This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for Atorvastatin Therapy for the Prevention of Atrial Fibrillation (SToP-AF) (NCT00252967), a randomized phase 3 study. The primary outcome is Time of Atrial Fibrillation Recurrence.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT00252967 |
| Phase | Phase 3 |
| Condition | Atrial Fibrillation, Inflammation |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Triple |
| Enrollment | 64 |
| Arms in the primary outcome | 2 |
| Lead sponsor | Emory University |
| Results posted | Yes |
2. Clinical Question
Population
The record describes people with atrial fibrillation who had undergone cardioversion. It does not report further eligibility criteria or a detailed description of the enrolled population in the primary outcome record. Enrollment was 64 participants, with 31 in the placebo arm and 33 in the atorvastatin arm.
Intervention
The atorvastatin arm received atorvastatin at a dose of 80 mg daily. The study rationale described in the registry focuses on the possibility that atorvastatin’s anti-inflammatory properties might affect the return of atrial fibrillation after cardioversion. The record does not report adherence or treatment exposure during follow-up.
Comparator
The comparison was placebo taken daily. The registry describes a randomized, parallel-group design with triple masking. Further details about who was masked or how treatment allocation was implemented are not reported in the registry record.
Primary question
The primary endpoint was time to atrial fibrillation recurrence, expressed as a median in days, upon recurrence and up to 12 months. The clinical question was whether the reported time to recurrence differed between the atorvastatin and placebo groups. The registry reports arm-specific medians and inter-quartile ranges, but the record does not report a hazard ratio, confidence interval, or significance test for this endpoint. The posted summaries therefore describe the observed distributions without providing a formal estimate of the comparative treatment effect.
3. Trial Design
SToP-AF was registered as a phase 3, randomized, triple-masked, parallel-group trial. Emory University was the sponsor. The registry lists a total enrollment of 64 participants. The primary completion date was 2008-11, the start date was 2005-10, and the study status is terminated. The reason for termination is not reported in the registry record.
Participants were assigned to daily atorvastatin or daily placebo. The atorvastatin dose was 80 mg daily. The registry identifies the endpoint as time of atrial fibrillation recurrence, measured in days, with a timeframe of upon recurrence, up to 12 months. The primary endpoint record gives 31 participants in the placebo arm and 33 in the atorvastatin arm. It does not explain whether these counts represent everyone randomized, everyone treated, or the population included in the endpoint analysis.
Randomization supports a comparison between groups by reducing the chance that measured and unmeasured baseline characteristics systematically determine treatment assignment. Masking can reduce the influence of treatment knowledge on participant behavior, clinical assessment, and outcome recording. These are general design principles; the registry’s labels do not give the allocation procedure or specify which people were masked. Those details are not reported in the registry record.
The trial’s brief description says the study aimed to test whether atorvastatin could reduce the rate of return of atrial fibrillation after cardioversion. The record also lists eight secondary outcome measures involving redox potential, reactive oxygen metabolites, isoprostanes, interleukin-6, interleukin-1, high-sensitivity C-reactive protein, and tumor necrosis factor alpha. The methods for measuring these outcomes, their analysis populations, and the planned statistical comparisons are not reported in the registry record.
The record reports termination but does not state why the trial ended or whether the termination affected enrollment, follow-up, or planned analyses. It also does not report the intended sample-size justification or a statistical analysis plan. As a result, the registry description supports a basic account of the design and observed arm summaries, but not a detailed reconstruction of how the trial was planned or conducted.
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
| Stage | Placebo | Atorvastatin |
|---|---|---|
| STARTED | 31 | 33 |
| COMPLETED | 31 | 33 |
| NOT COMPLETED | 0 | 0 |
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.
| Characteristic | Unit | Placebo | Atorvastatin | Total |
|---|---|---|---|---|
| Participants analysed | 31 | 33 | 64 | |
| Age, Continuous | years | 58.9 (SD 13.9) | 58.2 (SD 11.6) | 58.5 (SD 12.8) |
| Sex: Female, Male: Female | Participants | 5 | 6 | 11 |
| Sex: Female, Male: Male | Participants | 26 | 27 | 53 |
6. Primary Endpoint
The registry lists the primary outcome as Time of Atrial Fibrillation Recurrence, measured over Upon recurrence, up to 12 months.
