Implementation Effectiveness and Safety of Tenofovir Gel Provision Through Family Planning Services: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Implementation Effectiveness and Safety of Tenofovir Gel Provision Through Family Planning Services (NCT01691768), a randomized phase 2, phase 3 study. The primary outcome is Mean Number of Returned Used Applicators Per Month (i.e in 30 Days).

2. Clinical Question

The clinical question concerns implementation as well as product use. The trial compared gel delivery through existing family planning services with delivery through research clinics; the schedules and service settings differed. Its primary endpoint was the mean number of returned used applicators per month, a measure intended to reflect use during follow-up rather than a direct measure of HIV prevention.

The registry also lists secondary outcomes concerning HIV and pregnancy incidence, adherence, viral load and tenofovir resistance among HIV seroconverters, human papillomavirus incidence, detectable tenofovir levels in vaginal samples at 12 months of follow-up, and product acceptability. The record does not report numerical results for these outcomes. It therefore does not establish from the posted results whether the service-delivery approaches differed on the primary endpoint or on the secondary outcomes.

3. Trial Design

This was an open-label randomized controlled trial with a parallel-group model. The registry lists its phase as PHASE2 and PHASE3, enrollment as 372, and status as completed. The study start is recorded as 2012-10 and primary completion as 2015-04. The official title locates the study in KwaZulu-Natal, South Africa, and the listed sponsor is the Centre for the AIDS Programme of Research in South Africa.

Random allocation assigns participants to study groups by chance, which is intended to reduce systematic differences between groups at enrollment. In a parallel design, each participant is assigned to one group rather than crossing between both provision approaches. The trial was not masked: participants and study personnel were not blinded to the assigned service-delivery approach, as indicated by the registry’s “NONE” masking entry. The record does not report further details about the randomization sequence, allocation concealment, or procedures used to implement assignment.

The intervention group received 1% tenofovir gel through public-sector family planning services. Provision and monitoring were scheduled every 2-3 months, and Quality Improvement methodology was used to promote reliable service delivery. The control group received monthly provision and monitoring through CAPRISA research clinics. Thus, the contrast combines differences in service setting and provision schedule; the registry does not describe a design that isolates the effect of either difference on its own.

The study’s stated purpose was to assess the effectiveness of an implementation model integrating tenofovir gel provision into existing family planning services. The primary endpoint was a continuous measure of returned used applicators per month. The record describes both intent-to-treat and per-protocol analyses, but does not report the estimates, confidence intervals, or tests for either analysis. It also does not report a detailed statistical analysis plan, sample-size justification, or whether interim analyses were conducted.

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: 448 were assessed for eligibility and 382 were randomized. However, only 372 were eligible for study participation. Of the 10 who were ineligibly enrolled: 6 were HIV positive at enrollment, 2 were co-enrolled in another trial, 1 had no post-randomization follow-up data and 1 was pregnant at enrollment.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Intent to treat population and the per protocol population (excluding all subsequent data collected from participants who were not dispensed product for more than 120 days).

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: Intent to treat population was analyzed. All participants who were randomized and met the pre-enrollment eligibility criteria.

6. Primary Endpoint

The registry lists the primary outcome as Mean Number of Returned Used Applicators Per Month (i.e in 30 Days), measured over Between 2012 to 2015, up to 28 months.

The primary endpoint was the mean number of returned used applicators per month, measured in used gel applicators per month. It was intended to provide a common monthly scale for comparing use when the visit schedules differed. The intervention group had provision and monitoring every 2-3 months, whereas the control group had monthly provision and monitoring. For visits that were not monthly, the registry says the number of returned applicators was divided by the days since the previous visit and multiplied by 30. This estimates a monthly rate from the interval between visits.

The registry states that this approach assumes gel use was uniformly distributed over time for participants not seen monthly. In general, converting counts to a rate can make observations from different follow-up intervals more comparable, but the conversion relies on assumptions about how use occurred between visits. If use varied over an interval, a monthly rate derived from the total returned applicators would not show when use occurred.

