Brief Interventions in the Emergency Department for Alcohol and HIV/Sexual Risk: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Brief Interventions in the Emergency Department for Alcohol and HIV/Sexual Risk (NCT01351389), a randomized na study. The primary outcome is Number of Alcoholic Drinks Consumed Per Week.

2. Clinical Question

3. Trial Design

This was a randomized, parallel-group behavioral trial comparing Brief Motivational Intervention (BMI) with Brief Advice. The registry lists masking as none, so the trial was unmasked. It records enrollment of 400 and lists Brown University as sponsor. The study status is completed. The recorded start was 2011-05, and primary completion was 2013-11. The phase is listed as NA.

The two intervention groups are named in the record, but their arm descriptions are blank. Consequently, the registry information does not explain the practical differences between BMI and Brief Advice, such as their content, duration, or delivery. Those details are not reported in the registry record. The study summary describes the intended setting as an Emergency Department and frames the interventions around alcohol use and sexual risk taking.

The primary outcome was a continuous measure: number of alcoholic drinks consumed per week. Its description specifies standard alcoholic drinks consumed in the past month, with assessment at 3 month. The arm summaries report a mean and standard deviation for each group. The BMI summary includes 141 participants, with a mean of 9.1 and standard deviation of 11.5. The Brief Advice summary includes 161 participants, with a mean of 16.5 and standard deviation of 28.1.

The primary outcome description notes that 13 participants in BMI and 12 in Brief Advice did not complete the assessment. The registry record does not describe how missing outcome values were handled in the comparison, or which analysis population was used. It also does not report the trial's original sample-size justification, detailed statistical analysis plan, or interim analyses. These omissions limit what can be concluded about the planned and implemented analysis beyond the posted summaries.

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: 400 patients consented, 23 were lost prior to randomization, 5 withdrew from study prior to randomization, 45 did not completed the interventions

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: 13 and 12 participants did not complete the assessment in the BMI and BA groups respectively

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 Alcoholic Drinks Consumed Per Week, measured over 3 month.

The primary endpoint is the number of standard alcoholic drinks consumed in the past month, reported in drinks per week and assessed at 3 month. It is a continuous outcome, summarized by a mean and standard deviation for each randomized group. The endpoint therefore describes average reported consumption in the BMI and Brief Advice groups at the stated assessment, rather than a count of participants who crossed a threshold or experienced a particular event.

Alcohol consumption is relevant to the trial's stated focus on brief interventions for alcohol and sexual risk in emergency-department care. Measuring consumption numerically allows the registry to compare the group averages, while the standard deviation describes how spread out individual values were around each group mean. The posted BMI mean was 9.1 drinks per week, with standard deviation 11.5; the Brief Advice mean was 16.5, with standard deviation 28.1.

The outcome description identifies consumption during the past month, but the assessment timeframe is recorded as 3 month. The record does not provide additional detail about the instrument used to collect drinking reports or the exact assessment procedures. The outcome captures the reported consumption measure at that follow-up point; it does not directly measure every aspect of alcohol use or sexual risk taking. Although the trial summary mentions both topics, the primary endpoint described in the posted result is alcohol consumption.

7. Results Posted in the Registry

• Brief Advice: mean 16.50 versus 9.10 on control; difference 7.40 (95% CI 2.66 to 12.14), standardized effect d = 0.34; z-test p = 0.002.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary endpoint is a continuous variable: a numerical measure of alcohol consumption. For a continuous outcome, a group mean describes the average value among the participants included in that arm's summary. The posted mean is 9.1 drinks per week for BMI and 16.5 for Brief Advice. The standard deviations are 11.5 and 28.1, respectively. A standard deviation describes how dispersed individual measurements are around the mean. It is not the same as a confidence interval, and it does not indicate the precision of the estimated difference by itself.

What does the mean difference represent?

The comparison reports a mean difference of 7.4, defined in the direction Brief Advice minus BMI. This direction matters: a positive estimate means the reported mean is higher for Brief Advice than for BMI. The estimate expresses the gap between the two group means on the endpoint's scale, drinks per week. It is an aggregate contrast, not a statement that every participant in one group consumed more than every participant in the other.

For randomized parallel groups, a between-group comparison is generally used to assess whether the outcome summaries differ between assigned groups. General statistical interpretation: random allocation is intended to make groups comparable on average, including with respect to factors that may not have been measured. It does not guarantee identical groups in a particular trial, nor does the registry result describe the balance of participant characteristics. The record does not state whether the posted comparison used an adjusted or unadjusted model beyond labeling its p-value unadjusted; the detailed test and model are not reported in the registry record.

How should the p-value be read?

