A Unified Intervention for Young Gay and Bisexual Men's Minority Stress, Mental Health, and HIV Risk: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Unified Intervention for Young Gay and Bisexual Men's Minority Stress, Mental Health, and HIV Risk (NCT02929069), a randomized na study. The primary outcome is Any Condomless Anal Sex Acts.

2. Clinical Question

The trial was designed to test a transdiagnostic cognitive behavioral therapy intervention addressing pathways through which minority stress may relate to co-occurring mental, behavioral, and sexual health problems. The stated examples include depression, substance use, and condomless anal sex. The primary endpoint narrows that broad aim to a specific reported behavior, with particular conditions defining which acts count as the endpoint.

All arms included Voluntary Counselling and Testing. The comparison therefore concerns the added intervention assigned alongside VCT: ESTEEM or CMHT, compared with VCT alone, as well as the comparison between ESTEEM and CMHT. The registry record does not report the results needed to estimate the effect for any of these comparisons. It also does not report a statistical comparison of the arms for this primary outcome.

3. Trial Design

The registry record describes a randomized, parallel-group trial with single masking. It lists an enrollment of 254 and a completed status. The recorded start was 2016-09, and primary completion was 2020-06-16. The phase is listed as NA. The sponsor was Yale University.

The three intervention arms were ESTEEM, Community Mental Health Treatment (CMHT), and Voluntary Counselling and Testing (VCT). All arms included VCT. Participants assigned to ESTEEM received the ESTEEM intervention as well; the record describes it as individually delivered cognitive behavioral therapy based on the Unified Protocol, with 10 sessions. Participants assigned to CMHT received that treatment in addition to VCT. Participants assigned to the VCT-only arm did not receive a further intervention. The registry record describes VCT as a 45-minute session based on CDC guidelines.

Random assignment is intended to make treatment groups comparable, on average, with respect to factors that could influence outcomes. That purpose does not establish that the groups were identical at baseline, nor does it remove the need to examine how outcomes were measured and analyzed. The registry record does not report the method used to generate the allocation sequence or whether allocation was concealed.

Single masking means the record identifies the study as having one masked party, but it does not identify that party in the information. It also does not report whether masking was maintained throughout follow-up. Since the interventions differ in content, the record alone does not establish which participants or study personnel could have been unaware of assignment.

The parallel design assigns participants to an intervention arm and compares outcomes across arms rather than switching the same participants among treatments. The registry record does not report the planned follow-up schedule beyond the primary endpoint’s 8-month timeframe. Nor does it report a rationale for the enrollment target or a full protocol-level statistical analysis plan.

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: Intention to Treat. One participant in each condition did not provide data for this outcome measure; overall number of participants analyzed does not include these participants.

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: The total number of participants randomized was 254, but 1 participant dropped out following randomization and did not complete the baseline interview.

6. Primary Endpoint

The registry lists the primary outcome as Any Condomless Anal Sex Acts, measured over 8 Months.

The primary endpoint is the presence or absence of any condomless anal sex (CAS) during the assessment period, subject to two exclusions: the act is not counted if the participant was using PrEP or if an HIV-positive primary partner was known to have an undetectable viral load. The registry record specifies a timeframe of 8 Months and defines the measure as a count of participants. The outcome description says the Time-Line Follow-Back Interview (TLFB) assesses behavior during the previous 3 months.

A yes/no endpoint turns a reported behavior into a binary outcome: each participant with usable data is classified as having or not having the specified event. Counting participants in these categories can make an outcome easier to interpret than a measure of the total number of acts. However, it does not preserve how often the behavior occurred or describe differences in frequency among participants who report it.

The endpoint is relevant to the trial’s stated focus on sexual health and HIV risk. Its definition is deliberately narrower than all CAS: it excludes acts under the stated PrEP or partner viral-load conditions. Interpretation therefore depends on the operational definition, the participant’s reports, and the assessment window. The registry record does not report the arm-specific counts for this endpoint or a treatment-effect estimate. Consequently, the direction and precision of any difference between arms cannot be assessed from the recorded information.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does randomization contribute?

