Pilot Study to Compare Frovatriptan vs. Topiramate for Prevention of Migraine: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Pilot Study to Compare Frovatriptan vs. Topiramate for Prevention of Migraine (NCT00846495), a randomized phase 4 study. The primary outcome is Number of Migraine Attacks in Participants Using Frovatriptan in a Preemptive Treatment Paradigm vs. Daily Topiramate.

2. Clinical Question

The clinical question combines a comparison of medicines with a comparison of schedules: daily preventive treatment versus treatment during the pre-headache phase. The registry describes random assignment at Visit 2 and a 2-month treatment period before the exit visit. Participants first completed a baseline period in which they used their usual medication and recorded prodrome. The primary endpoint is limited to the attack count in the second treatment month, rather than a summary of all treatment months or a long-term outcome.

The study also listed secondary questions about monthly headache days, participants with greater than 50% reductions in attacks and headache days, quality of life, satisfaction, adverse events, and costs. Those topics broaden the planned comparison, but the primary result is specifically the count of migraine attacks in Treatment Month 2. The registry record does not report a primary result for those secondary questions in the registry record.

3. Trial Design

This was a randomized, single-blind, parallel-group Phase 4 pilot study. The registry record identifies Clinvest as sponsor and reports enrollment of 55. The study is marked completed, with a start date of 2009-08 and primary completion in 2011-01. The record describes recruitment at 2 investigative sites and a sequence of 3 visits.

At Visit 1, participants began a 1-month baseline period. During baseline they continued their usual medication and documented warning signs, described as prodrome before headache. At Visit 2, participants were randomized to a treatment group. The registry explains randomization in plain language as being like a flip of the coin. One group took topiramate daily; the other took frovatriptan during prodrome when the participant believed a disabling migraine would occur, before headache began. After a 2-month treatment period, participants exited at Visit 3.

The topiramate arm description states that the dose was titrated over 4 weeks to a maximum of 100 mg daily, with one dosage adjustment permitted and a minimum dose of 50 mg daily. The frovatriptan description specifies 5 mg for treatment during prodrome. These arm details describe the regimens recorded in the registry; they do not establish how consistently participants followed them.

Single-blind indicates that the study was masked, but the registry record does not say which people were blinded or how masking was maintained. Further details about allocation procedures, concealment, adherence, withdrawals, and the pilot study's original sample-size rationale are not reported in the registry record. The outcome-specific group sizes for the primary comparison are 20 for topiramate and 24 for frovatriptan. These are the sizes attached to the posted primary outcome, not a substitute for a participant-flow account.

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: Number of Units Analyzed is equivalent to number of migraine attacks reported during 2nd Treatment Month.

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 Number of Migraine Attacks in Participants Using Frovatriptan in a Preemptive Treatment Paradigm vs. Daily Topiramate, measured over Treatment Month 2.

The primary endpoint is the number of migraine attacks reported during Treatment Month 2. The registry labels it a mean, with standard deviation as the measure of spread, and gives the unit as migraine attacks. Its description frames the measure as a comparison between participants using frovatriptan preemptively during prodrome and participants taking topiramate daily. The registry's population note says the analyzed units correspond to migraine attacks reported during the second treatment month.

An attack count is used to represent the frequency of migraine events over a defined treatment period. Comparing group means provides a concise summary of the average reported count in each arm. It does not show how counts were distributed among individual participants: people in the same arm may have had different experiences, and the standard deviations give an indication of that spread.

The endpoint is specific to the second treatment month. It should not be read as a total across the full treatment period, as a measure of headache days, or as a direct measure of severity, function, or quality of life. The registry lists those latter topics among secondary outcomes, but they are distinct from the primary count. Nor does a group average describe what happened to every participant. The reported mean difference and its interval therefore address an average count contrast, not a guaranteed individual response or a broader judgment about either treatment paradigm.

7. Results Posted in the Registry

• Frovatriptan: mean 2.12 versus 1.35 on control; difference 0.77 (95% CI -0.14 to 1.68), standardized effect d = 0.49; z-test p = 0.096.

8. Statistical Interpretation

9. Statistical Methods Explained

The results page presents a comparison of a continuous outcome summarized by its mean and standard deviation. The endpoint is the number of migraine attacks reported during Treatment Month 2. Although attack counts are whole-number observations for individual participants, the registry characterizes the primary measure as a mean, so the posted group summaries and comparison are expressed as averages.

What does the mean difference describe?

The reported mean difference is 0.77 attacks, calculated in the direction frovatriptan minus topiramate. A positive value in this stated direction means the sample mean is higher for frovatriptan. The group means behind that estimate are 2.12 for frovatriptan and 1.35 for topiramate. The difference is a contrast between group averages; it is not the change experienced by a typical participant and should not be interpreted as an individual-level treatment effect.

