DATAS: The Dual Chamber & Atrial Tachyarrhythmias Adverse Events Study: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for DATAS: The Dual Chamber & Atrial Tachyarrhythmias Adverse Events Study (NCT00157820), a randomized na study. The primary outcome is CSAE-score Rate(Clinical Significant Adverse Events Score Rate).

2. Clinical Question

3. Trial Design

DATAS is a randomized, parallel study with no masking reported. Its stated enrollment is 354, and its status is completed. The record gives a start of 2000-11 and primary completion of 2005-05. Phase is listed as NA. The sponsor is Medtronic Cardiac Rhythm and Heart Failure.

Three programming groups are described. SC True received a single chamber ICD programmed as a single chamber. SC Simulated received a dual chamber ICD initially programmed as a single chamber. DC True received a dual chamber ICD initially programmed as a DDED. The two interventions are described as single chamber and dual chamber implantable cardioverter defibrillators. The registry summary says the study was designed to analyze the ability of dual chamber ICDs to reduce clinically significant adverse events compared with single chamber ICDs in a non-selected population with a conventional indication for ICD implantation. That statement describes the study objective, not a conclusion about the result.

The primary outcome is identified as an ITT outcome over a timeframe of 17 months. The listed outcome group sizes are 111 for SC True, 223 for SC Simulated, and 223 for DC True. These arm counts do not align with the stated overall enrollment, and the registry record does not explain the discrepancy. The one posted inferential analysis compares SC True with DC True using a Wilcoxon (Mann-Whitney) method. It does not include SC Simulated. Although a secondary outcome is listed as the number of each component of the CSAE, the registry record does not provide further component-level results in the registry record described for this analysis.

The record does not report details of the randomization procedure, allocation concealment, site structure, or the protocol’s planned statistical analysis beyond the named reported test. Those omissions limit how fully the trial’s design and analysis decisions can be assessed from the registry record.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Before assignment: The Dual chamber true (DC true) and Single chamber simulated (SC sim) arms crossed over after 8 months. All other crossovers were considered 'premature crossovers' and had to be authorized by an independent Adverse Events Advisory Committee. A 1-month wash out period was implemented after programmed crossover.

Randomized

Reasons for not completing this period:

1st 8 Month

Reasons for not completing this period:

1 Month Wash Out Period

2nd 8 Months

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: ITT

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 CSAE-score Rate(Clinical Significant Adverse Events Score Rate), measured over 17 months.

The primary endpoint is the clinical significant adverse-event score rate, reported in score/month. It is intended to combine clinically significant adverse events with specified high-priority outcomes into a score and relate that score to follow-up length. The registry describes assigning one point to each CSAE. Death was assigned a score equal to the maximum number of CSAEs in any individual patient in the entire study plus one. Premature authorized crossover was assigned a score equal to the maximum number of CSAEs in any individual patient in that period. The resulting score over the length of follow-up was expressed as a rate.

This construction gives death and premature authorized crossover special treatment rather than treating them as ordinary events. It reflects the study’s aim of combining adverse events and events indicating failure to remain on assigned therapy in one outcome. A rate also puts the accumulated score in relation to follow-up duration, rather than presenting only an unscaled score total. These are general reasons such a measure can be chosen; the registry record does not report additional justification for this specific endpoint, how the score’s clinical interpretation was established, or the exact patient-level calculation details beyond its description.

Because the endpoint is a score rate rather than a simple event count, interpretation depends on both the scoring rules and the follow-up exposure. The registry lists arm values of 0.112 for SC True, 0.128 for SC Simulated, and 0.075 for DC True. No arm-level dispersion or confidence intervals are reported for these values. The outcome is not directly interchangeable with the proportion of participants experiencing an adverse event, and the score-rate values alone do not show how much each component contributed.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry names the Wilcoxon (Mann-Whitney) method for the SC True versus DC True analysis. This is a rank-based comparison of observations from two groups. In general, a rank-based test orders the observations and evaluates whether values from one group tend to rank differently from values in the other. It does not require the same distributional assumptions as a standard comparison of means, but interpretation still depends on the study design, the observation structure, and how the reported effect measure was obtained.

What does the Wilcoxon (Mann-Whitney) label tell us?

The method name indicates the family of test used for the posted comparison. The record does not provide the test statistic, the rank distributions, or details about ties or other implementation choices. It also does not explain how the reported odds ratio of 0.31 was calculated alongside this rank-based method. Consequently, the method label is not enough to reconstruct the analysis or to translate its estimate directly into a score-rate difference. The odds ratio should be read as the effect measure the registry reports, with the reported 95% confidence interval and p-value, rather than as a recalculated comparison of the listed arm means or rates.

