An Observational Study of BOTOX® as Headache Prophylaxis for Chronic Migraine: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for An Observational Study of BOTOX® as Headache Prophylaxis for Chronic Migraine (NCT01686581), a registered phase not stated study. The primary outcome is Percentage of Participants Admitted to the Hospital for Headache.

2. Clinical Question

3. Trial Design

This was a prospective, non-interventional study intended to describe long-term, real-life use of BOTOX® for symptomatic treatment of adults with chronic migraine. The official title specifies a 24-month study, and the registry record gives a median of 21.20 months for the last-follow-up time point. That median does not establish that every participant was followed for the same duration.

The registry record reports enrollment of 641. The study began on 2012-07-23, had a primary completion date of 2016-10-12, and is listed as completed. Allergan is identified as the sponsor. The treatment was botulinum toxin Type A, administered according to physician prescription; treatment decisions lay with the physician. The study therefore records care decisions made in practice rather than assigning participants to treatment by a randomized study process.

The primary outcome concerns the percentage admitted to hospital for headache. The named secondary outcomes concern changes in migraine-specific quality of life, EQ-5D questionnaire scores, headache days, and treatment satisfaction. The record lists eight outcome measures overall, while its named secondary outcome list contains five titles. It lists zero analyses. The registry record does not report a randomized comparison, and its allocation information is not stated. Accordingly, the results should be read as a description of observed use and outcomes, not as a like-for-like comparison between assigned treatment groups.

The record does not report the study’s statistical analysis plan, a sample-size justification, or an interim-analysis plan. Its completion status and enrollment count do not establish what statistical conclusions were planned or whether the primary outcome was analyzed as intended.

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: Participants from the Safety Analysis Set, all participants who received at least one dose of BOTOX®, with data available for analysis at the given time-point.

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: Safety Analysis Set included all participants who received at least one dose of BOTOX®.

6. Primary Endpoint

The registry lists the primary outcome as Percentage of Participants Admitted to the Hospital for Headache, measured over Baseline (previous 3 months prior to first dose of BOTOX®) and Last Follow-up Visit (FU last) [median 21.20 months].

The primary endpoint is the percentage of participants admitted to the hospital for headache. It is a binary outcome at the participant level: for a specified measurement window, the participant either had a reported admission or did not. The outcome is framed as a percentage so readers can understand the frequency of this type of healthcare use in the analyzed group.

The registry record defines two measurement windows. Baseline covers the previous 3 months before the first BOTOX® dose. The last-follow-up value covers admissions since the previous visit, with the last-follow-up visit described as occurring at a median of 21.20 months. Because the windows are defined differently, the two percentages would need to be interpreted with their respective observation periods in mind. The record does not report either percentage, the numbers of participants contributing at each time point, or a statistical comparison.

Hospital admission is a healthcare-resource-use outcome, consistent with the study’s stated interest in resource utilization as well as patient-reported outcomes. It is not a direct measure of headache frequency, severity, or quality of life. Those are distinct concepts, and the registry lists separate secondary outcomes concerning headache days and questionnaire scores. The reason for selecting this endpoint beyond its stated role as the primary outcome is not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What kind of outcome is this?

The primary measure is a percentage of participants with an event. In general, a percentage for a binary outcome is based on the number of participants meeting the event definition relative to the number with usable data for that measurement. It describes how common the event was in that analyzed group during the specified window. It does not, by itself, describe how often events occurred per person or how long each participant was observed.

The registry record defines the event as admission to hospital for headache and names a Safety Analysis Set consisting of participants who received at least one dose and had data available at the time point. It does not report the number of participants analyzed or the event counts for either time point. Therefore, the primary percentages cannot be reconstructed from the record. The overall enrollment count is not a replacement for the time-point-specific denominators.

How should time-point comparisons be understood?

In general, a baseline-to-follow-up comparison describes change within an observed group. It is not equivalent to comparing randomly assigned groups. The study describes baseline admissions during the 3 months before the first dose and last-follow-up admissions since the previous visit. Those windows have different definitions. A percentage based on one period should not automatically be interpreted as directly comparable to a percentage based on another without considering how long each period covers and how the data were collected.

For a paired analysis, the same participants’ information at both time points can be important. The registry record does not state how many participants had data at both points, whether the analysis was paired, or how participants with a missing time point were handled. It also does not report whether the follow-up window was treated as a fixed period or accounted for in another way. These details are not reported in the registry record.

What do uncertainty and statistical tests add?

