Tilt-Table Study of the Clinical Efficacy of Midodrine in Symptomatic Orthostatic Hypotension: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Tilt-Table Study of the Clinical Efficacy of Midodrine in Symptomatic Orthostatic Hypotension (NCT01518946), a randomized phase 4 study. The primary outcome is Time to Onset of Syncope/Near Syncope While on Tilt Table.

2. Clinical Question

3. Trial Design

This was a Phase 4, randomized, double-blind, placebo-controlled crossover study of midodrine hydrochloride in male and female subjects with symptomatic orthostatic hypotension. The listed masking level is quadruple, and the listed model is crossover. Shire is identified as the sponsor. The registry lists enrollment of 24, a start date of 2012-05-14, primary completion on 2013-06-22, and study status as completed.

The intervention groups are Midodrine HCl and Placebo. In a crossover design, participants receive more than one treatment condition, so comparisons may be made within participants. The registry record does not report the treatment sequence, period lengths, washout, or the model details used to account for repeated measurements. The interpretation of an analysis therefore depends in part on protocol and analytic details that are not reported in the registry record.

The primary outcome was time to onset of syncope or near syncope while on a tilt table, measured in seconds at 1 hour post-dose. The posted primary outcome counts are 19 for placebo and 19 for midodrine. The outcome population was the Full Analysis Set, defined as randomized subjects who received at least 1 dose and had at least 1 measurement of the endpoint during tilt-table testing. These outcome counts should not be treated as a statement that 38 distinct participants were enrolled; the study is listed as having enrollment of 24 and uses a crossover model.

The registry reports one outcome measure and one analysis. It does not report secondary outcomes. The original protocol’s sample-size justification, planned interim analyses, and detailed statistical analysis plan are not reported in the registry record.

4. Participant Flow and Accounting

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

Open-label Phase

Reasons for not completing this period:

Randomized Phase (First Intervention)

Reasons for not completing this period:

Randomized Phase (Second Intervention)

Analysis population for the primary outcome, as defined in the registry: The Full Analysis Set was defined as all randomized subjects who received at least 1 dose of randomized investigational product and who had at least 1 measurement of the time to onset of syncopal symptoms/near syncope during tilt-table testing.

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: Randomized Safety Analysis Set defined as all subjects who received at least 1 dose of randomized investigational product.

6. Primary Endpoint

The registry lists the primary outcome as Time to Onset of Syncope/Near Syncope While on Tilt Table, measured over 1 hour post-dose.

The primary endpoint is the time, in seconds, from the start of tilt-table testing to the onset of syncope or near syncope. Testing followed a 30-minute supine period. The table was tilted from 0-90º within 30 seconds and maintained in that position for 45 minutes or until an endpoint occurred. A participant’s report of sufficient dizziness, lightheadedness, faintness, or feeling about to black out and a request to return the table to horizontal ended the test. An investigator’s observation that the participant was about to lose consciousness also counted as an endpoint.

Time to an event provides information about when symptoms arose under a defined physical challenge. A longer time means that the endpoint occurred later during the test, while a shorter time means it occurred sooner. The registry reports a least-squares mean for this continuous measure. The endpoint is useful for comparing response under standardized tilt-table conditions, but it is not equivalent to counting events over ordinary daily activity or measuring long-term symptoms. The registry lists the timeframe as 1 hour post-dose; further details about the timing of measurement within treatment periods are not reported in the registry record.

7. Results Posted in the Registry

• Midodrine HCl: mean 1,627 versus 1,106 on control; difference 521.00 (95% CI 3.17 to 1,039), standardized effect d = 0.64; z-test p = 0.049.

8. Statistical Interpretation

9. Statistical Methods Explained

The outcome is continuous: elapsed time in seconds until a defined symptom or investigator-observed endpoint during tilt-table testing. The registry identifies the reported outcome parameter as a least-squares mean and its dispersion as a standard error. A mean summarizes the central value of a set of measurements. A standard error describes the uncertainty in an estimated mean, rather than the spread of individual participant measurements. The registry reports a standard error of 186.8 for each treatment condition.

