The Dual Antiplatelet Therapy Study (DAPT Study): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for The Dual Antiplatelet Therapy Study (DAPT Study) (NCT00977938), a randomized phase 4 study. The primary outcome is MACCE (Death, Myocardial Infarction or Stroke) - Randomized DES ITT.

2. Clinical Question

The comparison is about treatment duration after the initial 12-month period, rather than a comparison of starting dual antiplatelet therapy with not starting it. The primary population is analyzed by randomized assignment, regardless of compliance after randomization. That distinction matters: the reported estimate describes outcomes according to assignment, not simply outcomes among patients who followed their assigned treatment throughout.

The registry gives a broader study purpose: assessing the duration, effectiveness and safety of dual antiplatelet therapy after coronary stent implantation. Its summary says a similar analysis was planned for a smaller bare-metal-stent cohort. The specific primary result discussed in the registry, however, is for randomized DES ITT patients.

3. Trial Design

The DAPT Study was a phase 4, randomized, parallel-group trial with enrollment recorded as 25682. The registry lists coronary artery disease as the condition and Baim Institute for Clinical Research as the sponsor. The recorded study start is 2009-10, the primary completion is 2014-06, and the status is completed.

The comparison of interest assigned patients to a 12-month or 30-month DAPT strategy after stent implantation. The intervention listings include placebo with aspirin and clopidogrel or prasugrel with aspirin. The arm descriptions specify that patients were compliant with 12 months of dual antiplatelet therapy and free of death, myocardial infarction, stroke, repeat coronary revascularization, major bleeding and stent thrombosis at 12 months after implantation. The registry text describing the arm populations is truncated, so it does not report additional detail about those descriptions.

The registry classifies masking as triple. Its brief summary describes the trial as double blind. Both descriptions appear in the registry record; further detail about who was masked is not reported there. Allocation is classified as randomized, and the model is parallel.

The primary analysis is not the full enrollment figure applied to every endpoint: it is specifically identified as randomized DES ITT. The registry also reports randomized BMS ITT outcomes and propensity-matched DES-versus-BMS secondary comparisons. Those are different analysis populations and should not be treated as interchangeable with the primary randomized DES comparison. The record’s description of the primary population says patients were analyzed according to randomized assignment, with non-events censored at 30 months or at last known follow-up, whichever was earlier.

Details such as the randomization procedure, the allocation ratio, site-level procedures, and the sample-size justification are not reported in the registry record. The posted sample-size illustrations are based on observed effects and specified assumptions; they are not the trial’s original design calculations.

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.

Treatment (12-30 mo. Post Procedure)

Reasons for not completing this period:

Observation (30-33 mo. Post Procedure)

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All randomized DES ITT patients; Patients were analyzed according to the treatment to which they were randomized (regardless of post-randomization compliance with the randomized treatment); Patients not experiencing the endpoint were censored at 30 months or at last known follow-up, whichever was earlier.

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 MACCE (Death, Myocardial Infarction or Stroke) - Randomized DES ITT, measured over 18 months (12-30 months post-index procedure).

The primary endpoint is MACCE: death, myocardial infarction or stroke. It is reported as a percentage of patients using a Kaplan–Meier estimate. The timeframe is 18 months, covering 12–30 months after the index procedure. The registry describes the coprimary efficacy endpoints as the cumulative incidence of MACCE and the cumulative incidence of ARC definite or probable stent thrombosis among randomized DES ITT patients during that interval.

A composite endpoint combines specified events into a single outcome. In general, this approach can capture the occurrence of several clinically important event types without requiring each component to serve as the sole primary measure. It also means the composite result does not, by itself, show how much each component contributed. Component-specific primary results are not reported in the registry record.

Kaplan–Meier estimation is a time-to-event method. In general, it uses event timing and accounts for follow-up that ends before the endpoint is observed by censoring those observations at the recorded end of follow-up. The registry says primary-population patients without the endpoint were censored at 30 months or their last known follow-up, whichever came earlier.

The posted estimates are 4.34% for DES 30-month DAPT and 5.92% for DES 12-month DAPT. The associated hazard ratio is 0.71, with a 95% confidence interval from 0.59 to 0.85 and p < 0.001. These are estimates for the specified population and follow-up period, not raw event counts or a claim that every patient had the same outcome.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does the time-to-event analysis estimate?

