Innovative Public-private Partnership to Target Subsidized Antimalarials in the Retail Sector (Aim 2): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Innovative Public-private Partnership to Target Subsidized Antimalarials in the Retail Sector (Aim 2) (NCT02461628), a randomized na study. The primary outcome is Number of Subjects With a Fever Who Receive a Malaria Test From Any Source.

2. Clinical Question

The study addressed whether a subsidy conditional on a positive malaria test, delivered through a partnership involving community health volunteers and private retailers, could affect malaria testing and treatment-seeking patterns. The stated hypothesis focused on increasing uptake of testing and the proportion of fevers tested before treatment. Those aims describe the question being studied, not a finding: numerical estimates for the primary outcome are not reported in the registry record.

The study took place in two Kenyan sub-counties described as having similar malaria burden but different access to health services. Community units, rather than individual participants, were assigned to intervention or comparison. This distinction matters when interpreting the question: the intervention was organized at community level, while the primary outcome counts participants with fever who received a test. The record does not report the number of community units assigned to each arm or the primary outcome results.

3. Trial Design

This was a randomized, parallel-group study with no masking, and the registry lists an enrollment of 40340. The study was conducted in two sub-counties in Kenya. Community units in each sub-county were assigned to an intervention or comparison arm, making this a cluster-randomized design rather than an individually randomized trial. The registry record does not report the number of community units or the allocation procedure beyond stating that the study was randomized.

In the intervention arm, trained community health volunteers offered eligible household members a free malaria rapid diagnostic test. Households in intervention community units were to be informed about the intervention and encouraged to contact a volunteer for febrile illness. A positive rapid diagnostic test qualified the person for a serialized voucher to purchase a quality-assured artemisinin combination therapy at a reduced fixed price from the retail sector. The comparison arm received standard community health volunteer visits.

The primary outcome was the number of subjects with fever who received a malaria test from any source. The stated outcome population comprised subjects with fever who completed a survey at 6 months, 12 months, or 18 months. Secondary outcomes concerned antimalarial use relative to testing, receipt of a correct artemether-lumefantrine dose, and correct treatment among subjects with fever. The registry record identifies the study as completed, with a start date of 2015-07 and a primary completion date of 2017-05-26. It does not report a masking procedure, which is consistent with the recorded masking status of none.

Cluster assignment can affect how uncertainty is estimated because people in the same community unit may have more similar outcomes than people in different units. The number of clusters and details of the sponsor's analysis are not reported in the registry record, so the strength and handling of this dependence cannot be assessed from the record's numerical results.

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: After enrollment, 7365 participants with a fever who received a malaria test from any source were approached to complete the survey. Only those 7365 participants are considered Started. Baseline data collection occurred before randomization of clusters to the arms.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Subjects with fever who completed the survey at 6, 12, or 18 months.

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: All participants who completed the survey.

6. Primary Endpoint

The registry lists the primary outcome as Number of Subjects With a Fever Who Receive a Malaria Test From Any Source, measured over 6 months, 12 months, 18 months.

The primary endpoint counts subjects with fever who received a malaria test from any source. It is a count of participants, not a direct measure of test accuracy, treatment success, or the proportion of all febrile illnesses tested. The endpoint is defined for subjects with fever who completed a survey at 6 months, 12 months, or 18 months. The registry record does not explain whether results were reported separately at each time point or combined.

This endpoint relates to the study's stated public-health question: whether a community-level partnership and conditional subsidy were associated with malaria testing among people with fever. Counting tests from any source allows the outcome to encompass testing beyond the volunteer's rapid diagnostic test. It does not, by itself, establish when testing occurred relative to treatment or whether a result led to appropriate treatment; those are distinct questions addressed in the study's stated aims and secondary outcomes.

The endpoint is a count, so its interpretation depends on the number of eligible people observed and on the analysis method used. The registry record does not report arm-level counts, an effect estimate, a confidence interval, or a statistical test for this outcome. Consequently, the record does not establish the size, direction, or precision of any difference between arms.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry record identifies a randomized parallel-group study in which community units were assigned to intervention or comparison. This is a cluster-randomized design: assignment occurred at the community-unit level, while outcomes were defined for people with fever. Statistical interpretation must keep those levels distinct. The record does not report the number of community units or the details of the analysis used to account for clustering.

Why does cluster assignment matter?

In general, outcomes from people in the same cluster can be correlated because they share local circumstances or exposure to the intervention. If an analysis treats each person as independent, it may understate uncertainty. Cluster-aware approaches can account for this dependence through methods such as cluster-level summaries, suitable regression models, or variance adjustments. These are general examples, not descriptions of the sponsor's method. The statistical method used by the sponsor is not reported in the registry record. The record explicitly cautions that a simple individual-level recalculation would be misleading.

