Hydrochlorothiazide as add-on to Olmesartan/Amlodipine in Hypertension: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Hydrochlorothiazide as add-on to Olmesartan/Amlodipine in Hypertension (NCT00902538), a randomized phase 3 study. The primary outcome is Change in Seated Diastolic Blood Pressure (SeDBP) of the Triple Combinations OM/AML/HCTZ 40/10/12.5 and 40/10/25 mg vs. OM/AML 40/10 mg.

2. Clinical Question

The clinical question is about differences in measured blood pressure over the randomized comparison period. It is not a question about long-term outcomes, nor does this endpoint alone describe every possible benefit or harm. The registry lists a broader set of secondary measures, including seated systolic pressure, ambulatory blood pressure, and achievement of a blood pressure goal.

3. Trial Design

This was a Phase 3, randomized, triple-masked, parallel-group trial with an enrollment of 2204. It was sponsored by Daiichi Sankyo and is recorded as completed. The registry lists essential hypertension as the condition. The study began in 2009-04, and primary completion was recorded in 2010-09.

The design included four treatment periods rather than a single uninterrupted comparison. In Period 1, participants received olmesartan 40 mg and amlodipine 10 mg in an 8-week, single-blind run-in. In Period 2, participants could be randomized to continue that combination or receive it with hydrochlorothiazide at 12.5 mg or 25 mg. Period 2 was double-blind and lasted 8 weeks. The primary comparison concerns this randomized period, using the week 8 measurement as baseline and the week 16 measurement as follow-up.

In Period 3, participants entering that period received the 12.5 mg hydrochlorothiazide combination in a single-blind period lasting 8 weeks. Period 4 was double-blind and involved a further treatment decision based on Period 3 response: responders continued the 12.5 mg combination, while non-responders could receive either the 12.5 mg or 25 mg combination. These later stages address questions distinct from the initial randomized comparison, so the primary result should not be treated as an estimate of every treatment decision across all periods.

The registry record does not report the original sample-size justification, full randomization procedure, masking roles, or detailed protocol for the later-period comparisons. Those details should not be inferred from the design labels alone.

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: The number of subjects entering Period 2 was only 808 because 1278 did not meet the entry criteria.

1-Single-blind, Run-in, Single-Treatment

Reasons for not completing this period:

2-Randomized Double-blind 3 Treatments

Reasons for not completing this period:

3-Single-blind, Single Treatment

Reasons for not completing this period:

4-Randomized Double-blind 2 Treatments

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The Full Analysis Set 1 included 806 randomized subjects who received at least 1 dose of double-blind study medication in Period II and provided at least 1 SeDBP measurement in Period II: 269 subjects in the OM/AML 40/10 mg group, 268 subjects in the OM/AML/HCTZ 40/10/12.5 mg group, and 269 subjects in the OM/AML/HCTZ 40/10/25 mg group.

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 Change in Seated Diastolic Blood Pressure (SeDBP) of the Triple Combinations OM/AML/HCTZ 40/10/12.5 and 40/10/25 mg vs. OM/AML 40/10 mg, measured over baseline (8 weeks) to 16 weeks.

The primary endpoint was change in seated diastolic blood pressure, measured in mm Hg, from the week 8 baseline to week 16. The registry describes the outcome as a least-squares mean with a standard error and states that three cuff blood pressure measurements were taken at each visit. The endpoint therefore concerns a continuous measure of blood pressure, rather than a count of participants reaching a threshold.

A change score expresses how the measured value at follow-up differs from the baseline value. Negative changes in the posted arm summaries indicate decreases from baseline. Comparing these changes across randomized groups asks whether the observed average change differed between the two-drug comparator and either triple combination during the randomized period.

Seated diastolic pressure is one way to quantify blood pressure in a standardized clinic measurement setting. Repeated cuff measurements at a visit can help characterize the measurement at that visit, but the registry record does not describe how the three readings were combined. It also does not provide the exact blood pressure goal used in the later goal-achievement outcome.