The reported unadjusted p-value is 0.002. In general, a p-value compares the observed result with what would be expected under a specified null model, assuming the model's conditions hold. A smaller p-value indicates that the observed result is less compatible with that null model, but it does not quantify the size of the difference, the probability that a hypothesis is true, or the likelihood that the result will reproduce. It also does not address whether the endpoint is important to participants or whether the analysis was affected by missing data or multiplicity.

The word unadjusted matters because it identifies the p-value as not adjusted for multiplicity in the posted comparison. The registry record lists six outcome measures and zero posted secondary outcome results, but it does not report a multiplicity plan or identify how outcomes were prioritized beyond the stated primary measure. Without that plan, the consequences of testing multiple outcomes cannot be fully evaluated from the posted information.

Why does variability matter?

General statistical explanation: when measurements vary widely within groups, it can be harder to estimate a group mean difference precisely, unless the sample contains enough information to offset that variability. Here, the Brief Advice standard deviation is 28.1, larger than the BMI standard deviation of 11.5. The comparison's confidence interval accounts for uncertainty in the difference under the calculation used for the posted result. The registry does not state the exact interval method. The unequal spreads and the outcome distribution cannot be evaluated fully from means and standard deviations alone.

Alcohol consumption data can, in general, be unevenly distributed, with many lower values and some much higher values. This is a general possibility for consumption measures, not a documented feature of this trial's participant-level data. If a continuous outcome has a distribution with unusually large values or unequal variance, the choice of model and the robustness of standard methods may matter. The registry record does not report distributional diagnostics, transformations, robust methods, or sensitivity analyses, so no conclusion about those choices is possible.

What does the confidence interval add?

The reported interval for the mean difference extends from 2.663 to 12.14. General explanation: a confidence interval conveys the precision of an estimate under the method used to construct it. A narrower interval indicates less uncertainty on the outcome scale than a wider interval, while the interval's location indicates which difference values are compatible with the analysis. It is not a guarantee that the true value lies within the interval, and its coverage interpretation depends on the method and assumptions. The registry record does not name the confidence-interval method.

The interval is expressed in drinks per week, the same scale as the difference estimate. That preserves a direct interpretation of the size of the contrast. It should be considered alongside the group means, standard deviations, incomplete assessments, and the record's limited description of analysis methods. The posted summary does not provide participant-level data, baseline measurements, or additional estimates that would permit a more detailed reanalysis.

What can planning calculations tell readers?

Sample-size planning calculations use assumptions about a target effect and a chosen power and significance level to illustrate how many participants might be required for a future study. The registry record's planning illustration uses the observed effect as its reference and displays scenarios for 100%, 75%, and 50% of that effect. It assumes a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation. The corresponding planning values are described in the sample-size reading. These calculations are not a report of the trial's original design rationale; the original design assumptions are not reported in the registry record.

General explanation: over-dispersion means that the observed variability is greater than a model assumes. If a planning or analysis method understates variability, it can overstate precision or understate the number of participants needed. The posted record does not identify an over-dispersion assessment or a model that accounts for it. The standard deviations are descriptive summaries, not evidence that a particular variance model was used. Thus, planning figures should be treated as illustrations tied to their stated assumptions, not as a reconstruction of the trial's protocol.

10. Confidence Intervals and Precision

The reported mean difference is 7.4 drinks per week, with a confidence interval from 2.663 to 12.14. The interval is on the same scale as the outcome, so it can be read as a range of plausible values for the difference in group means under the method used. Its lower and upper limits are both positive, consistent with the posted direction of Brief Advice minus BMI. The interval does not identify a single exact value for the underlying difference.

Confidence intervals describe uncertainty in an estimate. In general, a narrower interval indicates greater precision, while a wider interval indicates less precision. The interval reflects more than the difference between the two posted means: sample sizes and variation in the outcome contribute to the uncertainty. The group summaries report standard deviations of 11.5 for BMI and 28.1 for Brief Advice. The greater spread in Brief Advice is relevant context when considering precision, but it does not alone determine the interval's width.

The confidence level and calculation method are not reported in the registry record. No method should therefore be assumed when interpreting its exact coverage properties. The interval also does not account automatically for every possible source of uncertainty, such as missing outcome assessments, measurement limitations, or multiple outcome testing. The primary outcome description records 13 incomplete assessments in BMI and 12 in Brief Advice, but the method used to address missing values is not reported.

The p-value of 0.002 is a separate summary of compatibility with a null model, not a replacement for the interval. Together, the estimate and interval communicate direction, scale, and uncertainty more directly than a significance label alone. They do not establish the importance of the difference, provide an individual-level prediction, or settle questions about unreported analysis choices.

11. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The planning figures are participants per arm for this continuous outcome, not total event counts. Total event counts are used for time-to-event planning displays; that does not apply to this outcome. The calculations use a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect assumed for planning is a stated fraction of the observed effect.

