Antenatal Micronutrient Supplementation and Infant Survival: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Antenatal Micronutrient Supplementation and Infant Survival (NCT00860470), a randomized phase 3 study. The primary outcome is Infant Mortality Through 6 mo of Age.

2. Clinical Question

In practical terms, the trial asks whether infant mortality by the stated follow-up point differed between pregnancies assigned to the multiple micronutrient supplement and pregnancies assigned to iron and folic acid. Because the primary analysis population is live births, the endpoint does not represent every woman recruited into the trial. The registry record explicitly notes that some recruited women did not have pregnancies yielding live births. This distinction matters when interpreting what population the primary outcome describes.

The study was randomized and cluster-randomized. Thus, assignment and analysis involve a group structure, and the comparison should be interpreted in the context of that design rather than as a simple comparison of independent individual participants. The registry record does not report further detail about the clusters or the randomization procedure. It reports the sponsor's logistic regression analysis and the result for the primary comparison.

3. Trial Design

This was a completed phase 3, randomized, parallel-group trial. The registry record identifies the design as cluster-randomized and quadruple-masked. Its brief summary describes a community-based trial intended to examine newborn and infant survival and health, as well as other birth outcomes, in a rural setting in northwestern Bangladesh. Enrollment was 44567. The record gives a start of 2008-01 and primary completion of 2012-09.

The two arms compared different dietary supplements. One received iron (27 mg) and folic acid (600 ug); the other received multiple micronutrients. The brief summary describes daily antenatal and postnatal supplementation. The registry record does not report further operational details about how the supplements were distributed, how adherence was assessed, or how the cluster allocation was carried out.

Masking is recorded as quadruple, but the registry record does not identify which roles or groups were masked, or how masking was maintained. It also does not describe the number or definition of clusters, the method used to account for cluster assignment in the analysis, or whether any adjustment variables were included. Those details cannot be inferred from the design labels alone.

The primary outcome was infant mortality through 6 mo of age, with a timeframe of 6-months post-birth. The stated population for this outcome comprises all live births by arm, rather than all women recruited into the trial. Secondary outcomes include neonatal and post-neonatal mortality, still birth rates, preterm birth categories, and low birth weight. The registry record lists 10 outcome measures and 10 analyses, but does not report how these outcomes were prioritized relative to one another beyond identifying the primary outcome.

The registry record says the study is completed. It does not report a protocol, detailed statistical analysis plan, sample-size justification, interim analyses, or further implementation details. The design supports a randomized comparison, but the interpretation of the reported estimate depends on the analysis used for the cluster structure and the specific outcome population.

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

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Not all women recruited into the trial had pregnancies yielding live births. This represents all live births by arm.

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 Infant Mortality Through 6 mo of Age, measured over 6-months post-birth.

The primary endpoint is infant mortality through 6 mo of age. The registry record describes it as infant mortality to age 6 months (180 days from birth), with a timeframe of 6-months post-birth. The outcome is recorded as a count of participants with an event, and the unit is participants. In this setting, the event is death of an infant within the stated period after birth.

This endpoint addresses survival during early infancy, a central outcome in the trial's stated aim to examine newborn and infant survival and health. Measuring the outcome to a defined follow-up point makes the comparison specific: it concerns mortality through that point, not mortality across an unspecified duration or later childhood. The registry record does not report whether follow-up was complete for every live birth or how deaths were ascertained.

The outcome population requires particular care. The registry record states that not all women recruited into the trial had pregnancies yielding live births, and that the primary outcome represents all live births by arm. Thus the endpoint concerns infants born alive, not every person recruited or every pregnancy regardless of outcome. Other listed outcomes include still birth rates and neonatal mortality, but these are separate outcomes and should not be merged with the primary endpoint.

For a binary outcome, a risk ratio compares the probability of the event between groups. It gives a relative comparison, while the confidence interval communicates uncertainty around the estimate. Neither measure alone supplies the underlying number of deaths or the absolute event rates. Those counts and rates are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does the risk ratio describe?

A risk ratio compares the probability of an event in one group with the probability in another. For this trial, the event is infant mortality through 6 mo of age, and the comparison is multiple micronutrients relative to iron and folic acid. A ratio of 1 represents equal risks on this scale; values below 1 indicate a lower relative risk in the first group, and values above 1 indicate a higher relative risk. This is a general explanation of the measure. The registry record reports a risk ratio of 0.95.

