Combined Antioxidant and Preeclampsia Prediction Studies (CAPPS): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Combined Antioxidant and Preeclampsia Prediction Studies (CAPPS) (NCT00135707), a randomized phase 3 study. The primary outcome is Composite of Pregnancy-associated Hypertension and Serious Adverse Outcomes in the Mother or Fetus or Neonate.

2. Clinical Question

The registry describes the broader CAPPS project as having a randomized clinical trial and an observational cohort component. The cohort was intended to study potential predictors of preeclampsia; the primary randomized comparison concerns the vitamin and placebo groups. The registry record does not provide results for the observational cohort in the posted outcomes. The statistical question addressed by the primary comparison is therefore whether the frequency of the composite event differed between randomized groups, not whether a prediction marker identifies future preeclampsia.

3. Trial Design

CAPPS is described in the registry as a phase 3, randomized, triple-masked, parallel-group study. The record lists enrollment of 10,154 and status as completed. Its official title includes both a randomized clinical trial of antioxidants and an observational cohort study to predict preeclampsia. The brief summary describes the randomized part as multicenter and focused on low-risk nulliparous women, and says that participants were enrolled between 9 and 16 weeks of gestation.

The randomized comparison was between vitamin C and vitamin E capsules and matching placebo capsules. The detailed arm descriptions specify 1000mg of vitamin C and 400IU of vitamin E per capsule, taken twice daily between randomization and delivery. The brief summary describes the vitamins as daily, so the dosing frequency is not fully consistent across the registry descriptions. The placebo description lists its capsule ingredients and also specifies twice-daily use.

Follow-up was intended to continue through delivery. The brief summary describes monthly visits for study drug supply, weight and blood-pressure measurement, urine-protein measurement, and assessment of side effects, as well as blood and urine collection at two visits. The primary outcome timeframe is 20 weeks through discharge following delivery. The trial therefore evaluates events over pregnancy and the immediate period extending through discharge, rather than a short-term laboratory response.

The observational cohort was described as including patients also enrolled in the trial and as collecting potential biochemical and biophysical predictors, including uterine artery Doppler measurements and blood for a complete blood count. The registry record does not report results for that cohort. It also does not describe the trial’s original sample-size justification, detailed allocation procedures, the identities of the masked parties, or the methods used to handle missing outcome data.

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: Women who were no more than 15 weeks pregnant and who consented to participate in the study were given a supply of placebo and asked to return within 2 weeks. Those who returned, who had taken at least 50% of the placebo they were supposed to have taken, and who still met the eligibility criteria were randomly assigned to receive study drug.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The analysis was intent to treat.

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 Composite of Pregnancy-associated Hypertension and Serious Adverse Outcomes in the Mother or Fetus or Neonate, measured over 20 weeks through discharge following delivery.

The primary endpoint is a count of participants experiencing a composite of pregnancy-associated hypertension and serious adverse outcomes in the mother, fetus, or neonate. The registry defines the hypertension component as severe hypertension, blood pressure at least 160/110, or mild hypertension, blood pressure at least 140/90, at or after 20 weeks of gestation, in conjunction with one of the listed serious conditions. These include elevated liver enzymes, thrombocytopenia, elevated serum creatinine, eclamptic seizure, indicated preterm birth before 32 weeks because of hypertension-related disorders, fetal growth below the 3rd percentile adjusted for sex and race or ethnic group, fetal death after 20 weeks, or neonatal death.

The endpoint is assessed from 20 weeks through discharge following delivery. Because it combines hypertension with serious maternal or infant outcomes, it captures a broader clinical event than hypertension alone. A composite can increase the number of participants who meet an endpoint and can summarize several related complications in one comparison. However, its meaning depends on the components: a composite result does not by itself show which component accounts for the overall pattern. The registry record does not provide component-specific results for this primary endpoint, so the posted comparison cannot establish which individual components contributed most.

The primary analysis population is described as intent to treat, meaning the comparison is framed around randomized assignment rather than treatment actually received. The registry does not provide further detail on how participants with incomplete outcome information were handled.