For planning for 100% of the observed effect, the illustration gives 139 participants per arm for 80% power and 186 participants per arm for 90% power. For planning for 75% of the observed effect, it gives 247 participants per arm for 80% power and 330 participants per arm for 90% power. For planning for 50% of the observed effect, it gives 554 participants per arm for 80% power and 742 participants per arm for 90% power.

In plain language, the 80% power column indicates the number of participants per arm estimated to give an 80% chance of detecting the specified effect under the calculation assumptions, if that effect is present. The 90% column gives the corresponding planning target for a higher chance of detecting that effect. Higher power generally requires more participants. Likewise, planning for a smaller effect requires more participants, as the listed scenarios illustrate.

These are planning illustrations based on the observed effect, not observed power and not the trial's original design calculation. The original sample-size assumptions and justification are not reported in the registry record. The illustration should therefore be read as conditional on its stated assumptions, rather than as evidence about whether the completed study was adequately designed for a different target effect.

12. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = 7.400; pooled standard deviation σ = 21.974.
2. n = 2 × (1.96 + 0.842)² × 21.974² / 7.400² = 138.4, rounded up to 139 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
delta <- 7.4                      # observed difference in means
sigma <- 21.9736660573515                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(1351389)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 139 (80% power), 186 (90% power)
 75% of observed effect: n per arm = 247 (80% power), 330 (90% power)
 50% of observed effect: n per arm = 554 (80% power), 742 (90% power)
power.t.test: 139.4 
Simulated power at n = 139 per arm: 0.801 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

13. Multiplicity, Missing Data and Analysis Populations

The registry record lists six outcome measures, but no secondary outcome results are posted. It records zero analysis entries. The primary outcome comparison is nevertheless reported with group summaries, an estimated mean difference, a confidence interval, and an unadjusted p-value. The record does not explain the relationship between the listed outcomes and the reported analysis in a detailed statistical analysis plan.

Multiplicity arises when several outcomes, time points, or comparisons are examined. General statistical explanation: conducting multiple tests can increase the chance of obtaining at least one result that appears statistically notable by chance. Methods such as prespecifying a primary endpoint or adjusting for multiple tests can address particular multiplicity concerns. The registry identifies a primary endpoint, but the analysis plan, testing hierarchy, and any multiplicity adjustment are not reported in the registry record. The posted p-value is labeled unadjusted, so it should not be read as a multiplicity-adjusted result.

The primary outcome description says that 13 BMI participants and 12 Brief Advice participants did not complete the assessment. The registry does not say whether these participants were excluded, imputed, or handled using another method. The arm summaries report 141 participants for BMI and 161 for Brief Advice, but the record does not explain the relation between these summary counts and the number of randomized or enrolled participants. No analysis-population definition is reported, such as whether the comparison followed assignment regardless of intervention received.

These details matter because missing outcomes can affect the group comparison if the participants without assessments differ systematically from those assessed. General statistical explanation: an analysis based only on observed outcomes relies on assumptions about the missingness process, while imputation and model-based approaches rely on their own assumptions. Without the analysis method and sensitivity analyses, the effect of incomplete assessments cannot be established from the registry record.

The record also does not report interim analyses or the trial's original statistical analysis plan. Those items should not be inferred from the posted result. The absence of secondary results and detailed analysis information limits assessment of selective reporting and the full set of comparisons that may have been considered.

14. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

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 summaries list 184 participants at risk in BMI and 188 at risk in Brief Advice. The registry reports 0 serious adverse events in each group. It also reports 0 other adverse events in each group. These are the posted counts for the groups and categories in the record; they are not evidence that no adverse events of any kind occurred outside the reported categories or period.

The death fields are null, so death counts are not reported in the registry record. The record lists no serious adverse event terms. The adverse-event summaries do not provide event descriptions, severity gradations beyond the serious category, timing, relatedness assessments, or methods for collecting and adjudicating events. Those details are not reported in the registry record.

Safety counts should be interpreted with the at-risk group sizes and reporting scope in mind. The at-risk counts differ from the primary outcome summary counts, and the record does not explain that difference. A zero posted count is a count in the registry summary, not a guarantee of absence of harm. These data alone do not support a broad conclusion about safety or a comparison of risks beyond the specific reported categories.

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

The posted result describes a difference in average reported alcohol consumption between the randomized groups at the primary assessment. BMI had a mean of 9.1 drinks per week, while Brief Advice had a mean of 16.5. The comparison, defined as Brief Advice minus BMI, estimated a difference of 7.4 drinks per week, with a confidence interval from 2.663 to 12.14 and an unadjusted p-value of 0.002. These values summarize the observed group comparison and its reported statistical uncertainty.