The primary endpoint is analyzed as time to MACCE, rather than only as a count of whether an event occurred by a single cutoff. In general, time-to-event methods use information about when events occur and can retain observations for which the event has not been observed by the end of follow-up. The registry says these patients were censored at 30 months or their last known follow-up, whichever came earlier. Censoring is not the same as counting a patient as event-free indefinitely; it marks the point at which the analysis no longer has recorded follow-up for that patient.

The posted arm results are Kaplan–Meier estimates, and the comparison method is log rank. The Kaplan–Meier estimate describes cumulative incidence over time under the method’s assumptions. The log-rank test compares the event-time experience across groups. The registry reports the associated hazard ratio as 0.71, with a 95% confidence interval from 0.59 to 0.85 and p < 0.001.

How should a hazard ratio be read?

A hazard is the instantaneous event rate among people still at risk at a given time. A hazard ratio compares those rates between groups over the analysis period. In general, a hazard ratio is not the same quantity as a ratio of endpoint percentages, and it is not an absolute risk difference. The reported value of 0.71 is therefore interpreted as a relative time-to-event estimate, not as a percentage-point change. The confidence interval describes uncertainty around that relative estimate. Its interpretation depends on the model and analysis assumptions; those assumptions are not fully detailed in the registry record.

What does intention-to-treat mean in this record?

The registry specifies that randomized DES ITT patients were analyzed according to their assigned treatment, regardless of post-randomization compliance. In general, this preserves the randomized group comparison and estimates the effect of assignment to each strategy. It does not estimate the effect only among people who adhered to treatment, and it may not describe outcomes under perfect adherence. A separate per-protocol analysis is not reported in the registry record.

Why do some secondary estimates use other methods?

The posted secondary comparisons do not all have the same design or estimator as the primary analysis. The DES-versus-BMS comparisons are described as propensity matched, not as randomized comparisons. Propensity matching is a general method for making observed groups more comparable on recorded characteristics. It cannot guarantee balance on unrecorded factors, so these comparisons do not have the same protection against confounding as a randomized comparison.

Some secondary binary outcomes are accompanied by simple unadjusted recalculations of risk differences, confidence intervals and p-values. These calculations use the posted group summaries and should be distinguished from the registry’s own reported analysis. For instance, the posted recalculated comparison for randomized DES bleeding uses a risk difference and an unadjusted p-value, while the registry reports its own risk difference and confidence interval. Different estimators, comparison orientation and analysis methods can yield results that are not numerically identical. Neither recalculation replaces the sponsor’s reported analysis.

What do the planning illustrations mean?

The sample-size illustrations use a two-sided significance level of 0.05, equal allocation between the two arms and a normal approximation. They show planning requirements for detecting stated fractions of the observed effect. For time-to-event outcomes, the illustrations refer to total events using the Schoenfeld method; for other outcomes, they refer to participants per arm. These are hypothetical planning calculations based on the observed effect, not observed power, not the trial’s original sample-size justification, and not evidence that the trial was designed using those assumptions. The original sample-size justification is not reported in the registry record.

What cannot be concluded from the methods information?

The registry identifies the primary analysis population, endpoint, timeframe, censoring rule, log-rank method and reported hazard ratio. It does not report enough detail to reconstruct every analytic choice. Information about proportional-hazards assessment, covariate adjustment for the primary analysis, handling of all missing data, prespecified interim analyses and the full statistical analysis plan is not reported in the registry record. Those omissions limit how far an independent reader can assess model diagnostics or reproduce the sponsor’s analysis from the posted summaries alone.

10. Analyses Reported by the Sponsor

The primary randomized DES result is reported by the registry as a log-rank analysis: hazard ratio 0.71, 95% confidence interval 0.59 to 0.85, and p < 0.001. The posted primary results do not include a separate recalculated hazard ratio, so there is no primary recalculation to compare with the sponsor’s reported estimate.

