Protective Behavioral Strategies and Brief Alcohol Interventions: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Protective Behavioral Strategies and Brief Alcohol Interventions (NCT01168726), a randomized na study. The primary outcome is Drinks Per Week.

2. Clinical Question

3. Trial Design

The study is listed as randomized, single-masked, and parallel-group, with three behavioral intervention arms. Its recorded enrollment is 365, and its status is completed. The listed study start is 2010-01 and primary completion is 2010-09. The sponsor is the University of Missouri-Columbia. The registry summary identifies excessive college student drinking as the public health concern and describes the objective as improving prevention and treatment programs.

The three arms differed in their content. Protective Behavioral Strategies offered personalized feedback about use of protective behavioral strategies. Personalized Normative Feedback offered personalized feedback comparing the participant’s drinking with relevant norms. Alcohol Education offered educational information about harms associated with heavy drinking. The summary says participants would be randomized to one of two brief intervention conditions or an education-only control condition. The posted primary contrasts, however, designate Protective Behavioral Strategies as the reference group and compare it with each of the other arms.

Single masking is recorded, but which people were masked and how masking was implemented are not reported in the registry record. Further details about the randomization procedure, allocation concealment, intervention delivery, adherence, and follow-up are also not reported in the registry record. The primary endpoint is listed at 6 Months. The record lists two secondary outcomes, both at 6 months. The distinction in capitalization follows the registry entries; no separate timing or assessment schedule is described.

The enrollment figure is a study-level registry value, while the posted outcome summaries give separate arm sizes. The record does not describe the relationship between enrollment and the number of participants included in each outcome analysis, or whether any participants lacked 6-month outcome data. That information is important for understanding how closely the analyzed groups reflect the randomized groups.

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.

6. Primary Endpoint

The registry lists the primary outcome as Drinks Per Week, measured over 6 Months.

The primary endpoint is Drinks Per Week at 6 Months, summarized as a mean with a standard deviation in each arm. It measures the average reported weekly alcohol consumption for participants represented in the outcome summary. The posted means are 12.55 for Protective Behavioral Strategies, 9.57 for Personalized Normative Feedback, and 13.74 for Alcohol Education; the associated standard deviations are 8.96, 8.78, and 10.77, respectively.

A continuous measure retains gradations in reported drinking rather than reducing the outcome to a yes-or-no category. Comparing arm means can therefore describe differences in average reported consumption. The standard deviation describes how dispersed individual observations are around the group mean; it is not a measure of uncertainty in the mean itself. The uncertainty in a contrast is represented separately by its confidence interval.

Weekly drinks is relevant to the study’s stated focus on excessive drinking and alcohol-related problems. However, the endpoint alone does not capture all dimensions of alcohol use or harm. It also does not identify changes for individual participants, establish whether the observed group difference is clinically important, or demonstrate that an intervention caused a difference. The exact instrument, collection method, and definition used to derive weekly drinks are not reported in the registry record.

7. Results Posted in the Registry

• Personalized Normative Feedback: mean 9.57 versus 12.55 on control; difference -2.98 (95% CI -5.27 to -0.69), standardized effect d = -0.34; z-test p = 0.011.
• Alcohol Education: mean 13.74 versus 12.55 on control; difference 1.19 (95% CI -1.29 to 3.67), standardized effect d = 0.12; z-test p = 0.346.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Drinking Norms Rating Form

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Personalized Normative Feedback: mean -2.71 versus 1.52 on control; difference -4.23 (95% CI -5.03 to -3.43), standardized effect d = -1.38; z-test p = <0.001.
  • Alcohol Education: mean 1.20 versus 1.52 on control; difference -0.32 (95% CI -1.24 to 0.60), standardized effect d = -0.09; z-test p = 0.495.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Personalized Normative Feedback than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Alcohol Education and the control. The interval gives the range of effects that remain compatible with the data.

Protective Behavioral Strategies Scale.

