This page separates information reported in the ClinicalTrials.gov record from statistical interpretation. The ClinicalTrials.gov record reports that results are posted and that all three registered primary endpoints have results posted, but no formal statistical analyses, estimates, confidence intervals, or p-values are included in the ClinicalTrials.gov record.
1. Trial at a Glance
DAWN was a randomized, parallel-group clinical trial in ischemic stroke comparing the Trevo thrombectomy procedure with medical management. The registry lists 206 participants, two arms, single masking, and a treatment-oriented primary purpose.
| Feature | DAWN |
|---|---|
| Trial name | DAWN |
| Brief title | Clinical Mismatch in the Triage of Wake Up and Late Presenting Strokes Undergoing Neurointervention With Trevo |
| Condition | Ischemic Stroke |
| Phase | NA |
| Status | COMPLETED |
| Allocation | RANDOMIZED |
| Design model | PARALLEL |
| Masking | SINGLE |
| Primary purpose | TREATMENT |
| Enrollment | 206.0 |
| Lead sponsor | Stryker Neurovascular |
| Sponsor type | INDUSTRY |
| Start | 2014-07 |
| Primary completion | 2017-05-15 |
2. Clinical Question
The trial's central statistical question is whether randomized assignment to the Trevo thrombectomy procedure, compared with medical management, is associated with different 90-day functional outcomes and stroke-related mortality in participants with ischemic stroke.
Population
Participants enrolled in the DAWN trial with the registry-listed condition of ischemic stroke.
Intervention
Trevo Thrombectomy Procedure, classified in the registry as a device intervention.
Comparator
Medical Management, classified in the registry as an other intervention.
Primary question
How do the randomized groups compare on 90-day functional outcome and stroke-related mortality?
3. Trial Design
Randomized allocation is recorded in the registry.
The two intervention groups form parallel treatment arms.
The registry identifies the trial as single-masked but the ClinicalTrials.gov record does not specify which trial participants or personnel were masked.
The registry classifies the primary purpose as treatment.
Trevo Thrombectomy Procedure
- Intervention type: device
- Evaluated in a randomized parallel-group comparison
- Primary efficacy assessment at 90 days
Medical Management
- Intervention type: other
- Comparator in the randomized parallel-group design
- Primary efficacy assessment at 90 days
4. Endpoints
The ClinicalTrials.gov record lists three registered primary endpoints. All three have results posted in the registry, but the ClinicalTrials.gov record does not contain formal statistical-analysis records for them.
| Registered primary endpoint | Time frame | Type | Registry information |
|---|---|---|---|
| Weighted Modified Rankin Scale (mRS) Score, Lead Co-Primary Efficacy Outcome | 90 days | Other / unclear | The registry describes mRS as a scale for measuring the degree of disability or dependence in the daily activities of people who have suffered a stroke or other causes of neurological disability. |
| Functional Independence (mRS 0-2), Nested Co-Primary Efficacy Outcome | 90 days | Binary | The registry defines this as the number of participants with functional independence using the mRS framework. |
| Stroke-related Mortality, Primary Safety Outcome | 90 days | Other / unclear | The registered endpoint is stroke-related mortality assessed at 90 days. |
5. Planned Analysis
The ClinicalTrials.gov record contains no formal statistical analyses, so this page does not report a treatment-effect estimate, confidence interval, or p-value for any DAWN endpoint. The registry does indicate that results are posted and that each of the three registered primary endpoints has results posted, but the formal statistical-analysis records needed to reproduce an inferential comparison are not included in the ClinicalTrials.gov record.
Weighted Modified Rankin Scale
The lead co-primary efficacy endpoint is a weighted form of the Modified Rankin Scale assessed at 90 days. Because the mRS is an ordered disability scale rather than a continuous measurement with equal spacing between categories, an analysis should respect its ordinal structure or the prespecified weighting scheme.
For an ordinal outcome, a prespecified ordinal regression model can be used to compare the distribution of outcomes between randomized groups. Another possibility is an analysis specifically defined for a utility- or weight-based mRS outcome. The appropriate method depends on the statistical analysis plan and the exact definition of the weighting procedure. The ClinicalTrials.gov record does not identify the formal inferential method.
