A Study in Migraine Prevention: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study in Migraine Prevention (NCT01184508), a randomized phase 2 study. The primary outcome is Change From Baseline to 12 Week Endpoint in the Number of Migraine Attacks.

2. Clinical Question

The study therefore asks whether the recorded frequency of migraine attacks changed over the treatment period in the specified analysis population, with placebo as the comparator. Random assignment is intended to make the treatment groups comparable on average, while masking is intended to limit the influence of treatment knowledge on care, reporting, and assessment. Those design features support a fair comparison, but they do not supply the outcome estimate. Without a posted group result and an uncertainty measure, the magnitude, direction, and precision of the primary comparison cannot be evaluated from the registry record.

3. Trial Design

This was a phase 2, randomized, parallel-group, placebo-controlled study of LY2300559 in people with migraine. The official title describes it as double-blind, while the registry classifies masking as quadruple. The registry record does not explain which groups of people were masked or how masking was maintained. The study had two named arms, LY2300559 and placebo. The brief summary describes oral treatment with LY2300559 for 12 weeks; the dose, administration schedule, and corresponding placebo procedures are not reported in the registry record.

The recorded enrollment is 87. This is the study-level enrollment figure, not a reported count for the primary-outcome analysis. The safety record gives separate at-risk denominators for its adverse-event entries, but the registry record does not explain their relationship to enrollment or to the endpoint population. The primary endpoint population required baseline migraine activity, at least 1 dose of study drug, and post-randomization efficacy data, so it should not be assumed to include every enrolled or randomized participant.

The study start is listed as 2011-01 and primary completion as 2012-04. Its status is TERMINATED. The reason for termination, the timing of the decision, and whether termination affected data collection or follow-up are not reported in the registry record. A status of terminated alone does not explain why the study ended or indicate what its outcomes were.

The primary objective concerned change in migraine-attack frequency. Listed secondary outcomes address migraine severity, symptom duration, migraine days, global impressions, quality of life, interictal burden, and pharmacokinetics. The registry record does not provide enough detail to determine which measures were collected at every scheduled visit or how they were incorporated into the final analysis.

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: Prior to randomization, participants completed screening and washout followed by a 28-day baseline period to determine the number of migraine attacks participants experienced without the use of preventative medications. Participant Flow and results based on participants, post-randomization (12-week treatment and 4-week follow-up periods combined).

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Randomized participants (pts) who had at least 4 migraines and/or probable migraine headaches during the baseline period, who received at least 1 dose of study drug, and for whom had at least 1 post-randomization efficacy data for the specified endpoint was available.

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: Modified intent-to-treat 1 (mITT1) population: participants who received at least 1 dose of study drug.

6. Primary Endpoint

The registry lists the primary outcome as Change From Baseline to 12 Week Endpoint in the Number of Migraine Attacks, measured over Baseline and Month 3.

The primary endpoint is change from baseline to the 12 week endpoint in the number of migraine attacks. The record labels the measure as a mean and gives its unit as migraine attacks. It describes the timeframe as baseline and Month 3. A migraine was defined as a headache lasting 4 to 72 hours, based on International Headache Society diagnostic criteria.

Attack frequency was normalized to a 28-day month using the number of attacks and the number of days in the specified month. This normalization puts observations from months of different lengths on a common time scale. It is a way of expressing frequency, not a claim that each participant experienced the same observation period or the same number of attacks.

Attack frequency is relevant to a prevention study because it measures how often migraine attacks occur, rather than only their intensity or duration. The registry also lists secondary measures of severity, duration, migraine days, patient and clinician impressions, quality of life, interictal burden, and drug concentration. These address other aspects of migraine and treatment, but they do not replace the primary endpoint.

The registry record specifies standard deviation as the dispersion measure. A standard deviation describes variability among observations around their mean; it is not itself an estimate of the treatment contrast or a confidence interval. The primary group means, change estimates, and uncertainty around a comparison are not reported in the registry record, so the endpoint definition can be described but its observed result cannot be interpreted numerically.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry describes a randomized, parallel-group study with LY2300559 and placebo. In general, random assignment aims to balance measured and unmeasured characteristics across groups on average. A parallel-group design compares groups followed under their assigned interventions rather than having each participant receive both interventions in sequence. These are general design principles; the registry record does not report the randomization sequence method, allocation concealment procedures, or implementation details.

What does the endpoint mean?

The primary measure is a continuous outcome summarized as a mean: change in the number of migraine attacks from baseline to the 12 week endpoint. A mean describes the average value in a group, while the standard deviation describes the spread of individual observations around that group average. These answer different questions. The mean gives a location summary; the standard deviation describes variability. Neither alone supplies the between-arm treatment contrast or its precision.

The attack-frequency measure was normalized to a 28-day month. In general, normalization can make rates more comparable when the calendar period observed differs in length. The registry defines a migraine as a headache lasting 4 to 72 hours and says the definition was based on International Headache Society diagnostic criteria. The record does not report how attacks were verified, how diary data were handled, or how an incomplete observation period was treated.

How should a treatment contrast be read?