For some secondary binary outcomes, an unadjusted recalculation from posted group summaries is also available. That calculation is not the same analysis as the registry’s reported result. The reported analyses include a propensity-matched non-inferiority analysis for DES versus BMS and a reported risk-difference analysis for randomized DES bleeding. The recalculations use a direct risk-difference approach and unadjusted tests. These methods can differ in estimand, comparison orientation, handling of uncertainty and assumptions, so their results should not be treated as interchangeable.

The registry does not identify a discrepancy between its reported analysis and the posted summaries. The sponsor’s own analysis remains the relevant reported trial analysis; a simple recalculation is a supplementary check, not a replacement. Where the registry’s reported confidence interval is incomplete in the record, that missing limit is not reconstructed by the recalculation.

11. Multiplicity, Missing Data and Analysis Populations

The registry identifies MACCE and definite or probable stent thrombosis as coprimary efficacy endpoints for randomized DES ITT patients over 12–30 months after the procedure. It does not report how success across the coprimary endpoints was defined, whether an adjustment was used for testing more than one endpoint, or how the secondary endpoints were ordered for inference. A multiplicity plan is not reported in the registry record.

This matters because testing several outcomes can increase the chance of at least one nominally small p-value even when there is no true effect. In general, multiplicity procedures control or otherwise account for that issue, but the appropriate interpretation depends on the prespecified testing strategy. The reported secondary p-values should therefore not automatically be read as confirmatory findings. Prespecified interim analyses, stopping rules and any alpha-spending approach are not reported in the registry record.

The primary population is described as randomized DES ITT, with analysis according to assigned treatment despite post-randomization compliance. Patients without the endpoint were censored at 30 months or at last known follow-up, whichever was earlier. The registry does not report how missing covariates, withdrawals, loss to follow-up beyond this censoring description, or other incomplete data were handled. Imputation procedures and sensitivity analyses for missing data are not reported in the registry record.

The secondary results use distinct populations. Some are randomized DES ITT or randomized BMS ITT; others are propensity-matched DES-versus-BMS comparisons. Propensity matching addresses recorded characteristics but does not turn an observational comparison into a randomized one. The precise matching variables and diagnostic assessments are not reported in the registry record.

The registry reports that the primary analysis follows randomized assignment, but additional analysis-population details, including the full accounting of exclusions and endpoint availability, are not reported in the registry record. The record also does not report the full statistical analysis plan or sufficient information to verify all prespecified decisions.

12. 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 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: angina pectoris (2.1%), non-cardiac chest pain (2.0%), chest pain (1.7%), angina unstable (1.5%), osteoarthritis (1.1%).

The registry’s serious-adverse-event listing gives separate figures for DES 30-month DAPT, DES 12-month DAPT, BMS 30-month DAPT and BMS 12-month DAPT. The recorded at-risk and affected figures are, respectively, 2642 and 617; 2634 and 572; 842 and 144; and 845 and 150. These are safety-listing counts, not the primary efficacy estimates. The registry does not report deaths for these groups in the adverse-event summary.

The serious terms appearing most often in the listing include angina pectoris at 2.1%, non-cardiac chest pain at 2.0%, chest pain at 1.7%, unstable angina at 1.5%, osteoarthritis at 1.1%, and coronary artery disease at 1.1%. The record does not give an arm-specific breakdown for these terms in the registry record recorded. The listed percentages should therefore not be assigned to a particular treatment arm.

Serious adverse-event listings are not, by themselves, a causal comparison of treatment safety. Interpretation depends on the event definitions, follow-up, exposure, and the way affected patients and event counts are tabulated. The registry’s adverse-event summary does not provide enough detail to resolve those issues or to infer comparative safety from the counts alone.

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

The primary result is evidence of a difference in the estimated time-to-MACCE experience between randomized DES patients assigned to the 30-month and 12-month DAPT strategies during the specified 12–30-month period. The estimated hazard ratio is 0.71, with a 95% confidence interval from 0.59 to 0.85 and p < 0.001. The arm-specific Kaplan–Meier percentages are 4.34% and 5.92%, respectively.

Those results do not mean that every patient assigned to the 30-month strategy had a lower risk, or that the treatment effect was uniform over time. The confidence interval quantifies uncertainty in the estimated relative effect under the analysis; it does not describe variation in individual outcomes. The composite endpoint also does not show which of death, myocardial infarction or stroke accounts for the result.