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Personalized Normative Feedback: mean 12.14 versus 12.41 on control; difference -0.27 (95% CI -0.81 to 0.27), standardized effect d = -0.13; z-test p = 0.325.
  • Alcohol Education: mean 11.44 versus 12.41 on control; difference -0.97 (95% CI -1.47 to -0.47), standardized effect d = -0.49; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Personalized Normative Feedback and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Alcohol Education than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The secondary outcomes add information about drinking norms and protective behavioral strategies, rather than directly repeating weekly alcohol consumption. Across those outcomes, the posted contrasts do not show a uniform pattern in which one intervention is consistently lower or higher than the reference on every measure. Personalized Normative Feedback differs notably on the norms measure, while Alcohol Education has a contrast on the protective-strategies measure; the other posted secondary contrasts are less clearly separated from zero. This broad pattern is compatible with the interventions targeting different aspects of behavior and perceptions, but it does not establish a mechanism linking any secondary measure to the primary outcome.

The secondary results can help describe whether intervention content is reflected in measures related to that content. They cannot, on their own, explain why the weekly-drinks means differ, establish mediation, or demonstrate that a change in a secondary measure produced a change in drinking. The registry record does not provide an analysis of relationships among outcomes or participant-level changes. Each outcome is a separate result, with its own scale and interpretation.

The reported secondary p values are unadjusted, and the record does not report a prespecified hierarchy, multiplicity correction, or which outcomes were designated as confirmatory. When multiple outcome contrasts are examined, isolated small p values carry less weight than they would in a single prespecified test because the number of opportunities for an unusual result can affect interpretation. Secondary results should therefore be treated as supportive or exploratory descriptions unless a prospectively specified analysis plan says otherwise; such a plan is not reported in the registry record. Their overall pattern can contextualize the primary outcome, but does not resolve the primary outcome’s limitations or substitute for an analysis accounting for multiple comparisons.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Drinking Norms Rating Form
• Protective Behavioral Strategies Scale.

9. Statistical Interpretation

10. Statistical Methods Explained

Understanding the posted comparisons starts with identifying what is being estimated. For the primary outcome, the estimate is a difference between two arm means at 6 Months. The reference group is Protective Behavioral Strategies, with a posted mean of 12.55 drinks per week. Personalized Normative Feedback has a posted mean of 9.57, and Alcohol Education has a posted mean of 13.74. The reported estimates express each other arm’s mean relative to the reference mean. A negative estimate indicates a lower mean in the comparison arm; a positive estimate indicates a higher mean.

What does the standard deviation describe?

A standard deviation describes variability among observations within an arm. A larger standard deviation indicates more spread around that arm’s mean; a smaller one indicates observations are more concentrated. It is not the same as a standard error or a confidence interval. In the primary outcome, the posted standard deviations are 8.96, 8.78, and 10.77 for the three arms. Those values give context for the spread of weekly drinking measurements, while the contrast interval describes uncertainty about the difference between arm means.

How should a confidence interval be read?

A confidence interval gives a range of values compatible with the estimate under the statistical procedure used. It communicates precision as well as direction: a narrower interval represents greater precision than a wider interval on the same scale. The intervals posted for the primary contrasts are -5.266 to -0.694 for Personalized Normative Feedback versus Protective Behavioral Strategies and -1.286 to 3.666 for Alcohol Education versus Protective Behavioral Strategies. The first excludes zero; the second includes zero. The confidence level and interval-construction procedure are not reported in the registry record, so those technical details should not be assumed.

What does an unadjusted p value say?

A p value describes how unusual the observed data, or data more extreme, would be under a specified null hypothesis and the assumptions of the test. It is not the probability that the null hypothesis is true, and it does not measure the size or practical importance of a difference. The primary p values are unadjusted: 0.011 for the personalized normative feedback contrast and 0.346 for the alcohol education contrast. The registry record does not list the investigators’ inferential analyses or say whether these were the tests used in the original study. They should be interpreted as part of the posted unadjusted summaries, not as proof of a treatment effect.