How should the odds ratio and interval be read?

In general, an odds ratio compares odds between groups, and a value of 1 represents equal odds under the definition used for the analysis. The registry gives an estimate of 0.31 and a 95% confidence interval of 0.14 to 0.67. The record does not specify the exact event or ordering used to define the odds, so the direction should not be translated into a more detailed clinical claim than the registry supports. A confidence interval conveys uncertainty in an estimate under the analysis assumptions. It does not incorporate every possible source of uncertainty, such as endpoint construction, missing outcomes, model choice, or limitations in the registry reporting.

In general, a p-value assesses how compatible the observed data are with a null hypothesis under a statistical model or test procedure. The reported p-value is 0.003. It is not the probability that the null hypothesis is true, and it is not a measure of the size or clinical relevance of a difference. The p-value should be interpreted together with the effect estimate, interval, endpoint definition, and the limited scope of the pairwise comparison.

Why can over-dispersion matter for a count-like outcome?

General statistical context: count outcomes can vary more than a simple count model expects, a condition often called over-dispersion. If the analysis assumes less variation than the data contain, uncertainty estimates can be too small and tests can appear more decisive than warranted. A score rate that combines events and assigns special weights is not a plain count, but variation in event burden may still be important. The registry record does not report a dispersion estimate, a model for count variation, or an assessment of over-dispersion. It also does not report a mean, standard deviation, or other spread measure for the arm-level score rates.

What does the endpoint’s scoring rule imply?

The score assigns one point to each CSAE, while death receives a score based on the maximum number of CSAEs in any individual patient in the entire study plus one. Premature authorized crossover receives a score based on the maximum number of CSAEs in any individual patient in that period. General context: a composite or weighted endpoint can combine outcomes that differ in importance and frequency, but its overall result can be influenced by the scoring choices and by the component mix. The registry record does not report how often each scoring rule determined an individual’s score, nor does it provide a component breakdown of the primary endpoint. The listed secondary outcome concerns numbers of each CSAE component, but further results are not included in the registry record for this analysis.

What can the planning illustrations establish?

General context: sample-size planning illustrations describe how many participants per arm might be needed under stated assumptions to detect an effect at a chosen power and significance level. They are planning exercises, not observed power and not evidence that the trial had a particular probability of success. The stated illustration assumptions use a two-sided significance level of 0.05, equal allocation, a normal approximation, and an effect defined as a fraction of the observed effect. Any resulting figures should be treated as illustrative rather than as the trial’s own design justification. The registry record does not report the original sample-size justification or the planned target effect. The exact participant-per-arm outputs are not described in the registry entries addressed.

The registry record also does not report adjustments for multiple comparisons, prespecified interim analyses, a detailed missing-data procedure, or a full statistical analysis plan. The absence of those details does not prove that such procedures were not used; it means they cannot be assessed from the registry record. The available inference remains the single reported comparison, interpreted within the limits of the endpoint definition and record-level documentation.

10. Analyses Reported by the Sponsor

The registry record reports one sponsor analysis comparing SC True with DC True. It identifies the method as Wilcoxon (Mann-Whitney), gives an odds ratio of 0.31, a 95% confidence interval of 0.14 to 0.67, and p = 0.003. No discrepancy between the sponsor’s result and a recalculated result is flagged. The registry entries do not include a recalculated estimate for this count-like, specially scored endpoint, so there is no separate unadjusted recalculation to set beside the sponsor analysis.

The sponsor’s reported analysis therefore remains the inferential result available for this comparison. Its scope is limited to the two named groups; it is not a three-arm analysis and does not include SC Simulated. The registry record does not provide enough detail to independently reproduce the calculation, including the exact derivation of the odds ratio from the score-rate endpoint. The arm-level rate values are descriptive and should not be mistaken for a replacement analysis. The estimate, interval, and p-value can be reported as posted, while the method’s implementation and interpretation remain subject to the documentation limits.

11. Multiplicity, Missing Data and Analysis Populations

The registry identifies the primary outcome as an ITT outcome and gives a timeframe of 17 months. It does not describe how participants with missing follow-up or incomplete outcome information were handled. The registry record also does not report whether all randomized participants contributed to the primary analysis, how premature crossover was treated in the analysis beyond its special endpoint score, or whether any imputation or sensitivity analysis was performed. The ITT label identifies the stated analysis population, but does not by itself establish how missing observations were managed.

