Feasibility and Effectiveness of a Specialized Brief Intervention for Hazardous Drinkers in an Emergency Department.: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Feasibility and Effectiveness of a Specialized Brief Intervention for Hazardous Drinkers in an Emergency Department. (NCT03273283), a randomized na study. The primary outcome is Proportion of Risky Drinkers Measured by AUDIT-C.

2. Clinical Question

3. Trial Design

This was a randomized, parallel-group trial with no masking, listed as completed. The registry record gives enrollment as 200. The study took place in the emergency department of a tertiary hospital, and the stated aim was to assess the feasibility and efficacy of a Screening, Brief Intervention and Referral to Treatment programme in that setting. The intervention was a brief alcohol-use conversation based on motivational techniques; the comparator was informational leaflets about alcohol use.

The primary outcome was assessed at 6 weeks. The posted outcome counts are 20 participants meeting the risky-drinking definition among 72 in the intervention group and 39 among 81 in the control group. These outcome denominators do not equal the listed enrollment. The registry record does not explain the difference, so the reasons for participants not appearing in these outcome denominators are not reported in the registry record. It also does not report whether these outcome groups represent all randomized participants with available measurements or a particular analysis population.

The record lists no masking and does not describe any steps taken to limit awareness of treatment assignment during outcome assessment. The intervention itself involved a conversation and the control involved leaflets, so the record does not establish that participants or staff were unaware of assignment. Further details about randomization procedures, allocation concealment, follow-up procedures, and protocol deviations are not reported in the registry record. The registry record lists no posted analysis entries, although it does post primary outcome results and a group comparison.

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

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: At baseline, participants' demographic and enrollment-related characteristics were compared across treatment conditions using t- and chi-square tests according to the nature of each variable.

6. Primary Endpoint

The registry lists the primary outcome as Proportion of Risky Drinkers Measured by AUDIT-C, measured over 6 weeks.

The primary endpoint is the proportion of participants classified as risky drinkers by the AUDIT-C at 6 weeks. The registry describes AUDIT-C as a three-item measure covering frequency of alcohol consumption, amount consumed on drinking days, and frequency of binge drinking. Its score ranges from 0, indicating abstinence, to 12, indicating very high alcohol use. For this outcome, a score above 6 for men or above 5 for women meets the stated threshold.

Turning the score into a binary outcome makes the result a count of participants who cross the specified threshold, rather than a comparison of the full range of AUDIT-C scores. This provides a direct summary of how many participants in each group met the registry's definition of risky drinking at the assessment time. It also means that changes within the score range that do not move a participant across the threshold are not represented in this endpoint.

The registry identifies AUDIT-C as a tool for assessing alcohol consumption and describes the threshold as the main outcome. It does not report further justification for selecting the threshold, whether the measure was administered by a blinded assessor, or how the 6-week assessment was handled when a participant did not complete it. The result therefore concerns this defined threshold at the stated follow-up, rather than every dimension of alcohol use or the longer-term course of drinking.

7. Results Posted in the Registry

• Intervention: 27.8% (20/72) versus 48.1% (39/81) on control; risk difference -20.4 percentage points (95% CI -35.4 to -5.4), risk ratio 0.58 (0.37–0.89); unadjusted two-proportion z-test p = 0.010.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary endpoint is binary: each participant is either counted as meeting the AUDIT-C risky-drinking threshold or not meeting it. The registry posts counts by group and a comparison on the risk-difference scale. The posted estimates are unadjusted calculations from the group summary statistics. A risk difference expresses the contrast between the group proportions in absolute terms. Here, the reported estimate is -0.204, with a 95% confidence interval from -0.354 to -0.054. Its negative sign indicates that the intervention-group proportion was lower in the posted comparison.

What does a confidence interval describe?

In general, a confidence interval describes the statistical uncertainty around an estimate under the assumptions of the method used to calculate it. It is not a range containing the individual participants' outcomes, and it does not account automatically for all possible sources of bias. For the risk difference, zero represents no difference between group proportions. The posted interval does not include zero. The registry record does not identify the specific interval method, so the calculation procedure cannot be assessed from this record alone.

How should the p-value be read?