A relative measure does not give the event probability in either arm by itself. It also does not directly state how many deaths occurred or the absolute difference in risk. The registry record does not report the event counts or group-specific risks for this endpoint, so those quantities cannot be recovered from the stated estimate alone.

How should the confidence interval be read?

A confidence interval expresses uncertainty around an estimated effect under the statistical model used. In general, a wider interval reflects less precision, while a narrower interval reflects more precision. The reported 95% confidence interval is 0.86 to 1.06. Because it includes 1, the interval includes the no-relative-difference value for a risk ratio. It also includes values below and above 1, which leaves uncertainty about the direction and size of the relative comparison. A confidence interval is not a guarantee that the true effect lies within its limits, and it does not convey the practical importance of the effect on its own.

What does the p-value add?

The reported p-value is 0.36. In general, a p-value measures how incompatible the observed data are with a specified null model, assuming that model and the analysis assumptions hold. It is not the probability that the null hypothesis is true, nor is it a measure of the size or clinical importance of an effect. Here, the value is not small by commonly used significance conventions. That does not demonstrate that the risks are identical or that no difference exists. The estimate and confidence interval remain important for understanding the range of relative effects compatible with the analysis.

What does logistic regression mean in this record?

The registry record labels the method as logistic regression and the reported parameter as a risk ratio. It does not explain how the reported risk ratio was obtained from the regression, what covariates were included, or what link or variance adjustments were used. In general, regression can account for specified predictors and can support estimates adjusted for those variables, but details determine what the model estimates and how uncertainty is calculated. No particular adjustment or derivation can be assumed from the method label alone.

Why does clustering matter?

In a cluster-randomized trial, groups rather than isolated individuals are allocated. Participants in the same cluster may have more similar outcomes than participants in different clusters. As a general statistical principle, treating all participants as independent when outcomes are correlated within clusters can misstate the uncertainty of a treatment comparison. Cluster-aware methods account for this dependence in the analysis. The registry record notes that an individual-level re-analysis would be misleading and that only the sponsor's own analysis is shown. It does not report the cluster definitions, the correlation estimate, or the exact method used to account for clustering.

How should planning illustrations be distinguished from trial results?

Sample-size calculations are planning tools, not observed outcomes and not evidence that a trial had a particular level of power after completion. The planning assumptions described for the calculations use a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation; the effect to detect is expressed as a fraction of the observed effect. These are planning illustrations based on the observed effect, not the trial's own design. They should not be interpreted as observed power or as evidence that the trial was designed around those assumptions. The registry record does not report its original sample-size justification.

The planning description distinguishes participant counts per arm from total event counts for time-to-event calculations. This distinction is useful in general because different outcome types and models use different sample-size quantities. The primary endpoint in this trial is a binary participant outcome, while other outcomes include mortality and birth outcomes. The record does not provide additional details about how its original design calculations handled clustering or the different outcomes.

What analysis details are unavailable?

The registry record does not report the full statistical analysis plan, the covariates used, the handling of missing outcome data, the exact approach to cluster adjustment, or whether any interim analyses were conducted. It also does not report whether the result was adjusted for multiple comparisons. These details can affect how a statistical estimate should be interpreted. The available result can be described accurately as the sponsor's logistic regression analysis, but the method label does not substitute for the missing specification.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("Iron and Folate", "Multiple Micronutrient")
n      <- c(14142, 14374)
events <- c(764, 741)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  Iron and Folate  14142  764  5.4  5.0  5.8  5.0  5.8  5.0  5.8
ARM  Multiple Micronutrient  14374  741  5.2  4.8  5.5  4.8  5.5  4.8  5.5

11. Analyses Reported by the Sponsor

The primary result in the registry record is the sponsor's reported analysis: logistic regression, with a risk ratio of 0.95, a 95% confidence interval from 0.86 to 1.06, and p = 0.36. The registry record states that the trial was cluster-randomized and that a simple individual-level re-analysis would be misleading. For that reason, only the sponsor's own analysis is shown.

No separate unadjusted re-calculation is reported for this primary endpoint. That means there is no independent calculation from posted summary statistics to compare against the sponsor's estimate. The sponsor's result should be described on its own terms, without presenting it as corroborated by a second analysis.