7. Results Posted in the Registry

• Vitamins: 6.1% (305/4993) versus 5.7% (285/4976) on control; risk difference 0.4 percentage points (95% CI -0.5 to 1.3), risk ratio 1.07 (0.91–1.25); unadjusted two-proportion z-test p = 0.420.

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.

Pregnancy Associated Hypertension

• 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:
  • Vitamins: 29.2% (1457/4993) versus 26.6% (1322/4976) on control; risk difference 2.6 percentage points (95% CI 0.9 to 4.4), risk ratio 1.10 (1.03–1.17); unadjusted two-proportion z-test p = 0.004.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Vitamins than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Medically Indicated Delivery Because of Hypertension

• 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:
  • Vitamins: 10.3% (509/4952) versus 9.6% (473/4934) on control; risk difference 0.7 percentage points (95% CI -0.5 to 1.9), risk ratio 1.07 (0.95–1.21); unadjusted two-proportion z-test p = 0.250.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Vitamins and the control. The interval gives the range of effects that remain compatible with the data.

Aspartate Aminotransferase ≥100 U/Liter

• 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:
  • Vitamins: 0.7% (35/4993) versus 1.0% (48/4976) on control; risk difference -0.3 percentage points (95% CI -0.6 to 0.1), risk ratio 0.73 (0.47–1.12); unadjusted two-proportion z-test p = 0.147.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Vitamins and the control. The interval gives the range of effects that remain compatible with the data.

The posted secondary outcomes cover several related but distinct aspects of pregnancy-associated hypertension and maternal or infant complications. Across the available secondary comparisons, estimates do not all point in the same direction, and their precision varies. One secondary comparison has a p-value below 0.05, while other posted comparisons do not. This variation is not inconsistent with the primary composite result: a composite and its separate or related outcomes need not have identical event patterns or statistical precision.

Secondary outcomes are useful for describing the breadth of the results and identifying questions that may warrant further study. They do not replace the primary endpoint, which was the central randomized comparison specified in the registry record. In particular, a result for pregnancy-associated hypertension alone is not equivalent to the primary endpoint, which requires hypertension in conjunction with a listed serious outcome. The endpoint definitions and denominators can differ, so estimates should be interpreted in their own terms.

The registry lists 38 outcome measures and 38 analyses, with 27 secondary outcomes posted. When many outcomes are examined, the chance of observing at least one apparently unusual result can increase, even if no true differences exist across the outcomes. The registry record does not report a multiplicity adjustment or a prespecified hierarchy for interpreting the secondary tests. Accordingly, unadjusted secondary results carry less weight than a prespecified primary result: they are best treated as exploratory signals rather than as stand-alone confirmation.

Without a stated multiplicity strategy, a nominal p-value should not be read in isolation as decisive evidence. The overall pattern, the endpoint definitions, the confidence intervals, and whether the finding was planned all matter. The registry does not report enough detail to determine which secondary analyses were prespecified or how their interpretation was intended to be controlled.

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

• Preeclampsia (Mild, Severe, HELLP Syndrome, Eclampsia)
• Pregnancy Associated Hypertension
• Medically Indicated Delivery Because of Hypertension
• Aspartate Aminotransferase ≥100 U/Liter
• Creatinine ≥1.5 mg/dl (133 μmol/Liter)
• Antepartum Bleeding
• Premature Rupture of Membranes
• Placental Abruption

9. Statistical Interpretation

10. Statistical Methods Explained

What does a risk ratio describe?

General statistical explanation: a risk ratio compares the probability of an event in one group with the probability in a reference group. A value of 1 corresponds to equal event probabilities; values above 1 indicate a higher event probability in the group in the numerator, and values below 1 indicate a lower one. For the primary outcome, the sponsor reports the vitamins group relative to placebo. The reported risk ratio is 1.07, so the point estimate is above the equal-risk value. This is a descriptive direction, not proof that the true relative risk differs from 1.

The registry also contains an unadjusted risk-difference calculation based on the group counts. A risk difference expresses the absolute separation between event probabilities, whereas a risk ratio expresses their relative comparison. These measures can tell complementary stories, especially when events are uncommon, because a relative comparison can look substantial even when the absolute separation is modest. The posted primary risk-difference estimate is 0.004, with limits of -0.005 and 0.013. The sponsor’s analysis is presented as a risk ratio, and its estimates should not be mistaken for the risk difference.

