SEARCH SAPPHIRE Phase A: A Multisectoral Strategy to Address Persistent Drivers of the HIV Epidemic in East Africa: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for SEARCH SAPPHIRE Phase A: A Multisectoral Strategy to Address Persistent Drivers of the HIV Epidemic in East Africa (NCT04810650), a randomized na study. The primary outcome is Prevention Coverage Defined as Percentage of Follow-up Months That the Participant Used Either PrEP or PEP.

2. Clinical Question

3. Trial Design

This record describes SEARCH SAPPHIRE Phase A, a randomized, parallel-group study of a multisectoral strategy addressing persistent drivers of the HIV epidemic in East Africa. The registry lists the study as completed, with an enrollment of 2233. The recorded start date is 2021-04-15 and the primary completion date is 2023-10-16. The sponsor is the University of California, San Francisco. The phase is recorded as NA, and masking is listed as none.

The study encompasses several intervention areas, including dynamic prevention approaches for PrEP/PEP in outpatient clinics, antenatal clinics, and community households; a mobility dynamic treatment intervention; a healthy-living intervention for heavy alcohol users; and hypertension linkage and community interventions. Standard-of-care and control groups are also listed. The brief study description says the randomized trials assess effectiveness, fidelity, and cost of HIV and hypertension prevention and treatment interventions, with the objective of informing a population-based study of multisectoral, multidisease interventions.

The primary endpoint result is narrower than that overall program. It compares the outpatient-clinic Dynamic Prevention Intervention with the outpatient-clinic control group. The record explicitly says this primary outcome comparison is not applicable to the other arms. The endpoint summaries include 182 participants in the intervention group and 194 in the control group; they should not be mistaken for the overall enrollment. The registry record does not report the randomization procedure, allocation ratio, site details, or the practical content of the outpatient intervention and control beyond their labels.

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.

Only the first 6 of 14 groups are shown.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Participants who completed the Week 48 visit.

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.

Only the first 6 of 15 groups are shown.

6. Primary Endpoint

The registry lists the primary outcome as Prevention Coverage Defined as Percentage of Follow-up Months That the Participant Used Either PrEP or PEP, measured over 48 weeks.

The primary endpoint is prevention coverage, defined as the percentage of follow-up months in which a participant used either PrEP or PEP. The measurement period is 48 weeks, and the primary definition relies on self-report. The analysis population specified in the record consists of participants who completed the Week 48 visit.

This measure describes consistency of reported use across follow-up, rather than a clinical outcome such as HIV acquisition. Expressing coverage as a percentage of follow-up months gives a summary of how much of the observation period involved reported PrEP or PEP use. The mean is the registry’s selected summary for this continuous measure. A group mean captures the average coverage but does not show how values vary among individuals or whether results are concentrated around the mean.

The endpoint is relevant to a prevention intervention because it records reported use of prevention during follow-up. Its interpretation depends on what participants recall and report, and on the visit-completion population specified by the registry. The record does not report validation of self-reported use, the exact handling of partial or missing follow-up information, or whether other coverage definitions were analyzed. The posted values should therefore be read as summaries under the stated primary definition, not as direct measures of infection prevention or as proof of the intervention’s broader impact.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is continuous: each participant’s coverage is expressed as a percentage of follow-up months, and the posted group summary is a mean. A mean is the arithmetic average of the individual observations included in the summary. It gives a central description of the group but can conceal variation, skewness, or a concentration of values at particular points. The registry record does not report additional distributional summaries for this endpoint.

What do the confidence intervals describe?

The record reports a 95% confidence interval for each group mean. In general, a confidence interval describes uncertainty about a parameter, such as a population mean, under a specified statistical method. Across repeated use of the same method in comparable samples, intervals would capture the target parameter at the stated rate if the method’s assumptions hold. This general interpretation does not mean that a particular interval has a 95% probability of containing the fixed parameter.

The intervals accompanying the two means are group-specific. They do not directly estimate the difference between the intervention and control means. A formal comparison would ordinarily require an estimate of that contrast and its uncertainty, potentially accounting for the trial design and analysis choices. Neither a contrast estimate nor its confidence interval is reported in the registry record. The fact that separate intervals are reported does not supply the missing comparison test, and their visual separation should not be treated as a reported significance result.

What is the role of the mean?

For a continuous endpoint, the mean gives an intuitive measure of average coverage. Its usefulness depends on the distribution of individual observations and on the target population. In general, means can be sensitive to extreme observations, and two groups with the same mean can have very different distributions. The record reports the mean and its confidence interval but does not report standard deviations, medians, ranges, or individual-level distributions. Those features cannot be reconstructed from the posted summaries.

The endpoint is bounded by its percentage definition, but the registry does not describe the statistical method used to accommodate the scale or its possible boundary values. It does not state whether the confidence intervals were based on a normal approximation, a transformation, resampling, or another procedure. No assumption about the method should be inferred from the values alone.

Why can missing outcomes matter?

