Steering Together in a New Direction: Reducing the Risk of HIV/STD Among African American Men: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Steering Together in a New Direction: Reducing the Risk of HIV/STD Among African American Men (NCT02572401), a randomized na study. The primary outcome is Consistent (100%) Condom Use.

2. Clinical Question

The study addressed behavioral HIV and sexually transmitted disease risk. Its stated rationale was that randomized trials of interventions intended to reduce sexual risk behavior among heterosexually active African American men were limited, and that behavior change can diminish over time. In addition to evaluating STAND, the design included text messaging as a strategy intended to sustain intervention effects. This rationale describes the research question; it is not evidence that either intervention changed behavior.

The protocol description says participants would complete self-report measures using audio computer-assisted self-interviewing at baseline, immediately after intervention, and at follow-up. The primary outcome timeframe is given as baseline, 6 months, and 12 months post-intervention. The registry record does not report the primary outcome values at those assessments, or estimates comparing intervention assignments. Consequently, the research question can be described from the design, but the direction, magnitude, and precision of any observed difference cannot be determined from the registry record.

3. Trial Design

This was a randomized, single-masked, factorial trial with reported enrollment of 391. The four study groups combine assignment to STAND HIV Risk Reduction Intervention and assignment to text messaging: STAND without text messages, neither intervention, STAND plus text messages, and text messages without STAND. This arrangement allows the study to consider two intervention components within one trial design. The registry record does not report how the randomization sequence was generated, who was masked, or whether analyses focused on component effects, group-to-group contrasts, or an interaction between the components.

The registry lists the trial as completed. It gives a start date of 2011-10 and a primary completion date of 2013-06. The listed sponsor is the University of Pennsylvania. The phase is recorded as NA. These registration details describe the trial record and do not provide information about the conduct or outcome of individual assessments.

The planned participants were African American men aged 18 to 45 who reported recent unprotected intercourse with a woman. The brief description says the study would use self-report measures collected through audio computer-assisted self-interviewing at baseline, immediately post-intervention, and at 6 and 12 months after intervention. The primary outcome description identifies baseline, 6-month, and 12-month measurements; it defines the primary outcome population as participants with baseline data and at least one post-intervention assessment. The registry record does not report how many participants contributed outcome data at each time point, how many were assigned to each group for the primary analysis, or how missing assessments were handled.

Four group descriptions are listed, with 97 participants at risk in the HIV Risk Reduction Only group and 98 in each of the other three groups for the reported safety data. These are safety denominators, and they should not be treated as outcome-analysis sample sizes. The registry record does not report the allocation ratio, the number randomized to each group, or a sample-size justification. The enrollment total alone does not establish how much precision the primary comparison achieved.

The primary outcome was a binary participant-level classification of consistent condom use. Other listed outcomes were unprotected intercourse, multiple partners, insertive anal intercourse, and proportion of intercourse that was condom-protected. The registry lists five outcome measures and 15 analyses, but does not report the full analysis plan or the results of those analyses.

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: Participants providing primary outcome data at baseline and at least one post-intervention assessment.

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 Consistent (100%) Condom Use, measured over Baseline, 6 months, 12 months post-intervention.

The primary endpoint, titled consistent (100%) condom use, classifies a participant according to reported condom use during vaginal or anal intercourse with a woman in the previous 3 months. A participant needed to report at least one intercourse act. If the reported number of protected acts equaled the reported number of acts, the participant was classified as practicing consistent condom use. If the protected-act count was lower than the total act count, the participant was classified as not practicing consistent condom use.

This binary definition makes the outcome a count of participants meeting a specified behavior criterion, rather than a measure of the number of intercourse acts or the proportion of acts for an individual. The record also describes separate binary variables for consistent condom use with steady partners and casual partners, analyzed as repeated outcomes. The registry record does not report the values for these variables or how they relate to the primary endpoint analysis.

As a general statistical point, a clearly defined binary endpoint can make a behavioral outcome easier to classify and compare across randomized groups. The classification also simplifies varied patterns of behavior into two categories, so it does not convey how often condoms were used when use was less than consistent. The endpoint relies on participant self-report; the record says assessments were completed through audio computer-assisted self-interviewing. The registry record does not report a validation measure or other corroboration of those reports.

