Network Support for Alcohol Treatment 2: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Network Support for Alcohol Treatment 2 (NCT01129804), a randomized na study. The primary outcome is Proportion Days Abstinent (PDA).

2. Clinical Question

3. Trial Design

The registry identifies Network Support for Alcohol Treatment 2 as a randomized, parallel-group study with no masking. The sponsor is UConn Health, the listed condition is alcohol consumption, and the record marks the study completed. The record gives the start as 2009-09 and primary completion as 2015-12. The phase is listed as NA.

The study compared Network Support with Packaged Cognitive-Behavioral Treatment. The brief description says each treatment would last 12 weeks. Network Support aimed to shift participants’ social support toward people and activities associated with not drinking. The comparator focused on coping skills and other skills intended to help participants stop drinking, without specifically addressing social networks.

The record describes daily computerized telephone calls about participants’ experiences before and after treatment, as well as follow-up interviews every 3 months for 2 years following treatment. The primary outcome timeframe is also described as assessment every 3 months out to 2 years. The secondary Social Network Change outcome is described as assessed every 6 months out to 2 years, with mean change over 2 years reported.

The results record lists 96 participants in Network Support and 97 in Packaged Cognitive-Behavioral Treatment. The brief description instead says 160 patients will be assigned. The registry record does not explain the relationship between that description and the posted enrollment and arm counts. The original allocation procedure, details of concealment, recruitment flow, eligibility exclusions beyond the stated population, and reasons for any difference between enrollment and analyzed numbers are not reported in the registry record. The lack of masking is recorded, but the record does not say whether outcome assessors or analysts were masked.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

Analysis population for the primary outcome, as defined in the registry: Men and women aged 18-65 diagnosed with alcohol use disorder, moderate to severe.

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 Proportion Days Abstinent (PDA), measured over Every 3 months out to 2 years.

The primary endpoint is Proportion Days Abstinent, abbreviated PDA in the outcome description. It measures the share of days in an assessment period on which a participant reported zero drinks. The registry says PDA was calculated from daily drinking information collected in each 90-day timeline follow-back assessment, by counting days with no reported drinks.

As a general statistical matter, an abstinent-day proportion condenses repeated daily reports into a summary that can be compared across treatment groups and over follow-up. It retains information about the frequency of abstinent days, but does not by itself describe drinking quantity on days when drinking occurred, the pattern of drinking within the assessment window, or differences among participants that are hidden by a group mean. Those other details cannot be recovered from the arm means alone.

The reported outcome is a continuous measure summarized by its mean and standard deviation. The registry lists a mean of 0.67 and standard deviation of 0.33 for Network Support, and a mean of 0.73 and standard deviation of 0.33 for Packaged Cognitive-Behavioral Treatment. The comparison based on those summaries is an unadjusted mean difference. The registry also reports a separate mixed-model analysis of slope; the exact model specification and how that slope corresponds to PDA over follow-up are not reported in the registry record.

7. Results Posted in the Registry

• Packaged Cognitive-Behavioral Treatment: mean 0.73 versus 0.67 on control; difference 0.06 (95% CI -0.03 to 0.15), standardized effect d = 0.18; z-test p = 0.207.

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.

Social Network Change

• 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 errors.
• Result:
  • Packaged Cognitive-Behavioral Treatment: mean 0.30 versus 0.35 on control; difference -0.05 (95% CI -0.19 to 0.09), standardized effect d = -0.10; z-test p = 0.480.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Packaged Cognitive-Behavioral Treatment and the control. The interval gives the range of effects that remain compatible with the data.

The registry posts one secondary outcome, Social Network Change. Its group summaries and recalculated comparison are reported separately from the primary PDA analysis. Across the posted secondary outcome and primary endpoint, the registry does not give a complete set of model-based estimates that would allow a direct joint interpretation of treatment effects across outcomes. The sponsor’s reported primary analysis concerns a mixed-model slope, whereas no sponsor-reported analysis is listed for the secondary outcome.

The secondary result is relevant because the Network Support intervention was designed to change social support networks. Even so, a change in a network score is not interchangeable with the primary drinking outcome. A secondary measure can inform whether an intervention’s intended process changed, but its relationship to abstinent days would require an analysis of that relationship, which is not reported in the registry record. The posted secondary outcome should not be treated as confirmation of the primary result or as an explanation for it.

