Addressing the Continuum of Care Among High-risk Thai Men: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Addressing the Continuum of Care Among High-risk Thai Men (NCT05161689), a randomized na study. The primary outcome is Prevalence of HIV Testing in the Past 6 Months Among YMSM at Baseline.

2. Clinical Question

The primary endpoint concerns recent HIV testing at baseline. The posted results report 92 participants in the Mpowerment-based intervention arm and 48 in the standard-of-care arm answering yes. Each arm has 318 participants in the posted outcome counts. The unadjusted risk difference is 0.138, with a 95% confidence interval from 0.075 to 0.202; the unadjusted p-value is < 0.001.

The time point matters to interpretation. The endpoint is explicitly assessed at baseline, while the study description frames HUG-M+ as an intervention to be implemented and evaluated. A baseline comparison describes the groups before the intervention could have produced a subsequent outcome difference. The result therefore documents a difference in reported prior testing between the randomized groups at baseline; it does not by itself estimate the intervention’s effect on later testing or on the wider continuum of prevention and care.

The record describes two cities, with one assigned to the intervention condition and the other to control. It does not name the cities or give details of how participants were sampled or recruited. Those details, along with the endpoint’s wording and baseline timing, are important context for understanding exactly which clinical and behavioral question the posted comparison addresses.

3. Trial Design

This is a randomized, parallel-group study with no masking reported. The registry record lists 636 participants enrolled and a completed status. The study dates are 2021-11-27 to 2024-12-10 for start and primary completion, respectively. Mahidol University is listed as the sponsor, and the conditions include HIV/AIDS and STD.

The study description focuses on young Thai men who have sex with men and HIV prevention. It says that the investigators partnered with the Thailand Ministry of Public Health and aimed to finalize, implement, and evaluate HUG-M+, a multicomponent, multilevel community mobilization and combination intervention. The stated aim was to address the HIV continuum of prevention and care. One Northeastern Thai city was randomized to the intervention condition and one to the control condition, where standard of care was provided. The registry record does not name the cities.

The listed intervention is HUG-M+, while the outcome arm is titled Mpowerment-based Intervention. Its description says participants may be exposed to a multicomponent, multilevel community mobilization and combination intervention. The control arm is titled Standard of Care and is described as not exposed to the intervention. The registry record does not give further details about intervention delivery, adherence, or the content of standard of care.

The primary outcome is the count of participants answering yes to a question about HIV testing in the past six months, assessed at baseline. This is a binary response summarized as counts. Each arm has 318 participants in the posted outcome data. Although the study is listed as completed and reports 10 outcome measures, no secondary outcome results are posted in the registry record. The registry record does not report the trial’s original sample-size justification, detailed analysis plan, or the methods used for handling the city-level assignment in the outcome analysis.

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 Prevalence of HIV Testing in the Past 6 Months Among YMSM at Baseline, measured over Assessed at baseline.

The primary endpoint counts participants who answered yes when asked whether they had tested for HIV in the past six months. It is assessed at baseline, and the unit is participants. The endpoint therefore records a binary response for each participant and reports the number of yes responses in each arm. The posted result is 92 of 318 in the Mpowerment-based intervention arm and 48 of 318 in the standard-of-care arm.

Recent testing is relevant to the study’s stated focus on the continuum of HIV prevention and care. As a measure, it captures whether participants report a testing event within a specified recent period; it does not, by itself, describe other aspects of prevention or care. The registry record does not provide a fuller description of the survey instrument, how testing was verified, or how missing or ambiguous answers were treated.

The baseline timing limits the question answered by this endpoint. It measures testing history before the intervention could have changed testing behavior. Thus, the counts can describe how the randomized groups differed at baseline, but they cannot establish a change caused by HUG-M+. A later measurement would be needed to assess post-intervention testing, and the registry record does not post a later result for this primary outcome.

7. Results Posted in the Registry

• Mpowerment-based Intervention: 28.9% (92/318) versus 15.1% (48/318) on control; risk difference 13.8 percentage points (95% CI 7.5 to 20.2), risk ratio 1.92 (1.40–2.62); unadjusted two-proportion z-test p = <0.001.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is binary: each participant’s response is classified as yes or not yes to whether they tested for HIV in the past six months. The posted outcome is a count of participants answering yes, rather than a continuous measurement such as a score or a time-to-event outcome. The comparison uses the counts and arm sizes to estimate the difference between the two arms’ outcome probabilities.

What does the risk difference mean?