The primary outcome timeframe is baseline, 6 months, and 12 months post-intervention. The population for the primary outcome includes participants with baseline data and at least one post-intervention assessment. The endpoint description therefore indicates repeated measurement over time, but it does not specify the statistical model, the primary time point for a treatment contrast, or how data from participants with only some assessments were used.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry describes a randomized, single-masked factorial design. Randomization is a general method for assigning participants to study groups in a way intended to reduce systematic differences at assignment. In a factorial design, participants are assigned across combinations of intervention components. Here, the four described combinations include STAND and no STAND, crossed with text messaging and no text messaging. The record does not specify the randomization procedure, the identity of the masked party, or the statistical contrasts used to analyze the factorial arrangement.

What does the binary endpoint capture?

The primary endpoint is a binary classification: consistent condom use or not, according to the stated rule. The rule requires at least one reported intercourse act and compares the reported number of protected acts with the total number of acts. A general advantage of a binary endpoint is that it defines a recognizable participant-level event that can be counted across groups. A limitation is that it compresses different behavior patterns into the same category. Someone just below the consistency threshold and someone reporting fewer protected acts are both classified as not practicing consistent condom use under the described rule. The endpoint therefore does not retain all detail about frequency of use.

The primary outcome is measured across baseline and post-intervention follow-up, with the timeframe listed as baseline, 6 months, and 12 months post-intervention. The brief description also mentions an immediate post-intervention assessment. The record does not specify whether that immediate assessment contributes to the primary outcome analysis. A repeated-outcome analysis can account for measurements from the same participant being related to one another; as a general statistical matter, observations from one person are not automatically independent. The registry record does not name a repeated-measures model or say how it accounted for this relationship.

How should a treatment contrast be understood?

A treatment contrast is a statistical comparison between outcomes under different randomized assignments. In this trial, possible contrasts could involve the STAND assignment, the text-message assignment, or specific combinations of the two. In general, a factorial analysis may estimate component effects and may assess whether the effect associated with one component varies according to the other component. These are general possibilities, not claims about the analysis used in this trial. The registry record does not report the planned contrasts, whether an interaction was tested, or the model used to estimate effects.

For a binary outcome, statistical analyses often express group differences using a contrast appropriate to the analysis plan. The registry record does not provide an effect measure, model coefficient, or group result for consistent condom use. No contrast can be reconstructed from the available information. A confidence interval, in general, describes a range of values compatible with the data and the assumptions of the analysis; its width conveys precision. No confidence interval for the primary outcome is reported in the registry record, so its precision cannot be assessed.

What can p-values establish?

A p-value, in general, summarizes how compatible the observed data are with a specified null model, under the assumptions used for the test. It does not measure the size or importance of an effect, and it cannot replace an effect estimate and its uncertainty. The registry record does not report a primary-outcome p-value or state the significance threshold used for the outcome analysis. The sample-size assumptions describe an illustration using a two-sided significance level of 0.05, but those planning assumptions do not establish the actual analysis plan or a reported test result.

The record lists 15 analyses. That count alone does not show how many statistical tests were performed, which outcomes or time points they covered, or which were designated primary. In general, examining multiple outcomes or contrasts can increase the chance of at least one apparently unusual result if no multiplicity strategy is used. The registry record does not report whether a correction, hierarchy, or other approach was used. It would not be appropriate to infer one from the analysis count.

What do the planning illustrations represent?

The sample-size assumptions describe planning illustrations computed from an observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The effect to detect is described as a stated fraction of the observed effect. The record distinguishes participant-per-arm planning illustrations from total-event illustrations for time-to-event outcomes. These are illustrations, not observed trial outcomes and not observed power. The registry record does not report the trial’s own sample-size justification or establish that these assumptions were used to plan enrollment. No planning values are included in the record details described for the trial, so no sample-size figure can be interpreted.

In general, a planning calculation depends on assumptions about the effect to detect, variability or event frequency, allocation, and the analysis method. Changing an assumption can change the required sample size. A calculation based on an observed effect is especially dependent on that effect estimate; it should not be read as proof that a study had a particular probability of detecting a true effect. The record does not report an observed primary-outcome effect from which to judge the precision of these illustrations.

