Optimizing Individual Nutrition in Preterm Very Low Birth Weight Infants: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Optimizing Individual Nutrition in Preterm Very Low Birth Weight Infants (NCT02372136), a randomized na study. The primary outcome is Growth Velocity.

2. Clinical Question

The trial addresses whether the combined individualized and optimized nutrition strategy is associated with different growth outcomes than optimized nutrition alone in the described infant population. The registry summary also outlines questions about neurodevelopment, disproportionate growth, and metabolic-syndrome-related findings during early life. Those aims provide context for the broader study, but they do not change the interpretation of the posted primary growth result.

3. Trial Design

The registry identifies this as a randomized, parallel-group trial with quadruple masking. Its recorded enrollment is 120, and its status is completed. The phase is listed as not applicable. The record identifies the University of Texas Southwestern Medical Center as sponsor and gives the study start as 2016-01 and primary completion as 2023-12-31.

The registered groups were Individualized and Optimized Nutrition and Optimized Nutrition. The first group received both nutrition strategies; the comparison group received optimized nutrition. This design focuses the contrast on adding individualized nutrition to an optimized approach. It does not compare the registered strategies with a standard-fortification arm in the posted results.

The primary outcome is growth velocity. The registry specifies a measurement time of 36 (range 35-37) weeks postmenstrual age or discharge, whichever came first, and describes rate of weight gain and length velocity. The primary numeric result is expressed as a mean in g kg-1 day-1, with standard deviations. The primary population field is blank, and the record does not report further eligibility details for the analyzed primary outcome.

The record lists secondary outcomes spanning growth, blood pressure, hypertension or high systolic blood pressure, neurodevelopment, leptin, renal function, and comparison of weight with an expected value. Only a subset of those outcomes has numerical results in the posted information. The original sample-size justification, detailed allocation procedure, and full analysis plan are not reported in the registry record.

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: Neonates were excluded if they received comfort care only, had a major congenital abnormality, or were too unstable for the first 7 days to have an accurate length measurement using a length board or a caliper.

Overall Study

Reasons for not completing this period:

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 Growth Velocity, measured over 36 (range 35-37) weeks postmenstrual age or discharge (whichever comes first).

Growth velocity measures the pace of growth over time. The registry describes the primary outcome as rate of weight gain, expressed in g kg-1 day-1, and also refers to length velocity, expressed in cm week-1. The reported primary result is a mean growth-velocity value with a standard deviation for each group. A rate-based measure is useful in this setting because it expresses weight gain relative to body weight and time, rather than presenting only a single weight measurement.

The assessment was scheduled for 36 (range 35-37) weeks postmenstrual age or discharge, whichever came first. This makes the endpoint a measure of growth by a specified clinical milestone or discharge, rather than a fixed follow-up date for every participant. The result therefore describes the groups at the registry’s stated assessment point; it does not, on its own, describe growth after discharge or later development.

The posted primary means were 13.0 for Individualized and Optimized Nutrition and 13.1 for Optimized Nutrition, in g kg-1 day-1. The registry record does not report a separate numeric result for length velocity, even though the outcome description includes it. Interpretation of the posted primary comparison should therefore remain focused on the available growth-velocity result.

7. Results Posted in the Registry

• Optimized Nutrition: mean 13.10 versus 13.00 on control; difference 0.10 (95% CI -0.76 to 0.96), standardized effect d = 0.04; z-test p = 0.819.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Disproportionate Growth (Increased Fat Mass): BMI >90th Centile

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Optimized Nutrition: 1.8% (1/55) versus 2.0% (1/50) on control; risk difference -0.2 percentage points (95% CI -5.4 to 5.1), risk ratio 0.91 (0.06–14.15); unadjusted two-proportion z-test p = 0.946.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Optimized Nutrition and the control. The interval gives the range of effects that remain compatible with the data.

Blood Pressure

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Optimized Nutrition: mean 69.70 versus 69.40 on control; difference 0.30 (95% CI -2.46 to 3.06), standardized effect d = 0.04; z-test p = 0.832.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Optimized Nutrition and the control. The interval gives the range of effects that remain compatible with the data.