Unadjusted secondary results carry less weight because multiple outcomes and analyses create more opportunities for apparently noteworthy findings by chance. As a general statistical principle, interpreting several unadjusted tests without accounting for multiplicity makes each result less conclusive as stand-alone evidence. The registry record does not report a multiplicity adjustment, a hierarchy of outcomes, or a prespecified confirmatory role for Social Network Change.

Accordingly, the secondary outcome is best read as additional descriptive evidence about the intervention’s intended social-network target, not as a substitute for the primary analysis. Its statistical interpretation is limited by the absence of a sponsor-reported analysis and by the missing information about multiplicity and the broader analysis plan. The record does not report whether the secondary outcome was intended to test a mechanism, to support a confirmatory claim, or to serve another purpose.

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

• Social Network Change

9. Statistical Interpretation

10. Statistical Methods Explained

The results contain two kinds of primary-outcome evidence: arm-level descriptive summaries and a sponsor-reported mixed-model analysis. The distinction matters because they summarize the trial in different ways. The arm means and standard deviations describe the posted outcome summaries for each group. The recalculated comparison uses those summaries to estimate an unadjusted difference in means. The registry’s sponsor analysis reports a slope from a mixed model. A slope and a difference between group means are not automatically the same quantity.

What does a mean difference describe?

A mean is a summary of the observed outcome values in a group. For PDA, the posted means are 0.67 for Network Support and 0.73 for Packaged Cognitive-Behavioral Treatment. The unadjusted mean difference is reported as 0.06, defined in the comparison as Packaged Cognitive-Behavioral Treatment relative to Network Support. This is a difference in the posted group means, not an adjusted estimate from the sponsor’s model. It does not describe how every participant changed, and it may conceal variation in individual outcomes.

The posted standard deviations are 0.33 in both groups. As a general statistical concept, a standard deviation describes the spread of observations around a group mean. It is different from a standard error, which describes uncertainty in an estimated mean. The primary outcome record lists standard deviation as its measure of spread. The arm-level figures therefore give information about variability as well as the group averages, but do not show the distribution of individual observations.

What does the recalculated comparison add?

The unadjusted re-calculation combines the arm summaries to report a mean difference, confidence interval, and unadjusted p-value. It reports a difference of 0.06, a confidence interval from -0.033 to 0.153, and p = 0.207. The confidence interval expresses uncertainty around the difference under the assumptions used for that calculation. Since the interval includes zero, the summaries do not rule out zero difference at the conventional level associated with that interval. The p-value is also unadjusted: it does not account for any multiplicity across outcomes or analyses.

These recalculations are based on posted summary statistics and are not a substitute for the trial’s original analysis. They do not adjust for baseline values, repeated observations, or other model terms unless those are reflected in the summary-statistic calculation; the registry record does not indicate that they are. The precise adjustment structure of the sponsor analysis is not reported in the registry record.

What does a mixed-model slope mean?

As a general statistical explanation, a mixed model can analyze repeated observations while accounting for the fact that measurements from the same participant are related. A slope can describe change over time or a modeled trend, depending on how time and treatment are coded in the model. The registry reports a mixed-model slope of 0.1 and p <.04, but does not report the model equation, the time coding, the treatment-by-time terms, the covariance structure, the population represented by the estimate, or a confidence interval. Those omissions prevent a fuller independent interpretation of the reported slope.

The registry labels that analysis as superiority. This label identifies the analysis classification recorded by the sponsor; it does not fill in the missing model details. In particular, the record does not state whether the slope is a within-group trend, a between-group contrast in trends, or another model parameter. The reported p-value indicates evidence against the null hypothesis specified for that sponsor analysis, but the null hypothesis itself is not described in enough detail to reconstruct it.

When results from repeated-measures models are reported, the model’s target quantity and uncertainty are important alongside the p-value. A model can use information across follow-up assessments and may account for within-participant correlation. These are general properties, not claims about the particular specification used in this study, which is not reported in the registry record. The posted means, by contrast, provide a simple group-level summary and the unadjusted re-calculation provides a summary-based contrast.

Why can the analyses differ?

Different analysis methods can estimate different quantities and use different assumptions. A contrast of group means need not match a modeled slope, particularly when observations are repeated over time. Model adjustment, time coding, and handling of incomplete repeated measures can also affect an estimate. These are general possibilities; the registry record does not identify which, if any, accounts for the difference between the sponsor result and the re-calculation. Because the discrepancy is flagged in the record, the sponsor’s own analysis takes precedence in reporting the trial’s analysis, while the re-calculation remains a separate descriptive check on the posted summaries.

