Efficacy of Haloperidol vs. Metoclopramide for Treatment of Acute Headaches and Migraines in the Emergency Department: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Efficacy of Haloperidol vs. Metoclopramide for Treatment of Acute Headaches and Migraines in the Emergency Department (NCT02972502), a randomized phase 4 study. The primary outcome is Change in Pain Score According to the Numeric Pain Intensity Scale.

2. Clinical Question

The clinical question is a comparison of short-term pain-score change after treatment in the emergency department. The registry also lists two secondary outcomes: the need for additional medications in the emergency department and return to the emergency department or another healthcare provider for headache or migraine within 48 hours of discharge. Those outcomes are listed, but results for them are not reported in the registry record.

The registry’s stated aims include exploring another treatment option and the possibility of improving efficiency, including reducing emergency-department length of stay and return visits. Those aims should not be confused with the posted primary endpoint: the reported primary result concerns pain-score change at one hour, not length of stay or return visits. The available result therefore addresses a limited, short-term measurement rather than every aim described in the trial summary.

3. Trial Design

The registry describes a prospective randomized clinical trial with a parallel-group model and single masking. It identifies the study as phase 4 and lists OhioHealth as the sponsor. The trial compared haloperidol with metoclopramide for acute headaches and migraines in the emergency department. The registry lists normal saline and diphenhydramine as additional interventions and describes both treatment groups as receiving those components before the assigned study medication.

The registry reports enrollment of 66 participants. The study status is terminated, and the recorded primary completion date is 2017-04-24. The primary outcome population description says the study was terminated early, that information was entered for n=49 participants, and that data were not collected for 11 participants. The registry does not explain the relationship between the enrollment figure and the primary outcome information, or the disposition of all enrolled participants.

The posted primary outcome is continuous: change in pain score from baseline, prior to treatment, to one hour post treatment. The group summaries are means with standard deviations. The registry lists three outcome measures in total and two secondary outcomes, but it reports no analyses for the secondary outcomes. The registry record does not report the planned sample-size justification, the original design assumptions, or additional details about how randomization and masking were implemented. Those details should not be inferred from the general design labels.

The record gives no basis for describing the trial as a complete or fully analyzed study. Its early termination and incomplete outcome information are important context for interpreting the posted group results. The available evidence is limited to the reported primary endpoint summaries and comparison, plus the listed safety counts. The registry does not report reasons for termination or a detailed account of the analysis population.

4. Participant Flow and Accounting

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

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Study was terminated early. Information entered for n=49 participants. Data not collected for 11 participants.

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 Pain Score According to the Numeric Pain Intensity Scale, measured over Change from baseline (prior to treatment) to 1 hour post treatment (1 hour).

The primary endpoint is change in Numeric Pain Intensity Scale score from before treatment to one hour after treatment. The registry describes the scale as ranging from 0, no pain, to 10, the worst pain imaginable. Reporting change rather than only the final score is intended to express how the measured score differs from each participant’s baseline. For this endpoint, a negative change represents a lower score after treatment than before treatment.

The outcome is summarized as a mean for each treatment group, with standard deviation describing variability among participants. A group mean gives an average change, not the experience of every participant. The standard deviation indicates how spread out individual changes are around that average; it is not a confidence interval and does not, on its own, show how precisely the group mean is estimated.

A one-hour measurement provides a specified short-term time point for comparing the groups. It does not describe longer-term symptom changes. The registry also lists additional medication use and return for care within 48 hours as secondary outcomes, but results for those outcomes are not reported in the registry record. The primary endpoint should therefore be interpreted as a comparison of mean pain-score change at one hour, not as a measure of every dimension of headache care or the study’s broader stated aims.

7. Results Posted in the Registry

• Haloperidol (Haldol): mean -5.13 versus -2.86 on control; difference -2.27 (95% CI -3.67 to -0.87), standardized effect d = -0.92; z-test p = 0.001.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary endpoint is a continuous measure, and the registry reports each group’s mean and standard deviation. In general, a mean summarizes the average observed change, while the standard deviation describes variation among individual observations. These answer different questions: the mean concerns the group average, and the standard deviation concerns how dispersed the participant-level values are. Neither statistic alone gives the uncertainty around the between-group comparison.

What does the mean difference represent?

The reported mean difference compares the average change in the haloperidol group with the average change in the metoclopramide group. The estimate is -2.27 units, using the comparison’s stated direction. Because the endpoint is change from baseline, a negative value means a lower average change in the haloperidol group relative to the comparator group. This is an unadjusted comparison in the posted results; an adjusted analysis is not reported in the registry record.