The registry lists a full analysis set for the primary outcome: randomized participants who received at least one dose of double-blind study medication in Period II and provided at least one seated diastolic pressure measurement in that period. This definition makes clear that the endpoint analysis did not simply include everyone enrolled. The endpoint is useful for evaluating the randomized groups’ measured blood pressure changes, but it does not by itself establish the effects on outcomes beyond the measured period or population.

7. Results Posted in the Registry

• OLM 40-AML 10-Hydrochlorothiazide (HCTZ) 12.5: mean -7.10 versus -6.10 on control; difference -1.00 (95% CI -2.52 to 0.52), standardized effect d = -0.11; z-test p = 0.199.
• OLM 40-AML 10-HCTZ 25: mean -8.90 versus -6.10 on control; difference -2.80 (95% CI -4.32 to -1.28), standardized effect d = -0.31; 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.

Change in Seated Systolic Blood Pressure (SeSBP) of the Triple Combinations OM/AML/HCTZ 40/10/12.5 and 40/10/25 mg vs. OM/AML 40/10 mg

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • OLM 40-AML 10-Hydrochlorothiazide (HCTZ) 12.5: mean -8.60 versus -6.90 on control; difference -1.70 (95% CI -3.82 to 0.42), standardized effect d = -0.14; z-test p = 0.116.
  • OLM 40-AML 10-HCTZ 25: mean -10.50 versus -6.90 on control; difference -3.60 (95% CI -5.72 to -1.48), standardized effect d = -0.29; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between OLM 40-AML 10-Hydrochlorothiazide (HCTZ) 12.5 and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for OLM 40-AML 10-HCTZ 25 than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Number of Subjects Achieving Blood Pressure (BP) Goal at Week 16.

• 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:
  • OLM 40-AML 10-Hydrochlorothiazide (HCTZ) 12.5: 29.5% (79/268) versus 24.2% (65/269) on control; risk difference 5.3 percentage points (95% CI -2.2 to 12.8), risk ratio 1.22 (0.92–1.62); unadjusted two-proportion z-test p = 0.165.
  • OLM 40-AML 10-HCTZ 25: 41.3% (111/269) versus 24.2% (65/269) on control; risk difference 17.1 percentage points (95% CI 9.3 to 24.9), risk ratio 1.71 (1.32–2.20); unadjusted two-proportion z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between OLM 40-AML 10-Hydrochlorothiazide (HCTZ) 12.5 and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for OLM 40-AML 10-HCTZ 25 than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Change in 24-hour Diastolic Blood Pressure (DBP) Assessed by 24-hour Ambulatory Blood Pressure Measurement (ABPM).

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • OLM 40-AML 10-Hydrochlorothiazide (HCTZ) 12.5: mean -4.00 versus -2.10 on control; difference -1.90 (95% CI -3.34 to -0.46), standardized effect d = -0.22; z-test p = 0.010.
  • OLM 40-AML 10-HCTZ 25: mean -5.30 versus -2.10 on control; difference -3.20 (95% CI -4.64 to -1.76), standardized effect d = -0.38; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for OLM 40-AML 10-Hydrochlorothiazide (HCTZ) 12.5 than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for OLM 40-AML 10-HCTZ 25 than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The secondary outcomes extend the primary seated diastolic pressure question to seated systolic pressure, achievement of a blood pressure goal, and 24-hour ambulatory blood pressure. Other registered secondary questions concern non-responders in later treatment periods. These outcomes measure related but not identical aspects of blood pressure, and they can provide context for the primary endpoint rather than replacing it.

Across the secondary results with posted numerical summaries, the arm-level pattern is broadly aligned with greater blood pressure reductions in the triple-combination groups than in the two-drug group, while goal achievement is another way of representing response as a categorical outcome. The registry also reports sponsor analyses for the posted secondary comparisons. Their interpretation should follow the relevant outcome and sponsor method rather than treating every secondary measure as interchangeable with the primary endpoint.

