HIV Prevention Peer Navigation for Justice Involved Women: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for HIV Prevention Peer Navigation for Justice Involved Women (NCT04680390), a randomized na study. The primary outcome is Linkage to PrEP Services.

2. Clinical Question

3. Trial Design

The registry describes kINSHIP as a randomized, parallel behavioral trial comparing peer navigation with standard-of-care engagement practices. The allocation was randomized, and masking was listed as none. The registry reports enrollment of 30 and trial status as completed. The phase is listed as not applicable.

The active intervention was peer-led navigation intended to address intersectional stigma and improve initiation and engagement with PrEP services. Its stated components include increasing social support and self-efficacy in accessing PrEP services. The control condition was standard care, described as as-needed case management. The registry record does not provide a more detailed account of how often either form of support was delivered or the exact procedures used in each arm.

The primary endpoint was attendance at a PrEP services appointment, verified by medical records, at 4 months post enrollment. The primary analysis excluded one participant who did not consent to medical-record data extraction. The posted arm results report 4 participants with linkage among 13 in the active arm and 0 among 14 in control. These analysis denominators should be kept distinct from the registry’s overall enrollment figure.

The record lists Accepted PrEP Referral, Time to PrEP Linkage, Accepted PrEP Prescription, PrEP Initiation, PrEP Adherence, and PrEP Persistence as secondary outcomes. Details of the statistical plan, randomization procedure, concealment, and handling of other missing data 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.

Before assignment: Enrolled participants were only excluded from the analysis if they did not complete the baseline assessment or randomization.

Overall Study

Analysis population for the primary outcome, as defined in the registry: One participant did not consent to having their data extracted from their medical record so were not included it in the analysis.

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 Linkage to PrEP Services, measured over 4 months post enrollment.

The primary endpoint is linkage to PrEP services, defined as attendance at a PrEP services appointment and verified by medical records. The timeframe is 4 months post enrollment. The posted outcome is a count of participants meeting this definition, rather than a continuous measure or a time-to-event estimate.

This endpoint captures whether a participant reached a service appointment, an observable step between referral or interest and care. It is therefore distinct from accepting a referral, accepting a prescription, starting PrEP, adhering to it, or persisting with it. Those outcomes are listed separately as secondary measures. Appointment attendance alone does not establish that PrEP was prescribed, initiated, or continued.

Medical-record verification offers a defined source for determining whether an appointment occurred. In this trial, however, one participant did not consent to having data extracted from their medical record and was not included in the primary analysis. The registry does not report whether other endpoint information was missing or how any other missing data were handled. The endpoint gives a focused measure of service linkage, but its interpretation should remain limited to the appointment-attendance definition and the stated follow-up timeframe.

7. Results Posted in the Registry

• Active-kINSHIP Navigation Intervention: 30.8% (4/13) versus 0.0% (0/14) on control; risk difference 30.8 percentage points (95% CI 5.7 to 55.9); unadjusted two-proportion z-test p = 0.025.

8. Statistical Interpretation

9. Statistical Methods Explained

What does the risk difference describe?

A risk difference compares the proportion of participants with an event in one arm with the proportion in the other arm. Here, the event is attendance at a PrEP services appointment verified by medical records. The reported estimate is 0.308 for active kINSHIP navigation relative to standard care. It is an absolute contrast on the probability scale, not a relative measure such as a risk ratio. The estimate should be read together with its uncertainty interval and with the underlying counts: 4 participants with linkage in the active arm and 0 in control.

Because the control count is zero, the comparison is sparse. A zero observed count does not demonstrate that the underlying probability of linkage in similar participants is exactly zero. It means no control participant in this analyzed group met the endpoint. Estimates based on sparse counts can change substantially with a small change in observed outcomes, so a single point estimate should not be read as a highly precise description of the underlying contrast.

Why is Fisher’s exact test listed?

The registry reports a Fisher Exact test for the two-arm comparison. As a general statistical explanation, Fisher’s exact test evaluates evidence against no association using the observed table of event and non-event counts, while accounting for the table margins. It is commonly used when counts are sparse, because methods relying on large-sample approximations can be less dependable when event counts are very small. The method does not remove uncertainty or address design issues such as missing outcomes, unmasked assessment, or generalizability.

The posted p-value for the registry analysis is 0.041. A separate unadjusted recalculation from the summary counts gives p = 0.025. The p-values are not interchangeable: they are outputs of distinct analyses. A p-value describes how compatible the data are with a specified null model under the test’s assumptions; it is not the probability that the null hypothesis is true, nor a measure of the size or practical importance of the observed difference. The registry does not report further test specifications or an adjusted model.

How should the confidence interval be read?

The risk-difference interval runs from 0.057 to 0.559 around the estimate of 0.308. In general, a confidence interval conveys sampling uncertainty under the method used to construct it. It identifies a range of effect values compatible with the data and method, rather than a range containing a known fixed proportion of future results. The interval’s width is relevant: it signals that the observed estimate is not determined with fine precision. The registry does not identify the interval construction method in the available record, so further claims about its exact coverage procedure are not warranted.

What do the planning illustrations mean?

