Alcohol Use and Mental Health - Pilot Test of Video-assisted Drinking Topography: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Alcohol Use and Mental Health - Pilot Test of Video-assisted Drinking Topography (NCT03314454), a non-randomized phase 4 study. The primary outcome is Alcohol Topography: Sip Frequency.

2. Clinical Question

3. Trial Design

This was a Phase 4, non-randomized, single-masked, parallel-model study sponsored by the University of Florida. The registry lists alcohol use disorder as the condition and beer and a videotaped drinking session as interventions. Its overall purpose was to pilot a video-assisted alcohol topography procedure and explore its potential use as an indicator of alcohol use disorder.

The brief summary describes a process with phone prescreening, an in-person screening appointment, an initial alcohol drinking session with videotaping, and a follow-up appointment for retest. The primary measure's timeframe is stated as 60 days from phone screening to follow-up retest. The record lists a projected enrollment of 40 and an enrollment of 11. It states that 10 participants completed an alcohol administration laboratory session before the study ended early. The listed status is terminated.

Group descriptions include elevated mental status, non-elevated depressed mood, social drinking status, and heavy drinker status. The mental-status categories are based on higher or lower Patient Health Questionnaire scores. Social drinking is described as regular alcohol consumption with infrequent heavy drinking days; heavy drinking is described as regular consumption with frequent heavy drinking days. These labels describe classifications, not randomized assignments. The registry explicitly cautions that the arms are not a like-for-like control comparison.

The record gives a start date of 2017-10-12 and a primary completion date of 2022-12-15. It does not report the masking procedures, the details of allocation to the listed groups, or a protocol explanation for the termination.

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: As explained in recruitment details, enrollment for this pilot study ended early before social drinkers or participants with elevated mental status could be recruited, thus precluding group comparisons.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: We ended this study early, with only 10 of the projected 40 participants completing an alcohol administration laboratory session, before we could recruit social drinkers or participants with elevated mental status. For other classifications, subgroup sizes are too small for these results to have any meaning, thus we opted to present these results descriptively as a single group.

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: We ended this study early, with only 10 of the projected 40 participants completing an alcohol administration laboratory session, before we could recruit social drinkers or participants with elevated mental status. For other classifications, subgroup sizes are too small for these results to have any meaning, thus we opted to present these results descriptively as a single group.

6. Primary Endpoint

The registry lists the primary outcome as Alcohol Topography: Sip Frequency, measured over 60 days (from phone screening to follow-up retest).

The primary endpoint is mean number of sips per alcoholic drink, with the unit recorded as sips per drink and standard deviation as the dispersion measure. It is a continuous measure: each observation contributes a quantity, and the stated summary is its mean. The measure is one operational definition of alcohol topography, which concerns observable features of drinking behavior.

The study description frames this measure as a way to explore whether drinking behavior can be assessed noninvasively and whether an identified behavioral pattern could indicate alcohol use disorder. Sip frequency may describe how drinking is carried out during the laboratory session; it is not itself a diagnosis, nor does the record establish that it can distinguish diagnostic or drinking-status groups.

The stated timeframe runs from phone screening to follow-up retest. The registry record does not report the observed mean, standard deviation, number of observations contributing to this endpoint, or a comparison estimate. It also does not report how any repeated measurements were incorporated into the primary endpoint. Therefore, the measure's purpose can be described, but its observed level, variability, and precision cannot be assessed from the information recorded.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary endpoint is described as a mean of a continuous measure, with standard deviation used to describe dispersion. The registry record does not give the observed mean or standard deviation, nor does it report a statistical analysis for this endpoint. The concepts below are general explanations of how such summaries and comparisons are usually interpreted, not a description of an analysis confirmed in this study.

What does a mean describe?

For a continuous measure, a mean summarizes the central level across the observations included in its calculation. In this study, the measure is the number of sips per alcoholic drink. A mean can be useful for describing a group's typical level in an arithmetic sense, but it does not show the full distribution. Participants may have similar measurements or widely varying ones while sharing the same mean.

The standard deviation describes the spread of measurements around the mean. A smaller standard deviation indicates measurements more concentrated near the mean; a larger one indicates greater spread. It is not a measure of uncertainty in the mean. Uncertainty in an estimated mean is commonly expressed through a confidence interval, which gives a range of values compatible with the data and the statistical method used. The registry record does not report a confidence interval for the primary endpoint.

How should group contrasts be read?

A group contrast compares an outcome summary between groups. In a randomized study, random allocation can help balance both measured and unmeasured characteristics on average, supporting a causal interpretation under appropriate conditions. This study is listed as non-randomized, and the registry warns that its arms are not like-for-like control groups. A contrast between categories would therefore need to be interpreted cautiously: it may reflect differences in group composition as well as differences in the measured drinking behavior.

