Still Climbin': An Intervention to Improve Coping Among Black Sexual Minority Men: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Still Climbin': An Intervention to Improve Coping Among Black Sexual Minority Men (NCT04587869), a randomized na study. The primary outcome is Inadequate Healthcare Utilization.

2. Clinical Question

This is a comparison of a defined utilization outcome, not a direct measure of every objective named in the intervention description. The primary endpoint focuses on a particular combination of relatively low ambulatory care use and emergency department use or hospitalization. The registry also lists adaptive coping strategies (social support seeking) as a secondary outcome, but the primary comparison is inadequate healthcare utilization.

The comparison is between randomized groups, but the registry record does not report enough detail to establish how participants were recruited, what eligibility criteria were applied, or which people were masked. These details are not reported in the registry record. The record lists the study status as terminated; the reason for termination is not reported in the registry record.

3. Trial Design

The registry describes Still Climbin' as a randomized controlled trial with a parallel-group model and single masking. It lists an enrollment of 369 and the study status as terminated. The record gives a start date of 2021-02-09 and a primary completion date of 2025-03-20. The reason for termination is not reported in the registry record.

The intervention was an 8-session CBT group intervention addressing coping with discrimination and medical mistrust among Black sexual minority men. The comparator was a no-treatment control: participants assigned to that group would not receive the intervention. The registry identifies RAND as the sponsor. The identity of the party or parties masked under the single-masking designation is not reported in the registry record.

The primary outcome was inadequate healthcare utilization, measured from baseline to 12-month post-baseline. The description refers to follow-up assessments at 4, 8, and 12 months post-baseline. The registry reports outcome counts for the intervention and no-treatment control groups, alongside a logistic regression result. It also lists three outcome measures and three analyses, but the record does not provide a complete analysis plan or explain how those analyses relate to one another.

Randomization can support a comparison of groups by assignment, but it does not by itself resolve missing outcome data, guarantee that follow-up was complete, or explain the statistical model. The outcome was missing for participants who did not complete any follow-up survey, and nonresponse weights were used to address potential bias. Details of the allocation process, masking procedures, and the original statistical design assumptions are not reported in the registry record.

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

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The outcome is missing for all participants (n=53) who did not participate in any follow-up surveys (nonresponse weights were used to account for potential bias). The outcome was never missing among those who participated in any follow-up surveys.

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 Inadequate Healthcare Utilization, measured over baseline to 12-month post-baseline.

The primary endpoint is a binary indicator of inadequate healthcare utilization. A participant met the outcome definition if they reported fewer than 1 ambulatory visit and also reported at least 1 emergency department visit without subsequent hospitalization, or at least 1 hospitalization, in the past 4 months. The registry describes assessment at 4, 8, and 12 months post-baseline and gives the overall timeframe as baseline to 12-month post-baseline.

This endpoint combines low ambulatory care use with emergency or inpatient care. It is used to represent a particular pattern of healthcare utilization that the trial sought to address through an intervention intended to improve healthcare engagement. It should not be read as a comprehensive measure of access, quality, preventive care, or all healthcare use. Nor does the endpoint by itself establish why a participant used or did not use a type of care.

The registry reports participant counts for this endpoint: 41 in the intervention group and 39 in the no-treatment control group. These counts are reported alongside denominators of 148 and 168, respectively. The endpoint was missing for all participants (n=53) who did not participate in any follow-up surveys. The registry says nonresponse weights were used to account for potential bias and that the outcome was never missing among people who participated in any follow-up survey. The construction of those weights and the details of how information from the follow-up assessments was handled are not reported in the registry record.

7. Results Posted in the Registry

• Intervention: 27.7% (41/148) versus 23.2% (39/168) on control; risk difference 4.5 percentage points (95% CI -5.1 to 14.1), risk ratio 1.19 (0.82–1.74); unadjusted two-proportion z-test p = 0.360.

8. Statistical Interpretation

9. Statistical Methods Explained

The registry identifies the primary outcome as a count of participants meeting a binary definition. In statistical terms, each participant is classified as either meeting or not meeting the definition of inadequate healthcare utilization. The reported group counts are then used to compare the frequency of that outcome between the randomized intervention and control groups.