The record identifies the regression method and reported parameter, but does not provide the model specification, cluster adjustment details, covariates, or the derivation of the risk ratio. These omissions limit independent assessment of the analysis. They do not justify replacing the sponsor's cluster-trial analysis with an individual-level calculation that the record itself warns would be misleading.

The appropriate reading is therefore narrow: the posted result is the sponsor analysis for infant mortality through 6 mo of age among live births, comparing the multiple micronutrient arm with the iron and folic acid arm. Its estimate and uncertainty can be interpreted as reported, while the unreported analysis details remain unresolved.

12. Multiplicity, Missing Data and Analysis Populations

The registry record lists 10 outcome measures and 10 analyses. It identifies infant mortality through 6 mo of age as the primary outcome and lists secondary outcomes that include neonatal mortality, post-neonatal mortality, still birth rates, preterm birth categories, and low birth weight. The record does not explain how the other outcomes were prioritized, whether all were prespecified, or whether the listed analyses correspond one-to-one with the outcome measures.

When a study examines multiple outcomes, the chance of obtaining at least one apparently notable result can increase. In general, multiplicity procedures can account for multiple comparisons, but they may also change the interpretation of individual p-values and intervals. The registry record does not report whether a multiplicity adjustment was planned or used. The reported p-value for the primary outcome should not be assumed to have been adjusted, nor should it be treated as part of a specified family of tests without documentation.

The primary endpoint population is described as all live births by arm, because not all women recruited had pregnancies yielding live births. This clarifies the intended outcome population, but the record does not report the flow from recruited women to live births, the number of participants with outcome data, or how births or deaths with missing information were handled. Details of missing data, loss to follow-up, and any imputation or sensitivity analyses are not reported in the registry record.

The registry record also does not describe the analysis population for each secondary outcome, or whether those analyses followed intention-to-treat principles. The cluster-randomized design is noted, but the exact analysis handling of clustering is not reported. These gaps make it difficult to assess the full multiplicity strategy, missing-data assumptions, and analysis populations across all outcomes. They do not alter the stated definition of the primary outcome, but they limit how broadly the primary result can be interpreted.

13. Safety Results

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

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 adverse-event figures for both arms. For iron (27 mg) and folic acid (600 ug), the serious adverse event field records 0 affected among 22114 at risk, and the other adverse event field also records 0 affected among 22114 at risk. For multiple micronutrients, the serious adverse event field records 0 affected among 22346 at risk, and the other adverse event field also records 0 affected among 22346 at risk. No serious adverse event terms are listed.

These entries describe what is recorded in the registry; they should not be read as proof that no adverse events of any kind occurred. The record does not explain the event definitions, ascertainment procedures, reporting period, or completeness of reporting. Death counts in the adverse-event entry are listed as null for both arms, so those fields do not provide a death comparison.

The at-risk counts shown for adverse events should not be assumed to be the same as the primary endpoint population or used as denominators for infant mortality. The primary outcome concerns live births, while the adverse-event listing is a separate registry entry. The reported zero affected counts are a description of the adverse-event fields, not a stand-alone assessment of comparative safety.

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

The posted result describes infant mortality through 6 mo of age among live births in a comparison of multiple micronutrients with iron and folic acid. The reported risk ratio is 0.95, with a 95% confidence interval from 0.86 to 1.06 and p = 0.36. The estimate's direction is below 1, but its interval includes 1 and includes values on either side. The result therefore does not provide a precise conclusion about the direction of the relative difference.

The result does not establish that the two arms had exactly the same risk. A p-value of 0.36 is not proof of no effect, and a confidence interval that includes 1 does not demonstrate equivalence. Equivalence or non-inferiority conclusions require a suitable design, prespecified margins, and analysis details; those are not reported in the registry record. Conversely, the point estimate alone should not be treated as evidence of a definite reduction.

The relative estimate also does not tell readers the absolute mortality risk in either arm or the number of deaths. The registry record does not report those quantities. As a result, the practical size of any absolute difference cannot be described from the reported risk ratio and interval alone.