Adjustment can sometimes account for measured characteristics, but it cannot automatically remove bias from unmeasured factors or a poorly supported model. The registry record does not report an adjustment method, covariates, model form, or group-specific outcome estimates. It is therefore not possible to determine from the record whether any potential confounding was addressed.

Why does a small pilot require caution?

Small study groups generally provide limited information about the range of possible outcomes and make estimates more sensitive to individual observations. Subgroup analyses are especially difficult to interpret when the groups are small: estimates can be unstable, and uncertainty may be broad. The registry itself says subgroup sizes were too small for results in other classifications to have any, and it notes that recruitment did not reach social drinkers or participants with elevated mental status. These limitations are directly relevant to evaluating any attempted comparison across the planned classifications.

A confidence interval is useful because it communicates precision as well as an estimate. A wide interval indicates that the data are compatible with a broad range of values; a narrow interval indicates greater precision under the model's assumptions. An interval alone does not establish that a result is important, unbiased, or clinically meaningful. Since no primary estimate or interval is reported in the registry record, none of these properties can be assessed for sip frequency.

What do planning illustrations mean?

A sample-size planning illustration asks how many participants per arm might be needed under stated assumptions to detect a chosen effect with a chosen statistical procedure. The record describes its illustration as using a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation, with the effect to detect expressed as a fraction of the observed effect. It also specifies that the tables for non-time-to-event outcomes are in participants per arm. Those assumptions describe a planning exercise, not the trial's own design or an observed analysis.

Such an illustration depends on the effect used as its reference and on the assumptions made about the future study. It does not show that the pilot had a particular probability of detecting an effect. Observed power is not a substitute for reporting an estimate and its uncertainty. In this case, the record does not report the primary endpoint estimate needed to evaluate the stated effect reference, nor does it provide numerical planning results in the information recorded for the endpoint. Any planning figures should therefore be read as hypothetical design guidance, not as evidence that an effect was present or absent.

The registry gives no analysis entries for this record. It does not report a statistical model, test, confidence-interval method, treatment of repeated measurements, or handling of outliers. These details are not reported in the registry record, so no specific analytic procedure should be attributed to the 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("Study Sample")
n   <- c(10)
m   <- c(14.67)
sp  <- c(7.57)
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  Study Sample  10  14.670  7.570  9.255  20.085

11. Multiplicity, Missing Data and Analysis Populations

The registry record lists four outcome measures and zero analysis entries. It does not report a multiplicity strategy, such as whether one outcome was designated as the sole confirmatory test, whether other outcomes were exploratory, or whether any adjustment was planned for testing multiple outcomes. Without that information and without reported analysis results, the risk of interpreting chance findings across multiple outcomes cannot be evaluated. The primary endpoint is identified in the record, but that designation alone does not establish how multiple testing was handled.

The record describes a follow-up retest and gives a timeframe of 60 days from phone screening to follow-up retest. It does not report the number of participants with endpoint data at each assessment, missing-data reasons, or a method for handling missing observations. It also does not state whether all participants who entered the study were included in an analysis population or whether any eligibility, completion, or adherence rules were applied. These details are not reported in the registry record.

The enrollment figure is 11, while the record says 10 of the projected 40 participants completed an alcohol administration laboratory session. These figures describe different registry entries and should not be treated as interchangeable analysis denominators. The record does not provide an endpoint-specific denominator. It also states that recruitment ended before social drinkers and participants with elevated mental status were recruited, limiting the completeness of the intended group coverage.

Missing outcomes and incomplete recruitment can affect both the precision and the representativeness of results. If missingness is related to drinking behavior or group membership, complete-case summaries could differ from summaries for the intended population. This is a general concern, not evidence that a particular missing-data mechanism occurred in this study. The registry record does not report enough information to assess it.

12. 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 serious adverse event record lists the Study Sample group with 10 at risk and zero affected. It also lists 10 at risk and zero affected for other adverse events, with zero deaths. No serious adverse event terms are listed.

These figures describe what is recorded for the listed group; they are not an estimate of comparative safety between randomized arms. This study was non-randomized, and the record does not provide a control-group safety comparison. The number at risk in the adverse-event listing should not automatically be treated as the denominator for every outcome or as the number of participants completing the primary endpoint.

A zero count means that no events in the relevant category are recorded in this registry entry. It does not establish that the underlying risk is zero, particularly when the study is described as a small pilot that ended early. The registry record does not give event definitions, monitoring procedures, exposure duration for safety, or a detailed account of event collection. Those details are not reported in the registry record, so the safety figures should be read as limited descriptive reporting rather than a precise estimate of risk.

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

This study's stated aim was to pilot a video-assisted procedure for measuring alcohol topography and to explore whether a behavioral pattern might indicate alcohol use disorder. Its primary measure was mean sips per alcoholic drink. The record describes a screening and laboratory-session process that included videotaping and a follow-up retest, and it records the study as terminated.