Hypertension or High Systolic Blood Pressure

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Optimized Nutrition: 7.9% (3/38) versus 17.5% (7/40) on control; risk difference -9.6 percentage points (95% CI -24.2 to 5.0), risk ratio 0.45 (0.13–1.62); unadjusted two-proportion z-test p = 0.205.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Optimized Nutrition and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes span growth, blood pressure, hypertension or high systolic blood pressure, and other domains listed in the registry. The posted numerical results include outcomes at the primary assessment time and a later blood-pressure outcome. Across the available comparisons, estimates are uncertain, and the unadjusted results do not provide a consistent basis for claiming a clear difference across outcomes.

These results relate to the primary finding by adding context about other measured outcomes, not by changing what the primary growth-velocity comparison estimates. A secondary endpoint can differ in measurement scale, timing, denominator, and interpretation. Findings from one endpoint should not be treated as confirmation of another, especially when the registry record does not report a prespecified hierarchy or a full analysis plan.

Unadjusted secondary results carry less weight when several outcomes are examined. With multiple comparisons, the chance of observing a seemingly notable result by chance can increase. A multiplicity adjustment or a prespecified approach to interpreting secondary outcomes can address that issue, but neither is reported in the registry record. The posted p-values should therefore be regarded as unadjusted and exploratory rather than as stand-alone confirmatory evidence.

The record lists 14 secondary outcomes, while numerical results are posted for only a subset. It does not report numerical results for all listed neurodevelopmental, leptin, renal-function, and weight-comparison outcomes. Nor does it report why the available denominators differ across outcomes. The pattern should be interpreted as partial reporting of a wider set of registered measures, not as a complete account of every planned analysis.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Disproportionate Growth (Increased Fat Mass): BMI >90th Centile
• Blood Pressure
• Hypertension or High Systolic Blood Pressure
• Neurodevelopment
• Neurodevelopment
• Leptin
• Renal Function
• Comparison of Weight With Expected Value for Age and Gender

9. Statistical Interpretation

10. Statistical Methods Explained

The posted primary outcome is continuous: each group has a mean and standard deviation for growth velocity. A mean difference expresses the separation between group averages in the original measurement unit. For this comparison, the registry’s comparison values correspond to the Optimized Nutrition mean relative to the Individualized and Optimized Nutrition mean. Positive values indicate a higher group mean in the former group; negative values indicate a lower group mean.

What does a confidence interval describe?

A confidence interval gives a range of values compatible with the observed data under the assumptions of the method used to construct it. It is not a range containing a fixed proportion of individual infants’ measurements, nor does it assign a probability to the true effect after the data have been observed. The primary interval, approximately -0.759 to 0.959, includes zero and also includes differences in either direction. Its breadth is important: the estimate does not pin down the group difference precisely.

How should the p-value be read?

The unadjusted p-value for the primary comparison is 0.819. A p-value describes how unusual data at least as discrepant as those observed would be under a specified null model and its assumptions. It is not the probability that the null hypothesis is true, and it does not measure the size or clinical importance of an effect. The registry record does not report the sponsor’s statistical model or a full analysis plan. The posted p-value should therefore be read as an unadjusted comparison, not as evidence that all relevant sources of uncertainty have been addressed.

Why does variability matter?

The standard deviation describes spread among observations within a group. Larger variability makes it harder to estimate a group mean difference precisely when other features are similar. The primary group standard deviations are 2.6 and 2.1, and the pooled standard deviation used in the unadjusted summary is 2.368. The standardized difference, 0.042, expresses the mean difference relative to pooled variability. It is a scale-free description of the observed contrast, not an outcome measured directly in the infants.

For a continuous outcome, a simple comparison of group means is often summarized with a standard error, a confidence interval, and a test of a null difference. These calculations rely on assumptions about the sampling distribution and the independence of observations. The registry does not report whether the primary analysis used this simple approach, adjusted for covariates, or accounted for any design features beyond randomization. The displayed unadjusted statistics should not be mistaken for a fully specified trial analysis.

