Efficacy of Esomeprazole in Patients With Frequent Heartburn: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Efficacy of Esomeprazole in Patients With Frequent Heartburn (NCT01370538), a randomized phase 3 study. The primary outcome is Percentage of Heartburn Free 24 Hour Days During 14 Days of Randomized Treatment.

2. Clinical Question

This is a question about an average over the treatment period. It does not ask whether every participant experienced the same change, nor does it directly measure the severity or duration of each heartburn episode. The primary endpoint represents the share of observed 24-hour days without heartburn, averaged across the full analysis set in each group.

The registry lists additional questions about the number of participants with limited heartburn days and about days without heartburn during specified parts of treatment. Those outcomes provide other ways to describe participants’ experience, but the registry record does not report numerical results for all of them. The posted results therefore address the main average-day comparison more fully than the full set of listed questions.

The clinical question is bounded by the study’s 14-day treatment period and the enrolled population. The registry record does not establish how the outcome behaves over longer periods or in populations not described in the record. The comparison describes the observed trial data; it is not, by itself, a complete assessment of every possible benefit or harm.

3. Trial Design

The registry identifies this as a Phase III, randomized, double-blind, placebo-controlled, parallel-group trial. Its official title describes a multicenter study of 14-day treatment with esomeprazole 20 mg once daily in subjects with frequent heartburn. The recorded condition is heartburn, and the sponsor is AstraZeneca. The study status is completed.

Enrollment was 341. For the primary outcome, the posted esomeprazole group summary includes 162 participants and the placebo group summary includes 158. The primary outcome population is named as the full analysis set. The registry record does not explain the relationship between the enrollment count and the outcome-specific group counts, so the reason those counts differ is not reported in the registry record.

Random assignment is intended to make treatment groups comparable on average, while double masking is intended to reduce the influence of treatment knowledge on participants’ experiences and outcome assessment. These are general features of randomized, masked trials; the registry record does not describe the specific procedures used to generate or conceal assignment or maintain masking.

The study began in 2011-08, and primary completion is recorded as 2011-10. The primary endpoint was the percentage of heartburn-free 24-hour days during 14 days of randomized treatment, assessed from randomisation to day 14. The record also lists secondary outcomes concerning participants with few heartburn days and participants with days without heartburn over particular time windows.

Detailed eligibility criteria, site locations, the original sample-size justification, and the trial’s full statistical analysis plan are not reported in the registry record. The trial’s reported design therefore establishes the broad comparison and treatment period, but not all operational or analytical details needed to reproduce the original study conduct.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Full analysis set

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 Percentage of Heartburn Free 24 Hour Days During 14 Days of Randomized Treatment, measured over From randomisation to day 14.

The primary endpoint is the percentage of heartburn-free 24-hour days during the 14 days of randomized treatment. For each participant, the measure represents the proportion of days in the observation period without heartburn; the reported group result is a mean, with standard deviations also posted. The timeframe is from randomisation to day 14, and the stated population is the full analysis set.

As a general statistical rationale, a continuous measure of days without symptoms can reflect the overall pattern across a treatment period rather than reducing each participant’s experience to a single threshold, such as whether a specified number of symptom-free days was reached. A group mean offers a concise summary, while the standard deviation describes variation among participants around that mean. Neither statistic indicates that all participants had the same experience.

The endpoint is expressed as a percentage, so the difference between group means is naturally described in percentage points. The posted means were 46.79% in the esomeprazole group and 30.93% in the placebo group; the posted mean difference was 15.86 percentage points. This endpoint concerns days classified as free of heartburn. The registry does not report a separate measure of episode severity or duration in the primary outcome description.

7. Results Posted in the Registry

• Esomeprazole: mean 46.79 versus 30.93 on control; difference 15.86 (95% CI 9.04 to 22.68), standardized effect d = 0.51; z-test p = <0.001.

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.