The comparisons use the arm means, standard deviations, and sample sizes in the registry record to describe simple, unadjusted differences. A recalculation from summary statistics does not account for participant-level covariates, clustering, deviations from the assumed distribution, or missing observations unless those features are explicitly incorporated. The record says the discrepancy flag is false; there is therefore no indicated discrepancy that would call for a sponsor analysis to take precedence. Even so, the registry does not report the investigators’ analysis details, so the recalculated contrasts are not evidence that the original trial analysis used the same approach.

Why can variation affect precision?

In general, uncertainty in a mean difference depends on both the amount of variation in the outcome and the amount of information available in each arm. More variation tends to make a difference harder to estimate precisely, while more observations can improve precision. This general relationship does not establish that a particular sample size is adequate for this trial; the original design assumptions and sample-size justification are not reported in the registry record. The arm standard deviations and sizes inform the posted recalculations, but the record does not provide participant-level data for examining distributional shape or influential observations.

What can randomization contribute?

In general, random assignment aims to make comparison groups similar, on average, with respect to factors that could influence outcomes. It supports an interpretation of between-group differences as comparisons of assigned strategies, provided that the randomization was implemented appropriately and the analysis preserves that comparison. The registry identifies the study as randomized but does not describe sequence generation, concealment, treatment crossover, adherence, or the analysis population. It therefore does not allow a full assessment of those conditions.

In general, over-dispersion means that outcomes vary more than a particular statistical model assumes. When that happens, model-based standard errors may be too small if the extra variability is not accommodated, making intervals appear too narrow and tests too strong. The posted outcome is continuous and summarized with standard deviations, but the registry does not report a model, diagnostics, or an over-dispersion assessment. No claim about model adequacy can be made from the record. The same restraint applies to assumptions about normality, outliers, and transformations: those details are not reported in the registry record.

11. Confidence Intervals and Precision

The primary contrast with Personalized Normative Feedback is estimated at -2.98 drinks per week. Its confidence interval extends from -5.266 to -0.694. Because both endpoints are below zero, the interval is consistent with a lower mean in that arm than in Protective Behavioral Strategies under the posted contrast. The interval nevertheless covers a range of magnitudes; it does not imply that the true difference is exactly -2.98. The registry record does not report the confidence level or the method used to construct the interval.

The Alcohol Education contrast is estimated at 1.19 drinks per week, and its interval extends from -1.286 to 3.666. The interval includes zero, as well as values on either side. Thus the estimate is not precise enough to rule out a small difference in either direction under the posted calculation. This is not evidence that the arms are equivalent: an equivalence conclusion requires an appropriate design, a prespecified margin, and analysis, none of which is reported in the registry record.

Precision is distinct from magnitude. The point estimate indicates the center of the reported contrast, while the interval conveys uncertainty around that estimate. A confidence interval does not account automatically for every source of uncertainty, such as selective reporting, missing outcome data, or unmodeled design features. Those matters are not described in the registry record. The posted intervals should therefore be read as uncertainty ranges for the unadjusted contrasts rather than complete summaries of all uncertainty about the trial’s findings.

12. 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, not total participants and not total events. They are illustrations based on the observed effect, rather than the trial’s original design calculation. The stated planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, a normal approximation, and an effect to detect set to the stated fraction of the observed effect. The trial’s own sample-size justification and original design assumptions are not reported in the registry record.

The 80% power column describes the number of participants per arm that the planning calculation associates with an 80% chance of detecting the chosen effect under those assumptions. The 90% power column uses a higher probability of detection and consequently gives larger planning figures. These columns are not counts of participants enrolled in the trial, nor are they evidence about what the completed study was likely to detect in practice. This is not observed power.