What changes for binary outcomes?

A binary outcome records whether an event or condition was present. A risk difference subtracts the proportion in one group from the proportion in the other; its interpretation depends on which group is subtracted from which. A confidence interval for a risk difference can extend on both sides of zero, indicating that the observed data are compatible with effects in either direction. The secondary risk-difference estimates in the posted results are unadjusted. The registry does not report adjusted secondary models or a multiplicity plan.

Counts and denominators are also needed to understand a binary result. The denominators differ across posted secondary outcomes, and the registry record does not explain why. The denominators should therefore be read outcome by outcome. They should not be combined to infer a shared analysis population or the reason that some observations are absent.

For a continuous outcome, standard methods generally rely on an adequate approximation to the sampling distribution of the mean. For binary outcomes, methods must reflect the discrete nature of the data and can be sensitive to sparse event counts. These are general statistical considerations; the registry record does not identify the specific procedures used for every posted secondary comparison.

What is over-dispersion, and is it relevant?

Over-dispersion occurs when observed variability exceeds what a statistical model assumes. If it is ignored, standard errors can be too small and confidence intervals too narrow. This is a general concern in some count and clustered-data analyses. The posted primary endpoint is summarized as a continuous mean with standard deviation, and the registry does not report a count model or an over-dispersion assessment. It is therefore not possible to determine from the record whether over-dispersion was considered in any analyses.

The record lists no reported analyses for the posted secondary results. It also does not report the methods used to handle repeated measures, missing outcomes, or multiple comparisons. These are not details that can be recovered from the group summaries alone. In particular, an unadjusted estimate and p-value cannot reveal whether a more complex sponsor analysis was planned or conducted.

11. Confidence Intervals and Precision

Precision concerns how tightly the data estimate a group difference. For the primary growth-velocity comparison, the estimated mean difference is 0.1 g kg-1 day-1, with a 95% confidence interval of approximately -0.759 to 0.959. Because this range includes zero and extends to differences in either direction, the data allow a broad set of possible group contrasts. The point estimate alone should not be treated as the exact underlying difference.

The interval is reported for the unadjusted mean difference. It reflects sampling uncertainty under the method’s assumptions, but does not account for uncertainties that cannot be assessed from the registry record, such as the effects of missing data or deviations from the planned analysis. The record does not describe the analysis population, missing-data procedures, or any adjustment for baseline characteristics.

The standard deviations, 2.6 and 2.1, show that individual observations vary around their respective group means. A confidence interval for a mean difference is not an interval for individual growth measurements. It also does not indicate what range of differences would be clinically important; such a threshold is not reported in the registry record.

Secondary intervals are also broad for some outcomes and include values on either side of no difference. Because multiple outcomes are listed and no multiplicity strategy is reported, those intervals and their associated unadjusted p-values warrant particular caution. The available precision is about the estimates in the posted data, not a guarantee about results in another population or study.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The planning figures are illustrations based on the observed effect, not the trial’s original design assumptions. For the primary continuous outcome, the figures represent participants per arm. In general, a time-to-event calculation would be expressed as a total number of events, but these posted primary and secondary planning calculations are not time-to-event calculations.

The stated planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. Each planning scenario uses the stated fraction of the observed effect as the difference to detect. For example, planning for 100% of the observed primary effect gives 8,800 participants per arm for 80% power and 11,781 participants per arm for 90% power. These are planning illustrations, not observed trial results.

The 80% power column describes a hypothetical design’s chance of detecting the specified effect under the stated assumptions if that effect were the true one. The 90% column uses a higher target chance of detection, which generally requires a larger planned sample under otherwise similar assumptions. Smaller effects require more participants to distinguish them from random variation with the same target power; that is why planning for a smaller fraction of the observed effect yields larger figures.