The outcome is based on participant self-report. Audio computer-assisted self-interviewing is the collection mode named in the brief description. As a general measurement consideration, self-report can be affected by recall and reporting behavior; the registry record does not quantify such error or describe a validation procedure. It also does not report whether assessors or analysts were masked, so the meaning of single masking cannot be made more specific.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists five outcome measures and 15 analyses. The primary outcome is consistent condom use; listed secondary outcomes include unprotected intercourse, multiple partners, insertive anal intercourse, and proportion of condom-protected intercourse. The record does not explain how the 15 analyses map to outcomes, assessment times, partner categories, or intervention contrasts. It also does not identify which analyses were confirmatory or exploratory, or whether any multiplicity adjustment was used. A count of analyses does not by itself establish how statistical error across comparisons was managed.

The primary outcome population is defined as participants with outcome data at baseline and at least one post-intervention assessment. This inclusion description does not establish how missing outcome values were handled in the statistical analysis. The registry record does not report missingness by group or time, reasons for missing assessments, imputation methods, sensitivity analyses, or the number analyzed for the primary outcome. It is therefore not possible to assess whether the reported analysis, if any, retained all randomized participants or depended on available assessments.

In general, missing-data methods rely on assumptions about why values are absent and how the unobserved values relate to observed information. Different assumptions can lead to different estimates. A definition of the analysis population is useful, but it is not a substitute for describing the missing-data procedure and its assumptions. Those details are not reported in the registry record.

The record describes assignment to four intervention combinations, but does not report the analysis population for each comparison, the primary contrast, or the analysis plan. It does not report interim analyses or stopping rules. These omissions mean that neither a multiplicity strategy nor a complete account of analysis populations can be reconstructed from the registry record.

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 0% reporting threshold.

The safety record lists 97 participants at risk in the HIV Risk Reduction Only group and 98 in each of the No-Intervention No-Text Message Control, HIV Risk Reduction and Text Messaging, and Text Messaging Only groups. Serious adverse events affected 0 participants in each group. Other adverse events affected 0 participants in each group. The record separately lists 0 deaths in the first three groups and 1 death in this discussion Messaging Only group.

These figures should be read as reported counts, not as a complete account of the nature, timing, or cause of events. The registry record does not list serious adverse event terms, and it provides no further details about the death. The listed at-risk group sizes are safety denominators; the registry record does not state whether they are identical to the numbers included in the primary outcome analysis.

Zero reported affected participants does not establish that no event could have occurred outside the recorded categories or reporting period. General safety interpretation depends on how events were defined, collected, and followed. Those procedures are not reported in the registry record. These counts also do not answer the behavioral efficacy question: they are separate descriptive safety information, not estimates of changes in condom use.

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

The trial record describes an effort to evaluate STAND HIV risk reduction and text messaging using random assignment and a factorial arrangement. Its primary behavioral outcome has a specific definition based on self-reported condom use for every reported vaginal or anal intercourse act with a woman in the previous 3 months. This information clarifies what the study set out to measure, but it does not establish how participants’ behavior differed across assigned groups.

The registry record does not report the primary outcome results, intervention effect estimates, confidence intervals, or p-values. It therefore does not show whether consistent condom use was more or less common in one group, how large any difference was, or how precisely a difference was estimated. The record also does not report outcome counts by assessment time, so the behavior pattern across follow-up cannot be described from the registered results.

The stated motivation includes concern that behavior change may diminish over time and a plan to test text messaging as a strategy to sustain intervention effects. That motivation is not a result. The study’s factorial design makes the two behavioral components central to the research question, but the record does not report the comparisons or interaction analysis needed to describe their observed effects. No conclusion about an intervention’s benefit or lack of benefit follows from the design description alone.

The safety information is also limited to counts: no serious adverse events or other adverse events are listed as affecting participants in the four groups, and one death is listed in this discussion Messaging Only group. Further event details are not reported in the registry record. These safety counts do not establish the effect of either behavioral intervention on the primary outcome.

Thus, the registry record supports a careful account of the trial’s population, assignments, endpoint, follow-up, and posted safety counts. It does not support a conclusion about the primary behavioral result or its statistical certainty. Any stronger claim would require outcome estimates and analysis details that are not reported in the registry record.

