Azithromycin for Child Survival in Niger: Programmatic Trial (AVENIR): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Azithromycin for Child Survival in Niger: Programmatic Trial (AVENIR) (NCT05288023), a randomized phase 4 study. The primary outcome is Prevalence of Genetic Determinants of Macrolide Resistance From Population-based Samples.

2. Clinical Question

The study was designed to evaluate large-scale implementation in a programmatic setting while monitoring resistance and mortality. Its registered primary question is about resistance determinants, not a direct comparison of child mortality. Mortality appears in the study rationale: the registry summary says the MORDOR trial found that biannual azithromycin distribution to children aged 1-59 months reduced child mortality, and that World Health Organization guidance conditionally addressed the intervention in high-mortality settings. Those statements describe the background and rationale recorded for AVENIR; they are not results from AVENIR.

Because the registry says the study stopped before outcome measures were collected, it cannot establish whether the planned intervention changed the prevalence of resistance determinants in the target setting. Nor does the absence of an outcome estimate show that the groups had similar outcomes. The registry record does not report why the study terminated or whether any relevant outcomes were measured outside the listed registry results.

3. Trial Design

The registry describes a phase 4, randomized, parallel trial with no masking. One arm was assigned biannual oral azithromycin distribution to children aged 1-11 months through community health workers; the comparator was no additional intervention. The study title identifies Niger, and the stated purpose was to evaluate large-scale implementation in a programmatic setting while monitoring mortality and antimicrobial resistance.

The registry lists enrollment as 169707. This is the recorded enrollment figure, not evidence that the primary outcome was collected or that an analysis included that number. The record lists a start date of 2022-12-01 and a primary completion date of 2024-01-15, while the overall status is terminated. A listed primary completion date does not, by itself, establish that outcome collection or analysis was completed.

The primary outcome was prevalence of genetic determinants of macrolide resistance from population-based samples. The description specifies nasopharyngeal swabs from children aged 1-59 months and says measurement was intended after 1 year of distribution; the listed timeframe is 2 years. The registry does not explain the relationship between these timing descriptions. It also lists clinic-based resistance outcomes and all-cause clinic visits as secondary outcomes.

Although the record identifies randomization and a parallel model, details such as the randomization procedure, allocation concealment, clustering, and any stratification are not reported in the registry record. Masking is listed as none; the record does not specify which study personnel or outcome assessors, if any, were unmasked. The registry reports zero analyses and says the study stopped before outcome measures were collected. It does not report the reason for termination, a statistical analysis plan, or a sample-size justification.

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: The study stopped prior to the collection of outcome measures.

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: Inclusion criteria includes: age 1-59 months, primary residence in a study community, verbal consent of caregiver/guardian of children for study participation, weight ≥ 3.0 kg. Children were enrolled for participation. not guardians. The No Intervention arm data was not collected given the pragmatic design of the trial; therefore, the Programmatic Azithro 1-11 data was used to estimate the No Intervention population (33,953).

6. Primary Endpoint

The registry lists the primary outcome as Prevalence of Genetic Determinants of Macrolide Resistance From Population-based Samples, measured over 2 years.

The registered primary endpoint is the prevalence of genetic determinants of macrolide resistance in population-based samples. The description identifies determinants associated with Streptococcus pneumoniae, Streptococcus pyogenes, and Staphylococcus aureus, measured in nasopharyngeal swabs from children aged 1-59 months. The stated measurement was after 1 year of distribution, while the registry separately lists a timeframe of 2 years; it does not clarify how those entries relate.

In general, prevalence describes how common a specified characteristic is in a defined population or sample at a particular measurement time. Resistance determinants are genetic markers; measuring their prevalence can help describe the presence of resistance-associated material in the sampled population. This is distinct from measuring whether a child has a clinical infection or whether an antibiotic improves a clinical outcome. The registry does not report the precise laboratory procedures, threshold for detecting a determinant, or how multiple determinants were combined into the endpoint.

The registry states that the study stopped before outcome measures were collected. Therefore, it reports no primary endpoint values, group contrast, uncertainty estimate, or inferential test. The endpoint’s intended role can be described, but its observed value and any difference between assigned groups cannot be assessed from the registry results.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry records a randomized, parallel design and no masking. These are design descriptions, not an analysis plan. It reports zero analyses and no collected primary outcomes. The statistical methods actually used to estimate treatment effects are not reported in the registry record.

What does randomization contribute?

In general, random allocation is intended to make assigned groups comparable, on average, with respect to factors that could influence outcomes. This supports interpretation of a between-group contrast when outcomes are collected and analyzed appropriately. Randomization alone does not ensure balance in every feature, guarantee adherence, or resolve missing outcome data. The registry does not describe the allocation procedure, concealment, any grouping or stratification, or how allocation was implemented at the program level. It is therefore not possible to assess those design details from the record.