Secondary outcomes generally carry less confirmatory weight when many outcomes or comparisons are considered and the multiplicity strategy is not reported. A collection of unadjusted tests can produce apparently strong findings by chance, even when each test is considered separately. The registry record does not report whether secondary analyses were prespecified as confirmatory, how multiplicity was controlled, or the full statistical plan. Accordingly, the secondary measures are useful for understanding the breadth and consistency of the recorded results, but they do not independently resolve the primary question.

The registry lists nine posted secondary outcomes, including both the initial randomized comparison and questions involving non-responders. The latter refer to a selected group defined by response in an earlier period, so those later-period findings address a different population and treatment decision. Not all details needed to assess every secondary analysis are reported in the registry record. The overall pattern should therefore be read as contextual evidence, with attention to the distinct measurement methods, populations, and comparison periods.

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

• Change in Seated Systolic Blood Pressure (SeSBP) of the Triple Combinations OM/AML/HCTZ 40/10/12.5 and 40/10/25 mg vs. OM/AML 40/10 mg
• Number of Subjects Achieving Blood Pressure (BP) Goal at Week 16.
• Change in 24-hour Diastolic Blood Pressure (DBP) Assessed by 24-hour Ambulatory Blood Pressure Measurement (ABPM).
• Change in 24-hour Systolic Blood Pressure Assessed by 24-hour Ambulatory Blood Pressure Measurement.
• In Non-responders, the Change in Seated Diastolic Blood Pressure Associated With the Triple Combinations OM/AML/HCTZ 40/10/12.5 and 40/10/25 mg.
• In Non-responders, the Change in Seated Systolic Blood Pressure Associated With the Triple Combinations OM/AML/HCTZ 40/10/12.5 and 40/10/25 mg.
• In Non-responders, the Number of Subject Meeting Their Blood Pressure Goals Associated With the Triple Combinations OM/AML/HCTZ 40/10/12.5 and 40/10/25 mg.
• In Non-responders, the Change in 24-hour Diastolic Blood Pressure Assessed by 24-hour Ambulatory Blood Pressure Measurement.

9. Statistical Interpretation

10. Statistical Methods Explained

What does a least-squares mean represent?

A least-squares mean is an adjusted group mean estimated through a statistical model. In general, an analysis can use such a model to account for other measured factors included in the model, rather than merely subtracting one unadjusted group average from another. The registry identifies the primary outcome summaries as least-squares means and identifies ANCOVA as the sponsor’s comparison method. However, the covariates and model specification are not reported in the registry record. The precise adjustments behind the sponsor’s estimates therefore cannot be described.

The registry reports standard errors beside the arm-level least-squares means. A standard error describes the estimated uncertainty in a mean, rather than the spread of individual participants’ measurements. It is useful for understanding the precision of the mean estimate, but it is not a confidence interval by itself. The arm-level entries do not include confidence limits in the registry data considered for these comparisons.

What does ANCOVA do?

ANCOVA, or analysis of covariance, compares groups using a model that includes the outcome and one or more covariates. In a clinical trial, such a model can estimate group differences while accounting for baseline or other specified participant characteristics. This is a general description; the registry record does not state which covariates were included, how the model was fitted, or whether any assumptions were checked. The posted primary comparison p-values are identified as ANCOVA results, but the corresponding adjusted mean differences and confidence limits are not reported in the registry record.

The recalculated comparison is different in purpose and construction. It is an unadjusted comparison based on the posted group summaries. The standard deviations used in that recalculation are approximate, derived from the reported standard errors. It does not reproduce the sponsor’s ANCOVA model. Because the record marks a discrepancy between the recalculation and the sponsor result, the sponsor’s own analysis takes precedence when describing the primary statistical finding.

How should p-values be read?