In general, a confidence interval communicates the range of values reasonably compatible with the data and the statistical method used. A wider interval indicates less precision, while a narrower interval indicates greater precision under that method. A statistical test can assess how compatible the data are with a specified null hypothesis, but a p-value is not the probability that a treatment works or that a result is true. Neither a confidence interval nor a p-value can be interpreted without knowing the estimate, the analyzed sample, and the analysis method.

No confidence interval, p-value, or primary-outcome test is reported in the registry record. The record lists zero analyses. Thus, no inferential conclusion about the primary outcome can be drawn from the posted registry information. General statistical concepts cannot fill in the missing estimate or its uncertainty.

What do the planning assumptions mean?

The planning illustration is described as using a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It is framed around detecting a stated fraction of an observed effect. For non-time-to-event outcomes, such illustrations are expressed as participants per arm for a target power; time-to-event illustrations are described in terms of total events. In general, such calculations ask how much information a hypothetical comparison might need under specified assumptions. Their output depends on the assumed effect and analysis model.

These assumptions do not describe the design of this observational study: the registry record does not report randomized arms or equal allocation, and its allocation information is not stated. The planning illustration is not the study’s own sample-size justification. The observed primary effect needed as the calculation’s starting point is not reported in the registry record, and no planning output is reported in the record. Planning figures, when used, are not observed power and do not show what the completed study established.

What can be inferred from the study structure?

The record identifies a prospective, non-interventional study in which physicians made treatment decisions. This structure can describe experience in practice, but it does not provide the protection against confounding that random allocation can offer. In general, confounding occurs when other characteristics or events are associated with both treatment decisions and outcomes, making the observed relationship difficult to attribute to treatment alone. The registry record does not report an adjustment strategy for such factors.

The record lists eight outcome measures and five named secondary outcome titles, but does not describe a statistical analysis plan, outcome hierarchy beyond the named primary endpoint, or methods for handling multiple analyses. It also does not report how missing outcome data were addressed. These omissions prevent assessment of the analysis choices and their implications for precision or interpretation.

10. Multiplicity, Missing Data and Analysis Populations

The registry record identifies one primary outcome and lists five secondary outcome titles, with eight outcome measures overall. It lists zero analyses. No multiplicity adjustment, testing hierarchy, or method for controlling the chance of false-positive findings across multiple outcomes is reported in the registry record. In general, examining multiple outcomes can increase the chance of finding at least one apparently notable result by chance; methods for addressing that issue depend on the planned analyses. No particular method can be attributed to this study from the record.

The primary outcome population includes participants with data available at each specified time point. This makes the analyzed group potentially distinct from all enrolled participants, but the number with available data is not reported. The record does not give missing-data counts, reasons for missingness, imputation procedures, or sensitivity analyses. It is therefore not possible to judge how missing outcome information may have affected the reported result; in this case, the primary result itself is not reported.

The record names the Safety Analysis Set as participants who received at least one BOTOX® dose and had data available for the time point. It does not describe an intention-to-treat population, a separate per-protocol population, or how participants were handled if treatment or follow-up differed from expectations. Because this is non-interventional and physician-directed, randomized assignment is not described. The record does not report whether the same participants contributed to both primary-outcome time points.

11. 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.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: spinal disorder (0.2%), mental disorder (0.2%), depression (0.2%), abortion spontaneous (0.2%), psychosomatic disease (0.2%).

The registry record reports 633 at risk for the serious adverse-event category and 8 in the affected count. For other adverse events, it reports 633 at risk and 34 in the affected count. Deaths are reported as 0. These figures concern the safety reporting in the record, not the primary hospital-admission endpoint. The registry does not report a causal assessment in the information, so the counts should not be read as establishing that BOTOX® caused an event.

The listed serious adverse-event terms are spinal disorder, mental disorder, depression, abortion spontaneous, psychosomatic disease, and migraine; each is shown as 0.2%. Those term-level percentages should be read as reported, not combined into a new total. In general, event categories or terms may not correspond to mutually exclusive groups of people, so adding categories can misrepresent the number of affected participants.

The record does not provide further detail about the timing, duration, severity criteria, or individual circumstances of these events in the information reported. The count of 633 at risk is distinct from the total enrollment of 641. The reason for that difference is not reported in the registry record, and the safety denominator should not be replaced with the enrollment total.

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

The study’s intended contribution is a description of long-term, real-life BOTOX® use and outcomes among adults with chronic migraine. The primary outcome selected for that purpose concerns hospital admissions for headache, a measure of healthcare use. The registry record does not report the baseline or last-follow-up percentages, so it does not establish whether that outcome increased, decreased, or remained unchanged.