Returned used applicators are an indirect measure of product use. The measure does not itself establish correct use, adherence at every point in follow-up, or a biological effect. The registry record specifies a least-squares mean and a 95% confidence interval as the summary and uncertainty measure, but does not report the resulting mean estimates or intervals. Consequently, the size, direction, and precision of the primary comparison are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is continuous: it is a mean count expressed as used gel applicators per month. The registry identifies the summary as a least-squares mean. In general, a least-squares mean is a model-based estimate of a group mean, often accounting for variables included in the statistical model. Its interpretation depends on the model and covariates used. The specific model, adjustment variables, and handling of repeated observations are not reported in the registry record, so the precise adjustment represented by the stated least-squares mean cannot be described.

The registry also specifies a 95% confidence interval. In general, a confidence interval gives a range of values compatible with the data and analysis, while expressing sampling uncertainty. Its width gives information about precision, and its placement relative to a comparison value can help readers assess the estimate. The record does not report the interval values for the primary endpoint, so precision cannot be evaluated from the registry information.

The endpoint calculation addresses different intervals between visits. The control group had monthly visits, while the intervention schedule was every 2-3 months. For a non-monthly visit, the registry describes dividing returned applicators by the number of days since the prior visit and multiplying by 30. This standardizes the result to a monthly rate. The stated uniform-distribution assumption means the calculation treats use as spread evenly across the interval. In general, this is an assumption about the pattern of use, not a direct observation of daily behavior. Uneven use between visits could be obscured when a total is converted to a monthly rate.

Randomization and masking address different design issues. Randomization assigns participants to groups by chance and can help balance measured and unmeasured characteristics on average. Masking limits awareness of group assignment; this trial is listed as having no masking. In general, lack of masking may matter for outcomes affected by participant behavior, reporting, or study procedures. Returned applicators provide a concrete measure, but the record does not detail how returns were collected, checked, or assessed. These points identify considerations for interpretation; they do not establish that bias occurred.

The parallel-group structure means participants were assigned to the family-planning-service approach or the research-clinic approach. The intervention contrast includes both setting and schedule, as described in the registry. In general, when multiple aspects differ together, the comparison estimates the contrast between the approaches as implemented rather than the isolated effect of any one component. The record does not describe additional design elements that would separate the contribution of service setting from visit frequency or monitoring.

The registry describes an intent-to-treat population consisting of randomized participants who met pre-randomization eligibility criteria and had post-enrollment follow-up data. This definition is specific to the listed primary outcome. In general, an intent-to-treat approach aims to preserve the comparison created by random assignment, although the record’s eligibility and follow-up criteria determine who enters this stated analysis population. The registry does not report how many participants contributed outcome data or how missing observations were handled.

The per-protocol population is described as a subset of the intent-to-treat population. The registry says the per-protocol analysis excluded visits where no gel had been dispensed for >120 days. Excluding such visits can change which observations contribute to the analysis, so the per-protocol estimate answers a different analytic question from an analysis retaining all eligible follow-up. The record does not report the number of observations excluded, the resulting estimates, or further details about how the exclusion was applied across visits and participants.

The sample-size material is described as a planning illustration computed from the observed effect, rather than the trial’s own design calculation. Its stated assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation; the effect to detect is defined as a stated fraction of the observed effect. In general, such illustrations ask how many participants per arm would be needed under specified assumptions to detect an effect of a chosen size. They are planning exercises, not observed power and not evidence that the trial had a particular probability of finding a result. The registry record does not report the trial’s own sample-size justification.

For a continuous outcome, planning calculations depend on the assumed variability as well as the effect size and testing assumptions. The details and numerical outputs needed to assess those assumptions are not reported in the registry record. No planning value should be treated as an observed finding. The sample-size assumptions also do not substitute for the primary estimate and confidence interval, which are not reported.

General statistical interpretation should keep the estimate, its uncertainty, the endpoint’s measurement properties, and the analysis population in view together. Here, the record identifies the endpoint as a least-squares mean with a 95% confidence interval, but does not include the result values or enough modeling detail to reconstruct the analysis. Accordingly, the design and methods can be described, but the numerical evidence for the primary comparison cannot be evaluated from the registry record.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists multiple outcomes: the primary measure of returned used applicators per month and secondary outcomes concerning HIV incidence, pregnancy incidence, adherence, viral load and tenofovir resistance among HIV seroconverters, human papillomavirus incidence, detectable tenofovir levels from vaginal samples at 12 months of follow-up, and product acceptability. When a study examines multiple outcomes, there is a general possibility that some results may appear notable by chance. Statistical approaches to multiplicity can address that issue, but the registry record does not report whether a multiplicity adjustment was planned or used, how outcomes were prioritized beyond the stated primary endpoint, or how secondary analyses were interpreted.