A p-value is a measure of how compatible the observed data are with a specified null model, under the assumptions of the statistical test. It is not the probability that the null hypothesis is true, nor does it measure the size or clinical importance of a difference. A small p-value can be evidence against the null model, but interpretation also depends on the endpoint, analysis plan, uncertainty interval, and whether multiple comparisons were considered. The registry’s p-values alone do not reveal the sponsor’s full inferential framework.

The record gives a reported p-value for each primary dose comparison, but does not report a multiplicity adjustment. When multiple related hypotheses are tested, the chance of at least one apparently unusual result can increase if each is assessed separately. This is a general reason to interpret unadjusted comparisons cautiously. The original statistical plan and any prespecified hierarchy are not reported in the registry record.

Why do the recalculated estimates differ from the reported analysis?

The recalculated estimates are simple, unadjusted differences between the posted least-squares means. The registry’s own analyses use ANCOVA, and the model’s adjustment details are unavailable. The recalculated confidence intervals also rely on an approximate standard deviation inferred from the arm standard errors. Differences in model adjustment, variance estimation, and other analysis choices can produce differences in estimates or p-values. These recalculations should therefore be read as checks based on the posted summaries, not as a replacement for the trial’s analysis.

For the primary outcome, the recalculated mean differences compare each hydrochlorothiazide group with the olmesartan-amlodipine group. A negative difference indicates a more negative mean change in the triple-combination group. The recalculated confidence interval gives a range of values compatible with the summary-based calculation under its assumptions. It does not provide the sponsor’s adjusted confidence interval, which is not reported in the registry record.

What is the role of the analysis population?

The registry defines the primary Full Analysis Set 1 as participants who were randomized, received at least one dose of double-blind study medication in Period II, and provided at least one seated diastolic blood pressure measurement in Period II. The arm counts in that population are 269, 268, and 269. This definition includes a post-randomization treatment and measurement requirement. The record does not report how missing endpoint measurements were handled, whether any imputation was used, or whether sensitivity analyses assessed alternative assumptions about missing data.

In general, analysis populations matter because the set of participants contributing data determines the scope of the estimate. An analysis restricted by treatment receipt and availability of a measurement may differ from an analysis of all randomized participants. The registry description gives the eligibility conditions for this primary analysis, but does not provide enough detail to evaluate every potential effect of exclusions or missing observations.

Why can over-dispersion matter?

For count or binary outcomes, over-dispersion means that variability in the data exceeds what a simple statistical model expects. In general, ignoring extra variability can make standard errors too small and confidence intervals too narrow. The primary endpoint is continuous, and the registry does not report an over-dispersion assessment for it. This concept is relevant more broadly to trial analyses, but should not be treated as a reported issue in this study.

For the binary blood pressure goal outcome, the registry identifies the sponsor’s comparison method as Cochran-Mantel-Haenszel. In general, this method can compare categorical outcomes while accounting for stratification. The registry record does not state the strata used or provide the full analysis specification. Those details should not be assumed from the method name alone.

11. Confidence Intervals and Precision

The recalculated primary comparison for hydrochlorothiazide 12.5 mg has a 95% confidence interval from -2.524 to 0.524 mm Hg. This interval includes zero, so the summary-based estimate is compatible with a range that includes no difference as well as differences in either direction. The interval for hydrochlorothiazide 25 mg is -4.324 to -1.276 mm Hg; it is entirely below zero, indicating that the recalculated difference is compatible with a more negative change in that group than in the comparator under the calculation’s assumptions.

A confidence interval describes uncertainty around an estimate. In general, a narrower interval indicates greater precision, while a wider interval leaves more values plausible. An interval is not a guarantee that the true value lies within its limits, and its interpretation depends on the method used to construct it. These intervals are recalculated from posted summary statistics, not the sponsor’s ANCOVA model. The registry’s own adjusted confidence intervals are not reported in the registry record.

