The Use of Public Messaging for the Promotion of Firearm Safety to Veterans: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for The Use of Public Messaging for the Promotion of Firearm Safety to Veterans (NCT03417752), a randomized na study. The primary outcome is Number of Participants With Self-reported Injury Prevention Belief.

2. Clinical Question

The registry describes a randomized, double-masked, parallel study and lists the condition as injuries. Its stated purpose is to inform injury-prevention outreach aimed at Veteran populations. The principal interpretive question is therefore about the relationship between assigned messaging and reported beliefs or storage practices, not about clinical injury outcomes. The record does not report whether injuries were measured as trial outcomes, nor does it specify how the study’s stated aim of examining changes maps to the listed baseline primary measure.

3. Trial Design

The registry record describes a randomized, double-masked, parallel-group study with enrollment of 474. Participants were assigned to one of three listed messaging conditions: repeated firearm safety PSA exposure, a mix of general health promotion and firearm safety PSAs, or repeated exposure to a general health promotion video as an active control. Random assignment supports comparison of assigned groups, but the registry record does not report the allocation ratio, method used to generate the random sequence, or the people who were masked.

The exposure schedules differ in both message content and timing. The firearm safety PSA group received that message once per week for 3 weeks. The mixed-message group received the general health promotion video 1 week after randomization, then the firearm safety PSA at 2- and 3-weeks post-randomization. The active control received the general health promotion video once per week for 3 weeks. These descriptions establish the planned contrasts but do not report whether participants adhered to each schedule or how exposure was verified.

The registry lists a start date of 2015-12, primary completion of 2020-03, and a completed status. The phase is listed as NA. The trial is sponsored by Canandaigua VA Medical Center. Detailed eligibility criteria, sites, recruitment methods, and the timing of exit assessments are not reported in the registry record.

Randomization and masking describe design features, not the outcome findings. The registry record reports descriptive summaries by frequencies and weighted percentages or mean and standard deviation, as appropriate, for the primary measure description. It does not report a model, a prespecified contrast across the messaging groups, or inferential results. It also does not state whether the descriptive approach applied to every outcome.

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: Analyses were restricted to participants that completed both baseline and exit surveys. Descriptive statistics were tabulated by frequencies and weighted percentages or mean and standard deviation, as appropriate.

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: analysis restricted to participants with baseline and exit responses.

6. Primary Endpoint

The registry lists the primary outcome as Number of Participants With Self-reported Injury Prevention Belief, measured over baseline.

The listed primary endpoint is the number of participants with a self-reported injury-prevention belief at baseline. Specifically, it counts participants who responded to the statement that “safe storage of household firearms is effective for reducing injury or death.” The unit is participants, and the measure is described as a count rather than a continuous score.

Beliefs are relevant to a public messaging study because a PSA is intended to communicate information that could be associated with how people understand safe storage. In general, a baseline measure can characterize participants’ starting responses and help describe the study population. It cannot, by itself, show a change following exposure, because a change requires comparing measurements across time or otherwise defining a change outcome.

The study summary describes an aim to determine whether PSA exposure is associated with changes in related beliefs and safe storage practices. However, the listed primary measure is explicitly at baseline. The registry record does not explain how this baseline measure relates to a change-based primary question, report response categories, or provide the baseline count or weighted percentage. It also does not state a threshold for what would constitute a meaningful difference.

The record lists self-reported firearm storage behavior at baseline and exit among the secondary outcomes. Self-report can describe participants’ stated behavior, but it is not the same as an independently verified storage measure. The registry record does not report the wording of those items, their response options, or outcome results.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does the stated descriptive approach mean?

The registry says the primary-outcome description used frequencies and weighted percentages or mean and standard deviation, as appropriate. A frequency is a count of observations in a category. A percentage expresses the share of the analysis population represented by a category. Weighting can be used to make a descriptive summary reflect assigned weights rather than treating every observation as having identical influence. The record does not explain whether weights were used for this particular binary belief measure, how any weights were constructed, or what summaries were used for each secondary outcome.

A mean describes a center of numerical observations, while a standard deviation describes their spread around the mean. Those summaries are generally useful for numerical measures, but the primary endpoint is described as a participant count for a self-reported belief response. The registry’s phrase “as appropriate” does not identify which outcomes were summarized with which statistics.

What does the analysis population statement establish?

The primary outcome description says analyses were restricted to participants who completed both baseline and exit surveys. This identifies a complete-survey population for those analyses. It does not state that every randomized participant was included, and it does not report the size or characteristics of the complete-survey group. In general, restricting analysis to people with complete follow-up can change the composition of the analyzed population if completion is related to group assignment or outcome responses. Whether that occurred cannot be assessed from this record.