In general, a treatment contrast compares outcomes between the randomized groups. For a change endpoint, readers need to know whether the reported quantity is a within-arm change, a difference in changes, or a model-based contrast. The registry record identifies change from baseline as the endpoint but does not report the numerical analysis or specify the estimator used to compare arms. It is therefore not appropriate to infer that a particular adjustment, model, or test was used.

A confidence interval, when reported, describes a range of values compatible with the data and the statistical method under its assumptions. Its width conveys precision: wider intervals generally indicate less precision than narrower intervals. A p-value addresses compatibility of the data with a specified null model; it does not measure the size or clinical importance of an effect. Neither a confidence interval nor a p-value for the primary comparison is reported in the registry record. General explanations of these tools do not imply that the trial used any particular model or testing procedure.

Why can variability matter?

For a continuous endpoint, individual participants can have quite different attack patterns. Greater variability can make a group mean less precise, all else being equal. Standard deviation is one way to describe that variability, but it does not show the shape of the data. If a distribution is highly skewed or contains unusual observations, a mean and standard deviation may not fully describe it. The registry record does not report the outcome distribution or diagnostics, so readers cannot assess those features.

Over-dispersion is a general concern in count data: observed variation can exceed what a simple count model assumes. When relevant, a model that accounts for extra variation may avoid uncertainty estimates that are too optimistic. This is a general statistical point, not a description of the analysis in this study. The registry record does not report whether migraine-attack counts were modeled as counts, analyzed as continuous values, or handled with a method addressing over-dispersion.

What do the planning assumptions say?

The registry describes sample-size figures as planning illustrations based on an observed effect, not as the trial's original design justification. The stated assumptions include a two-sided significance level of 0.05, equal allocation between arms, a normal approximation, and an effect to detect expressed as a fraction of the observed effect. Such illustrations can show how a hypothetical target effect and statistical assumptions relate to planning requirements. They are not observed power and do not establish that the study was designed around those assumptions.

The registry record does not report the original sample-size justification or the effect estimate underlying the planning illustration. Without the observed effect and its uncertainty, the planning assumptions cannot be used to reconstruct the primary result. A target sample size is also not a guarantee of precision: actual precision depends on the analyzed data, variability, missingness, and analysis method.

What is not specified?

The record does not report the primary comparison estimate, its uncertainty, the statistical model, covariates, significance test, handling of repeated observations, or sensitivity analyses. It also does not report whether any interim analyses were planned or performed. These omissions prevent readers from independently reproducing or fully assessing the primary analysis. The study's randomized and masked design is relevant context, but design labels alone do not identify the statistical methods used or establish the reliability of an unreported result.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists 11 outcome measures and 10 analyses, including the primary measure and secondary measures of migraine symptoms, patient and clinician impressions, quality of life, interictal burden, and pharmacokinetics. Multiple outcomes or analyses create a general multiplicity issue: when many hypotheses are examined, the chance of at least one apparently unusual result can increase if each is interpreted without accounting for the full set. The registry record does not report whether multiplicity was controlled, how outcomes were prioritized in the analysis, or whether the secondary outcomes were intended to support confirmatory conclusions.

The registry identifies a primary endpoint, but its numerical result is not reported in the registry record. It also does not report whether analysis choices changed after the study began, whether any analyses were exploratory, or how outcomes with several categories or time points were summarized. The record's outcome and analysis counts alone cannot answer those questions.

The defined primary-outcome population required baseline migraine activity, at least 1 dose of study drug, and at least 1 post-randomization efficacy observation for the endpoint. This means the listed population is not simply everyone enrolled. The number included in the endpoint analysis, reasons for missing observations, and differences between those with and without endpoint data are not reported in the registry record.

Missing-data methods are also not reported in the registry record. In general, missing outcome data can affect a comparison if the likelihood of missingness is related to outcomes or differs between groups. Methods such as imputation or sensitivity analyses rely on assumptions, which should be described to judge their plausibility. No imputation method, missing-data assumption, or sensitivity analysis is specified for this trial. Readers therefore cannot determine how missing observations affected the primary or secondary analyses.

11. 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 4% reporting threshold.

Most frequent non-serious adverse event terms

The registry lists zero serious adverse events in the placebo group, with 45 participants at risk, and zero in the LY2300559 group, with 41 participants at risk. It also lists other adverse events for 32 participants in the placebo group out of 45 at risk and 30 participants in the LY2300559 group out of 41 at risk. These are counts as recorded, not percentages or estimates adjusted for differences in observation time.

The serious-event counts should be read alongside their listed at-risk denominators. They describe the registry entries, but the record does not report definitions, event-level details, duration of follow-up, or whether events were judged related to study treatment. The list of serious adverse-event terms is empty. The deaths field has no reported count; deaths are not reported in the registry record, so that field should not be interpreted as zero.

Counts of other adverse events do not establish causality. The registry record does not report the types or severity of those events in the information, or how adverse events were collected and coded. These safety entries should be interpreted as descriptive record data rather than as a complete comparative safety analysis.