| Element | Registry entry |
|---|---|
| Result type | Median |
| Measure of spread | Inter-Quartile Range |
| Unit | days |
| Endpoint classification | Time-to-Event Analysis |
The primary endpoint is the time from the relevant study reference point to atrial fibrillation recurrence, expressed in days. The registry describes the timeframe as upon recurrence, up to 12 months. It reports the median and inter-quartile range for each arm. The placebo arm’s median was 22.0 days, with an inter-quartile range from 7.0 to 70.0 days. The atorvastatin arm’s median was 29.0 days, with an inter-quartile range from 2.0 to 145.0 days.
Time-to-event endpoints are used when both whether an event occurs and when it occurs are relevant. A median time describes the midpoint of the reported event-time distribution: half of the observations lie on each side of the median, subject to how the analysis defines and handles observations without a recorded event. The inter-quartile range describes the middle half of the distribution and gives a sense of its spread. These are general explanations, not details of the trial’s analysis procedure.
The registry record does not explain how participants without a documented recurrence during follow-up were handled, whether or how censoring was used, or what rules defined recurrence. It does not report a comparative hazard ratio or confidence interval. The medians and ranges are therefore descriptive arm summaries; on their own, they do not quantify the uncertainty in a between-arm effect or establish whether the observed difference is compatible with chance variation.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The posted primary outcome consists of medians and inter-quartile ranges rather than a formal between-group treatment-effect estimate. The placebo arm had a reported median time to recurrence of 22.0 days, with an inter-quartile range from 7.0 to 70.0 days. The atorvastatin arm had a median of 29.0 days, with an inter-quartile range from 2.0 to 145.0 days. These summaries describe the recorded event-time distributions in the two arms. They do not, by themselves, show how precisely the groups differ.
The inter-quartile ranges are broad in both arms, indicating substantial spread in the middle portions of the reported distributions. The ranges overlap, but overlap between descriptive ranges is not a statistical test and does not settle whether a treatment effect exists. Nor should the medians alone be treated as the result of a time-to-event comparison. A formal analysis would need to account for the event-time data and any participants whose event status was not observed through the end of follow-up. The registry does not report the relevant analysis details.
The trial’s stated purpose was to test whether atorvastatin could reduce the rate of return of atrial fibrillation after cardioversion. The record does not provide a hazard ratio, confidence interval, p-value, or other inferential estimate for the primary endpoint. It also does not report a statistical test comparing the two medians. Without an effect estimate and its uncertainty, readers cannot determine from the registry results how compatible the observed summaries are with no difference or with effects of different magnitudes.
Enrollment was 64, with 31 participants in the placebo arm and 33 in the atorvastatin arm. Those figures provide context for the size of the trial, but they do not reveal how many participants experienced recurrence, how many completed follow-up, or how many contributed to a particular analysis. The event counts and endpoint analysis population are not reported in the registry record. The record also identifies the study as terminated but does not give the reason or its consequences for data collection and analysis.
The findings should therefore be read as a limited descriptive record, not as a definitive estimate of comparative effect. The arm-specific summaries are the available primary-endpoint results. Their interpretation is constrained by missing inferential statistics and by the absence of reported information on censoring, outcome ascertainment, missing observations, and the analysis population. No conclusion about treatment benefit or lack of benefit follows from these summaries alone.
The registry’s safety summary reports no serious adverse events in either arm. It reports other, non-serious adverse events in 26 of 31 participants assigned to placebo and 22 of 33 assigned to atorvastatin. These are counts from the registry record; the record does not provide event-specific detail or an inferential comparison. The serious-event summary should not be interpreted as proof that no serious events occurred outside the recorded reporting framework, and the other-event counts do not establish a causal difference between arms.
9. Statistical Methods Explained
The primary outcome is a time-to-event measure summarized with a median and inter-quartile range. The registry describes the event as atrial fibrillation recurrence and gives the timeframe as upon recurrence, up to 12 months. It reports the placebo arm median as 22.0 days, with an inter-quartile range from 7.0 to 70.0 days, and the atorvastatin arm median as 29.0 days, with an inter-quartile range from 2.0 to 145.0 days. The registry does not report a comparative hazard ratio, confidence interval, or significance test for the endpoint.