Functional Independence
Functional independence is defined using mRS 0-2 and is therefore a binary endpoint: each participant is classified as either functionally independent or not functionally independent at 90 days.
A binary endpoint of this type can commonly be compared using a risk ratio, risk difference, odds ratio, or an appropriate regression model. A two-group hypothesis test can also be specified. The choice among these approaches should be determined from the prespecified analysis plan rather than inferred from the endpoint definition alone. No formal comparison is in the posted results provided for this page.
Stroke-related Mortality
Stroke-related mortality is the registered primary safety outcome at 90 days. This is a binary outcome over a fixed follow-up period if each participant is classified according to whether stroke-related mortality occurred by 90 days.
For a fixed 90-day binary safety endpoint, the treatment groups can be compared using an absolute risk difference, risk ratio, odds ratio, or a suitable exact or asymptotic test depending on event frequency and the prespecified analysis plan. The ClinicalTrials.gov record does not specify which formal comparison was used.
6. Statistical Methodology
Randomization
Randomization is the central design feature for causal comparison in DAWN. By randomly assigning participants to the Trevo thrombectomy procedure or medical management, the trial establishes treatment groups whose differences in outcomes can be interpreted in relation to the assigned intervention, subject to the usual assumptions and limitations of randomized trials.
The exact statistical expression of this comparison depends on the endpoint and the prespecified analysis method.
Ordinal outcomes
The weighted mRS endpoint is based on an ordered disability scale. That structure is statistically important. Treating categories such as mRS 0, 1, 2, 3, 4, 5, and 6 as though the numerical distances between categories were necessarily equal can impose assumptions that are not inherent in an ordinal scale.
A method that preserves the ordered nature of the outcome can instead compare the probability of being at different points in the outcome distribution. If a proportional-odds model is selected, the model summarizes the treatment-group difference through a common odds ratio across cumulative thresholds. That assumption must be assessed or justified by the prespecified analysis strategy.
Binary outcomes
Functional independence at mRS 0-2 is a binary endpoint. Binary endpoints are naturally represented using a two-by-two table:
Medical management: independent / not independent
From this table, an analysis can quantify an absolute difference or a relative measure such as a risk ratio or odds ratio, depending on the prespecified statistical plan.
Fixed-time safety outcomes
Stroke-related mortality is defined over a 90-day time frame. A fixed-time mortality endpoint can be summarized as the proportion of participants experiencing the specified event by that time point. Because this is a safety endpoint, the direction and interpretation of the comparison should be stated explicitly rather than relying only on a generic statistical significance statement.
Confidence intervals
For any treatment-effect estimate, a confidence interval provides information about statistical precision. A narrow interval indicates that the estimate is more precisely estimated under the chosen model and sampling framework; a wide interval indicates greater uncertainty. The registry-reported DAWN statistical-analysis data contain no confidence intervals, so none are reproduced here.
7. Statistical Methods Explained
Why does randomization matter?
Randomization determines treatment assignment independently of the investigator's choice and the participant's baseline characteristics, within the limits of chance imbalance. This allows an outcome difference between randomized groups to be interpreted as evidence about the assigned interventions rather than simply an association between treatment choice and outcome.
Why is weighted mRS different from a simple yes/no endpoint?
Functional independence reduces the mRS to a threshold: mRS 0-2 versus values above 2. Weighted mRS retains more information about the ordered disability scale by assigning the categories a prespecified weighting scheme. The two endpoints therefore address related but not identical statistical questions.
What does “nested co-primary” mean?
A nested endpoint is defined within another outcome framework. Here, functional independence is based on mRS 0-2, while the lead co-primary endpoint is a weighted mRS score. The label signals that the binary endpoint is derived from the same underlying functional-status scale rather than being an unrelated clinical outcome.
How would functional independence normally be analysed?
Because functional independence is binary at 90 days, a standard analysis could compare the proportion functionally independent in each randomized group. The resulting effect could be expressed as an absolute risk difference, risk ratio, or odds ratio. Regression adjustment can also be used when prespecified. The ClinicalTrials.gov record does not identify the formal analysis used for this endpoint.
Why should mRS be treated as an ordered outcome?