There is no reported multiplicity procedure, missing-data method, or analysis-population definition in the registry record. Those details would help explain how the sponsor handled the repeated observations and how uncertainty was assessed. Without them, the reported mixed-model result can be quoted but not fully reproduced or evaluated. The re-calculation should likewise be understood as an unadjusted comparison, not as a replacement for the sponsor’s model or as an estimate of observed power.

11. Confidence Intervals and Precision

The recalculated primary mean difference is 0.06, with a confidence interval from -0.033 to 0.153. A confidence interval gives a range of estimates compatible with the observed data and the calculation’s assumptions. Its width reflects uncertainty: a wider interval leaves more possible values consistent with the data, while a narrower interval indicates greater precision under the same assumptions.

The interval crosses zero. This means the summary-based comparison does not precisely distinguish a positive difference from no difference or a difference in the opposite direction. The interval is not proof that the groups are equivalent. Equivalence requires a prespecified margin and an analysis designed to test whether the interval lies within that margin; such a margin and analysis are not reported in the registry record.

The recalculated p-value is 0.207. It is an unadjusted result based on the posted means and spread, and it does not measure the probability that a treatment is effective or that the null hypothesis is true. Its interpretation depends on the statistical model and assumptions used. It should be considered alongside the estimated difference and interval rather than in isolation.

Precision for the sponsor’s mixed-model result cannot be assessed as fully because the registry reports no confidence interval for its slope of 0.1. The p-value, p <.04, does not give the range of plausible slope values or show the estimate’s uncertainty in the units of the modeled outcome. The model specification is also not reported in the registry record, which limits evaluation of how the standard error and p-value were obtained.

The arm-level standard deviations are 0.33 for both groups. They describe spread in the outcomes, not precision of the between-group estimate. The reported arm sizes and standard deviations contribute to the recalculated comparison, but the resulting interval does not resolve questions about repeated measurements, missing outcomes, or model assumptions. Precision should therefore be considered separately for the re-calculated difference and the sponsor’s analysis.

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 sample-size estimates are planning illustrations based on the observed effect, not the trial’s own design calculation. The assumptions listed for these illustrations are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is the stated fraction of the observed effect. The original design assumptions for the trial are not reported in the registry record.

For continuous outcomes, the figures are participants per arm, not a total across the study. The planning values are therefore read as the number of participants needed in each treatment group under the stated assumptions. The display also distinguishes planning for 80% power from planning for 90% power. In plain language, the 80% column gives the per-arm planning estimate under a design that would have an 80% chance of detecting the specified effect if that effect and the assumptions were correct. The 90% column uses a higher target chance of detecting that effect, so it requires more participants per arm.

The rows correspond to planning for 100%, 75%, and 50% of the observed effect. As the assumed effect becomes smaller, more participants are needed to detect it with the same target power. For the primary outcome, planning for 100% of the observed effect gives 475 participants per arm for 80% power and 636 participants per arm for 90% power. Planning for 75% gives 845 participants per arm for 80% power and 1,131 participants per arm for 90% power. Planning for 50% gives 1,900 participants per arm for 80% power and 2,543 participants per arm for 90% power.

These values are not observed power. They are illustrative sample-size requirements generated from the observed effect under the listed assumptions. They do not establish what the original investigators planned or whether the completed study met an original target.

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 = 0.060; pooled standard deviation σ = 0.330.
2. n = 2 × (1.96 + 0.842)² × 0.330² / 0.060² = 474.9, rounded up to 475 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 <- 0.0599999999999999                      # observed difference in means
sigma <- 0.33                     # 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(1129804)
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 = 475 (80% power), 636 (90% power)
 75% of observed effect: n per arm = 845 (80% power), 1,131 (90% power)
 50% of observed effect: n per arm = 1,900 (80% power), 2,543 (90% power)
power.t.test: 475.8 
Simulated power at n = 475 per arm: 0.800 (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. Analyses Reported by the Sponsor

The registry’s sponsor-reported primary analysis is a mixed-model analysis of slope, recorded as 0.1 with p <.04 and classified as superiority. That is the sponsor’s own analysis and takes precedence over the recalculated comparison when describing the sponsor’s reported finding.

The recalculation is a separate, unadjusted comparison of posted arm means. It gives a mean difference of 0.06 for Packaged Cognitive-Behavioral Treatment relative to Network Support, a confidence interval from -0.033 to 0.153, and p = 0.207. It is not a reproduction of the sponsor’s mixed model. The registry record does not report enough detail about the model’s specification, slope definition, or treatment contrast to reconcile the two results or establish that they address the same estimand.