General statistical interpretation: a between-group mean difference preserves the units of the original outcome, which makes its scale clinically interpretable alongside the pain-score measure. It remains a group-level estimate. It does not imply that every participant had a change equal to the estimated difference, or that individual responses were uniform.

What does the confidence interval add?

The comparison has a 95% confidence interval from -3.667 to -0.874. In general, a confidence interval represents uncertainty in an estimated effect under the analysis model and its assumptions. It gives a range of estimates compatible with the observed data under that framework. It is not a statement that 95% of participants’ outcomes fall within those limits, and it does not include every source of uncertainty, such as uncertainty caused by missing data or limitations in the study design.

How should the p-value be understood?

The registry reports p = 0.001 for the comparison. In general, a p-value describes how unusual data at least as inconsistent with a specified null hypothesis would be if that hypothesis and the analysis assumptions were true. It is not the probability that a treatment is effective, and it does not measure the size or practical importance of an effect. The estimate and confidence interval are needed to understand direction, scale, and precision alongside the p-value.

Why do incomplete observations matter?

The primary outcome description states that information was entered for n=49 participants and data were not collected for 11 participants. General statistical principle: when outcome data are missing, the consequences depend in part on why they are missing and whether missingness is related to treatment assignment or outcome values. The registry does not report the reasons for missing data or an analysis addressing their potential impact. It is therefore not possible to determine from the record whether the missing outcome information could have changed the comparison.

The group summaries describe 26 participants in the metoclopramide arm and 23 in the haloperidol arm. The registry does not report whether the analysis followed an intention-to-treat approach, a per-protocol approach, or another prespecified population definition. The exact analysis-population rules are not reported in the registry record. This limits assessment of how participants were handled after randomization and how the posted comparison relates to the original randomized groups.

What can and cannot be inferred from the design labels?

The record labels the study randomized, parallel, and single-masked. In general, random allocation is intended to reduce systematic differences between groups, while masking can reduce some forms of bias in care or outcome assessment. The registry does not report details of the randomization procedure, who was masked, or how masking was maintained. Those implementation details cannot be inferred from the labels alone.

The trial was terminated early. Early termination can reduce the amount of information available for estimating an effect, but the registry does not report why the study ended early or whether a formal stopping rule was used. Interim analyses, stopping criteria, and the trial’s original statistical plan are not reported in the registry record. No particular rationale or analytic procedure should be assumed.

General statistical principle: when outcome variability exceeds what a simple model anticipates, over-dispersion can make uncertainty estimates too small if the model does not account for it. This trial’s primary outcome is continuous and is reported with standard deviations; the registry does not identify an over-dispersion issue or a specialized model. The available information does not justify claiming that over-dispersion was present or that a particular correction was used.

The registry lists two secondary outcomes and reports no analyses for them. It also does not report a multiplicity adjustment or a hierarchy of hypotheses. General statistical principle: testing several outcomes can increase the chance of at least one apparently notable result if each is evaluated separately. The record does not provide enough information to determine how multiplicity was considered in the trial’s original analysis plan.

The posted comparison is not labeled as recalculated, and the record does not flag a discrepancy between the reported comparison and the group summaries. The registry’s own reported analysis is therefore the relevant comparison to describe. A separate statistical plan, model specification, or additional diagnostic results are not reported in the registry record.

10. Confidence Intervals and Precision

The reported mean difference is -2.27 units, with a 95% confidence interval from -3.667 to -0.874. This interval is expressed on the same scale as the change in pain score and describes uncertainty around the estimated difference between group means. Its negative limits align with the direction of the reported estimate, with the comparison defined as haloperidol relative to metoclopramide.

In general, a confidence interval is more informative than a point estimate alone because it conveys a range of estimates compatible with the data under the analysis assumptions. A narrower interval indicates greater statistical precision than a wider interval on the same scale. Precision is not the same as validity: an interval can be calculated precisely under a model while remaining affected by missing data, early termination, or limitations in how the study was conducted.

The group standard deviations are 2.37 for metoclopramide and 2.59 for haloperidol. These describe variability in participant-level changes, not uncertainty in the group means or in the mean difference. The comparison’s confidence interval is the relevant posted measure of uncertainty for the between-group estimate. The p-value is p = 0.001, but a small p-value does not by itself establish high precision or resolve limitations in the data.

The registry reports outcome information for n=49 participants and says data were not collected for 11 participants. It does not report whether missingness affected the estimate or whether sensitivity analyses were conducted. Thus, the confidence interval captures statistical uncertainty under the reported comparison but cannot be taken as accounting for every uncertainty about the study. The registry record does not report additional precision analyses or adjusted estimates.