The arm-level primary summaries have standard errors of 0.55 mm Hg in each of the three groups. The recalculation uses these values to approximate variability for the difference, and consequently should not be read as an exact reconstruction of the sponsor’s model-based uncertainty. The registry reports the primary ANCOVA p-values, but not the adjusted effect estimates or their confidence limits. This limits direct comparison of the sponsor’s effect size and precision with the recalculated interval.

Precision is also distinct from clinical importance. An interval describes statistical uncertainty, not whether a difference matters to patients or changes care. The registry record does not provide a threshold for a clinically important difference in seated diastolic pressure. Therefore, the confidence intervals can characterize the numerical uncertainty in the posted and recalculated estimates, but cannot establish clinical importance on their own.

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. They report participants per arm for 80% power and participants per arm for 90% power, not total events. The planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. For each comparison, the effect to detect is a stated fraction of the observed effect. The registry record does not report the assumptions used for the trial’s original sample-size justification.

The 80% power column indicates the number of participants per arm that the planning calculation estimates would be needed to have an 80% chance of detecting the specified effect under the stated assumptions, if that effect were present. The 90% power column uses a higher target chance of detection, and therefore generally requires more participants per arm for the same effect. These are planning quantities, not guarantees about a future study’s result.

The rows for 100%, 75%, and 50% of the observed effect illustrate how the required sample size changes when the effect to detect is smaller than the observed estimate. Smaller effects need more participants to distinguish them from random variation under the same assumptions. For example, planning for 75% of the observed effect uses a less pronounced difference than planning for the full observed effect. The listed values depend on the particular outcome and comparison, so the relevant row should be read alongside its associated endpoint and arm contrast.

This is not observed power. It does not describe the probability that the completed trial’s result was correct, and it should not be used as a retrospective measure of whether the study succeeded. It is a forward-looking illustration generated from the observed effect and stated planning assumptions.

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 = -1.000; pooled standard deviation σ = 9.012.
2. n = 2 × (1.96 + 0.842)² × 9.012² / -1.000² = 1,275.0, rounded up to 1,275 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 <- -1                      # observed difference in means
sigma <- 9.01229896920205                     # 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(902538)
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 = 1,275 (80% power), 1,707 (90% power)
 75% of observed effect: n per arm = 2,267 (80% power), 3,035 (90% power)
 50% of observed effect: n per arm = 5,100 (80% power), 6,828 (90% power)
power.t.test: 1276 
Simulated power at n = 1,275 per arm: 0.804 (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. Analyses Reported by the Sponsor

The registry reports ANCOVA p-values for the primary comparisons: 0.119 for the 12.5 mg hydrochlorothiazide combination versus olmesartan and amlodipine, and < 0.001 for the 25 mg combination versus the same comparator. The corresponding adjusted estimates and confidence intervals are not reported in the registry record. The sponsor’s reported analysis is the primary inferential result for these comparisons.

The recalculated comparisons use posted arm-level least-squares means and standard errors to form simple, unadjusted estimates and approximate intervals. They give a mean difference of -1.0 mm Hg for the 12.5 mg comparison and -2.8 mm Hg for the 25 mg comparison. The associated recalculated p-values are 0.199 and <0.001. Because the registry record flags a discrepancy, the sponsor’s own analysis takes precedence; the recalculations are supplementary checks, not a correction to the sponsor’s result.

The distinction matters because the sponsor used ANCOVA, while the recalculation is not adjusted using the sponsor’s model. The model covariates and variance specification are not reported, and the recalculated standard deviations are approximate. The recalculated confidence intervals therefore should not be presented as the sponsor’s confidence intervals. The shared direction of the summary means is informative descriptively, but the inferential statements should be anchored to the registry’s reported analysis and considered alongside the missing model details and unreported multiplicity approach.

15. Multiplicity, Missing Data and Analysis Populations

The registry reports multiple primary dose comparisons against a common comparator and a set of secondary outcomes. It does not report whether the primary comparisons were adjusted for multiplicity, whether a testing hierarchy was prespecified, or how the family of secondary tests was handled. Without that information, the reported p-values cannot be interpreted as adjusted for the number of hypotheses tested. This matters because repeated testing can increase the chance of a result that appears unusual under a null model.