The registry does not report the primary endpoint's observed mean, standard deviation, group contrast, confidence interval, or test result. It therefore does not establish what sip frequency was observed, whether groups differed, or how precisely any difference was estimated. It also does not demonstrate that sip frequency is a valid indicator of alcohol use disorder. Establishing an indicator would require evidence about the measure's performance and relevance, which is not reported in this record.

The study ended before the planned recruitment was complete. The record says 10 of the projected 40 participants completed an alcohol administration laboratory session and that recruitment had not reached social drinkers or participants with elevated mental status. Subgroup sizes for other classifications were described as too small for results to have any. These limitations restrict the range of conclusions that can be drawn about the intended group classifications.

Because allocation was non-randomized, observed differences between groups, if reported elsewhere, would not by themselves establish that group classification caused a difference in drinking behavior. Differences could be associated with other characteristics, and the registry does not describe an adjustment plan. The record also contains no analysis entries, no reported missing-data method, and no multiplicity strategy.

Thus, the registry supports a description of the pilot's objective, design, early termination, and limited safety counts. It does not support a conclusion about the magnitude, precision, or statistical significance of the primary outcome, nor a claim that the method can identify alcohol use disorder. Lack of a reported result is not evidence that the measure showed no pattern; it means the registry record does not report that result.

14. Why This Trial Matters Statistically

Pilot studies can be important for assessing whether a measurement procedure is feasible and for learning how an outcome is defined and collected. This study's focus was video-assisted alcohol topography, with mean sips per alcoholic drink as one operational measure. Recording the intended measurement and its timeframe helps clarify what later studies might need to evaluate.

The statistical context matters because a measurement procedure and its clinical interpretation are separate questions. A measure can be recorded consistently yet still require evidence that it captures a meaningful behavior or distinguishes groups of interest. The registry describes an exploratory aim, not a validation result. It reports no primary outcome estimate or analysis result, so the statistical evidence for the measure's utility cannot be judged from this record.

The early termination and incomplete recruitment also matter for study planning. The intended classifications included social and heavy drinking status and elevated or non-elevated mental status, but the record says recruitment did not reach social drinkers or participants with elevated mental status. It further warns that other subgroup sizes were too small for results to have any. A pilot with limited group coverage cannot provide a stable basis for broad comparisons across those classifications.

Finally, the non-randomized design affects what comparisons can mean. Differences between groups may describe patterns in the participants observed, but they do not isolate the effect of an assigned intervention. The record's design information and its explicit caution about unlike groups are essential to prevent descriptive comparisons from being mistaken for causal evidence. Statistical interpretation must account for what was measured, who was included, how groups were formed, and what results were actually reported.

15. 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 records a projected enrollment of 40 and enrollment of 11, and states that 10 completed an alcohol administration laboratory session before the study ended early. It also says recruitment ended before social drinkers or participants with elevated mental status could be recruited. These features limit the extent to which the study can describe the intended classifications. The record explicitly states that subgroup sizes for other classifications were too small for those results to have any.

Allocation was non-randomized, and the registry cautions against treating the arms as like-for-like controls. Consequently, between-group differences would be vulnerable to differences in participant characteristics and other sources of confounding. The registry record does not report the method used to form groups or any statistical adjustment for group differences.

The primary endpoint is clearly named, but its observed value and uncertainty are not reported in the registry information. The record lists no analysis entries. It does not state how retest measurements were handled, give endpoint-specific denominators, describe missing observations, or explain whether any analysis population was defined. The missing-data and multiplicity approaches are also not reported.

The registry describes the study as a pilot exploring an indicator, but does not report validation metrics or evidence that the measure can classify alcohol use disorder. The record also does not explain the reason for termination, the masking procedures, or the sample-size rationale for the original plan. These details are not reported in the registry record, so no assumptions about them are warranted.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Alcohol Use and Mental Health - Pilot Test of Video-assisted Drinking Topography — NCT03314454.
• Registry record last updated 2024-12-27; status last verified 2024-12.
• Results first posted 2024-12-27.
• 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.

19. Record Summary

The University of Florida study, NCT03314454, was a Phase 4, non-randomized, single-masked, parallel-model pilot of video-assisted alcohol drinking topography. It listed alcohol use disorder as the condition and beer and a videotaped drinking session as interventions. The primary endpoint was mean sips per alcoholic drink, with standard deviation as the stated measure of dispersion.

The registry lists a projected enrollment of 40 and enrollment of 11. It says 10 participants completed an alcohol administration laboratory session before the study ended early, without recruitment of social drinkers or participants with elevated mental status. The study is listed as terminated. The registry record reports zero analysis entries and does not report a numerical primary-outcome result or its precision. Serious adverse events, other adverse events, and deaths are recorded as zero in the listed safety group, with 10 at risk for each event category.