What does the primary outcome count?

The primary outcome is the number of subjects with fever who received a malaria test from any source, among subjects who completed a survey at 6 months, 12 months, or 18 months. A count is not automatically a risk, proportion, or rate. To interpret a count comparatively, a reader would need to know the eligible population size and how follow-up and time points were handled. The registry record does not report those details or any arm-level outcome counts. General statistical practice would define the estimand and denominator clearly before choosing a model; the record does not report a model specification.

What can an estimate and confidence interval tell a reader?

In general, an effect estimate describes the observed contrast between groups on a specified scale, while a confidence interval expresses the uncertainty around that estimate under the analysis assumptions. A narrow interval indicates greater precision than a wide interval, but neither an estimate nor its interval establishes causation without a design and analysis that support the comparison. Here, the registry record does not report a primary-outcome estimate or confidence interval, so the size and precision of the contrast cannot be evaluated. No p-value or sponsor test result is reported either.

How should the planning illustrations be read?

The registry record's planning note describes illustrations based on the observed effect, with a two-sided significance level of 0.05, equal allocation, and a normal approximation. The stated effect to detect is a specified fraction of the observed effect. In general, a sample-size calculation estimates the number of participants or events required under assumptions about the effect, variability, allocation, and error rates. Such a calculation is a planning aid, not evidence that a study had a particular probability of detecting its observed result. Sample-size illustrations are never observed power. The numerical effect and planning outputs are not reported in the registry record, so their assumptions cannot be checked against actual outcomes.

The note distinguishes participants per arm for outcomes other than time-to-event outcomes, for which it describes total event counts. The primary endpoint is a participant count, not a time-to-event endpoint. This distinction describes the planning convention; it does not supply an observed result or identify the trial's original design assumptions. The registry record does not report the original sample-size justification.

What does the outcome structure imply?

The primary endpoint and secondary outcomes reflect related but distinct aspects of testing and treatment. In general, a count of people who received a test does not establish whether testing was followed by appropriate treatment. Likewise, counts related to antimalarial use or dose correctness do not establish testing uptake. Each outcome needs its own definition, analysis population, and effect estimate. The registry record does not report numerical results for the primary or secondary outcomes, nor does it specify how their uncertainty was estimated.

Because no sponsor model, cluster adjustment, confidence interval, or test result is reported in the registry record, the available information supports description of the trial question and design, not a quantitative efficacy interpretation. General methods can explain what would be needed for interpretation, but cannot substitute for the missing trial estimates.

10. Multiplicity, Missing Data and Analysis Populations

The registry record lists a primary outcome and secondary outcomes concerning antimalarial use, testing history, correct artemether-lumefantrine dosing, and correct treatment among subjects with fever. It does not report whether the analyses of these outcomes were adjusted for multiple comparisons, whether a hierarchy or gatekeeping procedure was used, or which outcomes were designated for formal hypothesis testing beyond identifying the primary outcome. The approach to multiplicity is therefore not reported in the registry record.

The primary outcome population is described as subjects with fever who completed the survey at 6 months, 12 months, or 18 months. The record does not report how missing survey data were handled, whether participants could contribute at more than one time point, or whether missingness differed between arms. It also does not report imputation, sensitivity analyses, or rules for excluding observations. These details matter because incomplete follow-up can change which participants contribute to an outcome estimate and may affect its interpretation.

The record does not specify whether the analysis used an intention-to-treat, per-protocol, or other analysis population. It does identify the primary outcome population in terms of fever and survey completion, but does not describe additional analysis-population rules. The number of outcome measures is listed as 5, and the record identifies 0 analyses; numerical outcome estimates are not reported in the registry record. Consequently, the available information does not establish how multiplicity, missingness, or analysis-population choices affected the study's results.

11. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

The registry record lists safety counts for the Malaria RDT & Conditional Voucher arm and the Comparison Arm. In the intervention group, the number at risk is 3729 for serious events and 3729 for other events; the recorded counts are 0 serious events, 0 other events, and 0 deaths. In the comparison group, the number at risk is 3625 for serious events and 3625 for other events; the recorded counts are 0 serious events, 0 other events, and 0 deaths.

These are the counts recorded in the registry, not comparative effect estimates. The record does not report a risk difference, confidence interval, or statistical comparison for safety outcomes. It also does not provide the event terms, follow-up duration for safety, or additional detail about how events were collected. The absence of recorded events in these groups should therefore be read as a description of the registry counts, not as proof that no events occurred outside the recorded safety information or as a complete estimate of safety.