The stated endpoint population includes participants who completed the Week 48 visit. In general, restricting summaries to people with an observed follow-up outcome can affect interpretation if completion is related to coverage or to group assignment. The registry does not report how many randomized participants lacked endpoint data, why data were missing, or whether imputation or sensitivity analyses were used. The endpoint-specific participant counts therefore describe the posted summaries, but they do not explain all movement from overall enrollment to the analyzed groups.

Randomization can support a fair comparison at assignment, but missing outcome data and post-randomization exclusions can complicate that comparison. In general, an analysis based on all randomized participants and an analysis limited to visit completers answer different questions. The registry identifies Week 48 completers as the primary outcome population; it does not report an intention-to-treat analysis or another analysis population for this endpoint.

What can be learned from planning illustrations?

Any sample-size illustrations associated with these results are planning examples based on the observed effect, not a record of the trial’s original design justification. The stated illustration assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation, with the effect to detect expressed as a fraction of the observed effect. General sample-size planning asks how much information may be needed to detect a chosen effect under specified assumptions. It does not show the probability that this completed study had of finding an effect, and it is not observed power.

The registry record does not report the protocol’s sample-size justification, the target effect used in planning, or the design calculations underlying the trial. Nor does it explain whether its planned design addressed multiple intervention areas or the particular outpatient endpoint. The planning assumptions should therefore be kept separate from the trial design and from the observed endpoint summaries.

10. Multiplicity, Missing Data and Analysis Populations

The study includes several intervention areas and the registry lists seven outcome measures. However, it does not report secondary outcome results, and it contains no analysis records. The available primary endpoint summaries concern only the outpatient-clinic PrEP/PEP comparison. The registry record does not report a prespecified hierarchy of outcomes, a multiplicity adjustment, or a plan for controlling the chance of false-positive findings across multiple comparisons.

In general, testing many outcomes or intervention comparisons can increase the chance that at least one result appears notable by chance. A prespecified analysis plan can clarify which results are primary and how multiple tests are handled. Without that information, the extent to which multiplicity was considered cannot be assessed. This is especially relevant to a multisectoral study with multiple intervention areas, but the record does not establish how many formal comparisons were planned or conducted.

Missing-data methods are also not reported. The primary endpoint population is described as participants who completed the Week 48 visit, and endpoint summaries are reported for 182 intervention participants and 194 control participants. The record does not state how many participants were randomized to these specific groups, how many lacked a Week 48 measure, why observations were unavailable, or whether any were imputed. It also does not report sensitivity analyses for missingness.

The analysis population is therefore defined only by the stated visit-completion criterion. Whether the results were analyzed according to randomized assignment, whether participants were excluded for other reasons, and whether any adjusted analyses were performed are not reported in the registry record. These omissions limit assessment of the analysis choices and of how well the posted group summaries represent all participants assigned to the outpatient comparison.

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.

Most frequent serious adverse event terms

Serious events with the highest arm-level rate: miscarriage (2.5%), chronic renal failure (2.0%), severe hypertension (2.0%), emergency caesarian section (2.0%), homicide (2.0%).

For the outpatient-clinic Dynamic Prevention Intervention group, the registry lists 7 participants affected by serious adverse events among 197 at risk. For the outpatient-clinic control group, it lists 0 affected among 206 at risk. The safety denominators differ from the primary endpoint group counts of 182 and 194; they describe the safety reporting groups and should not be substituted for the endpoint populations.

Across the other listed groups, the record reports serious adverse events in the antenatal-clinic dynamic intervention and control groups, the community-household dynamic intervention and control groups, the mobility intervention and control groups, the healthy-living intervention and control groups, and the hypertension linkage and community groups. The respective affected counts are 17, 2, 4, 1, 3, 2, 3, 1, 5, 0, 5, and 1. The registry lists no other adverse events in these groups. Deaths are listed in the mobility intervention and control groups, the heavy-alcohol-users control group, and the hypertension community intervention and control groups; the other listed groups have no deaths.

These are registry counts, not comparative risk estimates. They do not indicate the causes, timing, duration, or relationship to an intervention for each event. The record also lists serious-event terms and percentages, including Miscarriage at 2.5%, Chronic renal failure at 2.0%, Severe hypertension at 2.0%, Emergency caesarian section at 2.0%, Homicide at 2.0%, and Death of premature infant at 1.5%. The registry record does not give further context for interpreting those terms or their distribution by study group.

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

The posted primary results describe mean self-reported PrEP/PEP coverage among participants who completed the Week 48 visit. The dynamic intervention group mean is 47.5 percentage of months, and the outpatient-clinic control mean is 18.3 percentage of months. Their 95% confidence intervals describe uncertainty around the group means as reported; they are not a confidence interval for a treatment contrast.

The registry record does not report a formal statistical comparison, adjusted results, or a p-value for this endpoint. Consequently, the posted summaries alone do not establish whether the groups differed under a prespecified statistical test. They also do not show whether the result would remain similar under other assumptions about missing observations, population definitions, or analysis methods.