Number of Subjects Reporting Heartburn 2 Days or Less During the 14 Days Randomized Treatment Period

• 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:
  • Esomeprazole: 16.7% (27/162) versus 1.3% (2/158) on control; risk difference 15.4 percentage points (95% CI 9.4 to 21.4), risk ratio 13.17 (3.18–54.44); unadjusted two-proportion z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Esomeprazole than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The secondary outcomes broaden the description of heartburn by considering threshold counts and symptom-free days over selected parts of the treatment period. The numerical secondary result that is posted concerns the number of participants reporting heartburn on two days or less during the 14-day randomized treatment period. The registry lists additional secondary outcomes concerning days without heartburn and symptom frequency over other time windows, but does not report numerical results for all listed outcomes.

The available secondary result points in the same direction as the primary mean comparison: the esomeprazole group has a higher reported count meeting the stated low-heartburn threshold than the placebo group. This is a different summary of the experience, not an independent confirmation of the primary estimate. The primary outcome uses a mean percentage across the period, whereas the secondary threshold outcome classifies participants according to a specified number of heartburn days.

Secondary outcomes can help show whether the broad pattern depends on how the outcome is defined or on the time window considered. However, when multiple endpoints are tested, each additional test creates another opportunity for a result to appear statistically notable by chance. General statistical practice therefore treats unadjusted secondary p-values cautiously, especially when a complete testing hierarchy or multiplicity plan is unavailable.

The registry record lists no reported analyses and does not report whether secondary outcomes were prespecified in a formal hierarchy, whether adjustment was used, or how missing secondary observations were handled. The unadjusted results should therefore be read as descriptive supporting evidence, not as a set of equally definitive tests. The absence of posted numerical results for some listed outcomes also means that the overall secondary-outcome pattern cannot be fully assessed from the registry record.

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

• Number of Subjects Reporting Heartburn 2 Days or Less During the 14 Days Randomized Treatment Period
• Comparison of Number of Subjects With 0, 1, 2, 3 or 4 Days With no Heartburn Over Days 1 to 4 Between Esomeprazole 20 mg and Placebo
• Number of Subjects With Heartburn 1 Day or Less During the Final Week, Second Week, First Week of Treatment

9. Statistical Interpretation

10. Statistical Methods Explained

The registry describes a randomized, double-blind, parallel-group comparison of esomeprazole and placebo. In a randomized trial, chance determines the assigned group according to an allocation procedure. This aims to balance measured and unmeasured participant characteristics across groups on average, although chance differences can remain. Masking is intended to limit the influence of knowledge about assignment on reported symptoms, behavior, and assessment. These are general statistical concepts; the specific allocation and masking procedures are not reported in the registry record.

What does a mean difference represent?

The primary outcome is continuous: each participant contributes a percentage of heartburn-free days, and the arm summaries report means and standard deviations. A mean difference compares the average outcome in one group with the average in the other. Because the outcome is itself a percentage, the difference is expressed in percentage points. The posted comparison is 15.86 percentage points, favoring a higher average in the esomeprazole group. That group-level quantity does not describe the outcome for each participant separately.

A standard deviation describes how spread out participant-level values are around a group mean. The posted standard deviations were 33.32 for esomeprazole and 28.86 for placebo. As a general point, a larger standard deviation indicates greater variation around the mean, but it does not tell whether the distribution is symmetric, whether it contains outliers, or how many participants had a particular value. Those details are not reported in the registry record.

How should the p-value be understood?

A p-value measures how unusual data at least as incompatible with a specified no-difference hypothesis would be under the assumptions of the statistical test. The posted unadjusted p-value for the primary comparison is < 0.001. It is not the probability that the no-difference hypothesis is true, and it does not measure the size or practical importance of the effect. The registry record does not describe the full test specification or a multiplicity adjustment for the primary comparison.

“Unadjusted” matters because the posted p-value is not described as accounting for a family of multiple comparisons. When several outcomes or analyses are examined, the chance of observing at least one small p-value can increase. General methods for handling this include prespecifying a hierarchy of outcomes or adjusting inference across a family of tests. Whether such an approach was used in the trial is not reported in the registry record.

Why are confidence intervals useful?

A confidence interval combines an estimate with an indication of its statistical precision under a specified method. The primary mean difference has a 95% confidence interval from 9.035 to 22.68. A wider interval indicates less precision than a narrower interval on the same scale. The interval is not a range that contains every participant’s result; it concerns uncertainty in the estimated group difference. Its interpretation depends on the method and assumptions used to calculate it, details that are not fully described in the registry record.