Nor does the available record establish a causal treatment effect. Physicians prescribed treatment and made all treatment decisions; a randomized comparator is not described, and allocation is not stated. In general, an observed change over time in a non-interventional study may reflect treatment as well as other influences. A before-and-after measurement alone cannot identify which influence produced a change.

The study also lists outcomes concerning quality of life, EQ-5D scores, headache days, and satisfaction. Their titles indicate what was intended to be measured, not what the results were. The registry record does not report their outcome values or analyses. It is therefore not possible to characterize the direction, size, or precision of those results from this record.

Completion and enrollment are descriptive features of the study, not evidence of a particular outcome. Similarly, safety counts describe events reported in the record but do not establish causation. The appropriate reading is limited to what is reported: an observational study design, a physician-directed treatment context, the stated outcome definitions, and the available safety counts. The primary outcome’s numerical result and statistical uncertainty are not reported in the registry record.

13. Why This Trial Matters Statistically

This study is statistically useful as an example of the distinction between describing treatment in routine practice and estimating a causal treatment effect. A prospective observational design can collect information about treatment use and outcomes over time without assigning treatment. That design may address questions about experience in practice, but the lack of randomized assignment limits what a temporal comparison can establish.

The primary outcome also illustrates why an endpoint definition includes both an event and a measurement window. The event is hospital admission for headache, while the baseline and last-follow-up windows are defined differently. Understanding those windows is necessary before interpreting any percentages or comparing them. The record identifies the definitions but does not provide the values needed to assess the comparison.

The study’s stated interest extends beyond a single endpoint to resource utilization and patient-reported outcomes. The registry lists outcomes for quality of life, EQ-5D, headache days, and treatment satisfaction. That range highlights the distinction among clinical experience, healthcare use, and questionnaire-based outcomes; each requires its own definition, data completeness, and analysis.

Finally, the record demonstrates the limits of interpreting a completed study from registration information alone. It names enrollment, outcome measures, and safety counts, but lists zero analyses and does not report the primary outcome values. Statistical interpretation depends on estimates, denominators, uncertainty, and analytic details—not only on a study title, endpoint, or completion status.

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 registry record does not report the primary outcome values, denominators, confidence intervals, or statistical test. It therefore does not permit an assessment of the observed frequency of hospital admission for headache or the precision of any change. The record lists zero analyses, and the outcome titles alone cannot substitute for results.

The observational design limits causal interpretation. Treatment was prescribed by physicians, all treatment decisions lay with them, and a randomized comparator is not described. Allocation is not stated. Differences in participant characteristics, clinical decisions, or events during follow-up could be related to outcomes, but the record does not report an adjustment strategy or other analysis addressing those possibilities.

The primary outcome’s measurement windows differ: baseline covers the previous 3 months, and last follow-up covers time since the previous visit. The registry record does not give participant-level follow-up periods or explain how differing intervals were handled in the analysis. The median last-follow-up time is 21.20 months, but this does not establish uniform follow-up for participants.

Outcome availability is required for inclusion in the stated primary analysis population, yet the number analyzed at each time point and the handling of missing data are not reported. The enrollment of 641 should not be assumed to be the primary endpoint denominator. The safety figures use 633 at risk, and the record does not explain the difference from enrollment.

Finally, the registry record does not report a statistical analysis plan, multiplicity method, sample-size justification, or interim-analysis plan. These omissions leave important analytic decisions and the treatment of incomplete data unclear. The specific reason for each omission is not reported in the registry record.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: An Observational Study of BOTOX® as Headache Prophylaxis for Chronic Migraine — NCT01686581.
• Registry record last updated 2018-11-09; status last verified 2018-04.
• Results first posted 2018-11-09.
• 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.

18. Record Summary

NCT01686581 is a completed, prospective, non-interventional study sponsored by Allergan. It describes physician-directed BOTOX® use for adults with chronic migraine and reports enrollment of 641. The study began on 2012-07-23 and has a primary completion date of 2016-10-12.

The primary outcome is the percentage admitted to hospital for headache at baseline and last follow-up. The record defines the measurement windows and describes a median last-follow-up time of 21.20 months, but does not report the outcome percentages, analysis, or uncertainty. It lists eight outcome measures, five named secondary outcomes, and zero analyses.

Safety information includes 633 at risk in each adverse-event category, affected counts of 8 for serious adverse events and 34 for other adverse events, and 0 deaths. These figures do not replace the missing primary outcome results and do not establish causation.