General: Randomization uses chance to assign participants to study arms. Its purpose is to reduce systematic differences in measured and unmeasured characteristics between arms, so that an outcome contrast can more credibly be associated with assigned intervention. Randomization does not guarantee balance in a particular trial, and it does not prevent missing outcome data or errors in measurement. The record identifies the allocation as randomized but does not report the sequence-generation or concealment procedures.

The trial is listed as parallel, so participants were assigned to an arm rather than receiving each intervention in turn. The record describes single masking but does not identify who was masked. It is therefore not possible to assess from the registry record whether participants, clinicians, outcome assessors, or analysts were unaware of assignment.

How should a binary endpoint be understood?

General: A binary outcome places each observed participant into one of two categories, such as event present or event absent. A count of participants with the event is not the same as a count of event episodes. For this trial, the primary outcome is any qualifying CAS, not the number of CAS acts. The registry record says the outcome measure is a count of participants and defines the event using PrEP and partner viral-load conditions.

For a comparison between arms, an analysis would ordinarily need both the event counts and the number of participants with outcome data in each arm. Those quantities allow the event frequency to be characterized and the comparison to be estimated. The registry record does not report the primary arm-specific event counts, denominators, or effect estimate, so that comparison cannot be reconstructed from the information available.

What would a confidence interval add?

General: A confidence interval expresses uncertainty around an estimated effect. It helps readers distinguish a precise estimate from one that remains compatible with a broad range of effects. An interval must be interpreted alongside the effect measure and the study design; it is not a guarantee that the true effect lies within the reported limits. No confidence interval for the primary outcome is reported in the registry record, so its precision cannot be described.

What does the sample-size illustration mean?

The registry record describes a planning illustration computed from the observed effect, not from the trial’s own design. Its stated assumptions are a two-sided significance level of 0.05, equal allocation between two arms, a normal approximation, and an effect to detect defined as a stated fraction of the observed effect. The record does not report a sample-size justification for the actual trial design in the registry record.

General: A sample-size calculation estimates the number of participants needed under chosen assumptions about the effect, variability, significance level, power, allocation, and analysis. Changing an assumption can change the planning estimate. A calculation based on an observed effect is an illustration of what a future study might require under those assumptions; it is not evidence that the completed trial had adequate power, and it is not observed power. The planning assumptions described in the record refer to an equal allocation between two arms, so they should not be taken as a complete description of the three-arm trial’s design or analysis.

Why can over-dispersion matter?

General: Over-dispersion occurs when observed variability is greater than a model assumes. If a model underestimates variability, standard errors can be too small and confidence intervals too narrow, which can make evidence appear more precise than it is. The primary endpoint is binary, not a count of repeated event episodes, and the registry record does not report an over-dispersion model or an analysis that would allow its relevance to this endpoint to be assessed. The record also does not specify the statistical model, adjustment variables, test, or method used for the primary comparison.

How should the time window be read?

The registry lists an 8 Months timeframe for the primary outcome, while the outcome description says TLFB assesses behavior during the previous 3 months. These statements refer to the endpoint’s follow-up timing and the recall period used to assess behavior. The registry record does not provide further detail on the assessment schedule or how the recall period was aligned with the endpoint visit. General: A defined recall window helps specify what behavior contributes to an outcome, but it does not eliminate possible recall or reporting error. A yes/no measure also omits frequency information that a count of acts could retain.

For interpretation, the result would need to be considered together with the endpoint definition, arm-specific data, analysis population, and statistical method. The record describes the outcome as an intention-to-treat measure, but notes that participants without outcome data were excluded from the analyzed count. It does not report how missingness was handled analytically. Without the outcome data and method details, no primary effect estimate or uncertainty assessment can be independently examined.

