Metformin for Overweight & OBese ChILdren and Adolescents With BDS Treated With SGAs: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Metformin for Overweight & OBese ChILdren and Adolescents With BDS Treated With SGAs (NCT02515773), a randomized phase 4 study. The primary outcome is Change in BMI Z-score From Baseline to Month 6 in ITT Sample.

2. Clinical Question

The central question is whether the mean change in BMI z-score from baseline to Month 6 differed between participants assigned to metformin plus a healthy lifestyle intervention and participants assigned to the lifestyle intervention alone. The comparison concerns adding metformin to the shared lifestyle program; it is not a comparison of metformin alone with no intervention.

The registry describes a prospective, pragmatic, randomized trial involving children and adolescents with bipolar spectrum disorders treated with second-generation antipsychotics. The primary outcome was assessed in a modified intention-to-treat sample: participants who initiated randomized treatment and had at least one post-baseline assessment. The posted group summaries include 565.0 participants in the metformin-and-lifestyle group and 687.0 in the lifestyle-only group.

BMI z-score expresses BMI relative to age- and sex-adjusted reference values. The registry describes a score of 0 as the population mean and larger scores as heavier weights and worse outcomes. The reported outcome is change, so the group means describe movement in this measure across the stated timeframe, rather than the groups' endpoint BMI levels alone. The posted comparison estimates the difference between the group mean changes. It does not by itself describe individual responses, longer-term outcomes, or the other patient-centered outcomes mentioned in the trial summary.

3. Trial Design

The registry identifies the trial as randomized, open-label, and parallel-group, and lists it as Phase 4. The stated enrollment is 1565. Its status is completed; the recorded start is 2015-12 and primary completion is 2022-12. The sponsor is listed as Melissa Delbello. The trial summary characterizes the study as prospective and pragmatic.

Participants were assigned to one of two groups: metformin plus the healthy lifestyle intervention, or the healthy lifestyle intervention alone. The lifestyle program is described as counseling participants and families about healthy eating, physical activity, and sedentary activities. The metformin arm description says treatment began at 500 mg orally at night and was titrated gradually, but the remainder of that description is incomplete in the registry record. Further details of the treatment regimen are not reported in the registry record.

Open-label means that treatment assignment was not masked, according to the registry's masking field. The allocation method beyond the statement that the trial was randomized, the randomization sequence procedure, and allocation concealment are not reported in the registry record. The record also does not report the recruitment setting, eligibility criteria in detail, or how many sites participated. Those omissions limit what can be concluded about the practical conduct and generalizability of the study from the posted design information alone.

The primary outcome analysis was not based on every enrolled participant. The registry defines its modified intention-to-treat sample as participants who initiated randomized treatment and had at least one post-baseline assessment. The group counts for the primary outcome are 565.0 and 687.0. The registry's overall enrollment and those outcome-analysis counts describe different stages of the study; the detailed reasons some enrolled participants are not included in the primary sample 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: 1633 participants were screened. 68 did not meet criteria. Of the 1565 randomized patients, 228 patients did not have a visit after baseline and 38 patients, in the MET+LIFE group, did not start metformin per protocol by study day 137 post-randomization. These 266 patients were not included in the modified intent-to-treat (ITT) sample and results, giving an ITT sample size of 1299.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Modified intention-to-treat (ITT) sample consisting of patients who initiated randomized treatment and had at least one post-baseline assessment.

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: Modified intent-to-treat sample, which includes all randomized patients who had a post-randomization visit and started their randomized intervention.

6. Primary Endpoint

The registry lists the primary outcome as Change in BMI Z-score From Baseline to Month 6 in ITT Sample, measured over Baseline to Month 6.

The primary endpoint is change in BMI z-score from baseline to Month 6, measured in the modified intention-to-treat sample. BMI is calculated from weight and height, then converted to a z-score adjusted for age and sex using US Centers for Disease Control and Prevention growth charts. This standardization is intended to place a participant's BMI in context relative to age- and sex-specific reference values, rather than treating the same raw BMI as equivalent across children of different ages and sexes.

The registry explains that a z-score of 0 indicates the population mean and that larger z-scores correspond to heavier weights and worse outcomes. Because the endpoint is change, the group means express how the measure shifted over the interval, not the groups' absolute BMI z-scores at Month 6. The metformin-and-lifestyle group had a mean change of -0.08, while the lifestyle-only group had a mean change of 0.0.