What does a median time describe?
As a general statistical concept, the median is the central point of an ordered distribution. A median event time indicates the point by which half of the relevant observations have had the event, provided the calculation and analysis population support that interpretation. It is less affected by unusually long or short observations than a mean. It does not describe the full shape of the event-time distribution, and it is not itself an estimate of the treatment effect relative to another arm.
The inter-quartile range runs from the first quartile to the third quartile and covers the middle half of a distribution. It gives information about spread around the median. A broad range signals that observed times vary considerably, but it is not a confidence interval for the median. The registry labels the reported dispersion as an inter-quartile range and gives its lower and upper endpoints for each arm. It does not report confidence intervals for either median.
Why are time-to-event methods often used?
In general, time-to-event methods can incorporate both the time until an event and the fact that some participants may have no event recorded by the end of observation. Those participants may have incomplete event-time information rather than a known event time. Statistical methods can account for this through censoring, when the assumptions and data permit. The registry record does not say whether participants without recurrence were censored, how follow-up ended for them, or which method was used to estimate the medians.
A common comparative summary in time-to-event analyses is a hazard ratio. In general, a hazard ratio compares the instantaneous event rates between groups over follow-up, subject to the assumptions of the chosen model. It is not the same as a ratio of medians, and it does not directly state the probability that an individual will experience an event. The record does not report a hazard ratio for this trial’s primary endpoint, so no such comparative interpretation is available from the posted primary results.
What information would support an inferential comparison?
A comparative estimate, such as a hazard ratio or another prespecified measure, would ordinarily be accompanied by an uncertainty interval and details of the analysis. A confidence interval is a range of estimates that reflects sampling uncertainty under the statistical model and its assumptions. A narrower interval generally indicates greater precision than a wider one. A p-value, when reported, summarizes how unusual the observed data or more extreme data would be under a specified null model; it does not measure the probability that the null hypothesis is true. These are general explanations. The registry does not report a confidence interval or p-value for the primary endpoint.
For a time-to-event comparison, the analysis may also depend on whether the relative event rate remains sufficiently stable over time, how event times were defined, and how participants with incomplete follow-up were treated. These are general methodological considerations. The analysis model, its assumptions, and any checks are not reported in the registry record. Consequently, the arm-level medians and ranges cannot be used to reconstruct a sponsor analysis or evaluate its assumptions.
How should the arm summaries be read?
The placebo and atorvastatin summaries describe separate distributions. The medians are 22.0 days and 29.0 days, respectively, and the inter-quartile ranges extend from 7.0 to 70.0 days and from 2.0 to 145.0 days, respectively. Descriptive differences between these numbers do not substitute for an estimate of the between-arm effect. It would not be appropriate to infer a p-value, confidence interval, or hazard ratio from the summaries alone. The record gives no event counts or individual-level times with which to conduct a separate time-to-event analysis.
The registry reports 31 participants for placebo and 33 for atorvastatin in the primary outcome record. It does not explain how those denominators relate to randomization, treatment receipt, recurrence status, or follow-up completion. In general, specifying the analysis population matters because exclusions or missing outcomes can affect both the estimate and its interpretation. The trial’s analysis population and handling of missing endpoint data are not reported in the registry record.
What can be said about planning?
The registry record does not report the trial’s own sample-size justification, target effect, or statistical power calculation. A planning illustration based on an observed effect is not the same as the trial’s design justification and is not observed power. In general, sample-size planning depends on assumptions about the effect to detect, outcome variability or event rate, significance level, allocation, and analysis method. The assumptions and design rationale for this trial are not reported in the registry record. No conclusion about whether enrollment was adequate for the intended question can be drawn from enrollment alone.
Over-dispersion is a general issue for count outcomes: it occurs when variability exceeds what a simple count model expects, and ignoring it can make uncertainty estimates too small. That concept does not change the interpretation of the posted median-time summaries. The registry does not report a count-model analysis for the primary endpoint. More broadly, statistical methods must match the endpoint and the available data; the record’s sparse analysis reporting limits methodological appraisal rather than permitting a reconstruction of the analysis.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists eight secondary outcome measures: comparisons of redox potential for cysteine and glutathione, derivatives of reactive oxygen metabolites, isoprostanes, interleukin-6, interleukin-1, high-sensitivity C-reactive protein, and tumor necrosis factor alpha. Together with the primary endpoint, the record lists nine outcome measures. The registry does not report results for these secondary measures in the information summarized in this record.