The sample-size assumptions describe planning illustrations calculated from the observed effect, not the trial’s own design rationale. They use a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation; the effect to detect is described as a stated fraction of the observed effect. Such calculations ask what sample size a hypothetical future study might require under those assumptions. They are not observed power, do not show that the completed trial was adequately powered, and do not establish that the observed effect is a reliable planning target. The trial’s original sample-size justification and design assumptions are not reported in the registry record.

For general interpretation, the assumptions matter because planning estimates are sensitive to the effect size chosen and to the statistical model. If a future study aims to detect a smaller effect than the observed one, it would generally require more information, all else equal; this is a general principle, not a result calculated for this trial. The available record does not give the numeric planning outputs, so no particular planning sample size can be stated.

10. Confidence Intervals and Precision

The primary risk-difference estimate is 0.308, and its confidence interval is 0.057 to 0.559. The interval is important because the point estimate alone gives no indication of how uncertain the comparison is. Its endpoints remain above zero, but the interval covers a substantial range of possible absolute differences. The data therefore support a positive observed contrast while leaving considerable uncertainty about its magnitude.

The sparse event pattern is central to precision: 4 participants met the endpoint in the active arm and 0 in control. When event counts are small, the addition or removal of a small number of observed events can influence the estimate and its uncertainty. The zero control count is a sample observation, not proof of an underlying absence of linkage in the control population. The analysis denominators are 13 and 14, and one participant was excluded from the primary analysis because medical-record data extraction was not authorized.

A confidence interval is a method-based expression of sampling uncertainty, not a guarantee that the true effect lies within these particular endpoints. The registry record does not identify the interval construction method in the registry entries. Accordingly, the interval can be described by its reported limits, but its exact computational procedure cannot be reconstructed from the record.

The sponsor-reported Fisher Exact p-value is 0.041, while the separate unadjusted recalculation gives p = 0.025. Neither p-value communicates the range of plausible effect sizes as directly as the interval does. Precision should be judged from the interval and event pattern, not from a threshold reading of a p-value alone. The record does not report a separate precision target or a planned confidence-interval width.

11. Analyses Reported by the Sponsor

The registry lists a sponsor-reported Fisher Exact analysis with p = 0.041. The posted summary counts also permit a simple unadjusted recalculation, which gives p = 0.025 for the comparison. The recalculated value is not a replacement for the sponsor’s reported analysis: it is a check based on the published arm-level counts and a particular unadjusted calculation. Differences between test implementations or specifications can yield different p-values, and the registry record does not give enough detail to identify the reason for this difference.

The registry marks the analysis as a superiority comparison, but the reported result should still be interpreted with the observed counts and the uncertainty interval. A p-value alone does not quantify the size or precision of the risk difference. The sponsor’s result is the inferential analysis recorded for the trial; the recalculation helps readers see how a simple analysis of the posted counts compares, but it is not an adjusted analysis and does not account for additional design or data features that are not described.

The registry record does not report an adjusted effect estimate, covariate model, or further test details. The recalculation should therefore remain explicitly labeled as unadjusted rather than treated as an independent confirmatory analysis.

12. Multiplicity, Missing Data and Analysis Populations

The registry lists six secondary outcomes: Accepted PrEP Referral, Time to PrEP Linkage, Accepted PrEP Prescription, PrEP Initiation, PrEP Adherence, and PrEP Persistence. The record does not report whether statistical testing across outcomes was adjusted for multiplicity, whether any outcomes were prioritized beyond the designated primary outcome, or whether a hierarchical testing plan was used. Without that information, secondary-outcome p-values, if reported elsewhere in the registry, would need to be interpreted in light of the possibility of multiple comparisons.

For the primary endpoint, one participant did not consent to medical-record data extraction and was excluded from the analysis. The record does not report the reason for this lack of consent, its relation to treatment assignment, or whether the participant’s endpoint could be determined from another source. The registry entries do not describe other missing data, imputation, sensitivity analyses, or a specific missing-data strategy. It would therefore be inappropriate to assume that the identified exclusion was the only missing information or that missingness was unrelated to outcome.

The posted primary denominators are 13 for active kINSHIP navigation and 14 for control. The analysis-population definition beyond the stated medical-record exclusion is not reported in the registry record. In particular, the registry entries do not establish whether the analysis followed an intention-to-treat principle or another prespecified population rule. Details of protocol deviations and eligibility after randomization are also not reported.

These omissions limit how fully the primary analysis can be audited from the registry alone. They do not invalidate the posted summary, but they restrict conclusions about the handling of all randomized participants, the robustness of the result to missing data, and the interpretation of tests across the primary and secondary outcomes.

13. 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 registry reports no serious adverse events in either arm. For the serious-event summaries, the at-risk counts are 13 for active kINSHIP navigation and 15 for standard-care control; the affected counts are 0 in each arm. The registry also reports no other adverse events in the listed groups and no deaths.

These figures should be read as event counts within the stated safety denominators, not as proof that harms cannot occur. A zero observed count means that no event was recorded in that group under the registry’s reporting, not that the underlying risk is known to be zero. The safety denominators differ from the primary-outcome denominators, and the record does not explain the difference. They should not be substituted for the analysis populations used for linkage.

