Improving Uptake and Adherence to HIV Prevention Services With Pre-exposure Prophylaxis (PrEP), HIV Self- Testing and Adherence Support for Very High Risk HIV-negative Young Women Aged 15-24 in Kampala, Uganda: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Improving Uptake and Adherence to HIV Prevention Services With Pre-exposure Prophylaxis (PrEP), HIV Self- Testing and Adherence Support for Very High Risk HIV-negative Young Women Aged 15-24 in Kampala, Uganda (NCT04030520), a randomized na study. The primary outcome is Initiation of PrEP.

2. Clinical Question

3. Trial Design

The registry describes a randomized, parallel-group trial with no masking. The trial is listed as completed and has a total enrollment of 60. It began on 2019-07-08, and primary completion is listed as 2020-10-31. The phase is recorded as NA. The sponsor is the University of California, San Francisco.

Participants were assigned to an intervention arm or a comparison arm, with 30 participants in each. The intervention arm was offered counseling, an HIV oral fluid self-test, and PrEP. The comparison arm received counseling and condoms and was offered PrEP, without an offer of an HIV oral fluid self-test. Both interventions are classified as behavioral in the registry. The record does not explain the randomization sequence, allocation concealment, or whether assignment used any stratification.

The study's broader purpose was to develop and assess an intervention to enhance PrEP initiation and adherence among HIV-negative young women at high risk. Its stated aims also included formative research on social and sexual networks, documenting barriers and opportunities for PrEP uptake and adherence and repeat HIV self-testing, and developing a prevention intervention considered socially and culturally acceptable and feasible. These broader aims give context to the trial, but the posted primary outcome is a count of participants who initiated PrEP.

There are 2 outcome measures listed, but no secondary outcomes are posted. The registry record lists no analyses. The trial's original sample-size justification, planned analysis details, and any interim analyses are not reported in the registry record. The sample-size illustrations associated with the posted effect are planning calculations, not documentation of the trial's original design.

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: Number of participants who initiated PrEP using pharmacy pill count records.

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 Initiation of PrEP, measured over 6 months.

The primary endpoint is initiation of PrEP, measured as the number of participants who initiated PrEP during the 6 months timeframe. The registry says initiation was identified using pharmacy records indicating PrEP received; its population description also refers to pharmacy pill count records. The record does not explain how these two descriptions relate or how records were handled when information was unavailable.

This endpoint addresses a practical step in prevention-service uptake: whether participants began PrEP. A count is a binary-outcome summary in this context, because each participant contributes either as someone who initiated or as someone who did not. It is distinct from continued use or adherence after initiation. Although adherence appears in the study's broader aims, no adherence result is posted as a secondary outcome. The registry also does not post a primary result for taking or repeating an HIV self-test.

There were 18 initiators among 30 participants in the intervention arm and 21 among 30 in the comparison arm. These are the posted event counts and arm sizes. They describe initiation recorded for the trial outcome; they do not, on their own, establish why a participant initiated or did not initiate PrEP. The endpoint's usefulness for interpreting the intervention is therefore paired with the need to consider the group comparison, its uncertainty, and the limits of the registry's reporting.

7. Results Posted in the Registry

• Comparison: 70.0% (21/30) versus 60.0% (18/30) on control; risk difference 10.0 percentage points (95% CI -14.0 to 34.0), risk ratio 1.17 (0.80–1.70); unadjusted two-proportion z-test p = 0.417.

8. Statistical Interpretation

9. Statistical Methods Explained

Several statistical ideas help interpret a binary outcome such as PrEP initiation. The registry reports counts of participants with the outcome in each arm, together with a risk difference, risk ratio, odds ratio, confidence intervals, and an unadjusted p-value. These summaries answer related but distinct questions. The exact analysis method behind the registry's inferential results is not reported in the registry record.

What does a risk difference describe?

General: a risk difference subtracts the outcome risk in one group from the outcome risk in another. It uses the outcome's original scale, so it describes an absolute contrast. A value of zero represents no difference in risk on this scale. The registry's estimate is 0.1, with a confidence interval from -0.14 to 0.34. That interval includes zero and also permits values on either side, so the reported data do not establish a precise absolute difference.

How do relative measures differ?

General: a risk ratio compares outcome risks by division. A value of 1 represents equal risks on that scale. An odds ratio compares the odds of the outcome between groups, rather than comparing risks directly; its null value is also 1. The reported risk ratio is 1.167, with a confidence interval from 0.802 to 1.697. The reported odds ratio is 1.556, with a confidence interval from 0.534 to 4.532. Both intervals include their null value. Relative measures can look different from an absolute contrast because they use different scales; they should not be treated as interchangeable descriptions.