The categories represent increasing levels of disability, but an ordinal scale does not automatically imply equal numerical distances between adjacent categories. An ordinal analysis therefore avoids treating the mRS numerical labels as if they were necessarily measurements on an interval scale.
What would a confidence interval add to the analysis?
A point estimate alone describes the observed treatment comparison but not its statistical precision. A confidence interval adds an uncertainty range under the specified model and sampling framework. For DAWN, the ClinicalTrials.gov record does not contain the corresponding confidence intervals.
Why is a p-value not an effect size?
A p-value addresses compatibility of the observed data with a specified null hypothesis under the statistical model. It does not tell the reader how large the treatment effect is, whether that effect is clinically important, or how precisely it has been estimated. Those questions require an effect estimate and an uncertainty measure such as a confidence interval.
8. Safety Analysis
The registry supplies serious adverse-event counts by treatment arm. These are reported as affected participants divided by participants at risk.
| Safety measure | Trevo Thrombectomy Procedure | Medical Management |
|---|---|---|
| Serious adverse events | 41/107 | 47/99 |
These counts provide an important safety description, but they should not be interpreted as a formal statistical comparison because the ClinicalTrials.gov record does not contain a corresponding statistical-analysis record, confidence interval, or p-value.
Numerator
The first number represents participants affected by serious adverse events in the ClinicalTrials.gov record.
Denominator
The second number represents participants at risk for the reported serious-adverse-event measure.
Why the denominator matters
A count without its corresponding at-risk population can be misleading when comparing treatment groups. The registry data provide both components here.
Inference
A formal between-group safety comparison requires an analysis method and uncertainty estimate that are not reported in the ClinicalTrials.gov record.
9. What the Registered Endpoint Structure Tells Us
The primary endpoint structure gives useful information about how the trial was intended to capture clinical benefit and safety. The two efficacy outcomes both focus on functional status at 90 days, while stroke-related mortality provides a distinct safety outcome.
| Endpoint role | Clinical question | Statistical structure |
|---|---|---|
| Lead co-primary efficacy | What was the distribution of weighted functional disability at 90 days? | Ordered or weighted outcome |
| Nested co-primary efficacy | How many participants achieved functional independence at 90 days? | Binary outcome |
| Primary safety | How many participants experienced stroke-related mortality by 90 days? | Fixed-time mortality outcome |
This structure illustrates why a single summary statistic cannot fully represent a multidimensional clinical trial. Functional status and mortality capture different aspects of outcome, and the appropriate analysis depends on how each endpoint is defined.
10. Limitations and Interpretation Issues
- Incomplete statistical-analysis detail in the ClinicalTrials.gov record: the registry data identify the primary endpoints and state that results are posted, but no formal statistical analyses were posted.
- No numerical efficacy estimates reported: this page therefore does not report an effect estimate, confidence interval, or p-value for the three primary endpoints.
- Endpoint-specific methods matter: weighted mRS, binary functional independence, and stroke-related mortality have different statistical structures and should not automatically be analyzed with the same method.
- Ordinal-scale assumptions: a weighted or ordinal mRS analysis depends on how the scale is represented statistically. The appropriate model cannot be established from the endpoint name alone.
- Single masking: the registry identifies the study as single-masked, but the ClinicalTrials.gov record does not specify which individuals were masked.
- Safety denominators: serious adverse events are reported as affected/at-risk counts, but no formal inferential comparison is provided in the ClinicalTrials.gov record.
- Generalizability: the ClinicalTrials.gov record identifies ischemic stroke as the condition but does not provide a detailed baseline-characteristics table from which broader population generalizability could be assessed.
- Multiplicity: the registry identifies three primary endpoints, but the ClinicalTrials.gov record does not specify the alpha-allocation or multiplicity-adjustment procedure.