When a trial examines multiple outcomes or multiple comparisons, the chance of at least one apparently notable result can increase. Statistical approaches to multiplicity can control error rates or distinguish primary confirmatory questions from exploratory analyses. This is a general statistical consideration. Whether a multiplicity adjustment was planned or used, and how the primary and secondary outcomes were prioritized, are not reported in the registry record. The record does not report an analysis plan or interim analyses.
Missing-data methods are also important because the amount and pattern of missing outcome information can affect estimates. For this trial, the registry does not report how missing primary endpoint data were handled, how participants without recurrence during follow-up were treated, or how censoring was defined. It also does not report the number of participants with complete endpoint data or the reasons for any missing observations.
The arm counts in the primary outcome record are 31 for placebo and 33 for atorvastatin. The registry does not identify a named analysis population or explain whether the summaries use an intention-to-treat, per-protocol, or other population. Those terms describe different analysis choices in general; none can be attributed to this trial on the available record. The information needed to assess multiplicity, missingness, and analysis-population choices is therefore not reported in the registry record.
11. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| Placebo | 0 | 31 | 0.0% |
| Atorvastatin | 0 | 33 | 0.0% |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 5% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Placebo | 26 of 31 |
| Atorvastatin | 22 of 33 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Placebo | Atorvastatin |
|---|---|---|---|
| Atrial fibrillation | Cardiac disorders | 26 of 31 | 22 of 33 |
The registry reports zero serious adverse events among 31 participants in the placebo group and zero among 33 participants in the atorvastatin group. It reports other adverse events in 26 participants in the placebo group and 22 participants in the atorvastatin group. These are recorded counts, not a comparison adjusted for follow-up time or other factors. The registry record does not provide event-specific descriptions for the other adverse events, nor does it report an inferential test comparing the groups.
A count of zero serious events in a group does not demonstrate that the underlying risk is zero. In general, rare events can be absent from a limited observed sample even when the underlying event probability is not zero. The record does not give confidence intervals for the serious-event counts or describe the safety observation period in sufficient detail to assess exposure-adjusted rates. Death counts are not reported in the registry record.
The other adverse-event counts should likewise be read descriptively. The registry does not specify whether each participant could contribute more than one event, how severity or relatedness was determined, or whether events were collected consistently across the full follow-up period. These details are not reported in the registry record. The safety summary alone cannot establish that the treatments caused, prevented, or differed in the frequency of adverse events.
12. What the Result Does, and Does Not, Mean
The record offers a narrow but useful description of a randomized comparison: daily atorvastatin at 80 mg versus daily placebo, with time to atrial fibrillation recurrence as the primary endpoint. It reports arm-specific medians and inter-quartile ranges and a small total enrollment of 64 participants. It also records the study as terminated. These registry entries describe the trial and its posted results; they do not amount to a complete statistical evaluation.
The reported median time was 22.0 days in the placebo arm and 29.0 days in the atorvastatin arm. The middle half of reported times extended from 7.0 to 70.0 days for placebo and from 2.0 to 145.0 days for atorvastatin. The distributions therefore show substantial spread. Without a comparative estimate and uncertainty measure, the registry summaries do not establish the size or precision of a treatment effect. They also do not establish whether the observed arm summaries are statistically distinguishable.
The absence of a reported hazard ratio, confidence interval, or significance test is important. A median in one arm compared with a median in another does not provide the same information as a time-to-event model that accounts for follow-up and incomplete event observation. The record does not report event counts, censoring procedures, analysis populations, or endpoint missing-data methods, so those parts of the analysis cannot be assessed.
The safety record reports no serious adverse events in either arm and records other adverse events in 26 placebo participants and 22 atorvastatin participants. These counts are descriptive, and the record does not supply enough detail to assess event types, exposure, or comparative uncertainty. They should not be taken as proof of safety or harm.