The calculation is unadjusted: it compares the posted summary statistics without accounting for other participant characteristics or variables. The registry record does not report an adjusted model. It also does not give further details about how any participant-level repeated measurements, if collected, were handled in an analysis. The result should therefore be understood in the limited terms supported by the record: a difference between the reported group means for the specified month.

Why report standard deviations?

A mean alone conceals how much participants' outcomes vary. The standard deviation summarizes the spread of observations around the group mean. Here, the reported standard deviations are 1.31 for topiramate and 1.75 for frovatriptan. Larger spread means participants' counts were more dispersed around their respective group average. The standard deviation is not the uncertainty in the mean difference; that uncertainty also depends on the number of analyzed participants and the variation in their outcomes.

General statistical context: when observations are more variable, a group mean is generally estimated less precisely if the group size is unchanged. This is one reason a result with a visible difference in sample means can still have a broad interval. The reported outcome-specific group sizes are 20 and 24, so the estimates are based on a modest number of observations. The registry record does not report participant-level distributions, so the summaries cannot reveal whether the count data were skewed, whether a few observations were especially influential, or how individual values were arranged.

How should the p-value be read?

The posted unadjusted p-value is 0.096. General explanation: a p-value describes how incompatible the observed data, or more extreme data, would be with a specified null model if that model were true. It is not a measure of effect size, a probability that the null hypothesis is true, or a direct measure of clinical importance. The registry record does not provide a fuller statistical analysis plan for this endpoint, so the p-value should not be treated as evidence about procedures that are not described there.

The sample-size illustration separately states a two-sided significance level of 0.05 for its planning calculations. That planning assumption is not evidence that the trial's original design used the same threshold. The original design assumptions and sample-size justification are not reported in the registry record. The posted comparison's p-value is above the planning illustration's stated threshold, but a threshold comparison does not remove the need to consider the estimate and interval.

Why does over-dispersion matter in general?

General explanation: count outcomes can sometimes vary more than a simple statistical model expects. This is called over-dispersion. When it is present, a model that assumes too little variability can understate uncertainty and make evidence appear more precise than it is. Methods for count data can account for different distributional patterns, but the registry record does not report a count-regression model, a dispersion assessment, or any other model for the primary endpoint. The posted mean-and-standard-deviation comparison does not tell us whether over-dispersion was assessed or whether it would affect a different analysis.

It would be inappropriate to infer that a particular count model was prespecified or used. The available result can be described as a comparison of means with standard deviations and an unadjusted p-value. Questions about model choice, assumptions, covariate adjustment, and robustness remain unresolved in the registry record.

What can the analysis support?

The summaries support describing the observed average attack counts and the uncertainty around their contrast. The interval crosses zero, and the p-value is 0.096, so the posted analysis does not provide a precise estimate that excludes no difference. That does not prove the groups are identical; rather, the data leave a range of plausible contrasts. The interpretation is further limited by the pilot design, outcome-specific sample sizes, and absent details on missing data, participant flow, and the statistical plan.

General explanation: a pilot trial can help characterize feasibility and generate an estimate for later planning, but a small study is often poorly positioned to distinguish modest effects from random variation. Any sample-size figures shown alongside this result are planning illustrations based on the observed effect, not evidence that the original trial was designed or powered in a particular way. The registry record does not report the original sample-size rationale.

10. Confidence Intervals and Precision

The mean difference is reported as 0.77 attacks, with an interval that extends from below zero to 1.675. The interval crosses zero, so the posted estimate is compatible with no average difference as well as with a difference in either direction. Its breadth signals that the direction and size of the underlying contrast are not pinned down precisely by this pilot result.

General explanation: a confidence interval describes a range of values compatible with the observed data under a specified statistical procedure. It is not a guarantee that the true value lies within the particular interval, and it is not a range containing most individual participants' outcomes. Here it concerns the mean difference between groups, not the spread of individual attack counts. The registry record does not state the confidence level or method for this interval, so neither should be inferred from the values alone.

The group standard deviations, 1.31 for topiramate and 1.75 for frovatriptan, indicate variation among observed counts. The primary outcome lists 20 and 24 participants in the two arms. Those features help explain why precision is limited: the comparison combines variable outcomes with modest outcome-specific group sizes. The interval is more informative than the p-value alone because it displays the range of contrasts that remain compatible with the result, but it still cannot resolve which treatment paradigm has a smaller average count.

Precision is also specific to the endpoint and time window. This interval does not quantify uncertainty for monthly headache days, quality of life, satisfaction, safety, or cost. Nor does it describe outcomes over the full treatment period. The result should therefore be read as uncertain for the month-two attack-count comparison, without extending its precision to other outcomes or timeframes.