What would an outcome comparison show?

In general, an analysis of a binary prevalence endpoint might compare the proportion of samples with detected resistance determinants between assigned groups. The appropriate model and effect measure depend on the outcome definition, sampling structure, and design. For example, clustered assignment or correlated observations can affect the uncertainty calculation. The registry does not report whether assignment or sampling was clustered, how the endpoint was operationalized, or what model or contrast was planned. No prevalence estimates or group comparison are reported, so no numerical effect can be interpreted.

A confidence interval, in general, expresses the precision of an estimate under the assumptions of the analysis. A narrower interval indicates greater precision than a wider interval, but an interval does not correct for bias, incomplete outcome ascertainment, or a poorly specified analysis. No confidence interval is reported for the primary endpoint. Likewise, no p-value or other inferential result is reported. These absences mean that neither statistical significance nor the range of plausible effects can be assessed from the registry results.

What do the planning assumptions mean?

The registry includes a description of sample-size planning assumptions: a two-sided significance level of 0.05, equal allocation between arms, a normal approximation, and an effect defined as a stated fraction of the observed effect. It says that the illustrations are based on an observed effect and are not the trial’s own design. For time-to-event illustrations, the description refers to total events; for other illustrations, it refers to participants per arm. These are planning conventions, not observed trial results or observed power.

Those planning assumptions do not establish that AVENIR was designed using them. The registry does not report the trial’s sample-size justification, the effect used for any planning calculation, or a trial-specific statistical plan. In addition, the primary outcome is reported as uncollected, and no outcome analyses are listed. The record does not clarify how any observed-effect planning illustrations relate to this terminated trial. They cannot be used to infer the trial’s ability to detect an effect or to characterize an outcome that was not measured.

How does the implementation setting matter?

In general, programmatic evaluations may involve delivery processes, population sampling, or correlated observations that need to be reflected in analysis and uncertainty estimates. This is a general statistical consideration, not a statement about methods used in AVENIR. The registry says community health workers were to distribute the intervention, but it does not report whether delivery was organized by community, whether the assignment unit differed from the outcome unit, or whether the analysis accounted for clustering. It also does not report how population-based samples were drawn or how repeated or related samples, if any, were handled.

The intervention was not masked according to the record. In general, lack of masking can matter for outcomes affected by knowledge of assignment, especially when measurement or reporting involves judgment. The registry does not describe outcome-assessor masking, laboratory blinding, or the laboratory methods, so the possible relevance of masking to the resistance endpoint cannot be evaluated. No model checking, covariate adjustment, handling of protocol deviations, or sensitivity analyses are reported. These methods are not inferable from the randomized design or from the listed enrollment figure.

Overall, the statistical information describes design labels and generic planning assumptions, but not a completed analysis. The registry record does not report an estimate, an uncertainty interval, an inferential test, or enough analysis detail to reconstruct a result. With outcome collection explicitly reported as not having occurred, no statistical method can recover the missing primary endpoint comparison from the registry entries alone.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists five outcome measures and identifies a primary outcome concerning resistance determinants in population-based samples. It also names secondary outcomes concerning the prevalence and load of genetic determinants of macrolide resistance from clinic-based samples, and the number of all-cause clinic visits. The registry does not report how the remaining listed outcome measures are defined in the information available for this record, or how the outcomes were ordered for inferential testing.

In general, testing multiple outcomes or multiple comparisons can increase the chance of finding an apparently notable result by chance. Statistical plans may address this through a prespecified hierarchy or an adjustment procedure. No multiplicity strategy is reported in the registry record. Since no outcome analyses are listed and the study stopped before outcome measures were collected, there are no reported tests for which to assess adjustment or selective reporting.

The record does not provide an analysis population, a participant flow, or details of missing-data handling. It does state that the study stopped prior to collection of outcome measures. This indicates that the primary outcome data were not collected as registered; it does not describe the completeness of all other data or establish whether any follow-up information exists outside the registry. Methods such as imputation, complete-case analysis, or sensitivity analysis are not reported and should not be presumed. The reason for termination and the consequences for follow-up are not reported in the registry record.

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’s adverse-event entries list 0 serious events, 0 other events, and 0 deaths in the Programmatic Azithro 1-11 group, with 135754 at risk. For No Intervention, it lists 0 serious events, 0 other events, and 0 deaths, with 33953 at risk. No top serious-event terms are listed.

These are the figures entered in the registry, not a demonstrated absence of harm. The record does not report the safety observation period, event-collection procedures, completeness of reporting, or how the at-risk counts relate to exposure and follow-up. Without those details, the zero entries cannot establish that no events occurred during the study or allow comparison of safety risk between groups. The reported at-risk counts should be read as registry entries, not as proof that each person had identical observation or exposure. The study’s termination and lack of primary outcome collection are separately recorded; the registry does not explain how termination affected safety monitoring.

