during the session; Expect-No instructions were intended to create the perception that it would not be possible. The registry also lists a Behavioral Testing arm. Participants in that arm did not complete a scan session and were not assessed for the primary brain-activation outcome.

The registry reports an enrollment of 62 and gives the study status as completed. Its recorded start date is 2019-10-04, and its primary completion date is 2023-06-30. These record details describe the study timeline and enrollment, not the number of participants contributing usable data to every outcome. The brain-imaging analysis population is described separately as participants who completed the scan session and provided usable brain-imaging data. The number meeting that criterion is not reported in the registry record.

The primary outcome is brain activation during smoking-related and neutral cues, with a timeframe of approximately Day 7 (Visit 2). The record describes fMRI measurement and comparison of COPE values in two a priori regions of interest. The broader study summary also describes facial coding as a primary endpoint, while the outcome list records coded facial expressions among secondary outcomes. The registry record does not clarify this difference in endpoint labeling.

The record does not report how the random sequence was generated, how assignment was concealed, or whether any analysis was adjusted for baseline characteristics. It also does not report a statistical analysis plan, sample-size justification, or details of interim analyses. Randomization describes the assignment approach, but does not by itself answer those methodological questions.

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: Participants who completed the scan session and who provided usable brain imaging data were included in analyses. Participants in the Behavioral Testing Arm did not complete a scan session and were not assessed for brain activation during smoking-related and neutral cues.

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 Brain Activation During Smoking-related and Neutral Cues, measured over Approximately Day 7 (Visit 2).

The primary endpoint is brain activation during smoking-related and neutral cues, measured using blood-oxygen-level dependent signal from functional magnetic resonance imaging. The reported measure is a contrast of parameter estimate, or COPE. In this study, COPE values were calculated for each participant and extracted for two a priori regions of interest, with the aim of comparing activation patterns between the expectancy groups.

A contrast between cue types represents a difference in the measured brain response to smoking-related cues relative to neutral cues. A larger COPE value is described in the registry as indicating a greater difference in brain activation between those cue types. It should not be read as a direct measure of overall brain activity, nor does a contrast alone establish why a response differs. Interpretation depends on the regions studied, the cue comparison, the group estimates, and their uncertainty.

The endpoint is useful for testing whether the expectancy manipulation is associated with a change in the neural response to smoking-related cues relative to neutral cues. The registry describes the imaging analysis as using two a priori regions of interest, rather than presenting the endpoint as a general measure of all brain activity. The names of those regions and the numerical results for each are not reported in the registry record.

The outcome is reported as a mean, with standard error identified as the measure of dispersion. The timeframe is approximately Day 7 (Visit 2). The registry describes preprocessing and analysis software, but the resulting group means, standard errors, and between-group estimates are not reported in the registry record. Therefore the endpoint definition is available, while its observed magnitude and precision are not.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does the outcome measure?

The endpoint uses fMRI to measure blood-oxygen-level dependent signal while participants view smoking-related and neutral cues. The analysis converts the signal into contrast of parameter estimate values. The registry says that larger COPE values indicate greater differences in brain activation between the two cue categories. This is a cue contrast, rather than a direct count of neurons or a general score of brain function.

The record specifies two a priori regions of interest. In general, specifying regions in advance focuses an analysis on selected brain areas rather than treating every possible location as equally central. The registry does not name the regions, describe how they were selected, or report their individual estimates. It says the COPE values were extracted for comparison between groups.

What does the mean represent?

The outcome parameter is identified as a mean. In general, a group mean is a summary of the measured values for participants included in that analysis. It gives a central estimate, but does not show how variable the individual measurements were or how uncertain the group estimate is. The registry identifies standard error as the dispersion measure. A standard error describes uncertainty in an estimated mean; it is not the same thing as the spread of individual participant values.

The record does not provide the actual means or standard errors. It therefore is not possible to describe whether either expectancy condition had a higher mean COPE, or how precisely either mean was estimated. It also does not report a numerical between-group contrast. A numerical difference or a standardized effect should not be inferred from the endpoint description.

How should uncertainty be understood?

In general, a confidence interval expresses a range of effect estimates that are compatible with the data under the assumptions of the analysis. A narrow interval indicates greater precision than a wide interval, while neither interval alone establishes whether an effect is important. The registry record does not report a confidence interval for this endpoint. It also does not give a p-value or name the statistical test used for the primary comparison. No inferential result can therefore be summarized from the registry description.

General statistical interpretation also distinguishes evidence about a difference from evidence about its practical meaning. A small p-value, if one were reported, would not by itself describe the magnitude or importance of a difference; a confidence interval and the study context would also matter. Conversely, a result that is not statistically distinguishable from no difference would not establish that the conditions are equivalent. The record lacks the estimates needed for either kind of interpretation.

What does randomization contribute?