A standardized growth measure can be useful in a pediatric trial because body size changes with age and differs by sex. The registry's description makes the outcome interpretable as a relative position on those reference curves. It remains a measure of body size, however, and does not directly measure psychiatric symptoms, quality of life, treatment satisfaction, or adherence. Although those outcomes appear in the trial summary, the posted primary endpoint is the BMI z-score change. Further details about the endpoint's measurement schedule beyond baseline to Month 6 are not reported in the registry record.

7. Results Posted in the Registry

• Healthy Lifestyle Intervention (LIFE): mean 0.00 versus -0.08 on control; difference 0.08 (95% CI 0.05 to 0.11), standardized effect d = 0.28; z-test p = <0.001.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is continuous, and the registry reports each group's mean change and standard deviation. A mean is a measure of the group's average value; a standard deviation describes how spread out individual observations are around that average. These summaries make it possible to describe the average change in each group and to calculate an unadjusted contrast between group means. The posted standard deviations are 0.29 in both groups.

What the mean difference represents

The recalculated mean difference is formed by comparing the lifestyle-only group's mean change with the metformin-and-lifestyle group's mean change. The estimate is 0.08. Its sign depends on which group is subtracted from which, so interpretation should follow the stated order of the comparison. In endpoint terms, the group mean changes are -0.08 and 0.0, respectively. Since larger BMI z-scores are described as worse outcomes, the mean changes point in the direction of a lower average change in the group receiving both interventions.

This recalculation is unadjusted: it uses the posted group-level summaries rather than participant-level data or a fitted covariate-adjusted model. It cannot account for any baseline variables or other adjustments that may be included in the registry's ANCOVA. The adjusted model's covariates and implementation details are not reported in the registry record. Therefore, the recalculated estimate is a transparent summary-statistic comparison, not a reproduction of the complete sponsor analysis.

Why the confidence interval matters

A confidence interval gives a range of values compatible with the data and the method used to estimate the effect. It reflects uncertainty in the estimated mean difference; it is not a range containing a specified proportion of individual participants' outcomes. For this unadjusted comparison, the recalculated interval is entirely above zero. That means the interval does not include a zero mean difference under the calculation used. The interval does not remove uncertainty, and it does not establish clinical importance.

The interval pertains to the unadjusted comparison, while the posted ANCOVA result is a model-based analysis. The registry gives the ANCOVA estimate and p-value but does not report its confidence limits. These are distinct analyses, so the unadjusted interval should not be presented as if it were the ANCOVA interval.

How to understand the p-value

The unadjusted p-value is <0.001, and the registry's ANCOVA p-value is also <0.001. In general, a p-value describes how incompatible the observed data are with a specified null model, under the assumptions of the analysis. It is not the probability that the null hypothesis is true, and it does not measure the size or importance of an effect. The registry does not report all details needed to reconstruct the ANCOVA's null-testing procedure, so interpretation of that result is limited to the reported method, estimate, and p-value.

ANCOVA and the posted analysis

ANCOVA is a method for comparing group means while accounting for selected covariates. In general, covariate adjustment can improve precision or address chance baseline imbalances when the chosen model is appropriate. The registry names ANCOVA and reports a net mean difference of 0.08, with p <0.001. It does not report the covariates, model diagnostics, missing-data handling, or other implementation details. The posted ANCOVA result should therefore be described as reported, without inferring what adjustments it used.

The primary population is a modified intention-to-treat sample, defined in the registry as participants who initiated randomized treatment and had at least one post-baseline assessment. This is narrower than an analysis of everyone enrolled or randomized. General statistical interpretation depends on knowing how exclusions and missing measurements were handled; the relevant detailed procedures are not reported in the registry record. The group-specific analysis counts are available, but they do not explain the reasons for exclusion or the pattern of missing observations.

Standardized effect size

The reported standardized difference is 0.276, using the pooled standard deviation of 0.29. A standardized effect expresses a mean contrast in standard-deviation units, rather than in the endpoint's original units. It can assist comparisons across measures, but it is not a substitute for the original-unit mean difference or its confidence interval. Interpretation still depends on the measure, population, and context. The registry does not provide a threshold for a clinically important standardized or raw difference in BMI z-score.