What does a mean difference represent?

A mean difference compares the estimated average outcome in one treatment condition with that in the comparator condition. For this endpoint, a positive difference corresponds to a longer reported time to onset with midodrine than with placebo. The sponsor reports a mean difference in final values of 521.0 seconds. The arm summaries list least-squares means of 1106.0 seconds for placebo and 1627.0 seconds for midodrine. These treatment-condition estimates summarize the endpoint; they are not descriptions of the experience of every individual participant.

Least-squares means are model-based means. They can reflect the fitted analysis rather than only a direct arithmetic average of raw observations. The registry identifies the sponsor’s method as ANOVA, but it does not provide the full model specification, the terms included, or the way the crossover structure was handled. Those details matter when assessing exactly what the adjusted estimate represents.

What does ANOVA contribute?

Analysis of variance, or ANOVA, is a modeling approach for comparing mean outcomes across conditions. In a crossover study, a suitable analysis may need to account for the fact that the same participant can contribute measurements under more than one treatment condition. The registry reports ANOVA and a mean difference in final values, but the analysis description is not detailed enough to establish which within-participant or period effects were included. It is therefore not possible to reproduce or fully evaluate the sponsor’s model from the registry record alone.

The posted recalculation is different in scope. It is a simple unadjusted calculation from the treatment-condition summary statistics and treats the mean estimates and their standard errors as the available inputs. Its confidence interval and p-value are not a replacement for the sponsor’s model-based analysis. In particular, the summary does not include the participant-level paired information needed to independently reconstruct a crossover analysis.

Why do standard errors and confidence intervals matter?

A point estimate gives one estimate of the treatment contrast, but it does not show its precision. A confidence interval describes the uncertainty around the estimate under a statistical model and its assumptions. The sponsor reports a 95% confidence interval from 124.2 to 917.7 seconds. The recalculation reports a 95% confidence interval from 3.171 to 1039.0 seconds. The difference between these intervals reflects the different analytic inputs and approaches; it is not a contradiction that can be settled from the arm means alone.

A standard error is not the same as a standard deviation. The standard deviation describes variation among observations, while the standard error concerns the precision of an estimated quantity. For the recalculation, a pooled standard deviation of 814.3 is listed as approximate. That approximate value is not a directly reported measure of individual-level variability in the crossover analysis.

What does a p-value say?

A p-value measures how unusual the observed result, or a more extreme result, would be under a specified null hypothesis and analysis model. It does not give the probability that the null hypothesis is true, and it does not describe the chance that an individual will experience a particular outcome. The sponsor reports p = 0.013. The simplified recalculation reports an unadjusted p-value of 0.049. These values arise from different analyses and should be read alongside their corresponding estimates and confidence intervals.

General statistical interpretation also depends on the planned analysis and multiplicity strategy. If several outcomes or comparisons are tested, the chance of obtaining at least one small p-value can change. Only one outcome measure and one analysis are posted, and no multiplicity procedure is reported. Whether additional analyses were planned or conducted is not reported in the registry record.

Why can over-dispersion matter?

General explanation: over-dispersion occurs when observed variability is greater than a statistical model expects. When the model understates variability, standard errors can be too small and confidence intervals too narrow. This concern is especially relevant when an outcome’s distribution does not fit the model assumptions. The registry does not report a specific over-dispersion assessment for this continuous endpoint, so it cannot be concluded that over-dispersion occurred. It is a general consideration, not a reported finding from this trial.

The endpoint is time to a defined event, but the posted analysis treats it as a continuous measure and reports least-squares means. The test ended when the endpoint occurred or when the tilt period ended. The registry does not report how observations without an endpoint by the end of testing were handled in the analysis. That is a relevant methodological detail because a maximum test duration can affect the interpretation of time measurements.