What does a risk difference express?

A risk difference compares the probability of an outcome between groups by subtracting the control-group probability from the intervention-group probability. It is expressed on the absolute probability scale. A positive value means the outcome was more frequent in the intervention group in the observed comparison; a negative value means it was less frequent. The unadjusted recalculation is 0.045, with a confidence interval from -0.051 to 0.141. Because the interval spans values on either side of no difference, the estimate does not tightly pin down the direction of the difference.

What do the risk ratio and odds ratio express?

A risk ratio compares the probability of the outcome between groups as a relative measure. An odds ratio compares the odds rather than the probabilities. Both are relative measures, so their numerical values are interpreted differently from a risk difference. For the unadjusted recalculation, the risk ratio is 1.193 and the odds ratio is 1.267. Their confidence intervals are 0.817 to 1.743 and 0.763 to 2.106, respectively. For these measures, a value of 1 represents no relative difference. The intervals include that value, indicating uncertainty about whether the outcome was relatively more or less frequent in the intervention group.

What does the p-value add?

A p-value describes how compatible the observed data are with a specified null model, under the assumptions of the analysis. It is not the probability that the null hypothesis is true, nor a measure of the size or clinical importance of an effect. The unadjusted recalculation has p=0.36, while the sponsor's reported logistic regression result has p=.44. Neither value alone says that the groups are equivalent. The estimates and confidence intervals are needed to assess the direction, range, and precision of the comparison.

Why use logistic regression?

Logistic regression is a method for modeling a binary outcome. It commonly reports an odds ratio for a comparison. The registry lists the sponsor's method as logistic regression and reports an odds ratio of 1.23, with a 95% confidence interval from .73 to 2.07. The record does not state which covariates, if any, were included in that model, so its adjustment details are not reported in the registry record. The recalculated measures are simple unadjusted calculations from the posted summary statistics; they are not a replacement for a model using participant-level data.

Why do confidence intervals matter?

A confidence interval conveys uncertainty around an estimated effect. In repeated use of the same statistical procedure under its assumptions, intervals formed in this way have a stated coverage property. For interpreting this trial result, the practical point is that the interval marks a range of values compatible with the observed data and method. A wide interval signals limited precision; an interval that includes no difference does not demonstrate that the true effect is exactly zero. The risk-difference interval extends from -0.051 to 0.141, while the sponsor's odds-ratio interval extends from .73 to 2.07.

What is the role of nonresponse weighting?

The registry says that the outcome was missing for all participants (n=53) who did not participate in any follow-up surveys and that nonresponse weights were used to account for potential bias. Weighting is a statistical approach that can give different influence to observed participants to address differences associated with response. Its success depends on how weights are constructed and on assumptions about nonresponse and the information used. The registry says the outcome was never missing among participants who took part in any follow-up surveys, but does not report the weighting procedure, variables used to form weights, or sensitivity analyses. Those details are not reported in the registry record.

The distinction between randomized assignment and observed outcome data matters. Randomization defines the assigned groups, while missing follow-up means not every participant has a recorded outcome. A weighted analysis may seek to represent the target group despite nonresponse, but the result can still depend on unverified assumptions. The registry does not report the analysis population definition or a detailed account of whether all randomized participants were included in the sponsor's model. These analysis-population details are not reported in the registry record.

Finally, the registry lists three outcome measures and three analyses, but the multiplicity strategy is not reported in the registry record. When several outcomes or comparisons are assessed, statistical methods may account for the chance of false-positive findings across tests. Without a reported plan, it is not possible to tell whether or how that issue was addressed. No adjustment should be assumed from the available record.

10. Confidence Intervals and Precision

The unadjusted risk difference is 0.045, and its 95% confidence interval runs from -0.051 to 0.141. This interval includes no difference and spans values in both directions. It therefore leaves uncertainty not only about the size of the difference but also about whether the intervention group had a higher or lower frequency of the defined outcome than the control group.