What does the confidence interval add?

General statistical explanation: a confidence interval describes the range of estimates compatible with the observed data under a specified method and its assumptions. It conveys uncertainty around the point estimate, rather than only whether a test crosses a significance threshold. For the primary sponsor analysis, the 95% interval is 0.91 to 1.25. Since it includes 1, it is compatible with no relative difference as well as with relative differences in either direction. Its breadth also indicates that the result does not pin down the relative effect very precisely.

A confidence interval is not a range that is guaranteed to contain the true value, and it is not a probability statement about the parameter after the data are observed. Its coverage interpretation applies across repeated use of the method under its assumptions. The registry record identifies the interval as 95%, but does not give the detailed calculation method for that interval.

What does the chi-squared test assess?

The sponsor reports a chi-squared analysis for the risk ratio. General statistical explanation: a chi-squared test for a binary outcome assesses whether the observed distribution of events across groups is compatible with a null hypothesis of no group association. The test compresses the observed counts into a test statistic and corresponding p-value. It does not measure the clinical importance of a difference, and it does not establish the probability that the null hypothesis is true.

The primary p-value is 0.42. That value does not meet common criteria for declaring a statistically significant difference, and it is consistent with the interval including the null value. It should not be interpreted as proof that the two groups have identical risks. Statistical significance and precision are different: the test addresses evidence against a null value, while the interval shows a range of effect sizes that remain compatible with the data under the method.

How should counts and denominators be read?

The primary outcome is binary at the participant level: a participant either met the composite endpoint during the defined period or did not. The registry lists a group denominator and an event count for each arm. These counts are not event rates per unit of person-time, and the record does not describe a time-to-event analysis for the primary outcome. The risk ratio uses the event frequencies implied by the counts and denominators.

The primary analysis is described as intent to treat. General statistical explanation: an intent-to-treat approach preserves the comparison by randomized assignment, which is important because randomization is intended to make the groups comparable at baseline. The registry does not give enough detail to determine how incomplete follow-up, withdrawal, or missing outcome status was addressed. The listed denominators should therefore be read as the denominators posted for this outcome, not as proof that every randomized participant had complete follow-up.

Why can a composite endpoint be useful and difficult?

General statistical explanation: combining related outcomes can make a study-wide endpoint more inclusive and may allow a single comparison of a broader clinical question. But components can differ in frequency and severity. An overall composite can be driven by more common components, while a less common but serious component contributes fewer observed events. Without component-level results, the overall estimate cannot be assigned to a particular part of the composite. The registry’s description lists several maternal, fetal, and neonatal events, but primary component-specific counts are not reported in the registry record.

The endpoint also includes hypertension and a qualifying serious condition together; it is not simply a count of all cases of hypertension. The separate secondary outcome for pregnancy-associated hypertension is a different measure. Results for that secondary outcome should not be substituted for the primary composite result.

What can an unadjusted recalculation establish?

The simple recalculation uses the posted summary counts and denominators rather than individual participant data or a more detailed model. It provides a transparent check of the basic group comparison, but cannot reproduce analyses that depend on covariate adjustment, stratification, missing-data procedures, or other features not shown in the summary statistics. The registry flags no discrepancy between the sponsor result and the recalculation. The posted sponsor analysis remains the trial’s own reported result, while the simple calculation is a descriptive cross-check.

General statistical explanation: binary outcomes can also be compared with measures such as odds ratios, risk differences, or regression models. These measures are not interchangeable, and an odds ratio can differ from a risk ratio, particularly as outcomes become more common. The registry includes recalculated risk-ratio and odds-ratio summaries, but the sponsor’s primary report is the chi-squared risk-ratio result. The interpretation should retain the measure and analysis actually reported rather than silently switching metrics.

11. Confidence Intervals and Precision

The primary sponsor estimate is a risk ratio of 1.07, with a 95% confidence interval from 0.91 to 1.25. The interval is broad enough to include the null value of 1 and values on both sides of it. Thus, the data do not precisely determine the direction or size of the relative difference. The point estimate alone should not be treated as the answer: uncertainty around it is central to interpretation.