What can be said about the analysis population?

The primary outcome population is identified as the full analysis set. That label does not, by itself, define which randomized participants were included or how missing outcome observations were handled. The registry record does not report the operational definition of the full analysis set, the amount or pattern of missing primary outcome data, or the method used to address missingness. Those omissions limit assessment of whether the posted arm summaries represent all randomized participants or a particular subset.

The posted primary summaries include 162 participants assigned to esomeprazole and 158 assigned to placebo. The registry’s enrollment count is 341, but it does not explain the outcome-specific counts. A reader should not infer a particular exclusion, withdrawal, or missing-data process from the difference in those reported counts.

Why can over-dispersion matter?

As a general statistical concept, over-dispersion occurs when observed variability is greater than a model assumes. If an analysis uses a model that understates variability, its standard errors can be too small and its confidence intervals too narrow. This issue is relevant in some count and binary-data models, particularly when observations vary more than the chosen model allows. The primary endpoint is a continuous percentage summarized by a mean and standard deviation. The registry record does not identify a model that can be assessed for over-dispersion, so it is not possible to say whether this issue affected the trial’s primary analysis.

The posted secondary endpoint with numerical results is a count of participants meeting a threshold for heartburn days. Its arm counts are 27 and 2, and the posted risk-difference comparison is 0.154, with a 95% confidence interval from 0.094 to 0.214 and an unadjusted p-value < 0.001. A risk difference compares proportions between groups; it is distinct from the primary mean difference. The registry record lists no reported analyses and does not provide a detailed model or analysis specification for this secondary outcome.

Finally, the record does not report the trial’s original sample-size justification, interim analyses, or detailed statistical analysis plan. The planning illustrations based on observed effects are therefore not evidence of the assumptions used to design the trial. They answer a different question: how sample size might vary under stated planning assumptions if a specified fraction of the observed effect were targeted.

11. Confidence Intervals and Precision

The primary mean difference is 15.86 percentage points, and its 95% confidence interval extends from 9.035 to 22.68. The interval communicates uncertainty around the estimated difference: values within it are compatible with the observed data under the interval’s statistical method and assumptions. It is not a description of the range of individual participants’ outcomes.

The interval’s limits are on the same percentage-point scale as the mean difference. Its lower limit is above zero, which is consistent with a positive difference in average heartburn-free days under the comparison used. The unadjusted p-value is < 0.001. The p-value and interval are related summaries of statistical evidence, but they answer different questions: the p-value evaluates compatibility with a specified null hypothesis, while the interval displays a range of estimates compatible with the data and method.

Precision is influenced by the amount of information in the data and by variability. The posted standard deviations, 33.32 for esomeprazole and 28.86 for placebo, indicate substantial spread around the group means on the percentage scale. A confidence interval for the difference is more informative about the uncertainty in the group contrast than either standard deviation alone. The posted group counts are 162 and 158.

Statistical precision does not determine clinical importance. The registry record does not report a threshold for a meaningful difference, and the confidence interval cannot supply one. Also, the registry does not report the complete analysis method, missing-data handling, or a multiplicity procedure. Those details can affect how confidently the interval and p-value should be interpreted as representing the original planned analysis.

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 figures are planning illustrations based on the observed effect, not the trial’s original design calculations. The registry record does not report the original sample-size assumptions or justification. The illustrations assume a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The target effect is set at a stated fraction of the observed effect.

For the primary continuous outcome, the figures are participants per arm. Planning for 100% of the observed effect gives 61 participants per arm for 80% power and 82 participants per arm for 90% power. Planning for 75% of the observed effect gives 108 participants per arm for 80% power and 145 participants per arm for 90% power. Planning for 50% of the observed effect gives 243 participants per arm for 80% power and 326 participants per arm for 90% power.

Read each pair as the approximate number assigned to each arm under the stated assumptions to have the specified chance of detecting the target effect if that effect is real and the assumptions hold. The 90% power target requires more participants per arm than the 80% target at each stated effect fraction. The planning curve expresses the same general relationship: as the effect to be detected becomes smaller, the required sample size rises. These figures are not observed power; they are prospective illustrations calculated from the observed effect, not a report of the trial’s original design.

