Genotype-supported Versus Conventional Proton Pump Inhibitor Dosing: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Genotype-supported Versus Conventional Proton Pump Inhibitor Dosing (NCT02930824), a randomized na study. The primary outcome is Reflux Disease Questionnaire (RDQ).

2. Clinical Question

The clinical question concerns the comparative effect of making genotype information available to clinicians as part of dosing decisions. It does not isolate the effect of a particular PPI, dose, or prescribing algorithm. Because the endpoint population is defined by metabolizer result, the primary outcome does not directly summarize all adults and children enrolled. The registry does not report a separate primary-outcome result for participants with other metabolizer results, or a combined adult-and-pediatric estimate.

3. Trial Design

The registry describes a randomized, parallel-group comparative effectiveness study of genotype-supported versus conventional PPI management. Allocation is recorded as randomized, masking as none, and the model as parallel. The study was listed as completed, with enrollment of 185. The start is recorded as 2016-12 and primary completion as 2019-07-17. The sponsor is the University of Florida.

Adults and children presenting to gastroenterology clinics with GERD or dyspepsia symptoms could be recruited if they were starting PPI therapy or continued to have symptoms while taking a PPI. In the genotype-supported arm, physicians received CYP2C19 genotype information to assist with dosing. In the conventional arm, no genotype was provided to physicians for that purpose. The registry gives separate adult and pediatric arm descriptions, but the posted primary outcome results concern adults with a rapid or ultra-rapid metabolizer result.

The primary outcome groups contain 17 participants in adult genotype-guided treatment and 18 in adult conventional treatment. Pediatric outcome entries have group counts of 0 and no reported outcome values. Those entries do not establish that no children enrolled; they indicate that pediatric results are not presented in these outcome records. The relationship between the overall enrollment and the smaller adult outcome groups is not explained in the registry record.

The registry does not report the trial’s original sample-size justification, planned allocation ratio, detailed randomization procedure, or specific PPI dosing rules. It also does not report whether clinicians or participants received other information about the assigned strategy beyond the genotype availability described for each arm. Since masking is recorded as none, the registry does not describe a blinded comparison.

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

Analysis population for the primary outcome, as defined in the registry: Adult patients enrolled that have a CYP2C19 Rapid or Ultra-rapid metabolizer result.

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: Adult and children ages 5-100 with gastroesophageal reflux disease symptoms expected to take a proton pump inhibitor

6. Primary Endpoint

The registry lists the primary outcome as Reflux Disease Questionnaire (RDQ), measured over Change from baseline and 12 weeks.

The primary endpoint is the Reflux Disease Questionnaire (RDQ) mean score, expressed as change from baseline and at 12 weeks. The RDQ was developed to monitor treatment response over time. It assesses heartburn, regurgitation, and upper abdominal pain using 12 items across those 3 domains. Each symptom is assessed for frequency and severity over the previous week on a 6-point Likert scale, from 1 (did not have) to 6 (severe). The reported RDQ score is the mean response across the items and ranges from 1 to 6.

A repeated symptom measure can describe how participants’ reported symptoms change over the study period. Reporting change from baseline focuses the comparison on change in score rather than only on the score at a single time point. The registry describes the RDQ as psychometrically validated and designed to monitor treatment response, which supports its use as a symptom outcome in this study. The posted values are group means and standard deviations; they do not show individual response patterns or the distribution of scores.

The outcome timeframe is change from baseline and 12 weeks. The registry does not provide a separate baseline score and 12-week score for each group, a definition of the exact change calculation, or a threshold for interpreting the clinical importance of a score change. The reported endpoint should therefore be read as a group-level mean change, not as a count of participants who improved or as a measure of the proportion experiencing a particular degree of symptom change.

7. Results Posted in the Registry

• Adult Conventional Treatment: mean -0.30 versus -0.39 on control; difference 0.09 (95% CI -0.54 to 0.72), standardized effect d = 0.10; z-test p = 0.777.

8. Statistical Interpretation

9. Statistical Methods Explained

The registry reports a two-sided t-test for the primary adult comparison. A t-test assesses whether the observed difference between group means is large relative to the variability and uncertainty in the data, under the assumptions of the test. “Two-sided” means that departures in either direction are considered: the analysis does not test only whether one specified group has a higher mean. The registry reports a p-value of 0.78. The recalculation from the posted group summaries gives a p-value of 0.777. These are close, and the registry does not flag a discrepancy.

What does the mean difference represent?