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 entries are planning illustrations based on the observed effect, not the trial’s original design calculation. They give participants per arm, not a combined total. The assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set to the stated fraction of the observed effect. The trial’s own original sample-size assumptions are not reported in the registry record.

The 80% power column gives the illustrative number of participants per arm for an analysis designed to have an 80% chance of detecting the specified effect under the listed assumptions, if that effect is real and the assumptions hold. The 90% power column uses a higher target chance of detection, also under those assumptions. These are prospective planning quantities; this is not observed power.

The entries distinguish planning for 100%, 75%, and 50% of the observed effect. For the full observed effect, the illustration gives 19 participants per arm for 80% power and 25 for 90% power. Planning for 75% of the observed effect gives 34 participants per arm for 80% power and 45 for 90% power. Planning for 50% of the observed effect gives 75 participants per arm for 80% power and 100 for 90% power.

As a general principle, smaller effects are harder to distinguish from random variation, so detecting them with the same target power generally requires more participants. These figures should be read as conditional planning scenarios based on the observed estimate, not as evidence that the trial was designed with those targets or that its original assumptions matched them. The registry does not report its original sample-size justification.

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 = -2.270; pooled standard deviation σ = 2.475.
2. n = 2 × (1.96 + 0.842)² × 2.475² / -2.270² = 18.7, rounded up to 19 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 <- -2.27                      # observed difference in means
sigma <- 2.47541399933124                     # 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(2972502)
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 = 19 (80% power), 25 (90% power)
 75% of observed effect: n per arm = 34 (80% power), 45 (90% power)
 50% of observed effect: n per arm = 75 (80% power), 100 (90% power)
power.t.test: 19.7 
Simulated power at n = 19 per arm: 0.780 (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. Multiplicity, Missing Data and Analysis Populations

The registry lists one primary endpoint, change in pain score from baseline to one hour after treatment, and two secondary outcomes: need for additional medication in the emergency department and return for headache or migraine care within 48 hours of discharge. It reports no analyses for the secondary outcomes. Their results are not reported in the registry record, so no conclusions about those outcomes can be drawn from the posted primary comparison.

The record does not report a multiplicity adjustment, a hierarchy among outcomes, or a detailed statistical analysis plan. In general, multiplicity refers to the added chance of obtaining at least one statistically notable finding when several hypotheses are tested. Whether and how that issue was handled in the trial’s intended analyses is not reported in the registry record. The posted p-value should be understood as the reported comparison for the primary endpoint, not as evidence about unreported outcomes or a global test of every study question.

Missing primary outcome information is also important. The registry states that the study terminated early, that information was entered for n=49 participants, and that data were not collected for 11 participants. Reasons for missing data, the timing of missingness, and its distribution by treatment group are not reported in the registry record. The record does not describe imputation, sensitivity analyses, or other procedures for assessing the effect of missing observations.

Although the study is labeled randomized, the record does not specify whether the primary analysis included all randomized participants or followed another analysis population. It does not describe how withdrawals, deviations, or incomplete records were handled. The posted group sizes are 26 and 23, but the registry does not provide enough detail to establish the precise population rules behind the primary comparison. These gaps limit assessment of the robustness of the result.

14. Safety Results

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

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

For metoclopramide, the registry records 0 serious adverse events among 26 participants at risk, 0 other adverse events among 26 participants at risk, and 0 deaths. For haloperidol, it records 0 serious adverse events among 23 participants at risk, 0 other adverse events among 23 participants at risk, and 0 deaths. No serious adverse-event terms are listed.

These counts describe what is recorded in the registry for the stated groups and categories. Zero recorded events does not establish that an event is impossible or that no event occurred outside the registry’s captured reporting. The registry does not provide further detail about adverse-event definitions, collection procedures, duration of follow-up for safety, or whether all participants contributed safety information.

The counts should not be treated as a definitive comparative safety analysis. The study was terminated early, and its primary outcome information was incomplete; the registry does not explain whether the same completeness applies to safety reporting. The record also does not report a statistical comparison of adverse events. The figures are therefore best read as the posted event counts, not as proof of equivalent safety between groups.

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

The posted result describes a difference in average change in pain score at one hour between two treatment groups. Both group means were negative, and the haloperidol group’s mean change was more negative than the metoclopramide group’s. The registry comparison estimates a mean difference of -2.27, with a 95% confidence interval from -3.667 to -0.874 and p = 0.001. These are findings about the reported endpoint and comparison, in the participants represented by the posted analysis.