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 = 15.860; pooled standard deviation σ = 31.198.
2. n = 2 × (1.96 + 0.842)² × 31.198² / 15.860² = 60.7, rounded up to 61 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 <- 15.86                      # observed difference in means
sigma <- 31.1978391648117                     # 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(1370538)
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 = 61 (80% power), 82 (90% power)
 75% of observed effect: n per arm = 108 (80% power), 145 (90% power)
 50% of observed effect: n per arm = 243 (80% power), 326 (90% power)
power.t.test: 61.7 
Simulated power at n = 61 per arm: 0.797 (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 statistical inference is made across multiple outcomes, time windows, groups, or analyses. Even if each test uses the same nominal threshold, conducting more tests increases the chance that at least one will appear statistically notable by chance. This is a general statistical issue, not evidence that a particular result is invalid. A prespecified hierarchy or an adjustment method can help define how evidence should be interpreted across a family of comparisons.

The registry record lists a primary outcome and several secondary outcomes, but does not report a multiplicity strategy. It also lists no reported analyses. The posted unadjusted p-values should consequently be read as nominal results rather than as evidence that a stated family-wise error rate was controlled. Whether the original analysis plan specified a testing order or adjustment is not reported in the registry record.

Missing-data handling is also important for a 14-day outcome. The primary measure summarizes heartburn-free days over that period, but the registry record does not report how incomplete daily observations were treated, whether imputation was used, or how sensitivity to missingness was assessed. The label full analysis set is given for the primary endpoint, but its operational definition is not reported. The record does not provide enough information to determine whether all randomized participants contributed to the posted endpoint summaries.

The group counts for the primary outcome are 162 for esomeprazole and 158 for placebo, while total enrollment is 341. The registry record does not explain the relationship among these counts, so no particular reason for the difference can be assigned. It also does not report interim analyses or the detailed statistical analysis plan. These absences limit reconstruction of the original inferential strategy and make it important to distinguish the posted descriptive results from a fully documented analysis.

15. Safety Results

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

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 1% reporting threshold.

Most frequent non-serious adverse event terms

The serious adverse event counts are 0 for the esomeprazole group and 0 for the placebo group. The recorded numbers at risk were 162 and 158, respectively. No serious adverse event terms are listed. These are the serious-event figures recorded during the study; they do not establish that no serious events could occur outside the period or circumstances recorded.

Other adverse events were recorded for 18 participants in the esomeprazole group and 15 in the placebo group. These counts are descriptive. The registry record does not identify the specific events in this summary or state that they were caused by treatment. Death counts are not reported in the registry record.

Safety counts should be interpreted alongside the treatment period and the completeness of event reporting. The trial evaluated 14 days of randomized treatment, and the registry record does not provide further detail about event severity beyond the serious-event classification, event timing, or follow-up after treatment. The figures therefore describe the recorded events and should not be treated as a comprehensive assessment of longer-term safety.

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

The results describe a difference in the average percentage of heartburn-free days between the randomized groups during 14 days of treatment. The posted mean was 46.79% for esomeprazole and 30.93% for placebo, with a mean difference of 15.86 percentage points. The 95% confidence interval, from 9.035 to 22.68, gives a measure of statistical precision for that group comparison. These values summarize the trial data; they do not mean that every participant had an outcome equal to the group average.

The comparison is limited to the enrolled participants, the outcome definition, and the treatment period recorded. The registry record does not establish how the result would apply to populations with different characteristics, other treatment schedules, or longer periods. More detailed eligibility criteria and participant characteristics are not reported in the registry record, which further limits judgments about how closely another population matches this trial.

Statistical evidence and clinical importance are separate considerations. The unadjusted p-value is < 0.001, but a p-value does not establish that a difference is important to participants. The registry record does not report a threshold for clinical importance. The confidence interval describes uncertainty in the estimated group difference under the method used; it does not resolve whether that difference is worthwhile in a particular setting.

The result also does not prove a specific mechanism, identify which individuals might experience a particular outcome, or establish that the observed group difference would persist beyond the recorded period. The registry record does not report a complete statistical analysis plan, missing-data methods, or a multiplicity strategy. These omissions leave important uncertainty about how the displayed comparisons relate to the trial’s original inferential plan.