The group mean is the average change in RDQ score among participants in that outcome group. The reported comparison is the conventional-treatment mean relative to the genotype-guided mean, with an estimated difference of 0.091. A difference near zero describes group means that are close on the scale of the reported estimate; it does not mean that every participant had the same change or that the strategies are equivalent. The estimate is a summary of the observed group averages, not a description of individual outcomes.

The recalculated comparison is unadjusted: it uses the posted means, standard deviations, and group counts, without adjustment for other participant characteristics. An unadjusted result can be useful for describing the basic group contrast, but it cannot account for possible imbalances in prognostic characteristics. The registry does not report an adjusted primary analysis or the covariates that such an analysis might use. The sponsor’s reported test is the registry’s own inferential result; the recalculation is a separate check based on the posted summary values.

Why do the standard deviations matter?

A standard deviation describes the spread of participant measurements around a group mean. The posted standard deviations are 1.2 for the genotype-guided group and 0.58 for the conventional group. Greater spread means participant changes were more dispersed around that group’s average. It also affects the uncertainty of a comparison of means. A mean alone does not convey this variability, which is why the group counts and standard deviations are relevant to interpreting the estimate.

The recalculation uses a pooled standard deviation of 0.933 as part of its standardized effect calculation. Pooling combines group variability into a single summary for that calculation. This is a calculation from the posted summaries, not a separately reported trial estimate. The registry does not state the assumptions used for its original analysis beyond identifying a two-sided t-test. In particular, details of assumption checks and any prespecified analysis choices are not reported in the registry record.

What does the p-value say?

In general, a p-value describes how unusual results at least as discrepant as the observed result would be under a specified null model, if the model assumptions hold. It does not give the probability that the null hypothesis is true, the probability that the result occurred by chance, or the probability that a management strategy has a clinical effect. The reported value of 0.78 is not small evidence against the null comparison in the context of this test. That does not prove the group means are equal. The estimate and its interval are needed to understand which differences remain compatible with the data.

The endpoint is continuous, so the analysis compares mean score changes rather than event counts or proportions. A mean is informative about the group average, but it may not capture skewed distributions, clusters of very different responses, or the share of participants with large changes. The registry does not provide individual observations, distribution plots, or responder analyses. Those features cannot be assessed from the posted group summaries alone.

General statistical interpretation also depends on study conduct and analysis populations. Randomization can support a fair comparison when assignment is implemented as planned and follow-up and analysis are handled appropriately. The registry does not describe the detailed randomization procedure, missing-outcome handling, or whether the analysis followed intention-to-treat principles. Without those details, the reported comparison can be described, but its sensitivity to exclusions or missing data cannot be evaluated.

10. Confidence Intervals and Precision

The recalculated mean difference is 0.091, with a confidence interval that extends from a negative value to a positive value. A confidence interval gives a range of estimates that are reasonably compatible with the observed data under the method’s assumptions. It communicates uncertainty about the magnitude and direction of the mean difference more directly than the point estimate alone. Because the interval crosses zero, the data do not precisely establish which group had the more favorable mean change.

The interval is wide relative to the point estimate, so a range of materially different group contrasts remains compatible with the summary data. This is consistent with the modest reported group counts and the observed variability. A confidence interval is not a guarantee that the true value lies within the particular interval, nor does it assign equal probability to all values in that range. It also does not establish clinical importance: the registry does not report a threshold for a meaningful RDQ change.

The interval and p-value address statistical uncertainty under the analysis method, not every source of uncertainty. They do not resolve possible effects of missing outcomes, departures from the planned analysis, unmeasured imbalances, or differences in how clinicians implemented genotype-supported dosing. The registry does not report enough detail to assess those issues. The posted result is therefore best read as an estimate with substantial uncertainty, rather than as evidence that the strategies have identical outcomes or as a precise estimate of a clinically important difference.

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 figures are planning illustrations based on the observed effect, not the original trial design. For this continuous outcome, the entries give participants per arm, not a total event count. The stated assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. Each row considers an effect equal to the stated fraction of the observed effect. The trial’s own original sample-size assumptions and justification are not reported in the registry record.

The 80% power column describes the number of participants per arm that the illustration estimates would be needed to have an 80% chance of detecting the selected effect under its assumptions, if that effect were the true one. The 90% power column uses a higher target chance of detection and therefore gives a larger planning count for the same effect. These are prospective planning quantities, not descriptions of the probability that the completed study detected an effect.