What does the p-value say?

General: a p-value measures how unusual the observed data, or data more extreme, would be if a specified null model and its assumptions held. It does not give the probability that the null hypothesis is true, nor does it quantify the clinical importance of a difference. The registry reports an unadjusted p-value of 0.417. The statistical test, null specification, and calculation details are not reported in the registry record, so the value should be read as the posted result rather than as a reproducible analysis description.

What is the role of randomization?

General: random allocation is intended to create groups that are comparable on average at the point of assignment, including for factors that are not measured. It does not guarantee that every characteristic will be balanced in a particular trial, especially when the groups are small. Baseline characteristics and allocation procedures are not reported in the registry record, so the degree of observed balance and the implementation of randomization cannot be assessed from the available record. No masking was used, according to the registry.

What can be inferred from the posted analysis?

The record lists a primary endpoint and comparative estimates, but lists no analyses. That distinction means the numerical comparison is available while the full statistical analysis plan is not. It is not reported whether the primary comparison was adjusted for covariates, how missing outcomes were handled, or whether the analysis followed an intention-to-treat population. The term unadjusted appears with the p-value, but does not establish that every other estimate was calculated with the same method. The record also does not report model diagnostics or sensitivity analyses.

General: a small sample can produce wide uncertainty intervals because the data contain limited information about the outcome contrast. The same point estimate may be consistent with a range of plausible underlying effects. This is why interpreting the estimate together with its confidence interval is more informative than focusing only on whether a p-value crosses a conventional threshold. No claim of equivalence follows from a nonsignificant result; equivalence requires a defined margin and an appropriate analysis.

The binary outcome is reported as counts, with no dispersion measure. General: over-dispersion is a concern for outcomes whose variability exceeds what a statistical model assumes, particularly count outcomes modeled with a specified mean-variance relationship. It is not a separate summary needed to interpret these posted participant-level event counts. The registry does not report a model specification that would allow assessment of any such issue.

10. Confidence Intervals and Precision

Confidence intervals express uncertainty around an estimate under the method used to calculate them. General: a wider interval indicates that the data and analysis leave a broader range of values compatible with the estimate; a narrower interval indicates greater precision. An interval is not a range that guarantees where the true effect lies, and its interpretation depends on the statistical procedure and assumptions.

For PrEP initiation, the registry reports a risk difference of 0.1 with a confidence interval from -0.14 to 0.34. Because zero lies within that interval, the results are compatible with no absolute difference as well as with a difference in either direction. The endpoints also show that the estimate is not tightly constrained around its point estimate. The interval therefore communicates important uncertainty that is not conveyed by the point estimate alone.

The relative estimates are also imprecise. The risk ratio is 1.167, with a confidence interval from 0.802 to 1.697; the odds ratio is 1.556, with a confidence interval from 0.534 to 4.532. Each interval includes 1, the value representing no relative contrast. The registry does not state the confidence level or the calculation method for these intervals, so those details are not reported in the registry record.

The counts of 18 and 21 initiators come from arms with 30 participants each. These results should not be reduced to a binary conclusion based on the point estimate or p-value. The interval width matters: it indicates that the data do not precisely distinguish among several possible effect sizes and directions. The posted results do not establish that the arms are equivalent, nor do they rule out potentially meaningful differences. Any judgment about practical importance would require a context-specific threshold, which is not reported in the registry record.

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 original trial design assumptions. They are participants per arm, not total events. The planning assumptions listed are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is specified as a stated fraction of the observed effect. The trial's own original design assumptions and sample-size justification are not reported in the registry record.

The 80% power column gives the illustrative number of participants per arm for a design with 80% power to detect the stated effect under those assumptions. The 90% power column gives the corresponding number per arm for a design with 90% power. Power is a property of a planned design under specified assumptions; these columns are not observed power for the completed trial.

Rows labeled 100% of the observed effect, 75% of the observed effect, and 50% of the observed effect describe planning for progressively smaller effects relative to the observed effect. The table gives 356 and 477 participants per arm for the full observed effect, 644 and 862 per arm for 75% of the observed effect, and 1,471 and 1,969 per arm for 50% of the observed effect, at the listed power levels. Smaller effects generally require more participants to detect under the same planning assumptions, which is reflected in the increasing figures across those rows.