10. Multiplicity, Missing Data and Analysis Populations

The primary outcome is one part of a broader set of measures. The registry record lists secondary outcomes on minority stress using GRRS and IHS measures and reports 58 outcome measures overall. It does not report a multiplicity strategy, such as whether the primary comparison was designated as the sole confirmatory test, whether secondary analyses were adjusted, or how results across multiple outcomes were to be interpreted. When many outcomes or comparisons are examined, the chance of finding at least one apparently notable result by chance can increase. General: multiplicity procedures are used to control or describe that risk, but the appropriate approach depends on the analysis plan and scientific goals.

The primary outcome population is described as intention to treat. The record also says one participant in each condition did not provide data for the primary outcome and that the overall number analyzed excludes these participants. The number of participants assigned to each arm and the number with outcome data by arm are not reported in the registry entries. Neither are the reasons for missingness, patterns of missing data, or any imputation or sensitivity-analysis method.

General: An intention-to-treat analysis usually aims to analyze participants according to their randomized assignment, regardless of adherence. Missing outcome values create a separate problem: the analysis must specify how those values are handled, and exclusions can affect what the estimate represents. The registry record’s use of the intention-to-treat label alongside exclusion of participants without primary outcome data does not establish the precise analysis procedure. The method for reconciling that label with missing outcomes is not reported in the registry record.

Without an explicit multiplicity plan and missing-data methods, the primary and secondary findings cannot be assessed for those aspects of analysis. No claims about adjusted or unadjusted statistical significance, imputation, sensitivity analyses, or the handling of repeated measurements can be made from the registry information.

11. 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 5% reporting threshold.

Most frequent serious adverse event terms

Serious events with the highest arm-level rate: active suicidality (4.0%).

The registry record reports serious adverse event counts by arm: 4 for ESTEEM, 0 for Voluntary Counselling and Testing, and 1 for Community Mental Health Treatment. The corresponding at-risk counts are 100, 52, and 102. It lists deaths as 0 in each arm and other adverse events as 0 in each arm. Active Suicidality is identified as the top serious term, at 4.0%.

These are safety summaries, not the primary endpoint results. The record does not provide further detail in the registry entries on event timing, severity grading, relatedness to the intervention, or whether the counts refer to participants with events or event episodes. The term at 4.0% is reported as the top serious term; the record information does not provide further detail on its calculation or attribution by arm.

General: Adverse event counts are descriptive and should be interpreted in relation to the monitoring definitions, follow-up period, and number of participants at risk. A zero count means no event was recorded in that category in the registry summary; it does not establish that an event is impossible or that risks are identical across groups. Differences in counts alone do not establish a causal safety effect. The available figures do not support a fuller comparative safety analysis.

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

The registry record supports a description of the study’s design and question: young gay and bisexual men were assigned in a randomized, parallel-group trial to ESTEEM, CMHT, or VCT, with VCT included in all arms. The primary outcome was any CAS during the stated assessment period, excluding the specified PrEP or known undetectable-partner-viral-load circumstances. The record also provides selected safety counts.

It does not provide primary outcome counts by arm, the number analyzed in each arm, a treatment-effect estimate, or a confidence interval. Therefore, the posted information does not show whether the arms differed on the primary outcome, how large any difference might have been, or how precisely it could be estimated. The absence of these results is not evidence that the interventions had equal outcomes; it is a limit on what can be concluded from the registry record.

The trial’s stated aim was to test an intervention addressing minority-stress pathways and co-occurring mental, behavioral, and sexual health problems. That motivation does not establish a finding about those problems. The registry entries do not provide results for the listed minority-stress measures or a full set of results for other outcomes. Nor do they identify a statistical analysis plan or permit examination of the trial’s handling of multiple outcomes.

The sample-size assumptions described in the record are planning illustrations based on an observed effect, not observed power and not a report of the trial’s original design justification. They cannot fill the gap left by missing outcome estimates. Likewise, the completion status and enrollment figure describe the trial record but do not establish efficacy, safety, or precision.