General statistical explanation: precision concerns how tightly an estimate is determined by the data under the analysis method. A narrower interval usually indicates greater precision, while a wider interval leaves a larger range of effects compatible with the observations. Precision is not the same as clinical importance, and a narrow interval around a small effect can still describe a small difference. Conversely, a non-significant result can remain uncertain if the interval includes effects in either direction that would matter.

The primary comparison includes thousands of participants in each group, yet the composite event counts are relatively small compared with the group denominators. The amount of information for a binary endpoint depends in part on how many events occur, not only on the total number enrolled. The registry’s interval therefore remains the most useful guide to how well the relative comparison is resolved.

The primary confidence interval is a relative-effect interval. It should not be read as a range for the absolute difference in event frequency. The recalculated risk-difference interval is a separate measure, with its own scale and interpretation. Neither interval establishes equivalence, because the registry record does not report an equivalence margin or a noninferiority margin.

The listed interval and p-value describe sampling uncertainty under the analysis method. They do not account automatically for every possible source of uncertainty, such as missing outcomes, endpoint classification, or multiplicity across many outcomes. The registry record does not provide enough information to assess those issues fully.

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 sample-size calculations are planning illustrations based on the observed effect, not the trial’s original design assumptions. For binary outcomes such as the primary endpoint, the displayed figures are participants per arm. The planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect targeted is a stated fraction of the observed effect. The registry does not report the trial’s original sample-size justification or its design assumptions.

For the primary endpoint, planning for 100% of the observed effect gives 60,191 participants per arm for 80% power and 80,578 participants per arm for 90% power. Planning for 75% of the observed effect gives 106,199 participants per arm for 80% power and 142,169 participants per arm for 90% power. Planning for 50% of the observed effect gives 237,129 participants per arm for 80% power and 317,448 participants per arm for 90% power.

In plain terms, 80% power means that a study designed under the stated assumptions would have an 80% chance of detecting the targeted effect at the specified significance level if that effect were true. The 90% power column represents a higher chance of detecting that same targeted effect under those assumptions, and it requires a larger planned sample. Smaller effects are harder to distinguish from random variation, so the calculations require more participants as the target effect becomes smaller.

These are not observed power figures. They are hypothetical planning quantities calculated using the observed effect and the listed assumptions; they do not show the probability that the completed trial’s result is correct or that an effect exists.

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, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.057; treatment proportion p₁ = 0.061; average p̄ = 0.059.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.3337 = 0.6540.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.3337 = 0.2808.
4. n = (0.6540 + 0.2808)² / (0.0038)² = 60,190.1, rounded up to 60,191 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
p0 <- 0.0572749196141479                          # control proportion
p1 <- 0.0610855197276187                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^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 (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(135707)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 60,191 (80% power), 80,578 (90% power)
 75% of observed effect: n per arm = 106,199 (80% power), 142,169 (90% power)
 50% of observed effect: n per arm = 237,129 (80% power), 317,448 (90% power)
power.prop.test: 60190.1 
Simulated power at n = 60,191 per arm: 0.802 (20000 replicates)

The simulation draws many pairs of event counts at the planned sample size and counts how often the two-proportion z-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. Analyses Reported by the Sponsor

The sponsor’s reported primary analysis is a chi-squared comparison expressed as a risk ratio. It gives a risk ratio of 1.07, a 95% confidence interval from 0.91 to 1.25, and p = 0.42. The simple unadjusted recalculation from the posted group counts is close to the sponsor estimate. The registry flags no discrepancy between them.

The recalculation is based on the summary counts and denominators. It is useful as a check that the broad numerical interpretation is consistent with the posted data, but it is not an independent reconstruction of every step in the sponsor’s analysis. It cannot account for details not available in the registry record, such as participant-level data, possible analysis adjustments, or missing-data procedures. The registry identifies the sponsor’s method as chi-squared but does not give a complete statistical analysis plan.

For the primary conclusion, both the sponsor result and the simple comparison point to an estimate near the null value and substantial uncertainty around it. The sponsor’s reported result should be treated as the trial’s own analysis. The recalculation complements it by making the summary-count comparison more transparent; it should not replace the reported analysis or be presented as if it used additional participant-level information.