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

The registry record establishes the study question, design labels, endpoint definition, and selected safety counts. It does not report the numerical result for the primary endpoint. As a result, readers cannot determine from this record whether the mean number of migraine attacks changed differently between the LY2300559 and placebo groups, how large any contrast was, or how precisely it was estimated.

A study's design can strengthen the basis for a comparison without replacing its results. Randomization is intended to reduce systematic differences between groups, and masking is intended to limit influence from treatment knowledge. Whether those protections were implemented successfully is not described in detail. In any case, these features cannot establish an effect when the outcome estimates are absent.

The defined primary analysis population excludes people who did not meet the stated baseline, treatment, and post-randomization data criteria. The record does not give the size of that population or explain missing endpoint data. Thus, the study-level enrollment figure should not be treated as the number contributing to the primary result. The registry's TERMINATED status does not explain the reason for termination or its effect on the available data.

The safety counts describe the entries in the registry, but they do not show event types, attribution, or a full account of follow-up. They also do not answer the primary efficacy question. No inference about the presence or absence of an effect should be drawn from the lack of posted primary results, and the listed safety figures alone cannot characterize overall benefit or risk.

In general, a treatment result is most interpretable when it includes the group estimates, a clearly defined between-group contrast, a measure of uncertainty, and transparent information about analysis populations and missing data. Those elements are not reported for the primary endpoint in the registry record. The appropriate interpretation is therefore limited to what the record documents, without concluding that the interventions were equivalent or that either had a particular effect.

13. Why This Trial Matters Statistically

This trial illustrates why a clear endpoint definition and a reported endpoint result are both necessary for statistical interpretation. The registry specifies a mean change in migraine-attack frequency, the time frame, the attack definition, and the 28-day normalization. Those details help explain what investigators intended to measure. Without results by group and an estimate of the comparison, however, readers cannot assess the observed effect or its precision.

The record also distinguishes a study-level enrollment figure from an endpoint population with explicit eligibility conditions. Participants had to meet baseline migraine criteria, receive at least 1 dose, and have at least 1 post-randomization efficacy observation for the specified endpoint. This makes transparent reporting of analysis-population counts and missing observations especially important. Those counts and details are not reported in the registry record.

Another statistical issue is the breadth of outcomes. The listed measures address frequency, severity, duration, patient and clinician impressions, quality of life, burden, and pharmacokinetics. Such breadth can be useful for describing different dimensions of a study, but it makes clear reporting of outcome hierarchy and multiplicity methods important. The registry lists the number of outcomes and analyses but does not report how multiplicity was addressed.

Finally, the TERMINATED status and the safety entries should be interpreted separately from the primary endpoint. The termination reason is not reported in the registry record, and the safety counts do not provide the missing efficacy comparison. Together, these features make the record a useful example of how registration details can document design and measurement while leaving key inferential questions unresolved.

14. 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 most direct limitation is that the primary outcome result is not reported in the registry record. There are no arm-level endpoint summaries, treatment contrast, confidence interval, or p-value to assess. The reason for termination and any effect of termination on data collection or reporting are also not reported in the registry record.

The primary endpoint population has several conditions, but the number meeting them is not reported in the registry record. The enrollment figure therefore does not establish how many participants contributed to the primary analysis. The record also does not report the number with missing outcome data, reasons for missingness, or methods used to address it.

Although standard deviation is named as the dispersion measure, the endpoint distributions and analysis model are not reported in the registry record. Details such as covariate adjustment, repeated-measures handling, assumptions, and sensitivity analyses are also not reported. The outcome and analysis counts do not reveal how multiple comparisons were handled.

The safety entries are limited to counts and at-risk denominators. Event types, severity, attribution, follow-up duration, and details of how events were collected are not reported in the registry record. Death counts are not reported in the registry record. The empty list of serious-event terms should not be taken as a complete safety narrative.

Finally, the record does not explain the randomization or masking procedures, treatment dose, or dosing schedule. These omissions limit assessment of implementation and reproducibility. They do not establish that procedures were inadequate; they mean the relevant details are not reported in the registry record.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: A Study in Migraine Prevention — NCT01184508.
• Registry record last updated 2018-09-12; status last verified 2018-08.
• Results first posted 2018-09-12.
• 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.

18. Record Summary

NCT01184508 is a phase 2 randomized, quadruple-masked, parallel-group study of LY2300559 and placebo for migraine prevention. The brief summary describes oral treatment with LY2300559 for 12 weeks. Enrollment is recorded as 87, and the study status is TERMINATED. The study start is listed as 2011-01 and primary completion as 2012-04.

The primary endpoint is change from baseline to the 12 week endpoint in mean migraine attacks, normalized to a 28-day month. The primary-outcome result and its uncertainty are not reported in the registry record. The record lists secondary outcomes across symptoms, global impressions, quality of life, burden, and pharmacokinetics. Its safety entries list zero serious adverse events in each arm, along with counts for other adverse events; deaths are not reported in the registry record.

The termination reason, primary analysis details, missing-data methods, and multiplicity approach are not reported in the registry record. These omissions prevent a full statistical assessment of the primary comparison.