In general, random assignment aims to make comparison groups similar, on average, with respect to measured and unmeasured characteristics before the intervention. That rationale can support a causal interpretation of a between-group comparison when allocation, follow-up, measurement, and analysis are appropriately conducted. The registry labels the study randomized, but does not report the sequence-generation method, allocation concealment, adherence to assigned instructions, or the analysis strategy. Those details are important for evaluating how well the design protects the comparison from bias.

The study is also listed as unmasked. This means that masking is not reported as part of the design classification. In a behavioral instruction study, participants may be aware of the instructions they receive, but the registry does not describe which people involved in measurement or analysis knew assignments. It would therefore be inappropriate to assume that outcome assessors or analysts were either masked or unmasked beyond the registry's stated classification.

How were the imaging data processed?

The registry reports preprocessing with fMRIPrep (version 20.2.7) and use of the FMRI Expert Analysis Tool from FMRIB's Software Library (FSL; version 6.0.7.4) to characterize activation differences in response to smoking-related versus neutral cues. It states that participant-level COPE values were calculated and extracted for the regions of interest. These details describe the stated processing workflow; they do not provide the analysis settings, quality-control results, or full specification of the model. Those details are not reported in the registry record.

In general, imaging preprocessing is used to prepare collected signals for analysis, while a model estimates responses associated with task conditions. Choices made in processing and modeling can affect estimates, so reproducibility depends on sufficiently detailed reporting. The named software and versions identify part of the workflow, but do not establish that all relevant analytic decisions are documented in the registry.

What about over-dispersion?

Over-dispersion is a general issue for count outcomes when observed variability exceeds what a chosen count model assumes. It can lead to uncertainty estimates that are too small if not addressed. The primary endpoint is described as a continuous mean of imaging contrasts, not a count outcome. The registry does not report a count model for this endpoint, so over-dispersion is not a stated feature of its analysis.

Can planning calculations be interpreted as power?

General planning calculations based on an observed effect are illustrative rather than a report of the study's design justification. They do not represent observed power, and they cannot replace the actual estimate and uncertainty from the trial. The registry record does not report the trial's own sample-size justification or statistical plan. Without those details and without the endpoint results, the adequacy of the planned analysis cannot be assessed from the record.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists multiple outcomes spanning brain activation, facial expression coding, self-reported urge, self-reported affect, and nicotine dependence. The study summary describes fMRI and facial coding as primary endpoints, while the outcome listing classifies coded facial expressions among secondary outcomes. The record does not explain this difference in endpoint labeling or specify which hypotheses were treated as confirmatory for analysis.

When a study evaluates multiple outcomes, regions, cues, or comparisons, the chance of observing at least one apparently unusual result can increase. General approaches to multiplicity include prespecifying a primary comparison, adjusting inference across a family of tests, or clearly labeling analyses as exploratory. The registry does not report whether any such approach was used. The number of analyses identified in the record does not reveal the adjustment method or which outcomes were included in any family of tests.

The imaging analysis population is described as participants who completed the scan session and provided usable brain-imaging data. The registry notes that the Behavioral Testing arm did not complete a scan session and was not assessed for this outcome. The number of participants with usable data, reasons data may have been unavailable, and handling of missing imaging observations are not reported in the registry record. It also does not state whether the analysis followed assigned groups regardless of receipt of the intended instructions or used another analysis population.

For outcomes involving self-report or facial coding, the record does not describe missing-data procedures, coding reliability, or how incomplete assessments were treated. It does not report whether exclusions were specified before results were examined. These omissions limit assessment of selection and analysis decisions. They do not establish that missingness or multiplicity caused bias; they mean that the relevant methods are not reported in the registry 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 reports no serious adverse events, no other adverse events, and no deaths in any of the listed groups. For Expect-Yes, the listed at-risk count is 14 and the reported counts are zero for serious adverse events, other adverse events, and deaths. For Expect-No, the at-risk count is 13, with zero in each of those event categories. For Behavioral Testing, the at-risk count is 35, again with zero in each category.

These are registry counts, not evidence that no event could occur in a broader population or under different circumstances. The record does not provide event definitions, collection procedures, observation duration for safety reporting, or detail about whether all events were actively solicited. The zero counts should therefore be read as no events recorded in the listed categories for the reported groups, not as a guarantee of zero risk.

Safety reporting is distinct from the primary brain-imaging analysis. The at-risk counts listed for adverse-event reporting should not be used as the imaging denominators: the imaging population required a completed scan and usable brain data, and its size is not reported in the registry record. The safety figures also do not show whether participants experienced changes in urge or affect, which are separate study outcomes.

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

The registry describes an experiment on smoking expectancy: verbal instructions were intended to change whether participants perceived an opportunity to smoke during the session. The primary outcome focuses on the difference in fMRI response to smoking-related and neutral cues, with COPE values extracted from two a priori regions. This gives the reader a defined experimental question and a stated measurement approach.

However, the registry record does not report the numerical primary-outcome estimates or their uncertainty. It is not possible to say whether the Expect-Yes and Expect-No conditions differed, which condition had a larger mean, or how precise any estimated contrast was. Nor can the record establish the size or practical importance of a possible difference. Describing the study's intended manipulation is not the same as reporting its effect.