15. Multiplicity, Missing Data and Analysis Populations

The registry record lists 38 outcome measures and 38 analyses, and 27 secondary outcomes are posted. Testing multiple outcomes raises a general statistical concern: as the number of comparisons increases, the chance of at least one apparently noteworthy result can increase. Statistical approaches can address this issue by adjusting significance thresholds, defining a hierarchy of outcomes, or emphasizing a primary endpoint while treating other analyses as exploratory.

A multiplicity adjustment, a hierarchy for testing outcomes, and details of which secondary analyses were prespecified are not reported in the registry record. It is therefore not possible to establish from the posted information whether the secondary p-values were adjusted, whether a particular testing sequence was planned, or how the trial’s analysis plan intended secondary findings to be interpreted. The primary endpoint remains the main result to distinguish from the broader set of secondary comparisons.

The registry describes the primary population as intent to treat, but does not provide a detailed account of missing data or incomplete follow-up. The handling of participants with unknown endpoint status, withdrawals, or missing measurements is not reported in the registry record. The outcome-specific denominators are posted, but those alone do not reveal why they may differ or how missingness was addressed.

The registry also does not report interim analyses, stopping rules, a detailed statistical analysis plan, or the trial’s original sample-size justification. These are distinct issues: interim analysis procedures concern repeated review of accumulating data, while multiplicity concerns interpretation across multiple tests and outcomes. Neither should be inferred from the completed status or the reported p-values.

The primary analysis is identified as intent to treat, and the sponsor’s primary test is identified as chi-squared. Further detail on model specification, stratification, covariate adjustment, endpoint adjudication, or sensitivity analyses is not reported in the registry record. These gaps limit how completely the analysis can be reproduced from the registry summary.

16. 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 1% reporting threshold.

Most frequent serious adverse event terms

Serious events with the highest arm-level rate: fetal death (1.9%), neonatal death (0.5%), fetal birth defects (0.4%), renal and urinary disorders (0.2%), bleeding (0.2%).

The registry lists serious adverse event information for both groups. For vitamins, 5087 participants are listed as at risk and 108 as affected; for placebo, 5065 are listed as at risk and 173 as affected. It also lists zero participants with other adverse events in each group and one death in each group. These figures describe the registry’s posted safety fields and should be read separately from the primary efficacy endpoint.

The registry also lists the leading serious adverse event terms and percentages: fetal death, 1.9%; neonatal death, 0.5%; fetal birth defects, 0.4%; renal and urinary disorders, 0.2%; bleeding, 0.2%; and hemorrhage, 0.1%. The record does not provide enough detail to establish whether these terms overlap, how the percentages were calculated, or whether events were judged related to study treatment. The percentages should not be added together or interpreted as mutually exclusive categories.

These summaries do not by themselves establish comparative safety. The registry record does not report a statistical comparison for the serious adverse event figures, detailed event-level descriptions, or attribution by investigators. Differences in the listed counts should therefore be described without inferring causation or a treatment effect. The primary endpoint’s serious outcomes and the adverse-event reporting fields are also distinct summaries and should not be conflated.

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

The primary result describes the comparison of the vitamins and placebo groups for a composite of pregnancy-associated hypertension and serious maternal, fetal, or neonatal outcomes over the stated period. The sponsor’s estimate is close to the equal-relative-risk value, and its 95% confidence interval includes that value. The posted analysis does not show a statistically clear difference between groups on this composite.

This does not prove that the groups had identical outcomes. The confidence interval includes a range of relative effects, and the registry does not report an equivalence or noninferiority design with a prespecified margin. A failure to reject a null hypothesis is not the same as demonstrating equivalence. It also does not establish that the intervention caused any difference in the event counts; the estimates describe the randomized comparison and its uncertainty.

The result applies to the specific population, intervention descriptions, comparator, endpoint definition, and follow-up window in the registry record. It does not answer every question about preeclampsia, all pregnancy complications, or the observational cohort’s aim of predicting preeclampsia. It also does not identify effects on each component of the primary composite, because those component-specific primary results are not reported in the registry record.