The unadjusted risk ratio is 1.193, with a 95% confidence interval from 0.817 to 1.743. The unadjusted odds ratio is 1.267, with a 95% confidence interval from 0.763 to 2.106. Both intervals include 1, the value corresponding to no relative difference. The estimates should not be treated as exact measures of an effect: their intervals show that the data are compatible with a range of relative comparisons.

The sponsor's logistic regression result is an odds ratio of 1.23 with a 95% confidence interval from .73 to 2.07. That interval also includes 1. Its breadth reflects uncertainty around the estimate, and it does not rule out differences in either direction. The sponsor's p-value is .44. A p-value does not replace the interval as a description of precision.

These intervals quantify statistical uncertainty under the relevant analyses; they do not account for every source of uncertainty. In particular, the registry reports nonresponse weighting because outcomes were absent for people who did not participate in follow-up surveys. The details needed to assess the weighting assumptions are not reported in the registry record. Confidence intervals cannot, on their own, resolve potential bias from missing data, limitations of the endpoint definition, or the generalizability of the trial population.

11. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The sample-size figures are planning illustrations based on the observed effect, not the trial's original design assumptions. The registry record does not report the original sample-size justification or design assumptions. The planning calculations use a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set at the stated fraction of the observed effect.

The figures are participants per arm for 80% power and participants per arm for 90% power. They are not total events. The 80% power column describes the planned number per arm for a design with an 80% probability of detecting the specified effect under the planning assumptions, if that effect is truly present. The 90% power column describes the corresponding design with a 90% probability. These are planning concepts, not guarantees about the result of a particular study.

The rows describe planning for 100% of the observed effect, 75% of the observed effect, and 50% of the observed effect. For planning at 100% of the observed effect, the figures are 1,478 participants per arm for 80% power and 1,978 participants per arm for 90% power. For planning at 75% of the observed effect, they are 2,590 participants per arm for 80% power and 3,466 participants per arm for 90% power. For planning at 50% of the observed effect, they are 5,739 participants per arm for 80% power and 7,682 participants per arm for 90% power.

The pattern illustrates a general statistical principle: smaller effects are harder to distinguish from random variation, so detecting smaller effects requires more participants under otherwise comparable assumptions. These figures are planning illustrations and are not observed power for the completed trial. A sample-size calculation based on an observed effect should be interpreted cautiously because the observed estimate itself is uncertain.

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 = 0.232; treatment proportion p₁ = 0.277; average p̄ = 0.255.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6161 = 1.2075.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6153 = 0.5178.
4. n = (1.2075 + 0.5178)² / (0.0449)² = 1,477.5, rounded up to 1,478 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 <- 0.232142857142857                          # control proportion
p1 <- 0.277027027027027                          # 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(4587869)
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 = 1,478 (80% power), 1,978 (90% power)
 75% of observed effect: n per arm = 2,590 (80% power), 3,466 (90% power)
 50% of observed effect: n per arm = 5,739 (80% power), 7,682 (90% power)
power.prop.test: 1477.5 
Simulated power at n = 1,478 per arm: 0.803 (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. Analyses Reported by the Sponsor

The sponsor's reported analysis is a logistic regression comparison of the intervention and no-treatment control groups. It reports an odds ratio of 1.23, a 95% confidence interval from .73 to 2.07, and p=.44. The registry does not describe the model's covariates or other adjustment details.

The simple unadjusted recalculation from the posted summary statistics gives an odds ratio of 1.267, with a 95% confidence interval from 0.763 to 2.106, and an unadjusted p-value of 0.36. It also gives a risk difference of 0.045 and a risk ratio of 1.193. These recalculations are not the same as the sponsor's regression analysis: they use posted group summaries and do not reproduce a participant-level model.

The estimates are broadly similar in direction and uncertainty, and the registry record does not flag a discrepancy. The sponsor's analysis remains the reported trial analysis; the unadjusted calculations offer a transparent complementary comparison, not a substitute for it. The sponsor's odds-ratio interval includes 1, as do the recalculated relative-measure intervals, and the risk-difference interval includes no difference. Neither set of results establishes that the groups are identical or settles the size of any possible effect.