13. Why This Trial Matters Statistically

Statistically, this trial is notable for combining two behavioral intervention components in a randomized factorial arrangement. The four groups represent the combinations of STAND HIV Risk Reduction Intervention and text messaging, including each component alone and neither component. In general, this design can support questions about separate component assignments and whether the components’ effects vary together, provided that the analysis plan and contrasts are specified. The registry record does not report those details, so the design’s potential should not be confused with a reported component-effect analysis.

The primary endpoint also illustrates a common measurement choice in behavioral research: a complex pattern of reported acts is converted to a participant-level binary outcome. The definition is explicit about requiring at least one act and classifying participants by whether all reported acts were protected. That makes the outcome interpretable as a threshold-based behavior measure, while losing information about the degree of condom use among participants who do not meet the threshold. The registry also describes repeated outcomes and separate variables for steady and casual partners, raising questions about how repeated measurements and multiple related comparisons were handled. The analysis methods are not reported in the registry record.

The record gives a planned follow-up structure and defines an outcome population requiring baseline data and at least one post-intervention assessment. This highlights why analysis populations and missing-data methods matter in longitudinal trials. The stated population criterion does not explain how incomplete follow-up was addressed, and no group-specific analysis counts are reported.

Finally, the difference between an intervention rationale and an intervention result is important. The study was designed to investigate risk-reduction behavior and whether text messaging might help sustain effects, but the primary outcome results are not reported in the registry record. Statistical interpretation therefore centers on what can be evaluated from the record: design, measurement, and limits on inference, rather than a numerical estimate of intervention effect.

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 main limitation for interpreting this trial from the registry record is the absence of reported primary outcome results. No group-level outcome counts, effect estimate, confidence interval, or p-value for consistent condom use is reported. The record therefore does not permit assessment of the magnitude, direction, or precision of the primary comparison.

Although the study is described as randomized, single-masked, and factorial, important details are not reported in the registry record. These include the randomization procedure, the identity of the masked party, the planned factorial contrasts, and whether an interaction between STAND and text messaging was assessed. The record also does not report the trial’s sample-size justification, an interim-analysis plan, or a statistical analysis plan. The planning assumptions listed for illustrative sample-size calculations should not be taken as evidence of the trial’s own design calculations.

The endpoint relies on self-reported intercourse and condom use. The record describes audio computer-assisted self-interviewing but does not report validation, corroboration, or an assessment of reporting error. The outcome definition excludes classification of participants reporting no intercourse act from the two stated categories, but the analysis treatment of those participants is not reported in the registry record.

The primary outcome population requires baseline information and at least one post-intervention assessment. The record does not report how many participants had missing assessments, whether missingness differed by group, or how missing data were handled. It also lists 15 analyses without describing their relation to the outcomes and time points or any approach to multiplicity.

Safety information is limited to group counts. Serious and other adverse events are listed with zero affected participants in every group, and one death is listed in this discussion Messaging Only group. Serious adverse event terms and further details about the death are not reported in the registry record. These omissions limit interpretation of the study’s statistical results and safety information; they do not establish that unreported procedures or events did or did not occur.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Steering Together in a New Direction: Reducing the Risk of HIV/STD Among African American Men — NCT02572401.
• Registry record last updated 2019-10-15; status last verified 2019-10.
• Results first posted 2019-10-15.
• 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

The registry describes a completed, randomized, single-masked factorial trial of STAND HIV Risk Reduction Intervention and text messaging among African American men. Enrollment was 391. The four groups covered STAND alone, neither intervention, both interventions, and text messaging alone. The primary endpoint was binary consistent condom use, defined by reported condom use for every vaginal or anal intercourse act with a woman in the previous 3 months. The primary outcome timeframe was baseline, 6 months, and 12 months post-intervention.

The registry lists five outcome measures and 15 analyses. Serious adverse events and other adverse events are listed as affecting 0 participants in every group; 1 death is listed in this discussion Messaging Only group. Primary outcome estimates, uncertainty intervals, analysis details, and further safety-event details are not reported in the registry record. The registered design and definitions can be described, but the intervention effects cannot be determined from the posted results.