The primary endpoint includes intent-to-treat and per-protocol analyses. The record defines the populations but does not state whether one was designated as the principal analysis for inference, how the two analyses were compared, or whether any multiplicity procedure covered both. It also does not report whether interim analyses were conducted or whether any adjustment was made for them. These details are not reported in the registry record, so no specific analysis plan should be inferred.

The registry defines the intent-to-treat population as randomized participants who met pre-randomization eligibility criteria and had post-enrollment follow-up data. This definition means that the record’s stated population is not simply described as everyone randomized regardless of follow-up. It does not report the extent or pattern of missing outcome data, reasons for missingness, imputation methods, or sensitivity analyses. Missing-data assumptions can affect an estimate, particularly when follow-up availability is related to outcomes or group assignment; whether that occurred in this trial is not reported.

The per-protocol description excludes visits where no gel had been dispensed for >120 days. The record does not report how many visits this affected or how the exclusion influenced estimates. These omissions prevent assessment of the impact of missingness, exclusions, and multiple analyses on the reported findings. Since the numerical primary and secondary results are not reported in the registry record, there are no posted estimates with which to evaluate these issues empirically.

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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: caesarean section (2.6%), gastritis (1.1%), cardiac failure congestive (0.5%), crohn's disease (0.5%), cholecystitis (0.5%).

The registry reports safety figures by group. For the intervention group, the serious-event figures list 189 at risk and 13 affected; for the control group, they list 183 at risk and 17 affected. The other-event figures list 189 at risk and 166 affected in the intervention group, and 183 at risk and 181 affected in the control group. The record also lists deaths as 0 in the intervention group and 2 in the control group. These are counts as entered in the record, not evidence by themselves that an event was caused by the gel or by the service-delivery approach.

The listed serious-event terms include Caesarean section at 2.6%, gastritis at 1.1%, and cardiac failure congestive, Crohn's disease, cholecystitis, and oesophageal candidiasis at 0.5% each. The record does not provide further context for these terms in the safety information, including event timing, severity grading, attribution, or whether the same participant could contribute more than one event. Percentages and group counts should therefore be read as registry safety summaries, not as a causal comparison.

In general, safety counts need context about definitions, follow-up, exposure, and attribution. The record does not report enough detail to infer why the listed events occurred or to compare risk while accounting for differences in exposure time. No causal conclusion follows from the group counts alone.

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

This trial addresses an implementation question: how 1% tenofovir gel provision through public-sector family planning services compares with monthly provision and monitoring through research clinics. The randomized, parallel-group design is relevant to that comparison, and the registry specifies a primary measure based on returned used applicators per month. The endpoint was scaled to a monthly rate to accommodate the different visit schedules, using the uniform-use assumption stated in the record.

What the posted record does not establish is the result of that comparison. It does not report the least-squares mean estimates, their 95% confidence intervals, or a test for the primary endpoint. As a result, the observed direction, magnitude, and precision of the primary comparison are not reported in the registry record. The outcome definition alone cannot show whether one group returned more or fewer applicators per month.

The listed secondary outcomes broaden the study’s scope to incidence, adherence, biological measures, and acceptability. However, the record does not report their numerical results. Their listing should not be read as evidence that a difference was found, nor as evidence that no difference was found. The same restraint applies to safety: group counts and event terms are reported, but they do not establish causation or explain the events.

The comparison also reflects a package of service-delivery features. Family planning services used a 2-3 monthly provision and monitoring schedule with Quality Improvement methodology, while research clinics used monthly provision and monitoring. In general, this type of contrast can inform the comparison between the approaches as implemented; it does not, by itself, isolate the contribution of each component. The registry does not describe a component-specific analysis.