The unadjusted risk difference is 0.138, with the intervention arm subtracted from the control arm in the direction defined by the comparison. A risk difference is an absolute scale: it expresses how far apart the two outcome probabilities are. It is often useful when readers want to understand the size of a probability gap without expressing it as a ratio. The estimate does not adjust for participant characteristics, city, or other variables; the registry record does not report an adjusted analysis.

The confidence interval for this risk difference is 0.075 to 0.202. In general, a confidence interval gives a range of values compatible with the data and the method’s assumptions at the stated confidence level. It reflects sampling uncertainty for the estimate under that method, not uncertainty about every feature of the study design. The registry record does not specify the interval method used for the posted recalculation.

How do risk ratios and odds ratios differ?

The risk ratio is 1.917, with a confidence interval from 1.402 to 2.62. It compares the outcome probabilities by division, so a value above 1 corresponds to a higher observed probability in the intervention arm in this comparison. The odds ratio is 2.29, with a confidence interval from 1.549 to 3.386. Odds are a transformation of probability, and an odds ratio compares those odds rather than the probabilities themselves. For this outcome, the two ratios are different numerical summaries; the odds ratio should not be read as a risk ratio.

These ratio estimates and the risk difference are derived from the same posted counts. Considering more than one scale can help clarify whether a result is being described in absolute or relative terms, but it does not create independent evidence. Each estimate is unadjusted and pertains to the baseline data.

What does the p-value tell the reader?

The posted unadjusted p-value is < 0.001. In general, a p-value measures how unusual data at least as incompatible with a null hypothesis as the observed data would be, assuming the null model and its conditions. It is not the probability that the null hypothesis is correct, and it does not measure the size or practical importance of the observed difference. It also does not address whether a measured difference occurred after an intervention.

The result is labeled unadjusted because it is based on the basic comparison of the two arms’ reported counts, rather than a model controlling for other variables. The registry record lists no formal analyses, and no adjustment strategy, covariates, or model specification is reported. Thus, the posted significance result should not be described as an adjusted result.

Why does design matter for the uncertainty?

The trial description says two cities were involved, with one randomized to intervention and the other to control. Participants within the same city may share contextual influences. In general, outcomes from participants assigned through a shared cluster can be correlated, and an analysis that treats every participant as fully independent may understate uncertainty if it does not account for that correlation. The registry record does not state how the posted comparison handles city-level assignment, and the simple unadjusted recalculation does not incorporate a reported cluster adjustment.

With one city assigned to each condition, differences between cities and differences between conditions are linked in the described allocation. Randomization at the city level does not make the two cities identical, and a participant-level comparison cannot by itself separate a city effect from an intervention effect. This is a design limitation for interpreting any between-condition contrast. It is especially important not to treat the baseline difference as an intervention effect: the outcome was assessed at baseline, before an intervention effect on testing could be observed.

General statistical concepts help explain the posted estimates, but the registry record does not identify the calculation methods beyond the unadjusted summary comparison. The number of participants with missing endpoint responses, the analysis population, the handling of missingness, and any prespecified primary analysis are not reported in the registry record. Those details would matter in assessing how closely the posted counts represent all randomized or enrolled participants and whether the uncertainty estimate matches the intended design.

10. Confidence Intervals and Precision

The risk-difference estimate is 0.138, and its 95% confidence interval ranges from 0.075 to 0.202. This interval conveys the precision of the unadjusted estimate under the calculation’s assumptions. Because both limits are on the same side of zero, the interval is consistent with a positive baseline difference under that method. It does not establish that the intervention caused the difference, since the endpoint is assessed at baseline.

In general, a confidence interval is a range generated by a statistical procedure that would cover the underlying parameter at the stated rate over repeated applications, if the model assumptions hold. It is not a probability statement about the particular fixed parameter after observing the data. Nor does a narrow-looking interval account automatically for bias, unmeasured differences, clustering, or limitations in how the endpoint was collected.

The risk-ratio interval is 1.402 to 2.62, around an estimate of 1.917. The odds-ratio interval is 1.549 to 3.386, around an estimate of 2.29. Their scales differ from that of the risk difference, so the endpoints cannot be directly compared across measures. All three estimates summarize the same baseline counts rather than separate outcomes.

The registry record does not report the method used to construct these intervals or whether any clustering correction was applied. The two-city allocation is relevant because city-level similarity among participants may affect uncertainty. The displayed intervals should therefore be read as precision estimates for the posted unadjusted calculation, not as proof that all design-related uncertainty has been captured.