Missing-data methods describe how an analysis handles measurements that are unavailable. For example, an analysis may use available observations, make assumptions about missing values, or use a model that incorporates partial follow-up. The registry record does not report a missing-data method, reasons for incomplete surveys, or any sensitivity analysis. It therefore does not establish whether missing responses affected the estimates.

Why does a baseline endpoint need careful interpretation?

Random assignment occurs before the planned exposure, while the listed primary measure is explicitly at baseline. In general, baseline summaries describe starting characteristics or responses; they do not estimate a post-exposure effect unless the analysis is framed around a later outcome or a specified change. A change analysis would need to define which time points and responses are compared and how the groups enter that comparison. That analysis is not described in the registry record.

The study summary refers to changes in beliefs and storage practices, and the secondary outcomes name baseline and exit storage behavior. However, the record does not provide a defined change score, a repeated-measures model, or a between-group contrast. It would be inappropriate to infer a particular analysis from the study aim alone.

What is not established by the design labels?

Randomized allocation means participants were assigned to conditions by chance, according to the registry description. In general, this helps protect group comparisons from systematic differences that existed before assignment. It does not guarantee identical groups in the realized study, ensure adherence to assigned messages, or address bias from incomplete outcome data. The registry does not report the randomization procedure, allocation concealment, adherence, or an analysis accounting for deviations from assignment.

Double masking is the registry’s design label. Masking can reduce the influence of expectations on behavior or assessment, depending on who is masked and how. The record does not identify who was masked or how masking could be maintained while participants viewed different videos. No further detail should be inferred from the label alone.

How should planning calculations be read?

The registry’s sample-size description characterizes its calculations as planning illustrations based on an observed effect, not as the trial’s own design. It states a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with the effect to detect defined as a stated fraction of the observed effect. For time-to-event calculations, the description says the total number of events is used; other calculations are expressed as participants per arm.

In general, a planning calculation estimates the sample size associated with assumptions about an effect, variability, significance level, and desired power. It is not observed power, and it does not show that the trial achieved a particular chance of detecting an effect. The stated two-arm assumption is a planning assumption and should not be confused with the randomized study, which lists three messaging conditions. The registry record does not report a sample-size justification for the trial, the targeted effect, or a prespecified power calculation. No planning value or observed effect estimate is reported in the registry record, so no numerical sample-size interpretation is possible.

What can inferential statistics add?

In general, an effect estimate describes the size and direction of a contrast, while a confidence interval conveys the range of values compatible with the data and analysis assumptions. A p-value evaluates how unusual the data would be under a specified null hypothesis and model; it does not give the probability that the null hypothesis is true or measure practical importance. None of these inferential results is reported for the listed outcomes. The registry record therefore does not permit assessment of effect magnitude, uncertainty, or statistical compatibility with a null comparison.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists four outcome measures and zero analyses. The primary measure concerns a baseline belief about safe firearm storage, and the secondary outcomes concern self-reported firearm storage behavior at baseline and exit. The record does not provide results for these outcomes or specify a hierarchy of hypotheses, a family of confirmatory comparisons, or a multiplicity adjustment.

Multiplicity arises when several outcomes, time points, or group comparisons are examined. In general, examining multiple comparisons can increase the chance of at least one apparently noteworthy finding by chance. A prespecified hierarchy or adjustment can clarify how evidence is evaluated, but no such plan is reported in the registry record. It is also not reported whether the study used interim analyses, or whether any were planned.

The primary outcome description says analysis was restricted to participants who completed both baseline and exit surveys. The number of completers, the amount of missing outcome data, reasons for missingness, and the method used to handle missing values are not reported in the registry record. The record does not state whether analyses followed assignment regardless of exposure adherence or used another analysis population.

These omissions limit interpretation of both precision and risk of bias. Without outcome counts and denominators, readers cannot assess the completeness of the outcome data. Without a reported analysis plan, readers cannot determine which contrasts were intended as primary or how multiple outcomes were handled. The record also does not report whether the posted analysis population matches an intention-to-treat approach.

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 lists serious adverse events as 0 among 115 participants at risk in the Exposure to Firearm Safety PSA group, 0 among 133 at risk in the Exposure to a Mix of PSAs group, and 0 among 110 at risk in the Active Control group. The death count is 0 in each group. The other adverse-event count is also 0 in each listed group, with the same group-specific at-risk counts reported.

These figures are event counts in the registry record. They do not provide a comparative risk estimate or confidence interval, and they do not show whether the groups differed in event risk. The registry lists no top serious-event terms. It does not report event definitions, how events were sought, the observation window, or whether all randomized participants contributed to safety monitoring. Consequently, the zero counts should be interpreted as the recorded entries, not as evidence that the study established the absence of adverse events.

The at-risk figures are reported separately for each group; they should not be combined to create a new total. The record does not explain any difference between safety populations and the survey analysis population. These safety entries also do not substitute for outcome results on beliefs or firearm storage behavior.