The posted unadjusted p-value is 0.01. In general, a p-value describes how compatible the observed data are with a specified null model, assuming that model and the analysis assumptions are appropriate. It is not the probability that the null hypothesis is true, nor does it measure the size or practical importance of an effect. The registry record does not report a full statistical analysis plan, so the null model's details, any prespecification, and the approach to multiple testing are not available.

Why are risk ratios and odds ratios also reported?

A risk ratio compares the proportions meeting the outcome between groups. The posted estimate is 0.577, with a confidence interval from 0.373 to 0.892. An odds ratio compares odds rather than proportions; the posted estimate is 0.414, with a confidence interval from 0.211 to 0.814. Because these scales differ, a risk ratio and an odds ratio need not have the same numerical value. The registry record does not state which measure was designated as the main effect measure in an original analysis plan. These additional measures are useful descriptions of the same binary endpoint, but they do not add information about outcomes beyond the participants and assessment represented in the counts.

What does unadjusted mean in this comparison?

In general, an unadjusted comparison uses the group outcome summaries without accounting for other participant characteristics in a regression model or another covariate adjustment. The posted risk difference and p-value are unadjusted. The registry record does not report whether an adjusted analysis was performed elsewhere, which covariates might have been considered, or whether any subgroup analyses were planned. Random allocation is part of the recorded design, but it does not make details of the analysis or the completeness of follow-up unnecessary to evaluate.

What is not established by the posted method details?

The record describes randomized allocation, a parallel design, and no masking. It does not report the sequence-generation method, allocation concealment, the method used for the confidence intervals, or a detailed model specification. The registry record also does not explain how missing outcome measurements were handled. These omissions limit independent assessment of the analysis. They do not change the numerical values of the posted comparison, but they leave uncertainty about the procedures underlying it.

General statistical interpretation also distinguishes the estimate from the design and conduct of a trial. A confidence interval and p-value quantify sampling uncertainty under a statistical method; they do not correct for missing data, bias in measurement, deviations from randomization, or selective reporting. The record's posted comparison is therefore one part of the evidence needed to understand the result, alongside information about participants, follow-up, and analysis decisions.

10. Confidence Intervals and Precision

The risk-difference estimate is -0.204, and its 95% confidence interval runs from -0.354 to -0.054. This interval gives a range of values compatible with the estimate under the calculation method and its assumptions. Every value in the interval is negative, which is consistent with a lower observed risky-drinking proportion in the intervention group than in the control group. The interval is not especially narrow: the values it spans represent materially different magnitudes of group contrast. The point estimate should therefore be read together with the uncertainty range, rather than as an exact value.

The risk-ratio estimate is 0.577, with a confidence interval from 0.373 to 0.892. The odds-ratio estimate is 0.414, with a confidence interval from 0.211 to 0.814. Both intervals are below 1, the value corresponding to no ratio contrast. These ratio estimates use different scales and should not be interpreted as direct changes in the number of participants.

Confidence intervals describe statistical uncertainty under the method used; they do not include every uncertainty about trial conduct. The registry record does not report the interval calculation method, missing-data handling, or any adjustment for other factors. The number of participants represented in the outcome counts is also less than the listed enrollment, and the reason is not reported in the registry record. As a result, the intervals do not resolve uncertainty about who contributed outcome data or how unavailable measurements may affect interpretation.

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 sample-size figures are planning illustrations based on the observed effect, not the original trial design assumptions. They give participants per arm, not total events. The stated calculation assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The effect used for each illustration is the stated fraction of the observed effect. The original trial's sample-size justification and design assumptions are not reported in the registry record.

The 80% power column indicates the illustrated number of participants per arm for a design with an 80% chance of detecting the specified effect under the calculation assumptions, if that effect and the assumptions were correct. The 90% power column gives the corresponding planning target for a 90% chance. These are planning concepts, not guarantees that a future study will obtain a statistically significant result. This is not observed power.

Planning for 100% of the observed effect gives 88 participants per arm for 80% power and 118 for 90% power. Planning for 75% of the observed effect gives 161 participants per arm for 80% power and 215 for 90% power. Planning for 50% of the observed effect gives 370 participants per arm for 80% power and 495 for 90% power. The pattern illustrates a general principle: smaller effects are harder to distinguish from random variation, so they require more participants under otherwise unchanged assumptions.