The discrepancy should therefore be reported transparently rather than resolved by choosing the result that appears more favorable. The sponsor’s analysis is the trial’s reported analysis; the summary-statistic recalculation is useful as a check on the posted group summaries, but it does not override the sponsor’s model. Since the sponsor’s result lacks a confidence interval and model details, its magnitude and precision cannot be independently assessed from the registry entry. The reported p-value alone is not a measure of clinical importance.

15. Multiplicity, Missing Data and Analysis Populations

The registry lists a primary outcome and a secondary Social Network Change outcome, along with a sponsor-reported mixed-model primary analysis. It does not report a multiplicity strategy, such as adjustment across outcomes, time points, or analyses, or a prespecified hierarchy governing which tests would support confirmatory conclusions. Without that information, unadjusted p-values should be interpreted cautiously, particularly when more than one outcome or analysis is considered.

The record also does not report how missing outcome measurements were handled. It describes repeated follow-up assessments, but does not state how incomplete daily reports, missed interviews, or unavailable follow-up observations were treated in the sponsor’s analysis. It does not describe imputation, likelihood-based handling of incomplete observations, sensitivity analyses, or assumptions about why data might be missing. These are not details that can be inferred from the existence of a mixed model.

The analysis population is not defined in the registry record. It does not state whether the sponsor analyzed participants according to randomized assignment, included only participants with available observations, or used another population definition. No adherence-based analysis or exclusion rules are described. The record lists arm sizes for the posted summaries, but that alone does not establish the population used in the mixed-model analysis or whether all randomized participants contributed to it.

Other important elements of the statistical plan are also not reported in the registry record: the original sample-size justification, the model specification, the slope’s interpretation and coding, the planned contrasts, any interim analyses, and whether analyses were prespecified before outcome data were examined. No conclusions about these procedures should be inferred from the reported p-value or analysis label.

These reporting gaps do not by themselves show that the analyses were inappropriate. They limit the ability to reproduce the sponsor result, assess the effect of missing observations, and judge how the primary and secondary findings were selected and interpreted. The results should therefore be read with the distinction between reported estimates and unreported analytic details kept explicit.

16. 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.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: increased drinking from baseline (36.1%).

The registry lists serious adverse event figures for Network Support as 30 affected participants among 96 at risk, and for Packaged Cognitive-Behavioral Treatment as 35 among 97 at risk. It lists other adverse event figures as 5 among 96 at risk for Network Support and 5 among 97 at risk for Packaged Cognitive-Behavioral Treatment. The death count is 2 in each group.

The registry also lists Increased drinking from baseline as a top serious-event term, with 36.1%. That entry is not allocated to an arm in the posted summary, so it should not be interpreted as an arm-specific rate. The adverse-event record does not define the event-collection period or provide details about severity grading, relatedness, repeated events, or the method used to establish the at-risk population. Those details are not reported in the registry record.

These figures are descriptive counts in the registry. They do not establish that a treatment caused an event, and they do not by themselves support a comparative safety conclusion. Comparing event counts requires attention to definitions, follow-up, exposure, and the handling of participants with more than one event, details that are not reported in the registry record.

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

The results describe a randomized comparison of two behavioral treatments and a primary outcome based on the proportion of days with no reported drinks. The posted group means and the sponsor’s mixed-model result are different forms of evidence. The summary-based re-calculation reports a difference of 0.06 and an interval spanning -0.033 to 0.153, while the sponsor reports a slope of 0.1 with p <.04. The registry does not provide enough information to establish that these estimates target the same quantity.

Statistical significance and effect size answer different questions. A p-value addresses compatibility with a specified statistical null under a model; it does not say whether an observed difference is important to participants or clinically meaningful. A mean difference describes the contrast in the measured outcome, but its interpretation depends on the outcome definition, follow-up, and the population represented. The registry does not provide a threshold for a meaningful PDA difference.

The sponsor’s p-value is reported without a confidence interval or sufficient model detail to evaluate precision and direction independently. The recalculated interval is more informative about uncertainty in its own unadjusted contrast, but it does not replace the sponsor analysis. Neither result alone establishes that one treatment is superior in a broader population, nor does the summary comparison establish equivalence or no difference.

Randomization supports a comparative design, but a registry entry is not a complete account of trial conduct or analysis. The record does not report key details about missing data, analysis populations, multiplicity, model specification, or the original design assumptions. These omissions make it harder to assess the robustness of the reported results. The findings should be understood as the posted estimates and their stated limitations, rather than as medical advice or a guarantee of individual outcomes.