The row using 100% of the observed effect lists 1,653 participants per arm for 80% power and 2,212 participants per arm for 90% power. Planning for 75% of the observed effect gives 2,937 and 3,932 participants per arm, respectively. Planning for 50% of the observed effect gives 6,609 and 8,847 participants per arm. Smaller target effects require more participants because they are harder to distinguish from random variation. These calculations use the observed effect as an input, so they are sensitive to the uncertainty in that estimate. They are not observed power and should not be read as the trial’s actual design requirements.

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 = 0.091; pooled standard deviation σ = 0.934.
2. n = 2 × (1.96 + 0.842)² × 0.934² / 0.091² = 1,652.0, rounded up to 1,653 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 <- 0.091                      # observed difference in means
sigma <- 0.933530282250548                     # 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(2930824)
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 = 1,653 (80% power), 2,212 (90% power)
 75% of observed effect: n per arm = 2,937 (80% power), 3,932 (90% power)
 50% of observed effect: n per arm = 6,609 (80% power), 8,847 (90% power)
power.t.test: 1653 
Simulated power at n = 1,653 per arm: 0.802 (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 a two-sided t-test with p = 0.78. The recalculation from the posted means, standard deviations, and group counts gives p = 0.777 and an unadjusted mean difference of 0.091 for conventional relative to genotype-guided treatment. The reported and recalculated p-values are close, and the registry does not flag a discrepancy. The recalculation is a summary-statistic check; it does not replace the sponsor’s analysis or reproduce details that are not reported.

The sponsor’s p-value is the analysis result recorded for the trial. The recalculation gives a way to relate that result to the posted group summaries and an estimated interval. It is not an adjusted analysis and does not establish that the sponsor used the same computational details. The registry does not report the sponsor’s full analysis specification, including any assumption checks or handling of missing outcome data.

Both results indicate that this comparison does not show a clear mean difference by the reported test. Neither the sponsor’s p-value nor the recalculated result proves equality or rules out potentially meaningful differences. Interpretation should take account of the small reported outcome groups, the interval spanning values in both directions, and the lack of reported clinical-importance threshold.

14. Multiplicity, Missing Data and Analysis Populations

The registry lists 5 outcome measures and 5 analyses, but reports 0 posted secondary outcomes. The primary outcome is the RDQ mean change for adults with rapid or ultra-rapid CYP2C19 metabolizer results. The registry does not explain how the five analyses relate to the five measures or identify a hierarchy among all analyses. It also does not report whether the primary analysis was prespecified as the sole confirmatory test or whether adjustment for multiple comparisons was planned.

Multiplicity arises when several outcomes, comparisons, or analyses are considered: the more tests performed, the greater the chance that at least one result may appear notable by chance. General approaches include specifying a primary test in advance, controlling an overall error rate, or treating secondary analyses as exploratory. Which approach was used for this study is not reported in the registry record. The single posted p-value should therefore be interpreted as the result of its stated comparison, without assuming an unreported multiplicity adjustment or analysis hierarchy.

The record does not report missing-data counts or methods for handling missing RDQ measurements. It does not state whether participants with unavailable outcome data were excluded, imputed, or handled by another method. The registry also does not report whether the analysis population was intention-to-treat, per-protocol, or defined in another way. These omissions limit assessment of whether the analyzed participants represent everyone randomized and whether missingness could have influenced the comparison.

The pediatric outcome records have group counts of 0 and no reported values. The registry does not explain the absence of pediatric outcome results or provide a separate pediatric analysis. Nor does it give results for adults outside the rapid or ultra-rapid metabolizer subgroup specified for the primary outcome. These unreported analyses should not be inferred from the adult result. The registry’s adverse-event entries list zero counts across the recorded groups, but further detail about ascertainment and reporting is not available. These entries describe what is posted and do not establish that all possible events were captured.

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

The posted results describe average RDQ score changes for two adult treatment groups among participants with rapid or ultra-rapid CYP2C19 metabolizer results. The genotype-guided group mean change was -0.387 and the conventional group mean change was -0.296. The estimated mean difference was 0.091, and the reported test did not show a clear difference between the groups. The interval’s inclusion of values on both sides of zero leaves the direction and size of the underlying mean contrast uncertain.

This is not proof that genotype-supported and conventional management produce identical outcomes. A lack of a clear statistical difference can occur when the true difference is small, when the data are variable, or when there are too few observations to estimate the contrast precisely. The registry does not report an equivalence or non-inferiority margin, so the result cannot establish equivalence or non-inferiority. It also does not identify whether any difference would matter to patients, because a clinical-importance threshold is not reported.