The record does not list serious-event terms, and no detail is available about event definitions, ascertainment procedures, follow-up duration for safety, or completeness of reporting. The figures are therefore a concise account of what was posted, not a comprehensive characterization of safety or a basis for comparing rare harms precisely.

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

The result describes a difference in documented PrEP-service appointment attendance between randomized arms over the stated timeframe. In the analyzed groups, 4 participants in active kINSHIP navigation and 0 in standard care had the primary endpoint. The reported risk difference is 0.308, with a confidence interval from 0.057 to 0.559. These values summarize an observed contrast and its uncertainty; they do not establish a universal or precisely known effect.

The endpoint is service linkage, not PrEP use itself. Appointment attendance does not show whether a participant accepted a prescription, initiated PrEP, adhered to it, or persisted with it. Those are separately listed secondary outcomes, and their results are not reported in the registry record for this interpretation. The trial also does not establish an effect on HIV acquisition or other clinical outcomes.

The trial’s randomized design supports comparing the assigned groups, but the study is unmasked and the primary analysis excludes a participant who did not consent to medical-record extraction. The registry does not report the relationship of that exclusion to allocation or outcome. Sparse counts and the interval’s width further counsel against treating the point estimate as exact.

Nor does the result establish how the intervention operated, which elements were received, or whether the same pattern would occur in other populations or service settings. Generalization beyond the studied justice-involved women requires information not reported in the registry record. The result is best read as an early, bounded comparison of one service-linkage outcome, rather than as a conclusion about every step of PrEP care or the overall effectiveness of peer navigation.

15. Why This Trial Matters Statistically

This trial raises a statistical question about how to interpret a notable difference when event counts are sparse. The active arm recorded 4 linkage events and control recorded 0. The risk-difference estimate is 0.308, but the confidence interval from 0.057 to 0.559 illustrates that the estimate remains uncertain. The zero in control is consequential for the observed contrast, yet it cannot be equated with a zero underlying chance of linkage.

The example also shows why the test, effect estimate, and interval answer different questions. Fisher’s exact test supplies a test result for the sparse count comparison; the risk difference expresses the absolute contrast; and the interval conveys uncertainty about its magnitude. A p-value should not stand in for effect size or precision. The registry’s p-value of 0.041 and the unadjusted recalculation of p = 0.025 also illustrate that analysis choices matter, especially when results are based on limited counts.

Finally, the primary endpoint measures a step in access to prevention services, not the full path from referral through PrEP use. Distinguishing linkage from initiation, adherence, persistence, and clinical outcomes is essential when interpreting what a trial result can establish. The registry lists these additional outcomes, but their results and the approach to multiple testing are not reported in the available record. Clear separation of endpoint, estimate, uncertainty, and analysis plan is therefore especially important.

16. 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 primary comparison is based on a small set of observed events: 4 participants with linkage in the active arm and 0 in control. This sparse pattern limits precision and makes the estimated difference sensitive to the observed counts. The reported confidence interval is broad, so the point estimate should not be treated as a finely determined effect.

The trial was unmasked. The primary outcome is based on medical-record verification, but the registry does not report the procedures used to identify, extract, or adjudicate records. One participant did not consent to medical-record data extraction and was excluded from the primary analysis. The reason for non-consent and its relation to assignment or linkage are not reported. The record does not describe other missing-data handling or establish the analysis population beyond this exclusion.

The registry provides a sponsor-reported Fisher Exact p-value and a posted-count recalculation, but does not give further details of the statistical plan or test implementation. Adjustment for multiple comparisons across the secondary outcomes is not reported. The record also does not report the original sample-size justification, interim analyses, or a complete protocol-level account of analysis decisions.

The primary endpoint captures appointment attendance by 4 months post enrollment. It does not by itself measure PrEP initiation, adherence, persistence, or HIV outcomes. The registry entries do not give detailed results for the listed secondary outcomes. Finally, the record’s description of the intervention is incomplete, so delivery, participation, and fidelity cannot be assessed from these registry entries.

17. Related Tutorials

18. Related Statistical Calculators

19. Sources

• ClinicalTrials.gov: HIV Prevention Peer Navigation for Justice Involved Women — NCT04680390.
• Registry record last updated 2026-08-03; status last verified 2026-07.
• Results first posted 2026-08-03.
• 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.

20. Record Summary

NCT04680390 is a completed, randomized, parallel, unmasked behavioral trial comparing active kINSHIP peer navigation with standard-care engagement practices. The registry reports enrollment of 30 and lists the phase as not applicable. The primary outcome was medically verified attendance at a PrEP services appointment by 4 months post enrollment.

The posted primary counts are 4 of 13 in the active arm and 0 of 14 in control. The risk difference is 0.308, with a confidence interval from 0.057 to 0.559. The registry reports a Fisher Exact p-value of 0.041; the unadjusted recalculation from the posted counts gives p = 0.025.

One participant did not consent to medical-record data extraction and was excluded from the primary analysis. Six secondary outcomes are listed, but their results and the approach to multiplicity are not reported in the available record. Safety summaries report no serious or other adverse events and no deaths in the listed groups.