Finally, the Full Analysis Set definition requires at least one endpoint measurement after randomization and treatment. That definition makes the analyzed population explicit, but it is not the same as an intention-to-treat analysis of every randomized participant regardless of measurement availability. The registry does not report a missing-data method, sensitivity analysis, or participant flow detailed enough to assess the effect of missing measurements on the result.

10. Confidence Intervals and Precision

The sponsor’s estimated mean difference is 521.0 seconds, with a 95% confidence interval from 124.2 to 917.7 seconds. The interval is entirely above zero, indicating that under the sponsor’s reported model the data are compatible with a positive difference in time to onset. The interval also spans a substantial range of possible effect sizes, so the point estimate should not be treated as exact. A confidence interval is not a prediction interval for individual participants and does not mean that a stated proportion of participants experienced an effect within its limits.

The simplified recalculation gives the same point estimate, 521.0 seconds, but a wider 95% confidence interval from 3.171 to 1039.0 seconds. Its unadjusted p-value is 0.049, compared with the sponsor-reported p = 0.013. The contrast in precision reflects the fact that these calculations do not use the same statistical model. The recalculation is based on posted summary statistics and does not incorporate unreported participant-level pairing or model terms. It should not be treated as an alternative definitive analysis of the crossover data.

The registry reports a standard error of 186.8 for each treatment-condition mean. Standard errors describe precision of estimated means, not the full range of participant times. The posted outcome count is 19 in each condition. A modest number of outcome observations can leave an estimate sensitive to individual observations and modeling choices, but the record does not provide raw data to assess that sensitivity. The sponsor’s interval is the relevant precision statement for its reported ANOVA result; the simplified interval is useful only as a transparent summary-based recalculation.

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 planning figures are participants per arm, not total events. They are illustrations based on the observed mean difference and are not the original trial’s sample-size plan. The planning 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 to detect equal to a stated fraction of the observed effect. The figures do not account for details of the crossover structure, because the planning calculation is framed as equal allocation between two arms.

For the row planning for 100% of the observed effect, the illustration gives 39 participants per arm for 80% power and 52 participants per arm for 90% power. For planning for 75% of the observed effect, the corresponding figures are 69 and 92 participants per arm. For planning for 50% of the observed effect, they are 154 and 206 participants per arm.

In plain language, the 80% power column is the number of participants per arm indicated by the stated assumptions for an 80% chance of detecting the effect used in that row, if that effect and model assumptions were correct. The 90% power column gives the corresponding planning count for a higher target chance of detection. Smaller effects require more participants under these assumptions, which is why the counts rise as the target fraction of the observed effect gets smaller.

These figures are planning illustrations, not observed power. The trial’s original sample-size assumptions and justification are not reported in the registry record. The illustrations should therefore not be read as evidence that the completed study had a particular probability of detecting its result, nor as a reconstruction of the sponsor’s design choices.

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 = 521.000; pooled standard deviation σ = 814.330.
2. n = 2 × (1.96 + 0.842)² × 814.330² / 521.000² = 38.3, rounded up to 39 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 <- 521                      # observed difference in means
sigma <- 814.329500632269                     # 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(1518946)
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 = 39 (80% power), 52 (90% power)
 75% of observed effect: n per arm = 69 (80% power), 92 (90% power)
 50% of observed effect: n per arm = 154 (80% power), 206 (90% power)
power.t.test: 39.3 
Simulated power at n = 39 per arm: 0.796 (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 sponsor’s posted analysis is an ANOVA of final values, reporting a mean difference of 521.0 seconds, a 95% confidence interval from 124.2 to 917.7 seconds, and p = 0.013. The summary-based recalculation also gives a point estimate of 521.0 seconds, but its 95% confidence interval is from 3.171 to 1039.0 seconds and its unadjusted p-value is 0.049.

The recalculation is a simplified, unadjusted calculation from posted treatment-condition summaries. It does not reproduce the sponsor’s ANOVA, and the registry does not provide the participant-level data or full model specification needed to do so. Since the study uses a crossover model, the treatment observations may be related within participants; the posted arm summaries do not show that relationship. The two sets of inferential results should be reported with their methods distinguished. The sponsor analysis is the trial’s reported analysis, while the recalculation is a limited check based on aggregate information rather than a replacement analysis.

14. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when several hypotheses, outcomes, time points, or treatment comparisons are examined. If many tests are performed, the chance of obtaining at least one result that appears statistically notable can increase. Statistical plans may address this by defining a primary comparison in advance or by using an adjustment procedure. The registry lists one outcome measure and one analysis, and no multiplicity adjustment is reported. Whether other analyses were planned or performed, and whether any adjustment was specified in the protocol, are not reported in the registry record.

The trial’s primary outcome analysis population is defined as the Full Analysis Set: all randomized subjects who received at least 1 dose of randomized investigational product and had at least 1 measurement of time to onset of syncopal symptoms or near syncope during tilt-table testing. This definition establishes eligibility for the posted analysis, but does not include randomized subjects with no qualifying endpoint measurement. The registry does not report a missing-data method, a sensitivity analysis for missing values, or a detailed account of the number and reasons for missing outcome measurements.

That information is particularly relevant in a crossover study, where participants may contribute data in more than one treatment condition and incomplete periods can affect the comparison. The registry record does not report how incomplete treatment periods, sequence, period effects, or carryover were handled. It also does not give the detailed statistical analysis plan or participant-level data needed to assess those issues independently.

The sponsor reports an ANOVA result for final values, but the registry record does not specify the full model or any adjustment for multiple testing. The posted result can be described as the reported primary comparison, but the available information is insufficient to establish whether the analysis followed a prespecified hierarchy or how any unreported exploratory analyses were treated. These omissions do not establish that the analysis was inappropriate; they limit independent evaluation of its completeness and robustness.

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 5% reporting threshold.

Most frequent non-serious adverse event terms

The registry’s safety entries distinguish serious adverse events from other adverse events and list three groups. In the Midodrine HCl open-label phase, 0 of 24 participants were affected by a serious adverse event and 3 of 24 were affected by another adverse event. In the randomized placebo phase, 0 of 20 were affected by a serious adverse event and 1 of 20 by another adverse event. In the randomized Midodrine HCl phase, 0 of 20 were affected by a serious adverse event and 1 of 20 by another adverse event.

These are registry counts for the listed groups; they should not be combined to create a unique-participant total, particularly because the design includes crossover treatment conditions and an open-label phase. The record does not report the event terms in the serious-event list. The death counts are not reported in the registry record. The safety entries alone do not establish comparative risk or explain whether events were judged related to treatment. They should be read as descriptive counts with their listed denominators, not as a complete account of safety.

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

The posted efficacy result concerns the timing of syncope or near syncope during a specified tilt-table challenge. The sponsor’s reported mean difference is 521.0 seconds, with the midodrine condition having the larger least-squares mean time. Its 95% confidence interval is from 124.2 to 917.7 seconds, and the reported p-value is 0.013. These results describe a treatment contrast for this endpoint and under the sponsor’s reported analysis.

The result does not establish how long any effect would persist, whether the same contrast would occur outside a tilt-table test, or how participants’ day-to-day symptoms would change. The test has controlled positioning, timing, and stopping criteria. A delayed endpoint during that challenge is not by itself a complete measure of clinical benefit. No secondary efficacy outcomes are posted, so the registry does not provide other outcome measures to broaden the assessment.

The study’s crossover design can support within-participant comparisons, but the registry record does not report the sequence, period details, or full model specification. The arm counts and summary means alone cannot show how individual participants responded across conditions. The reported outcome population also requires at least one qualifying measurement, and missing-data handling is not reported. These factors limit independent assessment of the analysis.

The simplified recalculation produces a different confidence interval and p-value from the sponsor’s ANOVA, despite the same point estimate. That difference illustrates why a model-based result should not be reconstructed from summary means without the data and details used in the original analysis. Neither the p-value nor the confidence interval determines whether a change is important to participants; the registry does not report a threshold for clinical importance.