The registry identifies ANCOVA for the primary continuous endpoint and Cochran-Mantel-Haenszel for the blood pressure goal outcome. It does not report the full statistical analysis plan, model covariates, stratification factors, or detailed assumptions. The original sample-size justification is also not reported in the registry record. The planning illustrations based on observed effects do not fill that gap and are not evidence of the original design rationale.

The primary analysis population is described as randomized participants who received at least one dose of double-blind study medication in Period II and had at least one primary endpoint measurement in that period. That definition is informative, but details on missing data are not reported in the registry record. In particular, the record does not state how missing measurements were handled, whether participants with incomplete follow-up were excluded from particular analyses, or whether imputation or sensitivity analyses were used.

These omissions limit assessment of robustness. If missingness is related to treatment, outcome, or participant characteristics, the analyzed observations may not fully represent all randomized participants. This is a general concern, not a claim that such bias occurred in this trial. The registry also does not report interim analyses, stopping rules, or whether any adjustments were made for them. None should be inferred from the trial’s completed status or from the reported endpoint results.

Consequently, the most defensible reading distinguishes what is documented—the named methods, reported p-values, endpoint population, and treatment periods—from what is unavailable. The statistical plan, multiplicity control, and missing-data procedures are not reported in the registry record.

16. 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 3% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: ankle fracture (0.4%), atrial flbrillation (0.4%), erosive oesophagitis (0.4%), lobar pneumonia (0.4%), myocardial infarction (0.4%).

The registry reports serious adverse event counts with denominators of 2204 for the olmesartan-amlodipine group, 269 for the 12.5 mg hydrochlorothiazide combination, and 270 for the 25 mg combination. The corresponding serious adverse event counts are 18, 15, and 3. Other adverse event counts are also listed in the record, but these are not serious adverse events and should not be combined with the serious-event counts.

The denominators differ across the listed groups, and the registry record does not explain the observation periods or analysis populations underlying these safety entries. They should therefore be read as the recorded counts and denominators, not as directly comparable rates or as proof that a difference between groups exists. No rates or additional comparisons are calculated.

The registry lists serious-event terms including ankle fracture, atrial flbrillation, erosive oesophagitis, lobar pneumonia, myocardial infarction, and osteoarthritis, each with a reported value of 0.4%. The record does not provide enough detail in these entries to establish the precise context for each term’s percentage. Death counts are not reported in the registry record. These safety summaries are descriptive and do not establish causation.

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

The posted data describe changes in seated diastolic blood pressure during a randomized comparison of olmesartan and amlodipine with or without hydrochlorothiazide. The arm-level primary summaries show decreases from the week 8 baseline in all groups, with more negative mean changes in the groups receiving hydrochlorothiazide. The sponsor’s reported ANCOVA p-values differ between the two dose comparisons. Those results concern the recorded endpoint, population, and comparison period.

They do not establish what would happen to an individual participant, nor do they alone establish effects on health outcomes beyond blood pressure measurements. The registry record does not report a clinically important difference threshold for the primary endpoint, so statistical evidence should not be equated with clinical importance. The precision of the sponsor’s adjusted effect estimates cannot be assessed from the reported p-values because adjusted confidence intervals are not reported.

The treatment sequence also matters. The primary comparison followed a run-in and covers the randomized Period II comparison, whereas later periods involved further treatment and response-based decisions. Results from the initial randomized comparison should not be generalized automatically to the later selected groups or to a different duration of treatment. The analysis population was smaller than the overall enrollment and required treatment receipt and an available measurement, as specified in the registry.

The appropriate conclusion is limited to the statistical record: the posted summaries and sponsor analyses describe group-level blood pressure findings over the specified period, with different inferential results for the two hydrochlorothiazide comparisons. Interpretation is constrained by unreported analysis details, multiplicity procedures, missing-data handling, and adjusted effect intervals. These limitations do not erase the recorded results, but they define what can and cannot be concluded from them.