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 table is a planning illustration, not a record of the original trial's design calculation. Each row describes planning for a stated fraction of the observed effect. The table gives participants per arm for 80% power and participants per arm for 90% power; these are not total event counts. The registry's planning 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 the stated fraction of the observed effect.

For planning for 100% of the observed effect, The table shows 65 participants per arm for 80% power and 87 participants per arm for 90% power. For planning for 75% of the observed effect, it lists 116 participants per arm for 80% power and 155 participants per arm for 90% power. For planning for 50% of the observed effect, it lists 260 participants per arm for 80% power and 348 participants per arm for 90% power.

In plain language, the 80% column describes a design intended to detect the chosen effect under the stated assumptions with 80% power; the 90% column uses a higher power target. A smaller effect is harder to distinguish from random variation, so planning to detect a smaller effect requires more participants. These figures are not observed power. They are illustrative sample-size calculations based on the observed effect, not the trial's own design. The original sample-size justification and design assumptions are not reported in the registry record.

12. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = 0.770; pooled standard deviation σ = 1.566.
2. n = 2 × (1.96 + 0.842)² × 1.566² / 0.770² = 65.0, rounded up to 65 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
delta <- 0.77                      # observed difference in means
sigma <- 1.56633785809904                     # 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(846495)
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 = 65 (80% power), 87 (90% power)
 75% of observed effect: n per arm = 116 (80% power), 155 (90% power)
 50% of observed effect: n per arm = 260 (80% power), 348 (90% power)
power.t.test: 65.9 
Simulated power at n = 65 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. Multiplicity, Missing Data and Analysis Populations

The registry lists 8 outcome measures and identifies 6 posted secondary outcomes. These include monthly headache days, participants with greater than 50% reductions in migraine attacks and headache days, quality of life, satisfaction, adverse events, and cost. Looking at several outcomes creates a multiplicity question: when many comparisons are examined, the chance of at least one apparently notable result can increase. General statistical practice may address this through prespecified outcome hierarchies or adjustments, but the registry record does not report a multiplicity plan or adjustment for the primary comparison.

The registry record lists no posted statistical analyses. It does not describe whether the primary outcome was analyzed according to an intention-to-treat population, a per-protocol population, or another analysis set. The endpoint-specific group sizes are reported, but a participant flow account explaining how these relate to randomized participants is not reported in the registry record. The record also does not say how treatment discontinuation, nonadherence, or outcomes that were not recorded were handled.

Missing-data methods matter because results can be affected if outcome availability differs between groups or is related to participants' outcomes. General explanation: complete-case analysis, imputation, and other strategies make different assumptions about missing observations. No missing-data method or assessment is reported, so it is not possible to determine which approach, if any, was used. The same limitation applies to sensitivity analyses that might test how conclusions change under different assumptions.

The record does not report whether there were interim analyses, subgroup analyses, covariate adjustments, or a detailed statistical analysis plan. These details should not be guessed from the posted means and p-value. Without them, the result is best treated as the descriptive, unadjusted comparison shown, rather than as a complete account of all analyses conducted. The absence of those details does not show that a particular method was or was not used; it means that the registry record does not document it.

14. 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 serious adverse event figures are reported separately from the primary attack-count result. The record lists 28 participants at risk in the topiramate group and 27 in the frovatriptan group. It reports 0 participants with serious adverse events in each group. The deaths figures are not reported in the registry record.

These counts describe the serious adverse event entries recorded for the stated groups; they do not establish that no adverse events of any kind occurred. The registry also reports other, non-serious adverse events: 17 participants affected in the topiramate group and 5 in the frovatriptan group. Those figures are distinct from the serious adverse event counts and should not be combined with them when reading the serious-event summary.

General explanation: safety counts need context, including the number of participants at risk, the way events were defined and collected, and the duration of observation. The record reports the at-risk group sizes and gives the treatment period as 2 months, but it does not provide further details about event ascertainment or the nature of the other adverse events. A zero count in this study is a statement about the posted record, not proof that an event cannot occur or a basis for comparing safety definitively.

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

The result describes a narrow comparison: average migraine attacks reported in Treatment Month 2 among participants assigned to daily topiramate or frovatriptan during prodrome. The topiramate group mean was 1.35 and the frovatriptan group mean was 2.12. The estimated mean difference was 0.77 attacks in the direction frovatriptan minus topiramate. The interval crosses zero, and the unadjusted p-value is 0.096, so the posted estimate remains uncertain.

What the result does show is the observed difference between these group summaries in this pilot trial. It also shows that the available estimate does not precisely distinguish a difference from no average difference. It does not establish that the groups are equivalent, and it does not show that one treatment paradigm is superior. An uncertain result is not the same as evidence of no effect; it indicates that the information recorded is insufficient to settle the size or direction of the underlying contrast with precision.