Planning rows that use 100%, 75%, or 50% of the observed effect ask what sample size would correspond to detecting effects of those relative sizes. A smaller effect is harder to distinguish from random variation, so it generally requires more participants for the same power target. For example, the primary personalized normative feedback contrast lists 139 participants per arm at 80% power for 100% of the observed effect, and 248 participants per arm for 75% of that effect. The corresponding alcohol education figures are much larger, reflecting the smaller observed contrast used in that planning illustration. These calculations are conditional on the stated assumptions and are not a retrospective verdict about the trial’s success.

13. 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 = -2.980; pooled standard deviation σ = 8.867.
2. n = 2 × (1.96 + 0.842)² × 8.867² / -2.980² = 139.0, 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 <- -2.98                      # observed difference in means
sigma <- 8.86654287807236                     # 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(1168726)
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), 187 (90% power)
 75% of observed effect: n per arm = 248 (80% power), 331 (90% power)
 50% of observed effect: n per arm = 556 (80% power), 745 (90% power)
power.t.test: 139.9 
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.

14. Multiplicity, Missing Data and Analysis Populations

The registry record lists two primary arm contrasts for the primary outcome and two secondary outcomes, each with contrasts against the reference arm. The posted p values are unadjusted. A multiplicity strategy—such as defining a primary contrast, controlling an overall error rate, or specifying an ordered testing procedure—is not reported in the registry record. Without that information, the results should not be interpreted as if the several contrasts formed one prespecified confirmatory testing plan. The record also does not report whether any outcomes or analyses were added, omitted, or selected after examining results.

Missing-data handling is not reported in the registry record. The number of participants represented in each posted arm summary is given, but the record does not explain whether these are all randomized participants, participants with an observed 6-month measure, or another analysis population. The reasons for missing outcomes, any imputation method, and any sensitivity analysis are also not reported. If outcome availability differs between arms or relates to drinking, the observed summaries may not fully represent the randomized groups.

The analysis population is not reported in the registry record. There is no statement about whether the analysis followed intention-to-treat principles, included only participants who completed intervention activities, or used another definition. Adherence, protocol deviations, crossover, and exclusions are not described. Baseline outcome values and adjustments for baseline imbalance are not reported either. These gaps limit assessment of how the randomized design was carried through to the outcome analysis.

The record lists zero reported analyses and no reported analyses for the secondary outcomes. It does not include a statistical analysis plan, the original sample-size justification, interim analysis details, or specifications for the final model. These items are not reported in the registry record; their absence should not be replaced by assumptions about what investigators planned. As a result, posted summary contrasts can be described, but their relationship to any prespecified inferential framework remains unclear.

15. 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 adverse-event summary records zero serious adverse events for Protective Behavioral Strategies, Personalized Normative Feedback, and Alcohol Education. It also records zero other adverse events affecting participants in each arm. The corresponding numbers listed as at risk are 111, 121, and 133. No serious adverse-event terms are listed.

These figures describe what is recorded in the registry, not proof that no adverse experiences occurred. The registry record does not report event definitions, collection procedures, severity grading, causality assessment, or the period over which events were monitored. Death counts are not reported in the record; a null entry should not be read as a recorded zero. The posted counts therefore support a narrow description of the recorded adverse-event entries and do not permit a full comparison of safety or tolerability between the interventions.

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

The registry record describes a randomized comparison of three behavioral approaches and posts arm-level means for drinks per week at 6 Months. The unadjusted summary contrast for Personalized Normative Feedback is lower than the reference-arm mean, with an interval that excludes zero. The Alcohol Education contrast is higher by its point estimate, but its interval includes zero and allows differences in either direction. These findings describe the posted summaries; they are not, by themselves, a complete causal analysis.

The numerical size of a difference is not the same as its practical importance. The registry record does not specify what difference in weekly drinking would be meaningful to participants or researchers. Nor does it provide enough information about the measure, baseline status, missing data, adherence, or analysis population to determine whether the observed summaries represent the full randomized comparison. A p value cannot answer questions about importance, and an interval cannot correct for limitations not included in the calculation.