The study’s secondary outcome concerns social network change, the target addressed by Network Support. That outcome may be informative about the intervention’s intended focus, but it cannot by itself explain differences in abstinent days. A measured change in a proposed target and a change in a clinical outcome are distinct statistical questions; the registry does not report a formal analysis connecting them.

18. Why This Trial Matters Statistically

This trial illustrates why a registry result can contain more than one legitimate statistical description of the same study. The arm summaries describe average PDA in each treatment group, a simple re-calculation contrasts those means, and the sponsor reports a mixed-model slope. These quantities are not automatically interchangeable. Statistical interpretation depends on the estimand, the analysis method, and the assumptions behind each result.

It also shows why a confidence interval is useful alongside a p-value. The recalculated interval communicates a range of uncertainty for the unadjusted mean difference and includes values on both sides of zero. The sponsor’s reported p-value is not accompanied by an interval, so the precision and plausible magnitude of that slope are less clear from the registry record. A statistically significant label does not provide the missing magnitude information.

The endpoint itself is a repeated behavioral measure, based on reports of drinking across 90-day assessment periods and followed over time. Repeated observations raise methodological questions about within-participant dependence, incomplete follow-up, and the definition of change. A mixed model can address repeated measurements in general, but interpretation requires its specification and assumptions, which are not reported in the registry entry.

Finally, the planning illustrations show how sample-size requirements depend on the effect one aims to detect and the desired power. Smaller target effects require more participants under the listed assumptions. These calculations are distinct from the trial’s original design rationale, which is not reported. Together, the study’s results and reporting gaps make it a useful example of reading estimates, uncertainty, and analysis methods as separate parts of the evidence.

19. 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 does not define the sponsor’s slope estimate in enough detail to establish its direction or exact interpretation. It provides no confidence interval for that result and no model specification. This prevents a complete assessment of its magnitude, precision, and comparability with the unadjusted mean difference.

The primary outcome is based on reported drinking information collected through daily assessments and timeline follow-back. The registry does not describe validation procedures, completeness of daily reports, or how discrepancies between reports were handled. It also does not report the number of assessments available at each follow-up point or how missing observations were treated. These limitations are specific to interpreting an outcome built from repeated self-reported data.

The treatment descriptions are not equally complete. The Network Support arm description is truncated, while the comparator description is brief. The registry does not provide treatment fidelity measures, attendance, adherence, or the extent to which participants engaged in the activities described. Without such information, the posted results cannot be linked to treatment exposure or implementation in detail.

The record describes repeated follow-up through 2 years, but does not provide detailed follow-up completion or retention information. It also does not explain the difference between the 160 patients mentioned in the brief description and the enrollment of 193 or the posted arm counts. The analysis population and handling of any post-randomization exclusions are not reported in the registry record.

There is one posted secondary outcome, and the registry does not report its role in a prespecified testing hierarchy or any multiplicity adjustment. The serious-event summary also lacks detail about collection windows, event definitions, and relatedness. These gaps constrain independent interpretation of both efficacy-related and safety-related results. They do not establish that bias occurred, but they leave relevant sources of uncertainty unresolved.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Network Support for Alcohol Treatment 2 — NCT01129804.
• Registry record last updated 2021-11-02; status last verified 2021-10.
• Results first posted 2021-11-02.
• 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.

23. Record Summary

Network Support for Alcohol Treatment 2 is recorded as a completed randomized, parallel-group study without masking, sponsored by UConn Health. The registry describes a comparison of Network Support with Packaged Cognitive-Behavioral Treatment among men and women aged 18-65 with moderate to severe alcohol use disorder. Enrollment is listed as 193.

The primary outcome is Proportion Days Abstinent, derived from repeated 90-day timeline follow-back assessments. Posted means are 0.67 for Network Support and 0.73 for Packaged Cognitive-Behavioral Treatment. An unadjusted re-calculation gives a mean difference of 0.06, with a confidence interval from -0.033 to 0.153 and p = 0.207. Separately, the sponsor reports a mixed-model slope of 0.1 with p <.04. Because the results differ in analysis and the registry flags a discrepancy, the sponsor’s own analysis takes precedence as the reported trial analysis.

Social Network Change is the posted secondary outcome. Important details about the original design assumptions, analysis populations, missing data, multiplicity, and the mixed-model specification are not reported in the registry record. The posted estimates should be read with those limitations in mind.