The secondary outcomes broaden the descriptive picture but are less definitive when considered across many analyses without a reported multiplicity strategy. Safety summaries likewise require caution: the registry lists serious adverse event counts and terms, but does not provide enough information to make a causal or comprehensive safety assessment.

Overall, the posted data support a measured conclusion: the primary comparison is statistically inconclusive, with an estimate near no relative difference and uncertainty that remains meaningful. The record does not justify a claim of benefit, harm, equivalence, or absence of effect beyond what the stated estimates and their limitations can support.

18. Why This Trial Matters Statistically

This trial is statistically notable for pairing a large randomized comparison with a relatively infrequent composite endpoint. The enrollment is 10,154, while the primary event counts are 305 in the vitamins group and 285 in the placebo group. For binary outcomes, the amount of information depends not only on the number enrolled but also on how many participants experience the event. That helps explain why a large trial can still yield an interval that leaves uncertainty about the relative effect.

The primary estimate illustrates why point estimates, confidence intervals, and p-values should be read together. The risk ratio of 1.07 is above 1, but the 95% confidence interval from 0.91 to 1.25 includes the null value and the p-value is 0.42. The point estimate’s direction alone is not enough to support a firm conclusion; the interval shows that the data do not sharply resolve the range of plausible relative differences under the analysis.

The composite endpoint also raises a central design and interpretation issue. Combining hypertension with serious maternal, fetal, or neonatal outcomes creates a broader measure, but its result does not reveal which component contributes to the comparison. This makes component-level reporting important when interpreting the clinical meaning of a composite.

Finally, the registry’s secondary analyses show why multiplicity matters. Multiple outcome tests can generate isolated nominal findings, and the registry does not report an adjustment strategy. Distinguishing a prespecified primary test from secondary analyses is therefore essential to a statistically disciplined reading of the evidence.

19. 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 limitations follow from the registry information. The brief summary describes vitamin C and vitamin E as daily, whereas the detailed arm description specifies capsules taken twice daily. The more detailed arm description gives 1000mg of vitamin C and 400IU of vitamin E per capsule, but the difference in frequency across registry text makes the treatment schedule less clear from the record alone.

The primary endpoint is a composite with multiple maternal, fetal, and neonatal components. Component-specific primary results are not reported in the registry record, which limits the ability to determine what drove the overall comparison. The registry also posts several related secondary outcomes, but does not report whether multiplicity was controlled or which secondary analyses were prespecified.

The record identifies an intent-to-treat primary analysis but does not describe missing-data handling, losses to follow-up, or sensitivity analyses. It does not report the original sample-size justification, interim monitoring procedures, or a detailed statistical analysis plan. These omissions limit the ability to audit the analysis or assess whether all design assumptions were met.

The registry describes an observational cohort alongside the randomized trial, but does not post results for that cohort among the outcomes presented. The cohort’s predictive aims therefore cannot be assessed from the trial comparison. The safety summary also lacks detailed information on event attribution and overlap among adverse-event terms. These limits call for cautious interpretation of both the secondary and safety findings, without implying that unreported analyses were or were not performed.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Combined Antioxidant and Preeclampsia Prediction Studies (CAPPS) — NCT00135707.
• Registry record last updated 2019-02-21; status last verified 2019-02.
• Results first posted 2018-11-20.
• 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.

23. Record Summary

The registry record describes CAPPS as a completed phase 3, randomized, triple-masked, parallel-group study with enrollment of 10,154. The randomized comparison evaluated vitamin C and vitamin E capsules against matching placebo. The primary endpoint was a composite of pregnancy-associated hypertension and serious adverse outcomes in the mother, fetus, or neonate, assessed from 20 weeks through discharge following delivery.

The primary sponsor analysis reports a risk ratio of 1.07, with a 95% confidence interval from 0.91 to 1.25 and p = 0.42. The interval includes the null relative-risk value, so the posted result does not provide statistically clear evidence of a difference on the primary composite. Multiple secondary outcomes and safety summaries are also posted, but the registry does not report a multiplicity strategy, detailed missing-data methods, or results for the observational cohort.