The findings also do not establish that an intervention works or fails, that one strategy is generally preferable, or that results apply to populations beyond the study participants. The record focuses on college student drinking but does not report detailed demographic or eligibility information. It does not provide long-term outcomes beyond the stated 6-month endpoint, or evidence about persistence of any difference. The secondary measures describe related constructs but do not prove how or why the primary outcome may differ.

A careful reading separates three things: the observed arm summaries, uncertainty around the posted comparisons, and the broader claims a reader might wish to make. The first two are available in the registry record; many details needed for the third are not reported. The most defensible interpretation is therefore descriptive and bounded by the posted follow-up, outcomes, and unadjusted calculations.

17. Why This Trial Matters Statistically

This trial illustrates why a randomized label and a numerical contrast are only part of a statistical interpretation. The registry identifies random assignment and three parallel behavioral arms, while the posted primary result offers two comparisons against a common reference group. Comparing those estimates requires attention to direction, spread, interval width, and the fact that the p values are unadjusted. An estimate’s sign alone does not indicate certainty, and a non-significant result does not demonstrate equivalence.

The example also shows how endpoint choice shapes the question. Drinks per week summarizes average consumption, while drinking norms and protective behavioral strategies measure related but distinct aspects of the study’s aim. Results across these outcomes need not move together, and a secondary measure cannot automatically explain a primary outcome. The record’s limited analysis detail makes it especially important to distinguish descriptive summaries from conclusions based on a prespecified model.

Finally, planning illustrations based on observed effects are not a substitute for prospective design assumptions. Their values change with the effect size one intends to detect, underscoring the general statistical trade-off between smaller target effects and the amount of information needed. Missing information about the original sample-size rationale, multiplicity, and missing-data handling prevents a full audit of the trial’s statistical design. The record is therefore useful for teaching how to read contrasts and uncertainty while also showing why transparent methods matter.

18. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

Several trial-specific limitations follow from what the registry record does not report. Detailed eligibility criteria, participant characteristics, baseline drinking values, and outcome measurement procedures are not reported. Without those details, it is difficult to assess the starting comparability of the groups, the measurement context, or the extent to which the participants resemble other college student populations.

The registry identifies single masking but does not explain who was masked or how it was maintained. It does not describe allocation concealment, intervention adherence, protocol deviations, or contamination between arms. The analysis population and handling of missing 6-month observations are also not reported. These omissions make it harder to evaluate whether the posted group summaries preserve the benefits of randomization.

The record contains no reported analysis plan or inferential analyses. The posted comparisons are unadjusted recalculations from summary statistics, and the record does not say whether they match the investigators’ planned primary analysis. It does not report baseline adjustment, multiplicity handling, sensitivity analyses, or a meaningful-difference threshold. The confidence level and interval method are not reported either.

Follow-up is reported at 6 Months for the primary measure and the two secondary outcomes; longer-term outcomes are not reported in the record. Details of adverse-event definitions and monitoring are absent, and the death entries are not reported. These limits call for caution about causal attribution, completeness, generalizability, and safety interpretation without implying that the unreported procedures were or were not adequate.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Protective Behavioral Strategies and Brief Alcohol Interventions — NCT01168726.
• Registry record last updated 2017-07-13; status last verified 2017-06.
• Results first posted 2015-06-29.
• 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.

22. Record Summary

NCT01168726, titled Protective Behavioral Strategies and Brief Alcohol Interventions, is listed as a completed randomized, single-masked, parallel-group study sponsored by the University of Missouri-Columbia. The registry records enrollment of 365 and compares personalized protective-strategy feedback, personalized normative feedback, and alcohol education.

The primary outcome is mean drinks per week at 6 Months. The posted unadjusted comparison with Protective Behavioral Strategies is lower for Personalized Normative Feedback and higher by point estimate for Alcohol Education; their intervals differ in whether they include zero. Two secondary measures are also listed at 6 months. The registry record does not report the original analysis plan, analysis population, missing-data approach, or multiplicity strategy, so the posted summaries support a bounded interpretation rather than a complete statistical account of the trial.