These methods answer related but not identical questions: group means describe observed averages, the unadjusted calculation describes a direct contrast based on those averages, and ANCOVA describes the registry's model-based comparison. Reading them together clarifies the size, direction, and uncertainty of the primary result while keeping the distinct analysis methods separate.

10. Confidence Intervals and Precision

The recalculated unadjusted mean difference is 0.08, and its 95% confidence interval lies entirely above zero. This interval communicates the uncertainty in the summary-statistic estimate, rather than the range of participant-level changes. The registry's ANCOVA analysis also reports a mean difference of 0.08, but its confidence limits are not reported in the registry record. The recalculated interval should not be attributed to that adjusted analysis.

Precision depends partly on the variability of the outcome and the amount of information in the analysis. The posted standard deviation is 0.29 in each group, and the primary outcome counts are 565.0 and 687.0. These figures describe the observed analysis; they do not establish that the estimate is free from bias or that it applies outside the analyzed population. A narrow interval can provide more precise information about the estimated contrast, but precision alone does not determine whether that contrast matters to participants.

The confidence interval's position relative to zero aligns with the reported unadjusted p-value of <0.001. In general, a confidence interval and a hypothesis test are related ways of expressing uncertainty under corresponding statistical assumptions. Neither one resolves uncertainties from an open-label design, exclusions from the modified intention-to-treat population, unreported missing-data procedures, or the absence of posted secondary outcome results.

The registry reports no adjusted confidence interval for ANCOVA. That omission limits direct assessment of the precision of the sponsor's model-based estimate. The available unadjusted interval nevertheless describes the recalculated comparison based on posted summary statistics. The two should remain clearly distinguished when describing precision.

11. 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 sample-size table and curve are planning illustrations based on the observed effect, not the original trial design. The registry record does not report the trial's original sample-size assumptions or justification. The planning calculation assumes a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. It considers detecting the stated fractions of the observed effect. These assumptions describe the illustration only.

The displayed figures are participants per arm, not total participants, because this is a comparison of two groups. For the row planning for 100% of the observed effect, the table gives 207 participants per arm for 80% power and 277 participants per arm for 90% power. For 75% of the observed effect, it gives 367 participants per arm for 80% power and 491 participants per arm for 90% power. For 50% of the observed effect, it gives 826 participants per arm for 80% power and 1,105 participants per arm for 90% power.

In general, power is the chance that a study will meet its prespecified statistical criterion if an effect of the assumed size is present and the planning assumptions hold. The 80% and 90% columns show the per-arm sample sizes associated with those target chances in this planning calculation. The higher target requires more participants for the same assumed effect. Smaller effects are harder to distinguish from random variation, so planning to detect a smaller fraction of the observed effect requires more participants per arm.

These numbers are not observed power. They are prospective planning illustrations calculated from the observed effect and stated assumptions; they do not show the probability that the completed trial's result is true, nor do they replace the trial's original design rationale.

12. 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 = 0.080; pooled standard deviation σ = 0.290.
2. n = 2 × (1.96 + 0.842)² × 0.290² / 0.080² = 206.3, rounded up to 207 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 <- 0.08                      # observed difference in means
sigma <- 0.29                     # 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(2515773)
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 = 207 (80% power), 277 (90% power)
 75% of observed effect: n per arm = 367 (80% power), 491 (90% power)
 50% of observed effect: n per arm = 826 (80% power), 1,105 (90% power)
power.t.test: 207.2 
Simulated power at n = 207 per arm: 0.799 (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.

13. Analyses Reported by the Sponsor

The registry reports the sponsor's primary analysis as ANCOVA, with a net mean difference of 0.08 and p <0.001. Confidence limits for that result are not reported in the registry record. The separate recalculation uses the posted group means, standard deviations, and analysis counts to produce an unadjusted mean difference of 0.08, a 95% confidence interval entirely above zero, and an unadjusted p-value of <0.001.

The reported point estimate and p-value agree between the sponsor analysis and the summary-statistic recalculation. This agreement supports a consistent description of the direction and statistical evidence in the two reported approaches. It does not make the analyses interchangeable: ANCOVA may account for covariates, whereas the recalculation does not adjust for covariates, and the details of the ANCOVA model are not reported in the registry record.

The recalculated interval belongs only to the unadjusted calculation and should not be used as a substitute for an adjusted interval. The registry's ANCOVA result is the sponsor's reported primary analysis; the recalculation is an additional unadjusted check based on the posted summaries. No discrepancy is flagged between them in the registry record.