In short, the record defines a clinically focused randomized comparison, but the information available for this analysis is insufficient to judge the primary treatment contrast. Any interpretation should stay within that boundary: study design and outcome definition are documented, while primary comparative results and their uncertainty are not reported in the registry record.

13. Why This Trial Matters Statistically

This trial poses a statistical question that is more specific than its broad intervention rationale. It aims to address minority-stress pathways affecting mental, behavioral, and sexual health, while its primary endpoint focuses on whether participants reported any qualifying CAS in a defined assessment period. That distinction matters because a binary endpoint captures occurrence, not frequency, severity, or the wider set of experiences that the intervention is intended to address.

The three-arm structure also makes the choice of comparison important. ESTEEM and CMHT were each paired with VCT, while VCT alone served as an arm in its own right. A conclusion about one pair of arms would not automatically answer the comparison between the other pair. The registry record does not report primary outcome contrasts, so none of these comparative questions can be resolved from the available results.

From a biostatistical perspective, the record illustrates why a trial’s question, endpoint definition, analysis population, missing-data handling, and reported effect estimate all need to align. The primary outcome has explicit conditions for what counts as an event and a specified recall description. Yet one participant in each condition lacked primary outcome data, and the method used to handle missingness is not reported. Those details affect how readers understand the population represented by an estimate.

The record also lists many outcome measures, including minority-stress measures, but does not report a multiplicity strategy. When a trial examines a broad set of outcomes, transparent prespecification and reporting help readers distinguish confirmatory questions from exploratory analyses. In the absence of those details and the outcome results, the strongest statistical contribution of the registry record is to clarify what was intended to be studied—not what the study found.

14. 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 principal limitation for interpretation is that the registry record does not report the primary outcome results needed for an arm comparison. There are no arm-specific event counts, analysis denominators, effect estimates, confidence intervals, or primary-outcome tests in the registry record. This prevents assessment of the direction, magnitude, or precision of the primary comparison.

Additional analysis details are not reported in the registry record. These include the allocation-sequence method, allocation concealment, identity of the masked party, statistical model and test, covariate adjustment, and analysis-plan details. The record does not explain how the stated intention-to-treat population was handled when one participant in each condition lacked primary outcome data. Reasons for missingness, imputation, and sensitivity analyses are also not reported.

The endpoint is based on self-reported behavior assessed by TLFB, with a previous-3-month recall period and an 8 Months timeframe. General: Self-report and recall can introduce measurement error, while a binary any-event definition does not retain frequency information. The registry record does not report procedures that would establish the accuracy of recall or the timing details needed to assess how the endpoint window was implemented.

The trial’s broader focus includes mental, behavioral, and sexual health, but the primary outcome represents only the defined CAS measure. The record lists secondary minority-stress measures but does not report their results or a multiplicity plan. It also provides limited safety summaries without further detail on event timing, relatedness, or the precise meaning of the count fields. These omissions limit both the interpretation of findings and the ability to compare safety outcomes.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: A Unified Intervention for Young Gay and Bisexual Men's Minority Stress, Mental Health, and HIV Risk — NCT02929069.
• Registry record last updated 2022-10-18; status last verified 2022-09.
• Results first posted 2022-01-04.
• 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.

18. Record Summary

NCT02929069 is a completed, randomized, parallel-group trial with single masking, sponsored by Yale University. The record lists enrollment of 254, a start of 2016-09, and primary completion on 2020-06-16. The phase is listed as NA.

The arms were ESTEEM, CMHT, and VCT, with VCT included in all arms. The primary outcome was any qualifying CAS at 8 months, assessed using TLFB. The registry record does not report arm-specific primary outcome results or an effect estimate. It notes missing primary outcome data for one participant in each condition and states that these participants are excluded from the analyzed count.

Serious adverse event counts were 4 for ESTEEM, 0 for VCT, and 1 for CMHT; deaths and other adverse events were listed as 0 in each arm. Active Suicidality was listed as the top serious term at 4.0%. Statistical details for the primary comparison, missing-data handling, and multiplicity are not reported in the registry record.