11. Statistical Topics Not Specified in the Supplied Record
Several design and analysis topics commonly addressed in a full statistical analysis plan are not specified in the registry-reported DAWN trial data. They are therefore not assigned methods here.
| Topic | What the ClinicalTrials.gov record establishes | Interpretation |
|---|---|---|
| Non-inferiority margin | Not provided | No non-inferiority margin should be inferred. |
| Crossover | Not provided | No crossover effect should be assumed. |
| Factorial design | Not provided | The registry identifies a parallel design, not a factorial design. |
| Multiplicity procedure | Not provided | The existence of three primary endpoints does not establish a particular alpha-adjustment method. |
| Interim analysis | Not provided | No interim-monitoring procedure should be inferred. |
| Missing-data or imputation method | Not provided | No imputation approach is reported in the ClinicalTrials.gov record. |
| Stratification factors | Not provided | No stratified analysis factors are reported in the ClinicalTrials.gov record. |
| Bayesian methods | Not provided | No Bayesian model or posterior analysis is identified. |
12. Why This Trial Matters Statistically
DAWN is a useful statistical teaching example because the registered endpoints illustrate how a randomized clinical trial can examine the same clinical domain at different levels of measurement. Functional outcome is represented both as a weighted mRS endpoint and as a binary functional-independence endpoint, while mortality provides a separate safety outcome.
| Concept | How it appears in DAWN |
|---|---|
| Randomization | The registry identifies randomized allocation. |
| Parallel-group design | Two treatment arms are evaluated in parallel. |
| Ordinal outcome | Weighted mRS represents ordered neurological disability. |
| Binary outcome | Functional independence is defined as mRS 0-2. |
| Fixed-time safety outcome | Stroke-related mortality is assessed at 90 days. |
| Endpoint hierarchy | The registry distinguishes a lead co-primary efficacy outcome, a nested co-primary efficacy outcome, and a primary safety outcome. |
| Safety denominators | Serious adverse events are reported as affected participants divided by participants at risk for each arm. |
| Interpretation of uncertainty | Formal confidence intervals would quantify precision, but none are included in the ClinicalTrials.gov record. |
13. Statistical Interpretation: What Would a Complete Analysis Need?
A complete inferential presentation of the DAWN results would ideally connect each registered endpoint to four components: the observed outcome in each randomized group, the treatment-effect estimate, a confidence interval, and the hypothesis-testing result under the prespecified statistical framework.
Weighted mRS
The analysis should preserve or explicitly define the weighting and ordering of functional outcomes rather than treating the endpoint as an arbitrary continuous measurement.
Functional independence
The analysis should show the proportion with mRS 0-2 in each randomized group together with an appropriate comparative effect measure and uncertainty.
Stroke-related mortality
The analysis should report the 90-day mortality experience in each group and an appropriate safety comparison with uncertainty.
Serious adverse events
The registry-reported affected/at-risk counts should be accompanied by the statistical framework used to compare safety between randomized groups if a formal comparison was prespecified.
This is more than a reporting preference. Different endpoints answer different questions, so the statistical estimand must match the clinical question. An ordinal functional endpoint, a threshold-defined binary endpoint, and a mortality safety endpoint should not be reduced to the same generic interpretation.
14. Clinical Interpretation vs Statistical Interpretation
Statistical interpretation
DAWN is a randomized, parallel-group comparison with two principal efficacy outcomes centered on 90-day functional status and a separate 90-day stroke-related mortality safety outcome. The ClinicalTrials.gov record does not include the formal numerical inferential results.
Clinical interpretation
The trial was designed to evaluate a device-based thrombectomy procedure against medical management in ischemic stroke, with functional independence and the broader weighted mRS framework providing complementary measures of patient disability.
The distinction is important: statistical interpretation concerns the magnitude, uncertainty, and modeling of an observed treatment comparison, while clinical interpretation concerns what the endpoint represents for patients. Neither can be reconstructed completely from an endpoint name alone.
15. A Practical Framework for Reading the DAWN Results
| Step | Question to ask | Why it matters |
|---|---|---|
| 1. Identify the estimand | What difference between the randomized groups is being estimated? | Prevents confusion between a functional-distribution effect, a binary independence effect, and a mortality effect. |
| 2. Identify the analysis population | Which participants contribute to the efficacy and safety analyses? | The analysis population determines which randomized comparison the estimate represents. |
| 3. Examine the effect estimate | What is the magnitude and direction of the observed difference? | A p-value alone cannot communicate treatment-effect size. |
| 4. Examine uncertainty | How wide is the confidence interval? | Precision is essential for interpreting the stability of an estimate. |
| 5. Examine the endpoint | What clinical outcome does the statistic actually summarize? | Functional independence and mortality are not interchangeable outcomes. |
| 6. Check the analysis method | Does the statistical model match the scale and definition of the endpoint? | Ordinal, binary, and time-to-event outcomes require different modeling considerations. |
16. Trial Timeline
Trial start
The DAWN trial began in July 2014 according to the ClinicalTrials.gov record.
