A Risky Sex Prevention Intervention for Middle School Age Minority Girls: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Risky Sex Prevention Intervention for Middle School Age Minority Girls (NCT02103218), a non-randomized na study. The primary outcome is Adolescent Sexual Activity Index.

2. Clinical Question

3. Trial Design

The registry record describes a completed, non-randomized, parallel-group behavioral study. One arm received risky-sex-prevention education, activities, empowerment and racial-pride building; the other received general health education and activities. Because assignment was not randomized, measured or unmeasured differences between participants in the arms could contribute to outcome differences. The record itself cautions that the arms are not a like-for-like control comparison.

Masking was listed as none. The record does not explain whether outcome assessors were masked, how assignment was decided, whether participants could choose an arm, or what steps, if any, were used to make the groups comparable. Those design details are not reported in the registry record. The phase was listed as not applicable.

Enrollment was 212. The study start was recorded as 2012-11, primary completion as 2018-02-13, and status as completed. The sponsor was University of North Carolina, Greensboro. These registry entries describe the registry entry; they do not give the number analyzed for each outcome or explain participant flow over the study.

The primary measure was the Adolescent Sexual Activity Index, assessed at 12 months. The record says primary analyses were performed after multiple imputation for missing data and gives the numbers available before imputation for that outcome: n=46 in the Risky sex prevention arm and n=51 in the Health behaviors arm. The record does not describe the allocation process, follow-up procedures, intervention fidelity, or statistical model used to estimate the primary comparison.

Secondary outcomes included HIV knowledge, maternal and mother-teen communication measures, monitoring, self-esteem, racial pride and interpersonal empowerment. The record lists 19 outcome measures and 18 analyses, but does not explain how these measures and analyses map to one another. The completed status does not by itself indicate whether all planned outcomes were reported.

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: Analyses were performed after multiple imputation for missing data. Number at 12 months before imputation were n=46 for Risky sex prevention and n=51 for Health behaviors.

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 overall number of baseline participants (212) is based on both daughters and mothers in all arms.

6. Primary Endpoint

The registry lists the primary outcome as Adolescent Sexual Activity Index, measured over 12 months.

The primary endpoint was the Adolescent Sexual Activity Index for female adolescents at 12 months. It is a continuous score on a scale with a theoretical minimum of 0 and a theoretical maximum of 10. The registry description states that higher values indicate more activity, which it characterizes as a worse outcome. This direction matters when interpreting any arm difference: a higher score is not an improvement on the scale as described.

A continuous index can capture variation across participants rather than reducing sexual activity to a yes-or-no category. Such a measure may be useful when the study question concerns levels of activity, but the index’s exact construction, item content, scoring procedure and validation details are not reported in the registry record beyond the scale limits and cited description.

The registry lists the endpoint parameter as a least-squares mean and its dispersion as a standard error. A least-squares mean is generally a model-based mean; its interpretation depends on the model and covariates used. The record does not specify that model or list any endpoint means, standard errors, confidence intervals or between-arm estimate. Consequently, the endpoint’s definition and direction are available, but the observed result and its precision are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry record identifies the primary endpoint as a continuous score, names least-squares means as the parameter, and gives standard error as the dispersion measure. It also states that the primary analyses used multiple imputation for missing data. The model specification, contrast definition and inferential procedure are not reported in the registry record. The concepts below explain how these methods are generally understood; they should not be read as a claim that a particular unreported procedure was used.

What does a least-squares mean represent?

In general, a least-squares mean is a model-based estimate of a group mean. Unlike a simple arithmetic average, it may account for terms included in a statistical model. Its value and interpretation therefore depend on the model, the variables included, how categorical factors are represented and the population or covariate values to which the estimate is standardized. A least-squares mean is not automatically adjusted for every relevant difference between groups. The registry record does not identify the model terms or adjustments used for this trial, so the precise meaning of its planned least-squares mean cannot be established beyond the label.

For a comparison, the estimated difference between group means is usually the quantity that directly addresses whether the groups differ on the endpoint. A group mean alone does not give that contrast. With non-randomized allocation, adjustment may help account for measured characteristics, but it cannot guarantee removal of bias from unmeasured factors or from inappropriate model assumptions. The record does not report the primary contrast or the adjustment strategy.

How should a standard error be read?

In general, a standard error describes the sampling uncertainty of an estimate under the assumed analysis. It is not the same as the spread of individual scores. A smaller standard error indicates a more precisely estimated mean under that model, while a larger one indicates less precision. A confidence interval is often formed using an estimate and its standard error, with a method appropriate to the analysis. Since no endpoint estimates or standard errors are reported in the registry record, precision cannot be evaluated for this outcome.