These calculations use the observed effect as an input, so they do not show the probability that the completed trial detected an effect. This is not observed power. The trial’s own sample-size justification and original design assumptions are not reported in the registry record, so the planning illustrations should not be used to infer what the investigators planned before enrollment.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = 0.100; pooled standard deviation σ = 2.368.
2. n = 2 × (1.96 + 0.842)² × 2.368² / 0.100² = 8,799.6, rounded up to 8,800 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
delta <- 0.0999999999999996                      # observed difference in means
sigma <- 2.36761779751034                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(2372136)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 8,800 (80% power), 11,781 (90% power)
 75% of observed effect: n per arm = 15,644 (80% power), 20,943 (90% power)
 50% of observed effect: n per arm = 35,199 (80% power), 47,121 (90% power)
power.t.test: 8800.5 
Simulated power at n = 8,800 per arm: 0.800 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

14. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when a study assesses several outcomes or makes several statistical comparisons. Even if each comparison is considered separately, examining many results can increase the chance that at least one appears notable by chance. A prespecified hierarchy, adjustment procedure, or clearly labeled exploratory analysis can help readers understand how to interpret such findings. The registry record lists multiple secondary outcomes but does not report a multiplicity strategy or identify which secondary outcomes, if any, were intended to be confirmatory.

The record lists 18 outcome measures and 14 secondary outcomes, but numerical results are available for only a subset. It does not report the reasons that results are absent for the remaining listed outcomes. The record also lists no reported analyses for the posted secondary outcomes. Consequently, it is not possible to determine from the registry whether omitted outcomes were unmeasured, unavailable, analyzed but not posted, or treated in another way.

Missing-data handling is not reported in the registry record. There is no description of how participants with incomplete measurements were handled, whether any imputation was used, or whether sensitivity analyses were conducted. Denominators vary across the posted outcomes. That variation is visible in the record, but its causes cannot be inferred from the counts alone.

The analysis populations are not reported. The record does not specify whether the primary result used all randomized participants, participants with an available measurement, or another defined group. It also does not report how protocol deviations or treatment adherence were addressed. These omissions matter because the analysis population and missing-data method can affect both the estimate and its uncertainty.

The original statistical plan, interim-analysis arrangements, and sample-size justification are not reported in the registry record. Without those details, it is not possible to assess whether the posted outcomes and analyses match the original plan or whether any analysis decisions were made after seeing data. This uncertainty is separate from the numerical confidence interval: it concerns the transparency and interpretability of the analysis process.

15. 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.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: necrotizing enterocolitis (9.7%).

The adverse-event record presents serious events, other events, and deaths separately for groups labeled Experimental and Control. For serious events, the recorded at-risk groups are 62 and 58, with 6 and 5 affected, respectively. For other events, the recorded at-risk groups are 62 and 58, with 19 and 17 affected. The death counts are 3 and 4, respectively. These are registry counts and should be read as listed, not converted into rates without a defined analysis and denominator interpretation.

Necrotizing enterocolitis is listed as the top serious term at 9.7%. The registry entry does not clarify the denominator or provide a group-specific breakdown for that percentage. The adverse-event groups are labeled differently from the primary outcome arms, and the record does not explain how the safety groups map to the outcome populations. The at-risk counts also differ from some outcome denominators; the reasons are not reported.

These figures describe recorded safety events and deaths. They do not by themselves establish whether an event was caused by a nutrition strategy, nor do they provide an adjusted comparison of safety between groups. The registry record does not report detailed event definitions, exposure time, or a safety-analysis method.

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

The posted primary result describes similar observed group means for growth velocity: 13.0 g kg-1 day-1 for Individualized and Optimized Nutrition and 13.1 g kg-1 day-1 for Optimized Nutrition. The unadjusted mean difference is 0.1, and its confidence interval includes zero and allows differences in either direction. The result does not establish a clear difference in the primary measure between the registered groups.

That finding does not demonstrate equivalence or prove that the strategies have identical effects. The record reports no equivalence margin, and the interval still includes a range of possible differences. A non-significant p-value is not proof of no effect. The registry also does not report a threshold for what degree of change in growth velocity would be clinically meaningful.