The result does not establish that every participant experienced a similar change. Group means and a mean difference summarize averages, while individual outcomes can vary; the reported standard deviations describe that variability. Nor does the result establish what would happen in a different clinical setting or population. The record does not report enough information to assess the effect of the missing outcome data, the reasons for early termination, or all features of the analysis population.

The primary endpoint is measured one hour after treatment. It does not answer whether any group difference persists beyond that measurement time. The registry lists additional medication use and return for care within 48 hours as secondary outcomes, but their results are not reported in the registry record. The posted primary comparison also does not measure emergency-department length of stay, despite that being mentioned among the study’s broader aims.

The recorded absence of adverse events is limited to the counts and categories in the registry. It does not prove that either regimen has no risks or that their safety profiles are equivalent. The record does not report a statistical safety comparison or provide additional safety follow-up details.

Overall, the posted data support a description of the observed one-hour mean pain-score changes and the registry’s estimated between-group comparison. They do not resolve the unreported outcomes, missing-data questions, longer-term effects, or the study’s broader efficiency aims. The registry record does not report enough information to extend the result beyond those boundaries.

16. Why This Trial Matters Statistically

This trial illustrates why a treatment comparison should be read together with its endpoint, analysis population, and data completeness. The primary outcome is a continuous change score, so interpretation depends on both the direction of change and the between-group contrast. Negative values represent lower scores after treatment, while the mean difference expresses the contrast between group averages rather than individual responses.

It also demonstrates the distinction between statistical evidence and the scope of a conclusion. The registry reports a p-value and confidence interval for the one-hour endpoint, but those statistics do not answer every clinical or operational question raised by the trial’s aims. In particular, listed secondary outcomes have no reported analyses, and the primary endpoint is not a measure of length of stay.

Early termination and incomplete outcome information make transparency about analysis populations and missingness especially important. The registry states that data were not collected for 11 participants, but does not explain why, how missing observations were handled, or whether sensitivity analyses were conducted. Without those details, readers cannot fully assess how robust the estimate is to incomplete follow-up.

The planning illustrations offer a further statistical lesson: sample-size calculations depend on assumptions about the effect to detect, allocation, significance level, and model. The displayed scenarios are based on the observed effect and do not document the study’s original design process. They should not be mistaken for evidence about the trial’s prespecified target or its achieved power.

Finally, a registry result is a bounded record of specified outcomes and counts. Distinguishing reported results from unreported questions is necessary for clear interpretation, particularly when the study ended early and the available outcome information is incomplete.

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 study was terminated early, but the reason for termination is not reported in the registry record. The record does not state whether a prespecified stopping rule or interim analysis was used. Without those details, the circumstances that led to termination and their possible relationship to the observed estimate cannot be evaluated.

The primary outcome description reports information for n=49 participants and says data were not collected for 11 participants. It does not explain the reasons for the missing data, their distribution by group, or how missing observations were handled. The registry also reports enrollment of 66 participants but does not reconcile that figure with the primary outcome information or describe the disposition of all enrolled participants.

The record does not specify the original sample-size justification, the primary analysis population, or a detailed statistical analysis plan. It does not report whether the comparison was adjusted for any variables, or provide sensitivity analyses. The registry’s design labels identify randomization and single masking but do not describe implementation, allocation procedures, or who was masked.

The registry lists two secondary outcomes but reports no analyses for them. It does not report the planned approach to multiplicity, outcome hierarchy, or missing secondary-outcome data. The safety results are limited to the recorded counts; definitions, collection methods, and further follow-up details are not reported.

These limitations constrain the interpretation of the one-hour comparison and prevent broader conclusions about return visits, additional medication use, emergency-department length of stay, longer-term outcomes, or comparative safety.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Efficacy of Haloperidol vs. Metoclopramide for Treatment of Acute Headaches and Migraines in the Emergency Department — NCT02972502.
• Registry record last updated 2018-10-23; status last verified 2018-09.
• Results first posted 2018-10-23.
• 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 phase 4, randomized, parallel-group, single-masked trial of haloperidol versus metoclopramide for acute headaches and migraines in the emergency department. Enrollment is listed as 66, and the study status is terminated. The recorded primary completion date is 2017-04-24.

The primary endpoint is mean change in pain score from baseline to one hour after treatment. The posted group summaries include 26 participants in the metoclopramide group and 23 in the haloperidol group. The reported mean difference is -2.27, with a 95% confidence interval from -3.667 to -0.874 and p = 0.001. The primary outcome description notes early termination and incomplete data collection.

Two secondary outcomes are listed, but their results are not reported in the registry record. The safety counts record no serious adverse events, other adverse events, or deaths in either group. The record does not report the reasons for early termination, the original sample-size justification, or details of missing-data handling and analysis populations.