Because these figures use the observed effect as their starting point, they are sensitive to the uncertainty in that estimate. They should not be taken as evidence that the trial was originally planned around the observed result. The registry record does not report the original sample-size rationale.

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, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.481; treatment proportion p₁ = 0.278; average p̄ = 0.380.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6863 = 1.3451.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6710 = 0.5647.
4. n = (1.3451 + 0.5647)² / (-0.2037)² = 87.9, rounded up to 88 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
p0 <- 0.481481481481481                          # control proportion
p1 <- 0.277777777777778                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^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 (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(3273283)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 88 (80% power), 118 (90% power)
 75% of observed effect: n per arm = 161 (80% power), 215 (90% power)
 50% of observed effect: n per arm = 370 (80% power), 495 (90% power)
power.prop.test: 87.9 
Simulated power at n = 88 per arm: 0.806 (20000 replicates)

The simulation draws many pairs of event counts at the planned sample size and counts how often the two-proportion z-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 identifies a primary outcome and reports a comparison for it. It also lists two outcome measures, but does not provide a detailed account of all analyses or specify how any multiple comparisons were handled. The record lists no analysis entries. Whether additional outcomes, subgroup comparisons, or repeated analyses were prespecified or conducted is not reported in the registry record. Without this information, the p-value should be understood as the posted unadjusted result for the stated comparison, not as evidence that multiplicity was addressed.

Missing outcome data are also relevant. The registry lists enrollment of 200, while the primary outcome counts include 72 participants in the intervention group and 81 in the control group. The record does not explain why these denominators differ from enrollment, give follow-up completion information, or describe exclusions. It does not state whether the analysis followed an intention-to-treat approach, used only participants with observed outcomes, or applied any imputation method. The reasons for unavailable outcome measurements and their distribution between groups are not reported in the registry record.

Analysis populations are groups defined for statistical analysis, such as all randomized participants or participants with complete measurements. The registry record does not define the population used for the posted primary comparison. This makes it difficult to know how the estimate relates to all randomized participants. Randomized allocation supports a fair comparison at assignment, but missing outcomes can weaken that advantage if availability of measurements is related to group assignment or outcome status.

The record also does not report a full prespecified statistical plan, a protocol-level hierarchy of outcomes, or whether the posted endpoint was analyzed according to the original plan. Those details matter when interpreting isolated significance tests, especially when several possible outcomes or analyses may exist. Their absence does not establish that analyses were inappropriate; it means the record does not document the relevant procedures.

14. 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 serious adverse events in 0 of 99 participants at risk in the intervention group and 0 of 101 at risk in the control group. It also reports other adverse events in 0 of 99 and 0 of 101, respectively. The adverse-event groups are not the same as the primary-outcome denominators, and the registry does not explain their relationship to the outcome population.

Deaths are listed separately: 2 in the intervention group and 1 in the control group. The record does not provide further detail about these deaths, their timing, causes, or relationship to the intervention. The number of deaths should not be interpreted as an adverse-event rate or attributed to the assigned intervention from these figures alone.

Zero reported events in a group means no events of that category are recorded in the registry summary for the stated group at risk. It does not establish that an event is impossible or that all relevant harms were captured. The registry record does not provide event descriptions or additional safety-analysis details, so the figures are limited to the posted counts.

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

The posted results show fewer participants meeting the AUDIT-C risky-drinking threshold in the intervention group than in the leaflet control group at 6 weeks. The unadjusted risk difference is -0.204, with a 95% confidence interval from -0.354 to -0.054, and the reported p-value is 0.01. The confidence interval indicates statistical uncertainty around the estimated difference; it does not establish the exact size of a group contrast in other settings or at later times.

The endpoint is a threshold classification, not a full account of alcohol consumption. The result does not show how far individual scores changed, whether participants reduced consumption without crossing the threshold, or whether participants initiated specialized treatment. Although referral was described as part of the intervention when indicated, referral uptake and treatment engagement are not reported in the registry record.

The data do not establish that the observed contrast will apply to other emergency departments, patient groups, or follow-up periods. The trial is recorded as conducted in the emergency department of a tertiary hospital, and the posted primary assessment is at 6 weeks. Details needed to assess generalizability, including fuller eligibility criteria and participant characteristics, are not reported in the registry record.