Primary completion
The ClinicalTrials.gov record lists May 15, 2017 as the primary completion date.
Registry status
the ClinicalTrials.gov record identifies the study status as COMPLETED.
17. Serious Adverse Events: Statistical Perspective
The serious-adverse-event data provide a useful example of why both numerators and denominators are needed in clinical-trial safety reporting.
Reported serious adverse events
Trevo Thrombectomy Procedure vs Medical Management
The two fractions describe affected participants relative to the registry-reported at-risk populations. A formal comparison could be expressed as an absolute risk difference or a relative measure, but the choice should follow the prespecified statistical analysis plan. Because no statistical-analysis record is reported, this page does not assign a p-value or confidence interval to the comparison.
The key lesson is that a safety count is not itself a treatment-effect estimate. The affected count must be interpreted in relation to the number at risk, the definition of the adverse event, follow-up, and the statistical framework used for comparison. The ClinicalTrials.gov record provides the affected/at-risk counts but does not provide the remaining inferential details.
18. Why the 90-Day Time Frame Matters
All three registered primary endpoints use a 90-day time frame. This creates a common assessment horizon for the primary efficacy and safety outcomes while preserving distinct endpoint definitions.
Functional outcome
Weighted mRS is assessed at 90 days, providing an ordered measure of neurological disability.
Functional independence
mRS 0-2 defines the binary functional-independence outcome at 90 days.
Safety
Stroke-related mortality is assessed at the same 90-day time point.
Comparability
Using a common time frame helps place efficacy and safety observations within the same follow-up horizon.
19. Interpreting Functional Independence
Functional independence is defined in the registry as mRS 0-2. This converts the ordered mRS scale into a clinically interpretable binary threshold.
This representation makes the endpoint straightforward to summarize as a proportion, but it also discards information about differences among participants who remain above or below the threshold.
For example, two participants with mRS values on opposite sides of the threshold are classified differently, while two participants with substantially different mRS values that fall on the same side of the threshold receive the same binary classification. The weighted mRS endpoint can preserve more information about the underlying functional scale.
20. Interpreting the Weighted mRS Endpoint
The lead co-primary efficacy endpoint is the Weighted Modified Rankin Scale (mRS) Score. The registry describes mRS as a scale for measuring disability or dependence in daily activities after stroke or other causes of neurological disability.
Statistically, weighting an ordinal outcome can change the relative contribution of different mRS categories to the overall comparison. That makes the precise weighting rule part of the estimand: without knowing the weighting scheme and analysis procedure, one cannot reconstruct the intended treatment-effect measure from the endpoint name alone.
21. Sources
- ClinicalTrials.gov: DAWN, NCT02142283.
- Linked publication listed in the registry: PubMed record PMID 34850380.
- Linked publication listed in the registry: PubMed record PMID 34727737.
- Linked publication listed in the registry: PubMed record PMID 34281376.
- Linked publication listed in the registry: PubMed record PMID 31744425.
- Linked publication listed in the registry: PubMed record PMID 31303153.
Continue through the Clinical Biostats learning pathway
Explore the statistical concepts that underlie randomized comparisons, functional outcomes, and clinical-trial analysis.
22. Record Summary
DAWN is a completed randomized, parallel-group trial of the Trevo thrombectomy procedure versus medical management in ischemic stroke, with 206 participants and three registered primary endpoints assessed at 90 days. The lead co-primary efficacy endpoint is a weighted mRS score, the nested co-primary efficacy endpoint is functional independence defined as mRS 0-2, and the primary safety endpoint is stroke-related mortality. The ClinicalTrials.gov record also report serious adverse events as 41/107 for the Trevo thrombectomy procedure and 47/99 for medical management.
From a statistical perspective, the trial illustrates the importance of matching the analysis to the endpoint: weighted or ordinal functional outcomes require different considerations from binary functional independence, while safety outcomes require their own definitions, denominators, and inferential framework.