Accordingly, the registry supports a description of the sponsor-reported endpoint contrast, not a broad claim about treatment effects beyond the tested conditions. The small posted outcome counts and incomplete reporting of design and analysis details also call for restraint when generalizing the estimate.

17. Why This Trial Matters Statistically

This trial raises useful statistical questions because it combines a continuous time endpoint with a crossover design. The reported comparison is expressed as a mean difference in seconds, which is interpretable in the context of the tilt-table test. At the same time, a crossover comparison can depend on within-participant relationships, treatment sequence, and period effects. These features are not recoverable from treatment-condition means and standard errors alone.

The result also shows why estimates should be read together with their uncertainty and analysis method. The sponsor reports a mean difference of 521.0 seconds with a 95% confidence interval from 124.2 to 917.7 seconds. A summary-based recalculation retains the same estimate but produces a different interval and p-value. The distinction makes clear that aggregate summaries can support a basic check but cannot substitute for the original model and participant-level data.

Endpoint context is also important. Time to a defined symptom during a standardized tilt-table procedure is more specific than a broad description such as symptom improvement. Readers need to distinguish a response during this test from outcomes in other settings. The registry’s Full Analysis Set definition and its limited description of missing-data handling are relevant to understanding which measurements contribute to the reported result.

Finally, the record illustrates the difference between sample-size planning and observed trial evidence. Planning counts based on a fraction of the observed effect are useful for understanding how assumed effect size affects required enrollment, but they are not the original design justification or evidence of the study’s power. The original sample-size assumptions are not reported in the registry record. Overall, careful attention to endpoint, model, population, and uncertainty is essential when reading a concise registry result.

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 registry lists enrollment of 24, while the primary outcome summaries list 19 observations for each treatment condition. Because the study is described as a crossover, these condition counts should not be interpreted as independent groups or summed into a count of distinct participants. The record does not give enough participant-flow detail to explain the relation between enrollment and the outcome counts.

The treatment sequence, treatment-period lengths, washout, and handling of period or carryover effects are not reported in the registry record. Nor does the record provide the full ANOVA specification. These omissions prevent independent reconstruction of the sponsor’s crossover analysis. The summary-based recalculation cannot resolve those gaps because it does not include participant-level paired measurements.

The endpoint is specific to tilt-table testing, and the registry does not report a threshold for a clinically important difference. The test’s defined duration and symptom-based stopping criteria also mean that interpretation depends on how participants reaching the end without an endpoint were handled; that handling is not reported. Missing-data methods and sensitivity analyses are likewise not reported.

The registry posts one outcome measure and no secondary outcomes. It does not report the original sample-size justification, interim analyses, or a detailed statistical analysis plan. Safety entries are descriptive, with no serious adverse events listed in the groups, while death counts are not reported in the registry record. These limitations constrain conclusions about precision, robustness, broader clinical relevance, and safety comparison.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Tilt-Table Study of the Clinical Efficacy of Midodrine in Symptomatic Orthostatic Hypotension — NCT01518946.
• Registry record last updated 2021-06-08; status last verified 2021-05.
• Results first posted 2014-07-31.
• 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

This completed Phase 4 randomized, double-blind, placebo-controlled crossover study examined Midodrine HCl for symptomatic orthostatic hypotension. The primary endpoint was time to onset of syncope or near syncope during tilt-table testing, measured in seconds at 1 hour post-dose. The Full Analysis Set was defined by receipt of at least 1 dose and at least 1 endpoint measurement.

The sponsor-reported ANOVA gives a mean difference of 521.0 seconds, a 95% confidence interval from 124.2 to 917.7 seconds, and p = 0.013. A simplified summary-based recalculation has the same point estimate but a different confidence interval and unadjusted p-value. The registry does not report the full crossover model, missing-data methods, or the original sample-size justification, so the sponsor result should be distinguished from the limited recalculation and interpreted in the context of the controlled test setting.