The primary endpoint is a proxy for product use based on returned applicators. It is not a direct measure of every use episode, and the conversion for longer intervals assumes that use was uniform over time. The record therefore supports a description of what was measured and how the rate was derived, but not an unrestricted interpretation as actual use at each point in follow-up or as evidence of a biological effect.

Overall, the registry record documents a completed randomized implementation trial and sets out a primary endpoint, analysis populations, and other outcomes. It does not provide the numerical outcomes needed to judge the comparative findings. Conclusions about benefit, lack of benefit, statistical significance, or clinical effectiveness are not supported by the posted results described in the record.

13. Why This Trial Matters Statistically

The statistical challenge in this trial is not only comparing two randomized groups; it is defining a comparable measure when their service schedules differ. The intervention group had provision and monitoring every 2-3 months, while the control group had monthly provision and monitoring. Expressing returned used applicators as a monthly rate attempts to put observations on a common scale. The method also depends on the registry’s stated assumption of uniform gel use between visits.

This makes the endpoint’s construction central to interpretation. In general, a rate based on a longer interval can conceal variation within that interval, and the assumptions used to convert observations should be considered alongside the estimate. A reader evaluating implementation research benefits from knowing both what the measure captures and what it cannot directly observe. Here, the measure is returned used applicators, not a complete record of use or a direct clinical outcome.

The trial also compares delivery models rather than only a product against no product. One approach used existing public-sector family planning services and Quality Improvement methodology; the other used monthly provision and monitoring through research clinics. This is statistically important because the estimated contrast pertains to these approaches as described. It should not automatically be attributed to any single element, such as visit frequency or setting.

Finally, the distinction between intent-to-treat and per-protocol populations matters. The registry defines both and identifies a per-protocol exclusion related to time without product dispensed. Such choices can alter which follow-up contributes to an estimate, so transparent reporting of estimates, uncertainty, missingness, and exclusions is important. The registry record does not report the primary numerical results or these additional analysis details. The study is therefore informative about the design and measurement questions, while the posted record does not permit a statistical judgment of the observed comparative 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.

Several limitations follow from the design and endpoint description in the registry record. First, the comparison involved different delivery settings and schedules: public-sector family planning services with provision and monitoring every 2-3 months, versus monthly provision and monitoring through research clinics. Any group contrast would reflect those approaches together; the record does not report an analysis separating the effects of setting, schedule, monitoring, or Quality Improvement methodology.

Second, the primary outcome is based on returned used applicators. It is an indirect measure of product use and cannot establish use on particular days or correct use. To account for non-monthly visits, the registry describes dividing returns by the interval in days and multiplying by 30, assuming uniform use. That assumption may not describe every participant’s pattern, but the record does not report data that could assess its fit.

Third, the study was open-label. In general, awareness of assignment can affect behavior or reporting, although the record does not establish whether it did so. Further information on collection and assessment of returned applicators is not reported in the registry record.

The record defines intent-to-treat and per-protocol populations but does not report the amount or pattern of missing data, how missing observations were handled, or the numerical results for either population. It also does not report the analysis model, adjustment variables, multiplicity procedures, or sensitivity analyses. Finally, numerical results for the primary endpoint and listed secondary outcomes are not reported in the registry record. This limits assessment of the comparative findings, their precision, and their robustness.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Implementation Effectiveness and Safety of Tenofovir Gel Provision Through Family Planning Services — NCT01691768.
• Registry record last updated 2019-11-26; status last verified 2019-11.
• Results first posted 2019-11-26.
• 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, open-label, randomized, parallel-group trial of 1% tenofovir gel provision through public-sector family planning services compared with monthly provision and monitoring through CAPRISA research clinics. Enrollment is listed as 372, and the study is recorded as PHASE2 and PHASE3. The stated purpose was to assess an implementation model integrating gel provision into existing family planning services.

The primary endpoint was the mean number of returned used applicators per month, summarized as a least-squares mean with a 95% confidence interval. The record describes a monthly-rate conversion for non-monthly visits and specifies intent-to-treat and per-protocol populations. It also lists secondary outcomes concerning incidence, adherence, biological measures, and acceptability.

The registry record does not report numerical results for the primary endpoint or the listed secondary outcomes. It therefore supports description of the trial question, design, and measurement approach, but not a conclusion about the observed comparative effect.