In general, a randomized design can support causal comparisons when the conduct, outcome measurement, follow-up, and analysis are appropriate. The registry’s design labels alone do not verify those conditions. The available information does not justify a clinical recommendation or a claim that atorvastatin did or did not prevent recurrence. It establishes what the registry reports, while leaving important inferential and operational details unavailable.
13. Why This Trial Matters Statistically
This trial is statistically instructive because its primary endpoint concerns time to recurrence, yet the posted result is limited to arm-specific medians and inter-quartile ranges. Such summaries make the scale and spread of observed times easier to understand, but they do not answer the comparative question on their own. Readers need to distinguish descriptive summaries from an effect estimate and its uncertainty.
The contrast between the two reported inter-quartile ranges also illustrates why a single central value is incomplete. The placebo distribution extends from 7.0 to 70.0 days across its middle half, while the atorvastatin distribution extends from 2.0 to 145.0 days. The medians alone conceal this degree of variation. General statistical interpretation therefore considers both the location and spread of a distribution, along with the method used to compare groups.
The record also highlights the importance of reporting time-to-event methods. Event status and follow-up duration matter when some participants may not have a recorded recurrence by the end of observation. Without details on censoring, event counts, analysis populations, and the comparative model, an independent reader cannot evaluate the precision of the treatment comparison or whether the analysis appropriately used follow-up information.
Finally, the trial lists multiple secondary outcomes but does not report a multiplicity strategy or analysis plan. This matters because conclusions drawn from many outcomes depend on how questions were prioritized and how statistical uncertainty was handled. The record’s limited results are a reminder that design labels and descriptive statistics are not substitutes for transparent reporting of estimands, methods, missing-data procedures, and uncertainty.
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 registry identifies the study as terminated but does not report why it was terminated or whether termination altered recruitment, follow-up, or analysis. The planned enrollment target and the trial’s sample-size justification are not reported in the registry record. Consequently, the reported enrollment of 64 cannot be judged against a documented design target or a prespecified precision goal.
The primary outcome description does not state precisely how recurrence was defined or how it was ascertained. It gives a timeframe of upon recurrence, up to 12 months, but does not report the timing origin for the endpoint, monitoring procedures, adjudication, or follow-up completeness. It also does not describe how participants without a recorded recurrence were handled. These omissions matter when interpreting any time-to-event summary.
The posted primary result includes no event counts, hazard ratio, confidence interval, or significance test. It does not identify the analysis population or explain the relation between the reported arm counts and the randomized or treated populations. The number and reasons for missing endpoint observations are not reported in the registry record. The trial’s statistical analysis plan, allocation procedure, masking details, protocol deviations, and interim-analysis procedures are also not reported.
The record lists eight secondary outcome measures but does not provide their results or report a plan for handling multiple comparisons. The safety information is similarly limited: event-specific details, collection methods, relatedness, exposure, and death counts are not reported in the registry record. These limitations prevent a full assessment of the treatment comparison, outcome ascertainment, missing data, multiplicity, and safety reporting.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Atorvastatin Therapy for the Prevention of Atrial Fibrillation (SToP-AF) — NCT00252967.
- Registry record last updated 2014-09-03; status last verified 2014-08.
- Results first posted 2014-09-03.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
18. Record Summary
SToP-AF was registered as a phase 3, randomized, triple-masked, parallel-group study sponsored by Emory University. The record lists 64 participants, a start date of 2005-10, primary completion in 2008-11, and a terminated status. Participants were assigned to atorvastatin at 80 mg daily or placebo taken daily.
The primary endpoint was time to atrial fibrillation recurrence, in days, upon recurrence and up to 12 months. The placebo arm had a median of 22.0 days and an inter-quartile range from 7.0 to 70.0 days; the atorvastatin arm had a median of 29.0 days and an inter-quartile range from 2.0 to 145.0 days. The record does not report a comparative hazard ratio, confidence interval, or significance test.
The registry lists eight secondary outcome measures. It reports no serious adverse events in either arm and other adverse events in 26 placebo participants and 22 atorvastatin participants. Key details about endpoint analysis, missing data, multiplicity, and the reason for termination are not reported in the registry record.