The endpoint measures reported prevention use over follow-up, not HIV infection, transmission, or other clinical outcomes. A higher coverage summary does not, on its own, show what happened to those outcomes. Because coverage relies on self-report, the summary also reflects the measurement method stated in the record; validation or corroboration is not reported.

The result applies to the outpatient-clinic comparison, not to every intervention in SEARCH SAPPHIRE. The study includes approaches in antenatal clinics, community households, mobility, alcohol use, and hypertension. The primary outcome description says it is not applicable to the other arms. The outpatient result therefore should not be generalized to those distinct interventions or settings on the basis of this endpoint.

More broadly, the registry describes an objective of assessing effectiveness, fidelity, and cost to inform a population-based study. The posted primary endpoint is one component of that wider objective. The registry record does not report results on fidelity or cost, secondary outcome findings, or a broader population-based evaluation. The results should be read as descriptive evidence about the stated coverage measure and its specified completion population, with the limits of the reported analysis in view.

13. Why This Trial Matters Statistically

This study is statistically notable for bringing multiple intervention areas and health concerns into a single multisectoral program, while reporting a specific primary endpoint for only one comparison. The distinction matters: an overall enrollment of 2233 belongs to the broader program, whereas the outpatient coverage summaries concern 182 intervention participants and 194 control participants. Keeping those populations separate avoids attributing the focused endpoint result to all study participants or intervention settings.

The endpoint also illustrates the difference between reporting group summaries and reporting a comparative analysis. Means and confidence intervals describe the observed coverage summaries and their uncertainty, but the registry does not provide a direct estimate or test of the between-group contrast. For interpretation, readers need to distinguish descriptive evidence from inferential conclusions rather than treating separate group intervals as a formal comparison.

Another statistical issue is the stated Week 48 completion population. In general, a result among visit completers can differ in meaning from a result among everyone randomized, particularly when follow-up completion is related to outcomes. The registry does not report the missing-data pattern or sensitivity analyses, so the extent of this issue cannot be judged from the posted record.

Finally, multiple interventions and outcomes make clear reporting of prespecified comparisons, multiplicity handling, and analysis populations important. The registry lists several intervention areas and seven outcome measures, but reports no secondary results and no analysis records. These gaps limit the ability to place the primary summary within a broader inferential framework. They do not negate the descriptive information, but they define what can responsibly be concluded from it.

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 outpatient endpoint is based on self-reported use, as stated in the registry. No validation method, corroborating data source, or detail on how use in a follow-up month was classified is reported. The result therefore should be interpreted as a summary under the stated self-report definition rather than an independently verified measure of medication use.

The primary population is participants who completed the Week 48 visit. The record does not report the number assigned to each outpatient arm, the number who did not complete the visit, reasons for non-completion, or the missing-data approach. This prevents assessment of how the endpoint summaries relate to all randomized participants and whether missing observations could have affected the group means.

Arm descriptions are blank in the registry record, and there is little operational detail about what the dynamic intervention or outpatient standard of care involved. The record also does not report the randomization procedure, allocation ratio, study sites, or any adjustment variables. These omissions make it difficult to evaluate implementation and the exact analysis framework from the registry alone.

Although group means and confidence intervals are posted, a between-group estimate, formal test, and p-value are not reported. The confidence-interval method is not specified. No analysis records or secondary outcome results are posted, despite the registry listing seven outcome measures and multiple intervention areas. Multiplicity planning and sensitivity analyses are not reported.

The registry’s safety counts are presented across groups, but event timing, relatedness, and detailed event-level context are not reported. The serious-event terms are listed with percentages without an explanation of their group distribution in the record. These limitations restrict interpretation of both the efficacy-related endpoint and the safety information to the summaries explicitly posted.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: SEARCH SAPPHIRE Phase A: A Multisectoral Strategy to Address Persistent Drivers of the HIV Epidemic in East Africa — NCT04810650.
• Registry record last updated 2025-02-06; status last verified 2025-02.
• Results first posted 2025-02-06.
• 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

SEARCH SAPPHIRE Phase A is listed as a completed randomized, parallel-group study with no masking and an overall enrollment of 2233. It addresses HIV and other health concerns through several intervention areas, including PrEP/PEP prevention, mobility, alcohol use, and hypertension. The sponsor is the University of California, San Francisco.

The primary posted result concerns mean self-reported PrEP/PEP coverage over 48 weeks among Week 48 visit completers. The outpatient-clinic Dynamic Prevention Intervention group mean is 47.5 percentage of months, with a 95% confidence interval of 42.3 to 52.7, among 182 participants. The outpatient-clinic control group mean is 18.3, with a 95% confidence interval of 14.4 to 22.1, among 194 participants.

The record does not report a formal comparison analysis, secondary outcome results, analysis records, missing-data methods, or multiplicity procedures. These results are descriptive group summaries for the specified outpatient comparison.