These calculations use an observed estimate from a small trial as their starting point. They should be interpreted as illustrations of how sample size changes with the effect assumed for planning, rather than as evidence that a future study will achieve a particular result. The registry's planning assumptions do not establish that the completed trial was designed with those assumptions.

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 = 0.600; treatment proportion p₁ = 0.700; average p̄ = 0.650.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6745 = 1.3221.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6708 = 0.5646.
4. n = (1.3221 + 0.5646)² / (0.1000)² = 355.9, rounded up to 356 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 <- 0.6                          # control proportion
p1 <- 0.7                          # 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(4030520)
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 = 356 (80% power), 477 (90% power)
 75% of observed effect: n per arm = 644 (80% power), 862 (90% power)
 50% of observed effect: n per arm = 1,471 (80% power), 1,969 (90% power)
power.prop.test: 355.9 
Simulated power at n = 356 per arm: 0.800 (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. Multiplicity, Missing Data and Analysis Populations

The registry lists 2 outcome measures, but no secondary outcome results are posted. The primary endpoint is PrEP initiation. The study aims also mention PrEP adherence, uptake, repeat HIV testing by self-test, and formative research, but the registry record does not provide posted outcome results for those topics. No analysis entries are listed.

Multiplicity arises when researchers test or estimate multiple outcomes, comparisons, or subgroups. General: as the number of statistical tests increases, the chance of at least one apparently notable result can increase if no adjustment or prespecified testing strategy is used. The record does not report a multiplicity plan, a hierarchy of outcomes, or adjustments for multiple comparisons. It is therefore not possible to determine whether multiplicity was addressed in the trial's original analysis.

Information about missing data is also not reported in the registry record. The posted counts describe 30 participants in each arm and the corresponding initiation counts, but the record does not explain whether outcome status was available for every assigned participant, how unknown status was handled, or whether any participants were excluded from the analysis. A count denominator matching the arm size does not by itself document the missing-data procedures or analysis population.

The registry does not state whether the analysis followed intention-to-treat, per-protocol, or another population definition. General: intention-to-treat analysis ordinarily retains participants in the groups to which they were randomized, while other analysis populations may depend on treatment received or protocol adherence. These approaches can answer different questions, and the population definition should be prespecified and clearly reported. Without the definition and participant flow, the reader cannot verify which population generated the posted estimates.

Subgroup analyses, sensitivity analyses, and interim analyses are not reported in the registry record. Their absence from the record should not be interpreted as evidence that they were or were not conducted. The available results support interpretation of the posted primary comparison, but not a reconstruction of a comprehensive statistical analysis plan.

14. 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 serious adverse event figures report zero serious events among 30 participants at risk in the intervention arm and zero among 30 participants at risk in the comparison arm. The registry also reports zero other adverse events in each arm and zero deaths in each arm. No serious adverse event terms are listed.

These are counts as recorded in the registry. A zero count means that no events of the corresponding type were reported for the listed group and at-risk count; it does not establish that the underlying event risk is zero. General: when few participants are observed, uncommon events may not occur in the trial even when they could occur in a broader population. The registry does not provide event definitions, collection procedures, follow-up details for safety, or a separate safety analysis.

The adverse-event counts should be read separately from the primary initiation outcome. They do not provide evidence about PrEP adherence or other outcomes, and they do not allow comparisons of event rates beyond the reported counts. No medical conclusion about safety can be drawn from the registry figures alone.

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

The posted result describes PrEP initiation in a randomized trial of young women at high risk for HIV in Kampala, Uganda. There were 18 initiators among 30 participants in the intervention arm and 21 among 30 in the comparison arm. The registry reports a risk difference of 0.1, a confidence interval from -0.14 to 0.34, and an unadjusted p-value of 0.417. Relative estimates are also posted, and their confidence intervals include the null value for those measures.

Statistically, the central point is uncertainty. The risk difference interval includes zero and spans values in both directions. The data therefore do not precisely determine whether initiation was more common in one arm, nor how large an underlying difference might be. The p-value does not resolve that uncertainty, and it is not a measure of whether an intervention is worthwhile. The result does not establish equivalence between the arms or rule out a potentially important difference.

The comparison is between intervention packages. The intervention arm was offered counseling, an oral fluid HIV self-test, and PrEP; the comparison arm received counseling and condoms and was offered PrEP, without an offer of the self-test. Accordingly, the comparison cannot isolate the contribution of a single component from the package. The record does not provide component-specific analyses.