What does multiple imputation do?

In general, multiple imputation replaces missing values with plausible values generated under an imputation model, analyzes each completed dataset, and combines the results to reflect uncertainty from the missing values. The validity of the result depends on the assumptions and model, including how observed information relates to missingness and the outcome. Imputation does not recover values known to be true, and it does not automatically correct bias when the assumptions fail. The registry record confirms that multiple imputation was used for the primary analysis but does not report the number of imputations, variables in the imputation model, assumptions, diagnostics or sensitivity analyses.

The record gives n=46 for Risky sex prevention and n=51 for Health behaviors at 12 months before imputation. These are the numbers before imputation for the primary endpoint, not a description of the completed imputed datasets. The record does not give the post-imputation analysis counts or a participant flow account for the primary outcome.

Why does allocation matter?

In a randomized study, assignment is designed to make treatment groups comparable on average, subject to chance. In a non-randomized study, the process that places participants in groups may be associated with characteristics that also relate to the outcome. This creates a potential for confounding. Adjustment for measured characteristics can address some observed differences, but it does not establish that unmeasured differences are absent. The registry explicitly warns that the arms are not a like-for-like control comparison. The assignment mechanism and any adjustment are not reported in the registry record.

What can planning calculations say?

In general, a sample-size planning illustration asks how many participants per arm might be needed to detect a specified effect under stated assumptions. The registry’s stated illustration assumptions are a two-sided significance level of 0.05, equal allocation, a normal approximation, and a target effect expressed as a fraction of the observed effect. Such planning calculations are not observed power and do not establish that the original study was designed around those assumptions. The record says the illustration was computed from the observed effect, but the underlying effect and resulting planning counts are not reported in the registry record. No conclusion about the trial’s original sample-size justification can be drawn.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists 19 outcome measures and 18 analyses, and names a primary endpoint alongside several secondary measures. A study that assesses multiple outcomes can raise a multiplicity question: when many hypotheses are examined, the chance of at least one apparently notable result can increase. Statistical approaches may address this through prespecified outcome hierarchies or adjustments, but whether and how such methods were used is not reported in the registry record. The difference between the listed measure and analysis counts is not explained, so they should not be treated as a one-to-one inventory.

The secondary outcomes include HIV knowledge, maternal bond, mother-teen sexual-risk communication, open family communication, maternal monitoring, self-esteem, racial pride and interpersonal empowerment. The record does not give a multiplicity plan, identify which secondary analyses were confirmatory or exploratory, or report the results needed to judge the findings across outcomes. In general, interpretation of multiple results should account for the analysis plan and the possibility of selective reporting; those details are not available in the registry record.

For the primary endpoint, the record states that analyses were performed after multiple imputation for missing data. It reports n=46 before imputation for Risky sex prevention and n=51 before imputation for Health behaviors at 12 months. The imputation model, assumptions, diagnostics and sensitivity analyses are not reported. Handling of missing data for secondary outcomes is also not reported in the registry record.

The record does not define the analysis population, such as whether analyses followed an intention-to-treat principle, nor does it describe exclusions or participant flow. The stated enrollment should not be taken as the number contributing observed primary endpoint data. Without outcome results, an analysis-population definition and missingness details, the robustness and scope of any treatment comparison cannot be assessed.

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 adverse-event listing separates daughters and mothers by study arm. For Risky Sex Prevention-Daughters, the record lists 51 at risk for serious events, 0 seriously affected, 51 at risk for other events, 0 affected by other events and 0 deaths. For Healthy Behaviors-Daughters, the corresponding entries are 59, 0, 59, 0 and 0.

For Risky Sex Prevention-Mothers, the entries are 48 at risk for serious events, 0 seriously affected, 48 at risk for other events, 0 affected by other events and 0 deaths. For Healthy Behaviors-Mothers, they are 54, 0, 54, 0 and 0. No top serious-event terms are listed.

These are the registry’s recorded figures, not proof that no adverse experiences occurred. The record does not describe the event-collection period, definitions, ascertainment procedures or reporting thresholds. The figures also do not establish comparative safety, particularly given the non-randomized design. They should be read as a description of what the registry entry reports, with those reporting details unavailable.

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

The record establishes that a behavioral intervention combining education, activities, empowerment and racial-pride building was compared with general health education and activities in a non-randomized, parallel-group study. It identifies the Adolescent Sexual Activity Index at 12 months as the primary outcome and describes its scale direction. It also indicates that missing data were handled with multiple imputation for the primary analysis.