14. Multiplicity, Missing Data and Analysis Populations

The registry lists three outcome measures and three analyses. It identifies inadequate healthcare utilization as the primary outcome and adaptive coping strategies (social support seeking) as a secondary outcome. A multiplicity strategy, including whether any adjustment was planned or applied across outcomes or analyses, is not reported in the registry record. Without that information, the reported p-values should not be interpreted as if a particular multiplicity procedure had been used.

Missing data are described for the primary endpoint. The outcome was missing for all participants (n=53) who did not participate in any follow-up surveys. The registry says nonresponse weights were used to account for potential bias. It also states that the outcome was never missing among participants who participated in any follow-up surveys. The weighting method, variables used to construct the weights, and any sensitivity analyses are not reported in the registry record. The adequacy of the weighting assumptions therefore cannot be evaluated from the record.

The record reports outcome counts and denominators by group, but does not give a detailed definition of the analysis population used for the sponsor's logistic regression. It does not state whether the analysis followed an intention-to-treat approach, how assigned participants with missing outcomes contributed to the model, or whether additional exclusions were applied. These analysis-population and missing-data handling details are not reported in the registry record.

General statistical context: randomization helps make groups comparable at assignment, while nonresponse after assignment can affect the comparability of the observed outcome data. Weighting is one possible response to this issue, but it relies on assumptions about how response relates to measured information and the outcome. The posted description makes clear that weighting was used, but does not provide enough detail to judge those assumptions. Neither the presence of randomization nor the use of weights removes the need to consider uncertainty and possible bias from missing outcomes.

15. 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: stroke and other nervous system disorder (3.5%), cardiac-related hospitalization (2.9%), accidents (2.3%), general -surgery due to pre-existing conditions (2.0%), respiratory related hospitalization (1.8%).

The registry reports serious adverse events for the intervention group with 22 affected among 171 at risk, and for the no-treatment control group with 17 affected among 198 at risk. It reports deaths as 1 in the intervention group and 0 in the control group. Other adverse events are reported as 0 among 171 at risk in the intervention group and 0 among 198 at risk in the control group.

The groups at risk in these safety listings are not the same denominators as those reported for the primary outcome. The safety figures should therefore be read separately from the efficacy comparison, without substituting one set of denominators for the other. The registry also lists serious-event terms and percentages, including Stroke and other nervous system disorder at 3.5%, Cardiac-related hospitalization at 2.9%, Accidents at 2.3%, General -surgery due to pre-existing conditions at 2.0%, Respiratory related hospitalization at 1.8%, and Infection - hospitalization at 1.2%.

These registry figures describe reported events; they do not establish that an event was caused by the intervention. Details on event attribution, the definitions behind the listed categories, and comparative statistical testing are not reported in the registry record. The counts should not be interpreted as proof of either a safety difference or the absence of one.

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

The primary result is an uncertain comparison of the intervention and no-treatment control groups for the registry's defined indicator of inadequate healthcare utilization. The unadjusted risk difference is 0.045, with a 95% confidence interval from -0.051 to 0.141. The sponsor's logistic regression reports an odds ratio of 1.23, with a 95% confidence interval from .73 to 2.07 and p=.44. These results do not provide clear evidence of a group difference for this endpoint, but they also do not establish that the groups were equivalent.

The result applies to a specific composite pattern of reported healthcare use: fewer than 1 ambulatory visit together with an emergency department visit without subsequent hospitalization or a hospitalization during the stated period. It does not answer every question about healthcare engagement, preventive care, coping, stigma, or medical mistrust. Those are broader aims described in the study summary, and the primary endpoint alone cannot establish whether those aims changed.

The outcome data also need to be considered in light of follow-up nonresponse. The registry reports that the endpoint was missing for participants (n=53) who did not take part in any follow-up surveys, and that nonresponse weights were used. Since the weighting details are not reported in the registry record, the result cannot be evaluated fully for sensitivity to those choices.