The primary endpoint is initiation, not sustained adherence. Although adherence and repeat self-testing feature in the study aims, no secondary outcome results are posted. The result therefore cannot answer whether the groups differed in adherence, continued use, repeat testing, or other prevention outcomes. It also cannot establish effects in populations beyond the participants described in this trial.

The completed trial's sample size was 60. Its estimates are useful as the recorded results of this study, but the uncertainty intervals and the limited posted outcomes constrain interpretation. The record does not supply the detailed protocol, analysis plan, missing-data procedures, or baseline characteristics needed to assess all aspects of design and analysis. Interpretations should remain focused on the reported initiation counts and the uncertainty around the group comparison.

16. Why This Trial Matters Statistically

This trial illustrates why a clinical result should be read through both its estimate and its uncertainty. The posted initiation counts are straightforward, but the risk difference confidence interval extends across zero and permits effects in either direction. The result is not adequately described by selecting only the point estimate or the p-value. The interval shows that the available data leave substantial uncertainty about the magnitude and direction of the contrast.

It also highlights the difference between a study's broad aims and its posted endpoints. The registry describes goals relating to PrEP uptake and adherence, HIV self-testing, and formative research. Yet the posted primary comparison is initiation of PrEP, and no secondary results are posted. This distinction matters statistically: conclusions need to follow the outcome actually reported, rather than broader objectives that are not accompanied by results.

The study compares prevention packages that differ in more than one described feature. That design is relevant to evaluating a package, but it does not identify the effect of the self-test offer alone. Statistical interpretation must match the intervention contrast that was randomized. The record does not report component-level analyses that would separate contributions.

Finally, the planning illustrations show how assumed effect size affects the sample size required under stated assumptions. They are not the trial's documented original design rationale and do not change the uncertainty in the completed trial's outcome. Keeping design planning, observed estimates, and inferential precision distinct is central to responsible interpretation. The registry record makes the numerical comparison available, while leaving important details about methods and outcomes unreported.

17. 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 record provides limited detail for assessing trial conduct and analysis. Although it identifies the trial as randomized and unmasked, the randomization sequence, allocation concealment, baseline characteristics, participant flow, and protocol deviations are not reported in the registry record. These omissions limit assessment of how assignment was implemented and whether the groups were comparable in the realized sample.

The outcome descriptions refer to pharmacy records indicating PrEP received and pharmacy pill count records. The record does not explain how these sources were used together, what qualified as initiation, or how unavailable or inconsistent records were treated. The follow-up timeframe is 6 months, but details of outcome ascertainment and completeness are not reported.

The trial enrolled 60 participants, with 30 in each arm. The reported confidence intervals are broad enough to include effects in either direction for the risk difference, and the original sample-size justification is not reported in the registry record. The planning illustrations are based on the observed effect and should not be confused with the original design rationale.

The registry lists 2 outcome measures but posts no secondary results, and it lists no analyses. The analysis population, missing-data approach, adjustment strategy, confidence-interval method, and p-value test are not reported in the registry record. Multiplicity and interim-analysis plans are also not reported. These gaps prevent a full reconstruction of the prespecified analysis or assessment of whether the posted result reflects all randomized participants.

Finally, the reported comparison concerns the intervention package as described and the comparison package. It cannot isolate effects of counseling, condoms, PrEP offers, or self-testing offers individually. Results for adherence and repeat self-testing are not posted, so the registry record does not answer those broader study aims.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Improving Uptake and Adherence to HIV Prevention Services With Pre-exposure Prophylaxis (PrEP), HIV Self- Testing and Adherence Support for Very High Risk HIV-negative Young Women Aged 15-24 in Kampala, Uganda — NCT04030520.
• Registry record last updated 2025-07-15; status last verified 2025-07.
• Results first posted 2025-07-15.
• 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.

21. Record Summary

The registry record describes a completed randomized, unmasked, parallel-group trial sponsored by the University of California, San Francisco. It enrolled 60 young women aged 15-24 at high risk for HIV in Kampala, Uganda, with 30 participants in each arm. The intervention included counseling and an offer of an HIV oral fluid self-test and PrEP; the comparison included counseling, condoms, and an offer of PrEP without an offer of the self-test.

The primary endpoint was initiation of PrEP during the 6 months timeframe. The intervention arm had 18 initiators and the comparison arm had 21. The reported risk difference was 0.1, with a confidence interval from -0.14 to 0.34, and the unadjusted p-value was 0.417. The record posts no secondary outcome results and lists no analyses. Details of the original sample-size justification, missing-data handling, and statistical analysis plan are not reported in the registry record.