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

This registry record describes a randomized evaluation of public messaging about firearm safety. It identifies the message schedules, an active video control, the stated interest in beliefs and storage practices, and a baseline belief measure. These details clarify what the study was designed to examine, but they are not evidence of an observed effect.

The record does not report primary or secondary outcome counts, effect estimates, confidence intervals, or p-values. It therefore does not show whether exposure to the firearm safety PSA was associated with changes in self-reported beliefs or storage behavior. The absence of results in the registry record should not be interpreted as a null finding. It is also not possible to assess the precision or practical importance of any group difference.

The listed primary measure is at baseline, while the stated study aim concerns changes following message exposure. Without a reported post-exposure primary comparison or change analysis, the record does not establish how the primary measure addresses the study aim. The secondary outcome descriptions name storage behavior at baseline and exit, but no associated counts or analyses are reported.

Even if outcome results were available, self-reported beliefs and behavior would answer questions about participants’ stated responses, not independently verified firearm storage or injury outcomes. The registry does not list injury incidence as an outcome measure in the information. No conclusion about preventing injuries can be drawn from the available outcome descriptions.

The design labels and safety entries also need boundaries. Randomization is a design feature, not a result, and the record does not report adherence, masking implementation, or missing-data handling. The adverse-event entries list zero events, but do not include definitions or ascertainment details. They cannot support a broad claim about safety beyond the events captured in those fields.

13. Why This Trial Matters Statistically

The trial raises a useful statistical distinction between a communication intervention and the outcomes used to assess it. A public service announcement may be intended to influence beliefs or behavior, but the statistical question depends on the specific measurement, its timing, and the planned comparison. Here, the registry states an aim involving changes, while the listed primary measure is a baseline belief count. That mismatch makes clear why outcome timing and estimand definitions matter before interpreting a result.

The active control is also relevant to interpretation. Comparisons with a general health promotion video address a different question from comparisons with no video exposure: they concern the relative response to firearm safety messaging versus another form of health messaging. The mixed-message schedule adds a distinct sequence of exposure. Because the registry does not report the prespecified contrasts, the intended interpretation of comparisons among these conditions remains uncertain.

Complete-survey restriction illustrates another common issue in clinical and behavioral research. A randomized design can support fair comparisons at assignment, but a complete-case analysis may rely on a subset whose inclusion is related to study participation or outcomes. The record does not describe missingness or its handling, so the potential effect of that restriction cannot be evaluated.

Finally, statistical evidence requires results as well as a design description. Counts, denominators, effect estimates, and uncertainty measures are needed to distinguish a potentially important difference from a small or imprecise one. This record supplies the context for those questions but does not report the outcome analyses needed to answer them.

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 central limitation for interpretation is that outcome results are not reported in the registry record. There are no belief or storage-behavior counts, group comparisons, effect estimates, confidence intervals, or p-values. As a result, the magnitude and direction of any association cannot be assessed.

The listed primary outcome is a baseline belief measure, despite the study summary describing interest in changes. The registry does not explain whether this reflects the intended primary analysis, a reporting discrepancy, or a separate descriptive measure. It does not provide the belief item’s response options or specify what response counted as reporting the belief.

The registry describes analysis restricted to participants who completed both baseline and exit surveys but does not report the number included, missing-data patterns, or how missing responses were handled. The allocation ratio, randomization procedure, masking roles, adherence, and protocol deviations are also not reported. These gaps limit assessment of how closely the observed study conduct can be connected to the randomized design.

Although the record lists self-reported storage behavior at baseline and exit, it does not provide the exact assessment wording, results, or a definition of change. It does not list injury outcomes among the reported measures, so the record cannot establish effects on injury occurrence. The safety listings lack event definitions, ascertainment methods, and an observation window. Together, these limitations restrict interpretation to the trial description and the event entries, rather than evidence of effectiveness or a comprehensive safety assessment.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: The Use of Public Messaging for the Promotion of Firearm Safety to Veterans — NCT03417752.
• Registry record last updated 2020-08-19; status last verified 2020-08.
• Results first posted 2020-08-19.
• 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

NCT03417752 is a completed, randomized, double-masked, parallel study sponsored by Canandaigua VA Medical Center. The registry lists enrollment of 474, a start date of 2015-12, and primary completion of 2020-03. Participants were assigned to firearm safety PSA exposure, a mix of general health promotion and firearm safety PSAs, or an active control using a general health promotion video.

The study aimed to examine whether firearm safety messaging was associated with beliefs and safe storage practices. The listed primary measure is a baseline count of participants reporting a belief about safe storage; secondary outcomes concern self-reported storage behavior at baseline and exit. The registry reports a descriptive approach and says primary analyses were restricted to participants who completed both surveys. It lists zero analyses and does not report outcome results. Serious adverse events, other adverse events, and deaths are recorded as zero in each group, with no event terms listed.