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

The record supports a description of the study’s planned comparison: randomized assignment in a parallel design, with biannual oral azithromycin distribution to children aged 1-11 months versus no additional intervention. Its primary endpoint was intended to assess the prevalence of macrolide-resistance genetic determinants in population-based nasopharyngeal samples. The registry also records a programmatic implementation rationale and secondary outcomes related to resistance and clinic visits.

What the record does not support is an estimate of the intervention’s effect on resistance prevalence, mortality, clinic visits, or any other outcome. The registry states that outcome measures were not collected and lists zero analyses. It reports no group-specific outcome values, uncertainty intervals, or tests. An unreported result is not evidence that the intervention had no effect, that the groups were equivalent, or that a clinically important effect was absent.

The rationale cites prior findings from MORDOR and refers to conditional World Health Organization guidance. These statements explain why the programmatic study was proposed; they are not AVENIR findings. Likewise, the registry’s adverse-event counts are not enough to establish safety in the absence of information about observation and event ascertainment.

The appropriate interpretation is therefore about the evidence available in the registry, rather than a treatment conclusion: AVENIR is recorded as terminated before outcome collection, and the posted results do not provide an analyzed comparison. The reason for termination and the existence of any data beyond the registry record are not reported. No conclusion about the direction, magnitude, or precision of an outcome effect can be drawn from the registered results.

13. Why This Trial Matters Statistically

This trial is statistically notable because the registry records a randomized design but no outcome collection or analysis. Randomization can support a credible comparison when assigned groups have measured outcomes and the analysis reflects the design. When the primary outcome is not collected, that inferential advantage cannot produce an estimate. The distinction between a planned randomized comparison and a completed analysis is essential when reading a registry entry.

The endpoint also makes the distinction between a programmatic question and a clinical endpoint important. AVENIR’s stated purpose included implementation, mortality, and antimicrobial resistance, while its registered primary outcome was the prevalence of genetic resistance determinants in population-based samples. These are related areas of interest, but they are not interchangeable measures. A resistance-marker prevalence result would not, by itself, be a mortality result, and the registry reports neither as an analyzed AVENIR outcome.

The record illustrates why statistical interpretation depends on more than the study label or enrollment figure. Information about outcome ascertainment, allocation implementation, analysis populations, missingness, and the statistical plan would be needed to assess a completed comparison. These details are not reported, and the registry explicitly says the study stopped before outcome measures were collected.

Finally, the record’s planning assumptions should not be mistaken for findings or trial-specific design choices. Sample-size illustrations are not observed power, and no outcome effect is available in the registered results for evaluating precision. The principal statistical lesson is that a registry can describe a substantial planned study while still containing no outcome evidence from which to estimate an 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.

The most consequential trial-specific limitation in the registry record is that the study stopped before outcome measures were collected. No primary endpoint estimate or outcome analysis is reported. The record marks the study as terminated but does not give the reason, so the circumstances behind the missing outcomes cannot be assessed.

The endpoint timing is not fully explained: the primary outcome description says measurement was after 1 year of distribution, while the listed timeframe is 2 years. The record does not clarify whether these refer to different stages of the study or how outcome timing was intended to work. It also does not detail the sampling procedure, laboratory methods, endpoint threshold, or handling of multiple resistance determinants.

Although the registry identifies randomization, a parallel model, and no masking, it does not report allocation procedures, the unit of assignment, possible clustering, outcome-assessor masking, or the planned analysis. It gives no statistical analysis plan or trial-specific sample-size justification. The listed enrollment figure cannot establish how many participants had relevant exposure, follow-up, or outcome data.

The registry lists adverse-event counts but does not describe observation duration, ascertainment, or reporting completeness. It also does not describe an analysis population, participant flow, or missing-data approach. These omissions limit the interpretation of the safety entries and the design details; they do not change the central fact that the registered primary outcome was not collected.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Azithromycin for Child Survival in Niger: Programmatic Trial (AVENIR) — NCT05288023.
• Registry record last updated 2025-03-28; status last verified 2025-03.
• Results first posted 2025-03-28.
• 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 identifies NCT05288023 as Azithromycin for Child Survival in Niger: Programmatic Trial (AVENIR), a phase 4 randomized, parallel trial with no masking. The intervention was biannual oral azithromycin distribution to children aged 1-11 months by community health workers; the comparator was no additional intervention. Enrollment is listed as 169707. The sponsor is the University of California, San Francisco.

The listed start date is 2022-12-01, the primary completion date is 2024-01-15, and the status is terminated. The primary endpoint concerned prevalence of genetic determinants of macrolide resistance in population-based samples. The registry says the study stopped before outcome measures were collected and lists zero analyses. It does not report a primary outcome result or the reason for termination.