14. Multiplicity, Missing Data and Analysis Populations

The registry lists four outcome measures and four analyses, but secondary outcome results are not posted in the registry record. The trial summary names additional patient-centered domains, including adherence and satisfaction, psychiatric symptom burden, and quality of life. Their results, definitions, analysis populations, and time points are not reported in the registry record. The primary result therefore should not be treated as evidence about those other outcomes.

Multiplicity arises when several outcomes, time points, subgroups, or statistical tests are considered. In general, conducting many tests can increase the chance of at least one apparently noteworthy result by chance. A prespecified hierarchy, adjustment method, or other strategy can help address that issue. The registry's multiplicity plan is not reported in the registry record, so it is not possible to determine how the four listed outcome measures or analyses were prioritized or whether any adjustment was planned or used. The presence of four outcome measures and four analyses alone does not establish how many tests were confirmatory.

The primary population is described as modified intention-to-treat: participants who initiated randomized treatment and had at least one post-baseline assessment. The reasons for requiring treatment initiation, the number and characteristics of participants excluded from the primary analysis, and the balance of exclusions between groups are not reported in the registry record. The outcome counts are 565.0 for metformin plus lifestyle intervention and 687.0 for lifestyle intervention alone, while overall enrollment is 1565. The record does not supply a participant flow accounting that explains the difference between enrollment and the analyzed population.

Missing measurements can affect a result if the probability of missingness is related to participant characteristics or outcomes. In general, analysts may use methods that make assumptions about the missing-data process, but those methods need to be described to assess their implications. The registry does not report how missing outcome data were handled, whether any imputation was used, or whether sensitivity analyses were performed. It also does not report whether an interim analysis or a plan for early stopping was used. These omissions limit assessment of the analysis strategy but do not change the arithmetic of the posted group summaries.

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 2% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: abdominal pain (0.9%), pneumonia (0.4%), pyelonephritis (0.4%), urethral/kidney stone (0.4%), infection (0.4%).

The registry reports serious adverse event figures separately from the primary endpoint analysis. For metformin plus lifestyle intervention, 579 participants were at risk and 12 were recorded as affected by serious adverse events. For lifestyle intervention alone, 720 were at risk and 25 were recorded as affected. The record also lists 0 deaths in each group. These are the posted counts; they should be read with their stated at-risk denominators rather than treated as counts from the primary BMI z-score sample.

For other adverse events, the record lists 558 participants at risk and 352 affected in the metformin-and-lifestyle group, and 686 at risk and 339 affected in the lifestyle-only group. The registry presents the most frequent serious-event terms with percentages: abdominal pain, 0.9%; pneumonia, 0.4%; pyelonephritis, 0.4%; urethral/kidney stone, 0.4%; infection, 0.4%; and refractory headaches, 0.3%. These terms are listed as serious adverse events, not as a complete account of every event.

The counts alone do not establish that one group had a safer overall experience. The registry record does not provide enough detail about event collection, severity grading, duration of observation, or the relationship of events to treatment. These figures describe what was posted and should not be interpreted as medical advice or as a definitive comparative safety assessment.

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

The primary posted result describes a difference in mean change in BMI z-score over the interval from baseline to Month 6. The metformin-and-lifestyle group had a mean change of -0.08, compared with 0.0 in the lifestyle-only group. The unadjusted mean difference is 0.08 in the direction of the lifestyle-only mean minus the metformin-and-lifestyle mean. The registry's ANCOVA analysis reports the same point estimate and a p-value of <0.001.

This result does not mean every participant's BMI z-score changed in the same way. Group means summarize average changes and do not convey the distribution of individual responses. Nor does the result show the absolute BMI z-score at Month 6 for either group. It describes a single continuous endpoint in the specified modified intention-to-treat sample.

The posted result does not establish whether the observed difference is clinically important. The registry does not report a threshold for meaningful change in BMI z-score, and statistical evidence is not equivalent to clinical importance. It also does not establish effects on psychiatric symptoms, quality of life, adherence, satisfaction, or other outcomes named in the trial summary; secondary results are not posted in the registry record.