Those design and measurement details do not answer whether the intervention changed the primary outcome. The arm-specific means, between-arm estimate and uncertainty measures are not reported in the registry record. No claim about the size, direction, statistical significance or practical importance of an observed difference can be supported from the available result information. The study’s rationale is not evidence of an outcome effect.

Even if a difference were reported, non-randomized allocation would limit causal interpretation. Differences between participants or other group-level factors could contribute to an observed contrast. The record does not describe the assignment process, adjustment strategy or sensitivity analyses that might help assess those concerns. A statistical association between arm and outcome would not, by itself, establish that the intervention caused the difference.

The scale’s stated range and direction help explain what an endpoint estimate would mean, but they do not establish whether a difference is large or meaningful. The record does not supply a threshold for clinical or behavioral importance. Nor does the stated completion status establish that the intervention achieved the aims described in the rationale. The registry entry is informative about the question, design and endpoint, but it does not report the primary result needed to answer that question.

13. Why This Trial Matters Statistically

This trial raises several statistical issues that are central to interpreting behavioral intervention research. First, the study used non-randomized allocation. That can be useful in real-world settings, but it means that group comparability cannot be assumed from the design. Any interpretation of an arm contrast would need to consider how participants entered each arm and whether relevant differences were measured and addressed.

Second, the primary endpoint is a continuous score with a clearly stated direction. That makes it important to distinguish the outcome’s definition from its result: higher scores mean more activity and a worse outcome, but the registry record does not report the estimates needed to determine what was observed. The scale’s theoretical limits alone do not reveal the distribution of participant scores or the size of a group difference.

Third, the primary analysis used multiple imputation, while the record gives pre-imputation numbers for the outcome. This highlights how missing data affect both the analysis and its interpretation. The method’s assumptions and diagnostics matter, especially when the observed endpoint data do not represent everyone enrolled. Those details are not reported in the registry record.

Finally, the study includes multiple outcome measures, which makes transparent reporting of the analysis plan and multiplicity approach important. The record lists the measures and analyses but does not describe how findings were prioritized or adjusted. Together, these issues show why a registry description of design and methods cannot substitute for complete outcome estimates, uncertainty measures and analysis details.

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 most direct limitation of interpretation is that the registry record does not report the primary outcome estimates or a between-arm comparison. Without those results, the direction and precision of the observed outcome cannot be assessed. The record also does not report confidence intervals or p-values for the primary endpoint.

Non-randomized allocation limits the strength of causal conclusions. The assignment process and any methods used to account for differences between participants are not reported in the registry record. The lack of masking is recorded, but the record does not clarify which people involved in delivery or outcome assessment knew the arm assignments.

The primary endpoint analysis used multiple imputation, but details needed to judge that approach are absent: the imputation model, assumptions, number of imputations, diagnostics and sensitivity analyses are not reported. Pre-imputation outcome counts are listed, but participant flow, reasons for missingness and the analysis-population definition are not reported in the registry record. Handling of missing data for secondary outcomes is also unclear.

The record lists multiple outcomes and analyses without explaining the multiplicity strategy or how analyses were prioritized. It also does not describe the endpoint’s detailed scoring procedure beyond its theoretical limits and direction. Intervention duration, delivery, adherence and fidelity are not reported, limiting interpretation of what exposure the arm label represents. These limitations reflect missing information in the registry record and should not be read as evidence that the underlying study did or did not address these issues in other documents.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: A Risky Sex Prevention Intervention for Middle School Age Minority Girls — NCT02103218.
• Registry record last updated 2021-06-16; status last verified 2021-05.
• Results first posted 2019-10-18.
• 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 record describes a completed, non-randomized, parallel-group behavioral study sponsored by University of North Carolina, Greensboro. It compared an education, activities, empowerment and racial-pride intervention with general health education and activities. Enrollment was 212, masking was listed as none, the study start was 2012-11, and primary completion was 2018-02-13.

The primary endpoint was the Adolescent Sexual Activity Index at 12 months, a continuous scale for which higher scores indicate more activity and a worse outcome. The record states that primary analyses used multiple imputation and reports pre-imputation outcome counts of n=46 and n=51 for the two arms. It also lists secondary outcomes and adverse-event figures.

Arm-specific primary outcome estimates, a between-arm effect and measures of uncertainty are not reported in the registry record. The non-randomized design, missing-data details and unreported multiplicity approach limit what can be concluded from the available entry.