The study's terminated status is part of the registry record, but the reason for termination is not reported in the registry record. No conclusion about the cause of termination, or its implications for the statistical result, follows from the status alone. Overall, the posted numbers are best read as an imprecise estimate for a narrowly defined outcome, alongside unresolved questions about missing-data methods and the broader outcomes named in the trial description.

17. Why This Trial Matters Statistically

This trial illustrates why a clinical result should be read through both the endpoint definition and the statistical uncertainty. The intervention was intended to address coping with discrimination and medical mistrust and to promote healthcare engagement, while the primary endpoint captured a particular combination of ambulatory and acute care use. That makes the endpoint relevant to a specific utilization question, but it is not interchangeable with the intervention's wider goals.

It also illustrates the difference between a point estimate and the range of effects compatible with the data. The recalculated risk difference is 0.045, yet its confidence interval extends from -0.051 to 0.141. The estimate alone may look like a single answer, while the interval makes clear that the direction and magnitude remain uncertain. The same caution applies to the sponsor's odds ratio and its confidence interval.

Another statistical issue is outcome nonresponse. The registry says nonresponse weights were used because outcomes were missing for participants who did not attend any follow-up surveys. Reporting that approach is useful, but assessing it requires information about how weights were formed and what assumptions they rely on. Those details are not reported in the registry record.

Finally, planning calculations based on the observed effect highlight how uncertain estimates can inform future design without proving that the observed effect is stable. The planning figures are illustrations under stated assumptions, not evidence of the trial's original design or its observed power. Together, these features make the record a useful example of interpreting a randomized comparison without overreading a non-significant result or assuming that missing-data and design details are known.

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.

The registry describes the trial as terminated, but the reason is not reported in the registry record. The record lists enrollment as 369, but the detailed recruitment process, eligibility criteria, and reasons for nonparticipation are not reported in the registry record. The identity of those masked under the single-masking designation is also not reported.

The primary endpoint depends on participant reports and a specific definition combining ambulatory visits with emergency department visits or hospitalization. The registry does not report details on how those reports were collected or validated. The result therefore speaks to the defined reported outcome, not necessarily to all aspects of healthcare utilization.

Outcome data were missing for participants (n=53) who did not take part in any follow-up surveys. Nonresponse weights were used, but their construction and the assumptions behind them are not reported in the registry record. Sensitivity analyses for alternative assumptions about missing outcomes are also not reported. The registry states that outcomes were not missing among participants who took part in any follow-up surveys, which does not resolve uncertainty associated with those without follow-up.

The sponsor reports a logistic regression result, but the model covariates and analysis-population details are not reported in the registry record. The record lists multiple outcomes and analyses without reporting a multiplicity strategy. It also does not report the trial's original sample-size justification or design assumptions. Consequently, the planning figures derived from the observed effect should not be mistaken for the actual design rationale or an assessment of the study's power.

Finally, the record does not include results for every objective named in the study summary. The primary endpoint result cannot establish effects on coping, stigma, medical mistrust, preventive care, or other healthcare engagement measures. These limits constrain how broadly the primary comparison can be interpreted.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Still Climbin': An Intervention to Improve Coping Among Black Sexual Minority Men — NCT04587869.
• Registry record last updated 2025-11-13; status last verified 2025-10.
• Results first posted 2025-11-13.
• 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

The registry record describes a randomized, parallel-group, single-masked trial of an 8-session CBT group intervention for Black sexual minority men, compared with a no-treatment control. Enrollment is listed as 369, and the study status is terminated. The primary outcome is inadequate healthcare utilization over baseline to 12-month post-baseline, defined using reported ambulatory, emergency department, and hospitalization use.

The posted counts are 41 participants with the outcome in the intervention group and 39 in the control group. The unadjusted risk difference is 0.045, with a 95% confidence interval from -0.051 to 0.141. The sponsor's logistic regression reports an odds ratio of 1.23, a 95% confidence interval from .73 to 2.07, and p=.44. These estimates are uncertain and do not establish equivalence or a clear group difference.

The outcome was missing for participants (n=53) who took no follow-up surveys; the registry says nonresponse weights were used. The weighting details, original sample-size assumptions, and reason for termination are not reported in the registry record.