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 table and curve are planning illustrations based on the observed effect, not the trial’s original design calculation. The planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The planning effect is set to a stated fraction of the observed effect. The figures are participants per arm, not total participants or total events.

The columns for 80% and 90% power describe the number of participants per arm required, under those planning assumptions, to have the stated chance of detecting the selected effect if that effect were the true one. Power is a property of a planned design under assumptions; it is not a measure of how likely the observed result is to be true. The displayed figures are not observed power.

The rows consider planning for 100% of the observed effect, 75% of the observed effect, and 50% of the observed effect. A smaller effect is harder to distinguish from random variation, so it requires more participants per arm to reach the same target power. The figures increase as the planned effect becomes smaller, reflecting that general relationship.

These calculations use the observed baseline difference as their effect reference. Since the endpoint was measured at baseline, this is not a post-intervention treatment effect estimate and should not be taken as a direct forecast of the sample size needed for a future intervention evaluation. The registry record does not report the original sample-size justification or its assumptions.

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 = 0.151; treatment proportion p₁ = 0.289; average p̄ = 0.220.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.5860 = 1.1484.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.5777 = 0.4862.
4. n = (1.1484 + 0.4862)² / (0.1384)² = 139.6, rounded up to 140 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 <- 0.150943396226415                          # control proportion
p1 <- 0.289308176100629                          # 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(5161689)
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 = 140 (80% power), 187 (90% power)
 75% of observed effect: n per arm = 235 (80% power), 314 (90% power)
 50% of observed effect: n per arm = 495 (80% power), 662 (90% power)
power.prop.test: 139.6 
Simulated power at n = 140 per arm: 0.805 (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 lists 10 outcome measures, but no secondary outcome results are posted. It lists no formal analyses in the registry analysis records. The available primary comparison is therefore not accompanied by a posted analysis plan explaining how outcomes were prioritized or how multiple comparisons were handled.

When a study examines multiple outcomes or multiple analyses, the chance of at least one apparently notable result can increase if no plan addresses multiplicity. In general, multiplicity procedures can control that risk, but the choice depends on the analysis plan and the role of each outcome. The registry record does not report whether any multiplicity adjustment was planned or used. The single posted p-value should not be assumed to have been adjusted for other outcomes or analyses.

The record does not report the number of participants with missing responses for this endpoint, the reasons for missingness, or how missing responses were handled. It also does not specify whether the analysis included all enrolled participants, all randomized participants, or another analysis population. The outcome counts show 318 participants in each arm, but the registry record does not explain how those counts relate to missing or excluded observations.

In general, missing-data methods rely on assumptions about why measurements are absent. If missingness differs between arms or is related to the outcome, a complete-case comparison can differ from a comparison using all intended participants. Without a missing-data account, readers cannot assess that possibility from the registry record.

The primary outcome is explicitly assessed at baseline. The registry record does not provide a later primary-outcome comparison or an analysis of change over time. It also does not report adjusted analyses, covariate choices, model diagnostics, or a plan for the city-level assignment. These omissions limit the ability to evaluate the intended inferential strategy and its relationship to the randomized design.

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.

For the Mpowerment-based intervention arm, the registry record lists 318 participants at risk for serious adverse events and 318 at risk for other adverse events. It reports 0 serious adverse events, 0 other adverse events, and 0 deaths. For standard of care, it likewise lists 318 participants at risk in each category, with 0 serious adverse events, 0 other adverse events, and 0 deaths.

These are the adverse-event figures posted in the registry record. They should be read as reported counts, not as proof that no adverse events occurred outside the scope of what was recorded or submitted. The record does not provide adverse-event definitions, observation windows, collection procedures, or an explanation of whether the risk populations cover all enrolled participants throughout follow-up.

No serious adverse event terms are listed. The zero counts therefore do not support a detailed comparison of event types or timing, and the registry record does not give information about severity beyond its serious and other categories. These data describe the posted safety record; they do not establish a broader conclusion about safety.

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

The result establishes that the registry record reports different baseline counts of participants answering yes to HIV testing in the past six months: 92 in the Mpowerment-based intervention arm and 48 in standard of care, with 318 participants in each arm. The unadjusted risk difference is 0.138, with a 95% confidence interval from 0.075 to 0.202 and an unadjusted p-value of < 0.001. This is evidence of a baseline difference in the posted data under the unadjusted comparison.

It does not show that HUG-M+ increased HIV testing. The endpoint was assessed at baseline, so it predates any effect of the intervention on the measured behavior. The study description states an intention to evaluate a community mobilization intervention, but the posted primary result is not a post-intervention comparison. No later result for this primary endpoint is reported in the registry record.