The interpretation is specific to the trial's stated outcome population and comparison: live births in the two supplement arms, followed for the stated period. It does not automatically apply to pregnancies that did not yield live births, other populations, other supplementation regimens, or outcomes beyond those recorded. The cluster-randomized design also means the result must be understood as the sponsor's analysis rather than as an individual-level calculation that ignores clustering.

Finally, the result is one component of a broader trial that listed multiple outcomes. The registry record does not report the multiplicity strategy, missing-data procedures, or full analysis plan. Those omissions do not erase the reported estimate, but they limit the detail available for judging its robustness and its relationship to the other outcome analyses.

15. Why This Trial Matters Statistically

This trial raises a clear statistical question about a binary early-life outcome in a randomized comparison: whether infant mortality through 6 mo of age differed between two supplement strategies. The registry record makes the outcome timeframe explicit and identifies the population as live births by arm. Those definitions are important because they establish what event and population the reported risk ratio concerns.

The cluster-randomized design highlights a central issue in clinical-trial analysis: the unit of allocation may not be the individual whose outcome is measured. When participants are grouped, their outcomes can be correlated within groups. General statistical methods must reflect that structure to estimate uncertainty appropriately. The record's warning against a simple individual-level re-analysis is therefore an important part of interpreting the posted result.

The reported result also illustrates why an estimate, confidence interval, and p-value should be read together. The risk ratio gives a relative comparison, the interval describes uncertainty, and the p-value addresses compatibility with a null model under the analysis. None alone provides the underlying event risks or a complete account of practical importance. Here, the interval crosses the ratio value of 1, while the record does not give the event counts needed to describe absolute risks.

Finally, the registry record demonstrates the value and limits of registry reporting. It identifies the design, arms, primary outcome, and sponsor's estimate, while leaving important analytic details unreported. A careful statistical reading can distinguish what the result directly says from what would require further documentation, without overstating either the evidence or its limitations.

16. 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 trial-specific details needed for a fuller assessment are not reported in the registry record. Although the trial is identified as cluster-randomized and quadruple-masked, the record does not describe the clusters, the allocation procedure, the roles masked, or how masking was maintained. It does not give the model specification or explain how the reported risk ratio was derived from the logistic regression analysis. In particular, the exact method for accounting for clustering is not described.

The registry record does not report the number of infant deaths, the group-specific event risks, or absolute risk differences for the primary endpoint. It also does not report the number of live births contributing to the analysis or whether primary outcome follow-up was complete. The stated outcome population clarifies that the analysis concerns all live births by arm, but missing outcome data, loss to follow-up, and their handling are not described.

The record lists secondary outcomes and analyses but does not report a multiplicity strategy or a complete statistical analysis plan. It also does not report covariate adjustments, interim analyses, or the sample-size justification used for the trial's design. The planning assumptions described for illustrations are not a substitute for the original design rationale.

The adverse-event listing records zero affected in the serious and other event fields, but does not report event definitions, ascertainment procedures, or reporting completeness. Death counts in those fields are null. These limitations make it inappropriate to interpret the adverse-event entries as a complete safety assessment.

These gaps do not change the numerical result that the registry reports. They constrain independent evaluation of how the result was estimated, how robust it is to analytic choices, and how fully outcomes and adverse events were captured.

17. Related Tutorials

18. Related Statistical Calculators

19. Sources

• ClinicalTrials.gov: Antenatal Micronutrient Supplementation and Infant Survival — NCT00860470.
• Registry record last updated 2022-10-19; status last verified 2022-10.
• Results first posted 2015-05-27.
• 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.

20. Record Summary

The registry record describes a completed phase 3, randomized, cluster-randomized, parallel-group, quadruple-masked trial in rural northwestern Bangladesh. Enrollment was 44567; the record gives a start of 2008-01 and primary completion of 2012-09. The arms compared multiple micronutrients with iron (27 mg) and folic acid (600 ug).

The primary outcome was infant mortality through 6 mo of age among live births by arm. The sponsor's reported logistic regression result was a risk ratio of 0.95, with a 95% confidence interval from 0.86 to 1.06 and p = 0.36. The record cautions that a simple individual-level re-analysis would be misleading because of the cluster-randomized design.

Important details are not reported in the registry record, including the full statistical analysis plan, cluster-adjustment method, event counts, missing-data procedures, and multiplicity strategy. The result should be read as the sponsor's analysis for the stated outcome and population, with those reporting limitations in mind.