18. Why This Trial Matters Statistically

This trial illustrates why a clinical result needs more than a comparison of arm averages. The same primary endpoint has descriptive least-squares means, sponsor-reported ANCOVA tests, and recalculated unadjusted comparisons. Those quantities answer related but not identical statistical questions. Keeping them distinct prevents an approximate check based on summary statistics from being mistaken for the sponsor’s model-based analysis.

The two hydrochlorothiazide doses also share a comparator, making multiplicity relevant. The registry reports separate p-values but not the strategy used to interpret multiple tests. This is an important reporting issue: readers need to know whether a comparison is confirmatory, how related hypotheses are handled, and what estimates and uncertainty intervals accompany the tests.

The trial’s multiple-period design highlights another statistical distinction. The randomized Period II contrast addresses a different question from the later response-based treatment period. Population definitions, eligibility transitions, and treatment sequence all shape the estimand—the quantity a study analysis is intended to estimate. The registry gives useful design and population information, but does not report every analysis detail needed to reconstruct the estimand completely.

Finally, the planning illustrations demonstrate the dependence of sample size on the effect targeted and the assumptions used. They are not retrospective evidence about the power of the completed trial. Taken together, the record is a useful example of how to separate effect estimates, uncertainty, model-based tests, analysis populations, and planning calculations when reading a clinical trial.

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.

The registry record does not report the covariates or full specification for the primary ANCOVA, nor the adjusted effect estimates and confidence intervals. This limits assessment of the magnitude and precision of the sponsor’s adjusted comparisons beyond the reported p-values. The recalculated estimates are unadjusted and use approximate variability derived from standard errors, so they cannot substitute for those missing model-based results.

The multiplicity approach is not reported despite the primary comparisons involving two doses and the presence of multiple secondary outcomes. The record also does not report the original sample-size justification, missing-data procedures, interim analyses, or sensitivity analyses. These gaps make it difficult to evaluate the prespecified inferential framework and robustness to analytic choices.

The primary analysis population required receipt of double-blind medication and at least one Period II endpoint measurement. The record does not report how many randomized participants failed these conditions or how missing values were treated. The enrollment of 2204 therefore does not describe the population contributing to the primary outcome.

The study’s treatment periods and response-based progression add complexity. Period II was randomized, while later treatment decisions depended on response in Period 3. Findings from those later groups would apply to a different selected population, and the registry record does not include enough detail to assess all later-period estimates. For safety, the listed denominators vary, and the record does not explain their follow-up definitions. Serious adverse event counts should therefore be treated descriptively rather than as directly comparable rates.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Hydrochlorothiazide as add-on to Olmesartan/Amlodipine in Hypertension — NCT00902538.
• Registry record last updated 2019-01-11; status last verified 2012-02.
• Results first posted 2011-11-22.
• 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 describes a completed, randomized, triple-masked, parallel Phase 3 trial with 2204 enrolled participants with essential hypertension. The trial compared olmesartan and amlodipine with the same combination plus hydrochlorothiazide at 12.5 mg or 25 mg during the randomized Period II comparison.

The primary endpoint was change in seated diastolic blood pressure from the week 8 baseline to week 16. The sponsor reported ANCOVA p-values of 0.119 for the 12.5 mg comparison and < 0.001 for the 25 mg comparison. The registry does not report adjusted effect estimates or confidence intervals. Recalculated, unadjusted comparisons are supplementary and the sponsor’s analysis takes precedence where the record flags a discrepancy.

Secondary outcomes broadened measurement to seated systolic pressure, blood pressure goal achievement, and ambulatory blood pressure, among other registered questions. Interpretation is limited by unreported multiplicity procedures, model details, missing-data methods, and original sample-size assumptions. Safety entries report serious adverse event counts, but differences in listed denominators and missing context constrain direct comparison.