The result also does not establish that the observed baseline difference was caused by any particular participant or city characteristic. The record describes one city assigned to each condition and does not report an adjusted or city-aware analysis. Since the city and condition coincide in that design description, the posted participant-level contrast cannot by itself distinguish a city difference from a condition difference.

Nor does the result describe every part of the continuum of prevention and care. The endpoint is a yes/no report of recent HIV testing; it does not measure all prevention services, care engagement, or other outcomes. Although 10 outcome measures are listed, secondary results are not posted. The registry record also does not report missing-data handling or the analysis population.

The appropriate interpretation is therefore bounded: the posted counts document a statistically distinguishable baseline difference in reported recent testing between the two arms. They do not provide an estimate of intervention efficacy, a post-intervention change, or a general conclusion about treatment benefit or harm.

16. Why This Trial Matters Statistically

This study highlights why the timing of an endpoint is as important as its statistical significance. A small p-value and a confidence interval that does not include zero can describe a clear baseline difference, but they do not make that difference an intervention effect. The distinction is central when a trial’s stated purpose is evaluation and the available primary outcome is assessed before intervention exposure could alter it.

The trial description also makes the assignment level important. The registry record describes two cities, with one city assigned to each condition. When condition is assigned at a shared community level, participant outcomes may be correlated within city, and differences between cities can be difficult to separate from differences attributed to intervention. Statistical methods must reflect the randomization structure if uncertainty and causal interpretation are to be appropriate. The registry record does not report that analysis detail.

The contrast between the simple posted comparison and the described design is useful for biostatistical reading. Participant counts alone do not establish that participants are the independent randomization units. Similarly, the equal arm counts do not answer whether the analysis accounted for clustering, missingness, covariates, or multiplicity. Those are separate questions about design and analysis, and their details are not reported in the registry record.

Finally, the result shows why baseline measures should be distinguished from outcomes measured after assignment. Baseline data can describe initial group differences and may inform subsequent analyses, but they cannot demonstrate a change attributable to the intervention. Clear reporting of time points, analysis populations, and design-aware methods is necessary to connect a trial’s stated objective to the statistical evidence posted for its outcomes.

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 most consequential limitation for interpretation is that the primary endpoint is assessed at baseline. The posted comparison therefore cannot estimate the effect of the intervention on subsequent HIV testing. The registry record does not post a later result for this endpoint, so no conclusion about post-intervention testing can be drawn from these counts.

The study description identifies two Northeastern Thai cities, with one assigned to intervention and one to control. This leaves city and condition linked in the described assignment. The record does not report a method to separate city-level differences from intervention differences, and the posted recalculation is unadjusted. The registry record also does not say whether the analysis accounts for similarity among participants within a city.

Other analysis details are not reported in the registry record: the original sample-size justification, formal statistical analysis plan, analysis population, missing-data handling, outcome collection and verification procedures, and any adjusted estimates. The record lists 10 outcome measures but posts no secondary results and no formal analysis records. These gaps limit assessment of prespecification and of how representative the counts are of all participants intended for analysis.

The posted primary outcome is a self-reported yes/no measure of testing in the past six months, as described in the registry record. The record does not report how responses were verified or how nonresponses were treated. The adverse-event record lists zero events in its categories, but does not explain collection procedures or observation windows. These reporting limits should be kept distinct from the reported numerical results.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Addressing the Continuum of Care Among High-risk Thai Men — NCT05161689.
• Registry record last updated 2025-12-26; status last verified 2025-12.
• Results first posted 2025-12-08.
• 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

NCT05161689, titled Addressing the Continuum of Care Among High-risk Thai Men, is listed as a completed randomized, parallel-group study with no masking and 636 participants enrolled. Mahidol University is the sponsor. The registry record gives a start date of 2021-11-27 and a primary completion date of 2024-12-10.

The study description concerns YMSM in two Northeastern Thai cities, with one city assigned to HUG-M+ and the other to standard of care. The primary outcome counts participants reporting HIV testing in the past six months at baseline. The posted arm counts are 92 of 318 for the Mpowerment-based intervention arm and 48 of 318 for standard of care.

The unadjusted risk difference is 0.138, with a 95% confidence interval from 0.075 to 0.202 and p < 0.001. The baseline timing means this is not an estimate of intervention effect. Secondary outcome results, formal analysis records, and details of the original sample-size justification are not reported in the registry record.