The endpoint is a group mean change on a symptom questionnaire. It does not show how many participants improved, worsened, or remained unchanged, nor does it describe individual experiences. The result is not a comparison of a single PPI drug or a fixed dosing regimen. It evaluates a management strategy in which physicians either received genotype information to assist with dosing or did not receive it for that purpose.

The primary outcome result applies to the specified adult metabolizer subgroup. It should not be extended automatically to children, adults with other metabolizer results, or all people with GERD or dyspepsia. The registry does not report outcome estimates for those groups. Overall enrollment of 185 does not alter the fact that the posted primary comparison uses the smaller adult group counts. The findings are descriptive of the reported comparison and do not constitute medical advice.

16. Why This Trial Matters Statistically

This study raises a statistical question that is distinct from simply comparing two medicines: how should a trial evaluate a treatment strategy that makes genetic information available to clinicians? The intervention is a decision-support approach, while the comparator leaves that information unavailable for dosing. Such strategy comparisons can include variation in clinicians’ choices, so the estimated group difference reflects the assigned management approach as recorded, rather than a fixed drug-dose contrast.

The endpoint also highlights the difference between a mean change and the pattern of responses across participants. Average RDQ change is a compact summary of symptom scores, but it cannot indicate whether the average reflects broadly similar changes or a mix of large improvements and worsening. The posted standard deviations add information about spread, while the lack of participant-level results limits further distributional assessment.

Another statistical lesson is that a non-small p-value should not be read as evidence of no effect. Here, the mean difference is small relative to the uncertainty conveyed by the confidence interval, and the interval includes values in either direction. A careful interpretation reports both the estimate and its precision, without turning a test result into a claim of equivalence.

Finally, the distinction between total enrollment and the outcome population matters. The registry records 185 enrolled participants, but the primary analysis concerns 17 and 18 adults in the posted outcome groups. Clear reporting of analysis populations, missing data, and results for additional groups would help readers understand how the primary estimate relates to the broader study. Those details are not reported in the registry record.

17. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The primary outcome results are limited to adults with rapid or ultra-rapid metabolizer results, and the adult outcome groups are small. Pediatric group entries contain zero participants and no outcome values, while results for adults with other metabolizer results are not reported. The registry does not explain why the posted outcome population is narrower than the overall enrollment of 185 or how the outcome groups were selected from the enrolled population.

The registry provides group means and standard deviations but not individual data, separate baseline and 12-week scores, or distributions of change. It therefore cannot show participant-level patterns or the proportion meeting a particular response threshold. No threshold for a clinically meaningful RDQ change is reported, making it difficult to assess the clinical importance of the difference from the posted summary alone.

Details of the original sample-size justification, dosing algorithm, specific PPI use, and detailed randomization procedures are not reported in the registry record. The record also does not describe missing-outcome handling, the analysis population, adjustment for participant characteristics, or methods for addressing multiple analyses. These omissions limit evaluation of robustness and of how closely the analysis follows the study’s original plan.

Masking is recorded as none. The registry does not provide further information about how knowledge of assignment may have affected symptom reporting or clinical decisions. The adverse-event records list zero counts across groups but do not provide additional ascertainment details. Finally, the comparison concerns availability of genotype information as part of a dosing strategy; it does not isolate a particular medicine, dose, or prescribing rule.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Genotype-supported Versus Conventional Proton Pump Inhibitor Dosing — NCT02930824.
• Registry record last updated 2024-05-16; status last verified 2024-05.
• Results first posted 2021-06-21.
• 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.

21. Record Summary

The registry describes a completed randomized, parallel-group comparison of CYP2C19 genotype-supported and conventional PPI management, sponsored by the University of Florida. Overall enrollment was 185. Adults and children with GERD or dyspepsia symptoms could be recruited when starting PPI therapy or continuing to have symptoms during current PPI treatment.

The primary outcome was RDQ mean change from baseline at 12 weeks among adults with rapid or ultra-rapid metabolizer results. The posted adult group means were -0.387 for genotype-guided treatment and -0.296 for conventional treatment. The sponsor-reported two-sided t-test p-value was 0.78; the unadjusted recalculation gave a mean difference of 0.091 and p = 0.777. The confidence interval spans values on both sides of zero, indicating uncertainty about the direction and size of the mean difference.

The record does not report pediatric primary outcome values, secondary outcome results, the original sample-size justification, or details of missing-data handling and the analysis population. The findings therefore describe the posted adult subgroup comparison rather than all enrolled participants.