The listed enrollment is 200, while the primary outcome denominators are 72 and 81. Because the record does not explain this difference or describe missing-data handling, the comparison cannot by itself show how the result would look if outcomes had been available for every enrolled participant. The statistical result is a description of the posted analysis, not a substitute for information about follow-up and analysis populations.

The posted figures also do not demonstrate that the intervention is safe or unsafe. The safety summary records no serious or other adverse events in the stated groups at risk and lists deaths in both groups, without further details. The registry record does not report the relationship of those deaths to the intervention. Overall, the result is a bounded finding about one binary outcome at one stated follow-up in the participants represented by the registry counts.

16. Why This Trial Matters Statistically

This trial raises a statistical question about testing a behavioral programme in an emergency-department setting, where the recorded intervention and comparator differ in format. The primary endpoint translates an alcohol-use questionnaire into a binary threshold, making the comparison easy to express as group counts and proportions. That clarity comes with a trade-off: the threshold result does not preserve all information in the underlying AUDIT-C score.

The posted comparison includes an absolute measure, a risk difference, and relative measures, a risk ratio and an odds ratio. Seeing more than one scale helps readers recognize that the same group data can be expressed in different ways. The absolute contrast is useful for understanding the difference in group proportions, while the relative measures describe the contrast as ratios. The scales answer related but distinct questions and should not be conflated.

The confidence interval is important because the point estimate alone can suggest more precision than the data support. Its range conveys uncertainty in the estimated contrast, while the unadjusted p-value summarizes compatibility with a null model under statistical assumptions. Neither measure resolves issues that require design and conduct details, such as how missing outcomes were handled or whether the analysis followed a prespecified plan.

The difference between listed enrollment and the outcome denominators is especially relevant to interpretation. A randomized design is informative, but an unclear analysis population limits assessment of how the posted result relates to all enrolled or randomized participants. The registry record does not document follow-up completeness or the reasons for missing outcome measurements.

Finally, the planning illustrations show how sample-size requirements depend on the effect a future study aims to detect. They are not a reconstruction of the trial's original planning, which is not reported in the registry record. This distinction matters because planning from an observed result and designing a study before results are known are different tasks.

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.

The registry record gives limited information about the trial's conduct and analysis. It identifies randomized allocation, a parallel design, and no masking, but does not describe sequence generation or allocation concealment. It does not report fuller eligibility criteria, participant characteristics, or the specific recruitment and follow-up procedures. These omissions restrict assessment of how representative the analyzed participants were and how assignment was implemented.

The primary outcome denominators, 72 in the intervention group and 81 in the control group, are below the listed enrollment of 200. The reason is not reported in the registry record. The record does not describe missing outcome data, exclusions, protocol deviations, or the analysis population. Thus, the posted estimate cannot be assumed to represent all enrolled participants.

The result is limited to a binary AUDIT-C threshold at 6 weeks. The record does not report the distribution of scores, changes in score, or outcomes at other times. It describes referral when indicated but does not report referral completion or treatment initiation. The stated study aim includes feasibility, but a separate feasibility outcome is not reported in the registry record.

The record does not provide a detailed statistical analysis plan, the method used to calculate confidence intervals, or an account of multiplicity. It lists no analysis entries. The posted comparison is unadjusted, and any adjusted or secondary analyses are not reported in the registry record. Safety information is also limited to event counts and deaths, without descriptions or attribution details.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Feasibility and Effectiveness of a Specialized Brief Intervention for Hazardous Drinkers in an Emergency Department. — NCT03273283.
• Registry record last updated 2020-05-15; status last verified 2020-05.
• Results first posted 2020-05-15.
• 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, parallel-group trial with no masking and enrollment of 200. It evaluates a brief motivational conversation about alcohol use, with referral when indicated, against informational leaflets in an emergency-department setting. The primary outcome is the proportion meeting the AUDIT-C risky-drinking threshold at 6 weeks.

The posted counts are 20 of 72 in the intervention group and 39 of 81 in the control group. The unadjusted risk difference is -0.204, with a 95% confidence interval from -0.354 to -0.054 and p = 0.01. Outcome denominators, missing-data handling, the original sample-size justification, and a detailed statistical analysis plan are not reported in the registry record.