The result applies to the posted measurement and comparison, not automatically to every outcome in the trial’s broader aims. The registry describes interests in neurodevelopment, disproportionate growth, and findings suggestive of metabolic syndrome during the first 3 years of life, but numerical results for all such outcomes are not reported. The primary result does not answer those questions.

The comparison is between individualized plus optimized nutrition and optimized nutrition alone. It does not estimate the effect of either strategy compared with standard fortification, and the registry summary does not support conclusions about populations beyond those described. Details about the analyzed population and the statistical model are not reported in the registry record.

Finally, posted group summaries cannot resolve questions about missing data, multiplicity, or the analysis population. Those limits affect how confidently the evidence can be generalized or interpreted, even when the numerical estimates are read accurately. The statistical result is a description of the observed comparison; it is not medical advice or a recommendation about treatment.

17. Why This Trial Matters Statistically

This trial is statistically interesting because its primary question concerns an incremental strategy: individualized nutrition was added to optimized nutrition, while the comparison group received optimized nutrition alone. That makes the group contrast narrower than a comparison with standard fortification. The interpretation depends on keeping that specific contrast in view.

The primary outcome is a rate of weight gain at a milestone-based assessment, and the posted group means are close. Yet the interval around their difference is broad enough to include values in both directions. This illustrates why an estimate and its precision should be read together: a small point estimate does not by itself establish that meaningful differences have been excluded.

The record also illustrates the distinction between a registered outcome set and a complete set of reported results. A range of secondary domains is listed, while numerical results are available for only some of them. When many outcomes are considered, the absence of a reported multiplicity plan makes isolated unadjusted findings harder to interpret as confirmatory evidence.

Finally, a completed randomized trial can still leave important statistical details unclear in the registry record. The analysis population, missing-data procedures, original sample-size rationale, and detailed analysis plan are not reported. These omissions do not erase the posted estimates, but they limit independent evaluation of how those estimates were generated and how fully they answer the broader research questions.

18. 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 primary outcome description includes both rate of weight gain and length velocity, but the posted primary numeric result is a single growth-velocity mean in g kg-1 day-1. A separate result for length velocity is not reported in the registry record. The primary population field is blank, so the precise population contributing to that summary is also unclear.

The registry’s summary describes extremely low gestational age neonates and small-for-gestational-age preterm infants, but it does not provide a detailed analyzed-population description alongside the primary result. The record does not report whether both groups of infants were represented in the posted comparison or whether any subgroup results were assessed.

Denominators differ across the primary, secondary, and safety results. The registry does not explain the differences, so they cannot be attributed to follow-up, missing measurements, exclusions, or another cause. This limits assessment of completeness and makes it important not to treat every outcome as if it involved the same analysis population.

The registry record does not report the full analysis plan, covariate adjustment, missing-data handling, original sample-size justification, or multiplicity approach. It lists many secondary outcome domains, but numerical results are not posted for all listed measures. These reporting gaps limit interpretation of the wider set of aims and the strength of conclusions from secondary comparisons.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Optimizing Individual Nutrition in Preterm Very Low Birth Weight Infants — NCT02372136.
• Registry record last updated 2024-07-03; status last verified 2024-06.
• Results first posted 2021-05-20.
• 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.

22. Record Summary

The registry record describes a completed randomized, parallel-group trial with quadruple masking and recorded enrollment of 120. It compared Individualized and Optimized Nutrition with Optimized Nutrition. The primary outcome was growth velocity at 36 (range 35-37) weeks postmenstrual age or discharge, whichever came first.

The posted primary means were 13.0 and 13.1 g kg-1 day-1, respectively. The unadjusted mean difference was 0.1, with a 95% confidence interval of approximately -0.759 to 0.959 and an unadjusted p-value of 0.819. The interval includes zero and allows differences in either direction.

Numerical results are available for only a subset of the listed secondary outcomes. The record does not report the full analysis plan, analysis populations, missing-data methods, multiplicity strategy, or original sample-size justification. These limitations constrain interpretation of the broader outcome set and the precision of conclusions beyond the posted primary comparison.