The endpoint is an attack count, not headache days, symptom severity, quality of life, satisfaction, or cost. The registry lists those as separate secondary outcome topics, but the primary result cannot stand in for them. It is also limited to Treatment Month 2 and cannot be assumed to represent the whole treatment period or later experience. The group average does not describe what happened to each participant, and the standard deviations show that counts varied within both groups.

The study's pilot size and the absence of a reported analysis plan, missing-data approach, and multiplicity handling further limit interpretation. The registry record does not provide a basis for extrapolating beyond the enrolled study population or for making a clinical recommendation. The appropriate statistical reading is a preliminary, imprecise comparison that can inform questions for further investigation, while leaving important uncertainties unresolved.

16. Why This Trial Matters Statistically

This trial is statistically interesting because it compares two different treatment schedules as well as two medicines. One group took topiramate daily, while the other used frovatriptan when prodrome suggested that a disabling migraine would occur. The primary outcome therefore asks about attack frequency under distinct treatment paradigms, over a defined month, rather than comparing two daily regimens.

The trial also illustrates why a pilot result needs to be read through both effect size and precision. The observed mean difference is 0.77 attacks, but the reported interval includes zero and reaches 1.675 on the positive side. The p-value is 0.096. Those results leave considerable uncertainty despite a visible difference between the sample means. A reader who focuses only on the means could overstate what the evidence establishes; a reader who focuses only on the p-value could incorrectly treat the result as proof of no difference.

The sample-size illustration makes the role of assumptions visible. It gives different planning requirements for detecting the observed effect and smaller fractions of it, under stated assumptions of equal allocation, a two-sided significance level of 0.05, and a normal approximation. It is a useful demonstration of how the target effect influences planning, but it is not a reconstruction of the original trial design and not observed power.

Finally, the record demonstrates the limits of registry result summaries. Outcome summaries and safety counts are available, but detailed analysis methods, missing-data handling, participant flow, and multiplicity procedures are not reported. These omissions constrain how confidently the posted comparison can be interpreted. They also show why a numerical result is only one part of evaluating a trial: the endpoint, design, uncertainty, and reporting completeness all matter.

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.

This is a pilot study with an outcome-specific sample of 20 participants in the topiramate arm and 24 in the frovatriptan arm. The registry reports overall enrollment of 55, but it does not provide a participant-flow explanation for the relationship between enrollment and the primary outcome group sizes. Withdrawals, exclusions, and missing outcome records are not described in the registry record.

The primary endpoint covers Treatment Month 2 only. It does not establish whether the attack-count contrast was similar in the other treatment month, across the complete treatment period, or after treatment ended. The primary outcome is also distinct from the listed secondary outcomes concerning headache days, reductions in attacks and headache days, quality of life, satisfaction, adverse events, and cost. Results for those topics are not included in the primary comparison.

The record describes the study as single-blind but does not identify who was blinded or describe how masking was maintained. It also does not report details of allocation concealment, adherence, protocol deviations, or the statistical model. The posted comparison is unadjusted; no adjusted analysis, missing-data method, multiplicity plan, or analysis population is reported. The confidence-interval level and method are not stated in the registry record.

General explanation: a small pilot estimate can be sensitive to the outcomes of a limited number of participants, particularly when individual counts vary. The reported standard deviations indicate variation, but the group summaries do not show the individual observations or distribution shape. The pilot result should therefore not be treated as a precise estimate for a broader population. The registry does not report the original sample-size justification, so the study's intended precision and design assumptions cannot be reconstructed from the planning illustration.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Pilot Study to Compare Frovatriptan vs. Topiramate for Prevention of Migraine — NCT00846495.
• Registry record last updated 2012-01-13; status last verified 2012-01.
• Results first posted 2011-08-09.
• 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 record describes a completed randomized, single-blind, parallel-group Phase 4 pilot study sponsored by Clinvest, with enrollment of 55. It compared daily topiramate with frovatriptan taken during prodrome for migraine prevention. Participants had a 1-month baseline period and a 2-month treatment period, with 3 visits described.

The primary endpoint was the number of migraine attacks during Treatment Month 2. The posted means were 1.35 for topiramate and 2.12 for frovatriptan, with an unadjusted mean difference of 0.77. The reported interval crossed zero, and the unadjusted p-value was 0.096. The endpoint-specific group sizes were 20 and 24, respectively.

The registry lists secondary outcome topics and adverse event counts, but it does not report a detailed statistical analysis plan, missing-data methods, or original sample-size justification. The primary result is therefore best read as an imprecise pilot comparison for a specific endpoint and timeframe.