The finding applies to the randomized DES ITT population and the registered timeframe. It is not, on its own, a result for the propensity-matched DES-versus-BMS population, the randomized BMS population, or people outside the study population. The registry’s primary analysis is assignment-based and does not isolate outcomes among only those who remained compliant after randomization.

Statistical evidence does not by itself establish the practical importance of an effect, resolve the balance of benefits and harms, or answer questions beyond the defined endpoint and follow-up. The secondary outcomes and adverse-event listings provide additional context, but they involve other populations, endpoints and analysis methods. Their interpretation depends on the limitations described in the registry record, including incomplete reporting of multiplicity and missing-data methods.

14. Why This Trial Matters Statistically

This trial poses a statistically important question about duration: after an initial period of dual antiplatelet therapy following stent implantation, how do assigned strategies of 12 months and 30 months compare over the subsequent period? Framing the question by treatment assignment and a defined follow-up interval makes the estimand clearer than a general comparison of medication use.

The primary result also illustrates why endpoint and analysis scale matter. MACCE is a composite, and the registry reports both Kaplan–Meier percentages and a hazard ratio. The percentages describe estimated cumulative incidence, while the hazard ratio compares time-to-event rates. They are related summaries of the same primary comparison but are not the same statistical quantity.

The record includes both randomized analyses and propensity-matched comparisons. Keeping those designs separate is essential: randomization supports an assignment-based comparison, whereas matching can address only observed differences captured by the matching process. Likewise, the simple unadjusted recalculations of selected secondary summaries should not be mistaken for the registry’s own reported analysis.

The trial is also a useful example of how a statistically clear primary estimate can coexist with limits on interpretation. The registry identifies the outcome, population, timeframe and principal comparison, but does not report the full multiplicity plan, missing-data strategy or original sample-size justification. Those omissions constrain independent evaluation and underscore why a p-value alone is not a complete account of a clinical trial result.

15. 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 primary result is limited to randomized DES ITT patients and the 12–30-month period after the index procedure. The registry also identifies stent thrombosis as a coprimary endpoint, but a detailed randomized-DES estimate for that endpoint is not reported in the posted results. The primary composite result does not provide component-specific estimates for death, myocardial infarction or stroke.

The registry record does not report a full statistical analysis plan, the original sample-size justification, multiplicity procedures, interim-analysis details, or all missing-data methods. It also does not provide enough information to assess every model assumption or reproduce the primary analysis from summary figures alone. The treatment-arm descriptions are truncated, and the record gives different masking descriptions: triple masking in the design classification and double blind in the brief summary.

Some secondary comparisons are propensity matched rather than randomized, and their matching variables and diagnostics are not reported in the registry record. Secondary outcomes also use different populations, endpoints and analysis methods. The available detailed secondary results do not cover every outcome listed in the registry. The simple recalculations are unadjusted and cannot substitute for the trial’s reported analyses.

The adverse-event summary gives group-level at-risk and affected figures and a list of common serious terms, but it does not provide enough arm-specific detail to support a complete comparative safety assessment. These limitations do not erase the posted primary result; they define the bounds of what the registry record can establish on its own.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: The Dual Antiplatelet Therapy Study (DAPT Study) — NCT00977938.
• Registry record last updated 2017-06-09; status last verified 2017-05.
• Results first posted 2015-10-22.
• 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.

19. Record Summary

The completed phase 4 DAPT Study was randomized and parallel, with enrollment recorded as 25682. Its primary analysis compared 30-month with 12-month DAPT among randomized DES ITT patients for MACCE from 12 to 30 months after the index procedure. The registry reports Kaplan–Meier estimates of 4.34% and 5.92%, respectively, with a log-rank hazard ratio of 0.71, 95% confidence interval 0.59 to 0.85, and p < 0.001.

The record identifies definite or probable stent thrombosis as a coprimary endpoint and lists secondary outcomes spanning randomized and propensity-matched populations. Detailed methods for multiplicity and missing data, the original sample-size justification, and complete results for every listed secondary outcome are not reported in the registry record. Secondary recalculations are unadjusted supplements, not replacements for the registry’s analyses.