Interpretation is bounded by the measurement interval and population definition. The primary outcome concerns baseline to Month 6, and the sample includes participants who initiated randomized treatment and had at least one post-baseline assessment. Detailed follow-up beyond this endpoint, reasons for exclusions, and missing-data methods are not reported in the registry record. The open-label design is another context for interpreting a pragmatic trial, although the primary outcome is a calculated measure based on weight, height, age, and sex.

Overall, the posted information supports describing a statistically distinguishable average change between the analyzed groups under the reported analyses. It does not, on its own, settle the practical importance, durability, or broader effects of the difference.

17. Why This Trial Matters Statistically

This trial offers a useful example of why the treatment contrast must be stated precisely. Both randomized groups received a healthy lifestyle intervention; the comparison is metformin added to that intervention versus the intervention alone. That distinction is essential when interpreting the mean difference, since the result does not isolate a contrast between metformin and no care.

The primary endpoint also illustrates the role of age- and sex-standardized measures in pediatric research. Raw BMI is converted to a z-score using growth charts, allowing change to be interpreted relative to reference values rather than as an uncontextualized raw BMI change. The registry defines larger scores as heavier weights and worse outcomes, which helps clarify the direction of the reported group means.

Statistically, the record places a model-based ANCOVA result alongside enough group-level summaries to describe an unadjusted recalculation. The point estimates and p-values agree, but the analyses are not identical. Distinguishing adjusted sponsor analyses from unadjusted calculations is important: an unadjusted confidence interval should not be passed off as the uncertainty interval for a covariate-adjusted estimate.

The trial also highlights the difference between a completed study's enrollment and the population used for its primary analysis. The modified intention-to-treat definition requires treatment initiation and at least one post-baseline assessment. Without a detailed accounting of exclusions and missing data, readers cannot fully assess how that analytic population relates to all enrolled participants.

Finally, the trial summary mentions several patient-centered outcomes beyond the primary endpoint, while no secondary results are posted in the registry record. A careful statistical reading keeps conclusions tied to the endpoint and time point actually reported, and separates statistical evidence from clinical importance and from unreported outcomes.

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.

Several limitations specific to the available trial record affect interpretation. The study is open-label, so participants and study personnel were not masked to assignment according to the registry's masking entry. The primary outcome is based on weight, height, age, and sex rather than a subjective rating, but the open-label design may still be relevant to other outcomes described in the trial summary. The record does not report details that would allow a fuller assessment of how treatment knowledge could have influenced study conduct or other measurements.

The primary analysis uses a modified intention-to-treat population rather than all enrolled participants. Its definition requires initiation of randomized treatment and at least one post-baseline assessment. The registry does not give a full participant-flow explanation, reasons for exclusion, or details of how missing outcome data were handled. Those omissions make it difficult to judge the potential effect of exclusions and missingness on the result.

Only the baseline-to-Month-6 primary endpoint result is posted in detail. Although the trial summary names adherence, satisfaction, psychiatric symptom burden, and quality of life, secondary outcome results are not posted in the registry record. The multiplicity plan and the relationship between the four listed outcomes and four analyses are also not reported. The record does not state whether interim analyses or sensitivity analyses were performed.

Details of randomization procedures, allocation concealment, ANCOVA covariates, and model diagnostics are not reported in the registry record. Those gaps limit independent evaluation of the sponsor's adjusted analysis. The unadjusted recalculation is based on posted summary statistics and cannot recover those missing model details. The record also does not report a clinically meaningful threshold for BMI z-score change, which limits interpretation of practical importance.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Metformin for Overweight & OBese ChILdren and Adolescents With BDS Treated With SGAs — NCT02515773.
• Registry record last updated 2025-01-14; status last verified 2024-12.
• Results first posted 2025-01-14.
• 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 describes a completed, randomized, open-label, parallel-group Phase 4 trial with stated enrollment of 1565. It compared metformin plus a healthy lifestyle intervention with the lifestyle intervention alone in children and adolescents with bipolar spectrum disorders treated with second-generation antipsychotics. The primary outcome was change in BMI z-score from baseline to Month 6 in a modified intention-to-treat sample.

The group mean changes were -0.08 and 0.0, respectively, with a posted mean difference of 0.08. The registry reports an ANCOVA p-value of <0.001; the unadjusted recalculation also gives p <0.001 and a confidence interval entirely above zero. Secondary outcome results are not posted in the registry record. Details of missing-data handling, multiplicity, and several analysis procedures are not reported.