The result does not show that every participant in the groups had outcomes matching the group averages. The standard deviations, 11.5 for BMI and 28.1 for Brief Advice, indicate variation among participants. The endpoint concerns reported alcohol consumption during the past month at the 3 month assessment. It does not, by itself, summarize sexual risk behavior, long-term outcomes, or other measures named in the study's rationale.

Randomization is a strength of the design for comparing assigned groups, but the registry record is unmasked and does not describe the interventions in detail, the outcome collection instrument, or the analysis population. It also does not report the method used for missing assessments. Because 13 BMI participants and 12 Brief Advice participants did not complete the assessment, the handling of those outcomes is important to interpreting the estimate. The record does not provide enough information to resolve that issue.

A statistically notable p-value does not establish practical importance, explain the mechanism behind the difference, or predict an individual participant's outcome. Nor does the posted result alone establish whether the difference would be similar in other settings or populations. The result should be read as the reported group-level estimate for this endpoint and assessment, together with its interval and the limitations of the registry information—not as medical advice or a recommendation.

16. Why This Trial Matters Statistically

This trial offers an example of how a clinical study can report a continuous behavioral endpoint as group means, standard deviations, and a mean difference. Those summaries make the direction and scale of the comparison interpretable: the posted contrast is in drinks per week, with Brief Advice minus BMI as its direction. The confidence interval adds a measure of statistical precision that a p-value alone cannot supply.

The variability differs across the reported groups, with a larger standard deviation for Brief Advice than for BMI. That matters statistically because outcome spread is one component of uncertainty in a group comparison. Readers should distinguish the spread of individual measurements from the precision of the estimated contrast. The confidence interval concerns the latter; the standard deviations describe the former.

The record also illustrates why outcome and analysis reporting need to be read together. The outcome description records incomplete assessments in both groups, but the missing-data method is not reported. The trial lists six outcome measures while posting no secondary outcome results, and its p-value is labeled unadjusted. Without a stated multiplicity plan, analysis population, and detailed model, the reported estimate can be described but some aspects of its statistical context remain uncertain.

Finally, planning calculations derived from the observed effect are not interchangeable with the trial's original design assumptions. Their values depend on the stated assumptions and on the effect size chosen for planning. Distinguishing a planning illustration from observed power or protocol justification prevents retrospective planning numbers from being mistaken for evidence about the study's design adequacy.

17. 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 limitations are specific to the information reported for this trial. First, the primary endpoint has incomplete assessments: 13 participants in the BMI group and 12 in the Brief Advice group did not complete the assessment. The registry record does not explain how those missing values were treated or whether any sensitivity analysis was performed. The arm-summary counts are 141 and 161, while the adverse-event at-risk counts are 184 and 188; the record does not explain the differing counts or define the analysis populations.

Second, the trial was unmasked. The registry record does not describe how the self-reported or otherwise collected consumption measure was administered, so the possible influence of awareness on reporting cannot be assessed from the posted information. The arm descriptions are blank, leaving the content and delivery of BMI and Brief Advice unspecified.

Third, the record does not report baseline alcohol-consumption summaries, changes from baseline, participant characteristics, the model used for the comparison, or the confidence-interval method. It does not give a detailed analysis plan, missing-data strategy, multiplicity approach, interim-analysis plan, or original sample-size justification. These are reporting gaps in the registry information, not evidence that a particular method was or was not used outside what is recorded.

Finally, the primary endpoint is alcohol consumption at one recorded assessment timeframe. Although the study rationale also refers to sexual risk taking, the posted primary result is not a result for that behavior. No secondary outcome results are posted. These limits constrain interpretation to the reported endpoint and comparison.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Brief Interventions in the Emergency Department for Alcohol and HIV/Sexual Risk — NCT01351389.
• Registry record last updated 2025-06-19; status last verified 2025-06.
• Results first posted 2025-06-19.
• 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.

21. Record Summary

The registry record describes a completed randomized, unmasked, parallel-group behavioral trial sponsored by Brown University. It lists enrollment of 400 and compares Brief Motivational Intervention (BMI) with Brief Advice. The recorded study start was 2011-05, and primary completion was 2013-11.

The primary endpoint was number of alcoholic drinks consumed per week, described as standard drinks consumed in the past month and assessed at 3 month. The posted means were 9.1 for BMI and 16.5 for Brief Advice. The reported mean difference, defined as Brief Advice minus BMI, was 7.4, with a confidence interval from 2.663 to 12.14 and an unadjusted p-value of 0.002.

The outcome description notes 13 incomplete assessments in BMI and 12 in Brief Advice. The registry record does not report how missing outcomes were handled, the detailed statistical model, or the original design assumptions. It lists six outcome measures, with no secondary outcome results posted.