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

The registry record establishes that the study tested a community-level approach linking volunteer-delivered malaria testing to a conditional retail subsidy for antimalarial treatment. It also identifies the comparison condition, the primary outcome, and the planned survey time points. Those details clarify what the trial was designed to examine. They do not show whether the intervention changed testing or treatment outcomes.

For the primary outcome, the registry record does not report counts by arm, an estimated contrast, a confidence interval, or a p-value. It is therefore not possible to describe the size, direction, or precision of a difference in testing from this record. The listed secondary outcomes likewise have no numerical results in the registry record. Statements about the stated hypothesis should not be mistaken for evidence that it was supported.

The cluster-randomized design is relevant to any future interpretation of results. In general, statistical uncertainty should reflect the community-unit assignment and potential similarity among participants within a unit. The record does not describe the sponsor's model or other adjustments, so that aspect of the analysis cannot be evaluated from the registry information. The planning illustrations, based on fractions of an observed effect, are not observed power and do not fill the gap left by absent outcome estimates.

The safety record lists zero serious events, zero other events, and zero deaths in each arm's recorded group. These counts are descriptive and do not quantify a comparative safety effect. Overall, the record supports a description of the trial's question, setting, and design, but not a conclusion about the intervention's impact or comparative safety.

13. Why This Trial Matters Statistically

This trial is statistically notable because the intervention was assigned at the level of community units while the primary outcome was defined for individual subjects with fever. That setup makes the assignment structure important to estimation and uncertainty. A participant-level count does not make the randomized units individual participants: analyses must respect the cluster allocation and possible correlation within communities.

The outcome also links a community strategy to testing from any source. This broad definition is relevant to a program that combined volunteer testing and a retail voucher, but it differs from outcomes about whether a volunteer performed the test, whether treatment followed a positive result, or whether dosing was correct. The trial's secondary outcomes address some related behaviors, yet their numerical results are not reported in the registry record.

The contrast between intervention and standard volunteer visits illustrates why an intervention's delivery pathway should be distinguished from its outcome. The study aimed to evaluate a targeted subsidy partnership and its community-wide effects, but a stated aim or hypothesis is not a measured effect. The registry record describes the question and design without reporting estimates that would establish the observed impact.

Finally, the cluster design limits what can be learned from a simple re-analysis of participant counts. In general, valid uncertainty estimates depend on accounting for how assignment occurred. The sponsor's analysis details and numerical estimates are not reported in the registry record, so the intervention's statistical effect and its precision cannot be assessed. The record therefore matters as a clear illustration of the information required to interpret a community-randomized clinical study, while leaving its central outcome question unanswered numerically.

14. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The most important limitation for statistical interpretation is that numerical results for the primary and secondary outcomes are not reported in the registry record. There are no arm-level testing counts, estimates of effect, confidence intervals, or test results with which to assess direction, magnitude, or precision. The record describes the primary endpoint and its population but does not say whether observations at 6 months, 12 months, and 18 months were analyzed separately or combined.

The community-unit assignment creates a further interpretive constraint. The record does not report the number of units, the sponsor's statistical model, or how clustering was handled. It explicitly cautions against a simple individual-level recalculation. Without these details and the numerical sponsor estimates, the analysis cannot be independently evaluated from the registry record.

Other important details are also not reported: the original sample-size justification, missing-data procedures, multiplicity approach, and analysis-population rules beyond the primary outcome population description. The listed safety counts are limited to the recorded event categories and do not include event terms or a safety follow-up description. These gaps limit assessment of the results and their uncertainty; they do not, on their own, establish that the study was poorly conducted.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Innovative Public-private Partnership to Target Subsidized Antimalarials in the Retail Sector (Aim 2) — NCT02461628.
• Registry record last updated 2019-10-29; status last verified 2019-10.
• Results first posted 2019-10-29.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

18. Record Summary

The registry record describes a completed randomized, parallel-group, unmasked cluster trial in two Kenyan sub-counties. Community units were assigned to an intervention combining volunteer-delivered malaria rapid diagnostic testing with a conditional fixed-price retail voucher for quality-assured artemisinin combination therapy, or to standard community health volunteer visits. The listed enrollment is 40340; the study start was 2015-07 and primary completion was 2017-05-26.

The primary outcome counts subjects with fever who received a malaria test from any source among survey completers at 6 months, 12 months, or 18 months. Numerical primary and secondary outcome results, sponsor analysis details, and confidence intervals are not reported in the registry record. The safety groups list 0 serious events, 0 other events, and 0 deaths in each arm. These counts are descriptive and do not establish a comparative safety effect.