Safety information is similarly bounded. The registry records no serious adverse events in either group and records other adverse events for 18 esomeprazole participants and 15 placebo participants. The record does not report death counts. These data are descriptions of events recorded during the study, not a guarantee about risks beyond the observed treatment period or evidence about causation.

17. Why This Trial Matters Statistically

This trial illustrates why the endpoint scale matters. The primary measure is a percentage of heartburn-free days, summarized as a mean, so the principal contrast is a difference in percentage points. That answers a different question from a threshold outcome that counts participants meeting a symptom-day criterion. Both can be informative, but they describe different aspects of the data and should not be treated as interchangeable.

It also illustrates the value of reading an estimate together with its confidence interval. The mean difference of 15.86 percentage points is more informative when considered alongside the interval from 9.035 to 22.68 than when read in isolation. The interval communicates precision, while the p-value < 0.001 addresses compatibility with a no-difference hypothesis under the comparison used. Neither quantity alone determines clinical importance.

The listed secondary endpoints raise a common multiplicity issue. Multiple outcomes can offer useful perspectives, but unadjusted testing across outcomes does not have the same interpretation as a single prespecified primary comparison. The registry record does not report a multiplicity approach or detailed analysis plan, so the primary and secondary evidence should be distinguished rather than combined indiscriminately.

Finally, the record demonstrates the limits of registry-level summaries. It identifies the design, primary endpoint, arm summaries, and some event counts, but does not report the full analysis population definition, missing-data procedures, original sample-size assumptions, or detailed inferential methods. Statistical interpretation is strongest when the estimate, uncertainty, endpoint definition, design, and reporting gaps are considered together.

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 treatment period was 14 days, and the primary endpoint covers that period from randomisation to day 14. The results therefore describe a short observation window; longer-term outcomes are not reported in the registry record. The record also does not report detailed eligibility criteria or participant characteristics, limiting assessment of how well the enrolled group represents other populations with heartburn.

The primary outcome is a mean percentage, which summarizes group experience but can conceal variation in individual outcomes. Standard deviations are posted, but the registry record does not provide participant-level results or a fuller description of the outcome distribution. The primary population is named as the full analysis set, but its definition and the handling of incomplete daily observations are not reported.

The registry lists secondary outcomes, but numerical results are not reported for all of them. The posted p-values are unadjusted, and a multiplicity strategy is not reported. The record also does not include a detailed statistical analysis plan or reported analyses, making it difficult to determine whether the displayed comparisons match the original planned inferential procedures.

Enrollment was 341, while the primary outcome summaries include 162 participants in the esomeprazole group and 158 in the placebo group. The registry record does not explain the relationship between enrollment and those outcome counts. No reason for the count difference should be inferred from the registry summary alone. The original sample-size justification and interim analysis plan are also not reported.

Finally, the serious adverse event counts are zero in both groups, but death counts are not reported and the record contains limited detail on other adverse events. These safety summaries do not establish the absence of unrecorded events or describe outcomes after the study period.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Efficacy of Esomeprazole in Patients With Frequent Heartburn — NCT01370538.
• Registry record last updated 2013-04-11; status last verified 2013-03.
• Results first posted 2013-04-11.
• 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

NCT01370538 is a completed Phase III randomized, double-blind, parallel-group, placebo-controlled trial sponsored by AstraZeneca. It enrolled 341 participants with frequent heartburn and evaluated esomeprazole 20 mg once daily over 14 days. The primary endpoint was the percentage of heartburn-free 24-hour days from randomisation to day 14 in the full analysis set.

The posted group means were 46.79% for esomeprazole and 30.93% for placebo. The posted mean difference was 15.86 percentage points, with a 95% confidence interval from 9.035 to 22.68 and an unadjusted p-value < 0.001. The registry also reports one numerical secondary outcome and event counts, while numerical results are not reported for all listed secondary outcomes.

The record does not report the detailed statistical analysis plan, missing-data methods, multiplicity strategy, or original sample-size justification. The findings should be read as summaries of the recorded trial outcomes and treatment period, with those reporting limits in mind.