The endpoint is also specific. A contrast between smoking-related and neutral cues does not by itself describe responses to every kind of cue, everyday smoking behavior, or clinical outcomes. Likewise, an fMRI contrast does not directly measure subjective urge. The registry lists self-reported urge, affect, and facial expression outcomes, but their numerical results are not reported in the registry record.

Randomization is relevant to the design, but a label of randomized does not answer every question about implementation or analysis. The registry does not report the randomization procedure, allocation concealment, detailed analysis plan, or handling of missing imaging data. It describes the analyzed imaging population as those with completed scans and usable data, while the size of that population is not stated.

The record reports zero serious adverse events, other adverse events, and deaths in each listed group. Those observations should be understood within the recorded study groups and safety reporting, not generalized as proof that such events cannot occur.

The most defensible reading is therefore descriptive: the registry documents a completed randomized study, its expectancy conditions, the intended imaging endpoint, and the stated data-processing workflow. It does not report enough outcome information to draw a result-based conclusion about the primary comparison or the secondary outcomes.

13. Why This Trial Matters Statistically

This study illustrates why a clinical-trial record should be read at more than one level. The design classification, intervention description, outcome definition, analysis population, and numerical results answer different questions. A randomized label describes the intended assignment design; it does not reveal the size or precision of an outcome estimate. A detailed endpoint definition helps clarify what was measured, but cannot stand in for the observed data.

The primary imaging measure is a contrast between responses to smoking-related and neutral cues, extracted for a priori regions of interest. That specificity matters statistically because the result concerns a defined comparison and selected brain regions, not an unrestricted claim about brain activity. The distinction between a cue contrast and a general activity level is essential to interpreting what a COPE estimate would mean.

The record also underscores the role of analysis-population reporting. Usable imaging data were required for inclusion in the primary endpoint analysis, and the Behavioral Testing arm did not undergo scanning. Without the number of usable scans or information about unavailable data, readers cannot determine the endpoint's analysis denominator or evaluate the potential effect of exclusions.

Multiplicity and endpoint classification are relevant because the study spans imaging, facial coding, and self-report outcomes. The brief summary and outcome listing describe primary status differently for facial coding. Without an analysis plan or multiplicity procedure, readers cannot establish how the investigators distinguished confirmatory findings from other analyses.

Finally, this record is a useful example of the difference between a completed study and a fully interpretable reported result. Completion status, enrollment, and named software do not disclose the estimated effect. Statistical interpretation depends on estimates, uncertainty, and transparent methods, none of which can be replaced by the study rationale alone.

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.

Several trial-specific reporting gaps constrain interpretation. The record describes the intended expectancy manipulation and the primary fMRI endpoint, but it does not provide primary-outcome estimates, standard errors, confidence intervals, or p-values. The observed direction, magnitude, and precision of the comparison therefore cannot be assessed.

The imaging analysis includes participants who completed a scan and provided usable brain data. The record does not report how many met those criteria, why data might have been unusable, or how missing observations were handled. It also does not report the randomization sequence method, allocation concealment, adherence to instructions, or detailed analysis population. These omissions make it difficult to evaluate the implementation and analysis of the randomized comparison.

The names of the two a priori regions of interest are not reported, and no region-specific results are given. The processing workflow identifies software and versions, but the record does not include all model settings, quality-control details, or a full analytic specification. The statistical analysis plan, multiplicity approach, interim analyses, and sample-size justification are not reported in the registry record.

There is also an endpoint-labeling ambiguity: the brief summary describes facial coding alongside fMRI as a primary endpoint, while the outcome listing classifies coded facial expressions among secondary outcomes. Numerical findings for the listed secondary measures are not reported. The record does not explain the difference in labeling or how confirmatory analyses were distinguished from other analyses.

These limitations concern what is documented in the registry record. They do not establish that the study was conducted improperly or that results were biased. They do mean that the record alone cannot support a detailed assessment of the primary estimate or of several choices that affect its interpretation.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Examining Brain Responses Linked to Emotion in Individuals Who Smoke Cigarettes — NCT04310735.
• Registry record last updated 2025-03-07; status last verified 2025-03.
• Results first posted 2025-03-07.
• 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

NCT04310735 is a completed randomized, unmasked, parallel-group study sponsored by Penn State University. It examined verbal instructions intended to alter whether people who smoke cigarettes perceived an opportunity to smoke during an experimental session. The registry reports enrollment of 62, a start date of 2019-10-04, and a primary completion date of 2023-06-30.

The primary outcome is brain activation during smoking-related and neutral cues, measured by fMRI and summarized as COPE values in two a priori regions of interest. The imaging analysis population consisted of participants with completed scans and usable brain data; the Behavioral Testing arm was not assessed for this outcome. The registry record does not report the numerical primary-outcome results, analysis denominator, statistical plan, or missing-data methods.

Zero serious adverse events, other adverse events, and deaths are recorded for each listed group. These safety counts are separate from the unreported imaging results.