Only one inferential analysis is reported in the record: SC True compared with DC True. A third arm, SC Simulated, is listed, but there is no reported inferential comparison involving that group in the analysis. The registry also lists one secondary outcome, the number of each component of the CSAE. The registry entries do not provide a component-level inferential analysis or a set of secondary outcome results. It is therefore not possible to assess the full family of comparisons that may have been considered.

Multiplicity arises in general when several outcomes, groups, time points, or analysis choices are tested. If many tests are conducted, the chance of at least one apparently noteworthy finding can increase unless the analysis accounts for that testing structure. The registry record does not report a multiplicity adjustment, a hierarchy of hypotheses, or which comparisons were prespecified. It also does not report interim analyses, stopping rules, or whether the reported analysis was selected from a larger set of analyses.

These gaps should be treated as limits on interpretation, not as evidence that a particular method was or was not used. The result can be described as the posted comparison, but the record does not allow a complete audit of missing-data handling, analysis populations, or control of repeated testing. Those issues matter especially when interpreting a single p-value without a complete prespecified analysis plan.

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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: hospitalization or prolongued hospitalizations (29.7%), invasive interventions (9.9%), death (9.0%).

The adverse-event record lists serious events among 50 of 111 participants at risk in SC True, 49 of 223 in SC Simulated, and 44 of 223 in DC True. For other adverse events, the listed counts are 13 of 111, 8 of 223, and 3 of 223, respectively. Death counts are listed as null for all three groups, so death counts are not reported in the registry record for this safety listing.

The leading serious-event terms are hospitalization or prolongued hospitalizations at 29.7%, invasive interventions at 9.9%, and death at 9.0%. These are presented as terms and percentages; the registry entries do not specify their denominators or explain whether participants could contribute more than one term. They should not be interpreted as mutually exclusive categories or as participant-level risks without that information.

In general, adverse-event counts and percentages describe what was recorded under the registry’s definitions. They do not, on their own, establish causality or provide a full comparison of safety between groups. The record does not provide a statistical comparison for these safety figures, detailed event timing, or an explanation of how the serious-event terms relate to the primary CSAE score. They therefore give context but do not replace the primary endpoint analysis.

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

The posted results establish that the registry contains a primary score-rate outcome, three arm-level values, and one inferential comparison between SC True and DC True. The reported values are 0.112 score/month for SC True, 0.128 score/month for SC Simulated, and 0.075 score/month for DC True. For SC True versus DC True, the reported odds ratio is 0.31, with a 95% confidence interval from 0.14 to 0.67 and p = 0.003. This is the scope of the reported statistical evidence.

The result does not provide an inferential comparison for SC Simulated, even though that group has a listed score-rate value. Nor does the reported pairwise analysis establish an overall difference among all three groups. The registry record does not give arm-level confidence intervals or dispersion for the score rates, component-specific primary outcome results, or an explanation of the odds ratio’s precise derivation. These omissions make it difficult to translate the reported estimate into an absolute difference in the endpoint’s original units.

In general, statistical evidence and clinical interpretation are distinct. A p-value does not establish the importance of an effect, while an effect estimate and confidence interval describe an association under the analysis assumptions, not certainty about its cause or future reproducibility. The endpoint itself combines events using specific weights, including special scores for death and premature authorized crossover. Its meaning therefore depends on those choices and on the component events observed.

The safety listings provide counts and leading serious-event terms, but they do not show a formal comparison or resolve how the events contributed to the primary score. The record also does not report enough about missing data, multiplicity, randomization procedures, or the full analysis plan to assess those aspects independently. This means that conclusions should remain tied to the reported comparison and the documented endpoint rather than extending to unreported groups, outcomes, or mechanisms.

The trial’s objective was to analyze whether dual chamber ICDs could reduce clinically significant adverse events compared with single chamber ICDs in a non-selected population with a conventional ICD indication. That is the stated question, not a conclusion. The registry result should be described as a statistical finding for a specified pair of programming groups, without converting it into treatment advice or a claim that a device strategy works or fails.

14. Why This Trial Matters Statistically

DATAS illustrates several important issues in clinical-trial statistics. First, a composite score can bring multiple clinically significant events into one endpoint, but special weights affect what the resulting summary represents. Here, ordinary CSAEs receive one point, while death and premature authorized crossover receive scores tied to maximum event counts. The score rate is therefore not simply a count of participants with an event. Understanding its construction is central to interpreting the reported result.

Second, the registry lists three programming groups but reports only one inferential comparison. The distinction between the full design and the specific contrast matters: a pairwise result for SC True and DC True does not answer every possible question about the three-arm trial. Descriptive values for the third group do not fill that inferential gap.

Third, the reported method and effect measure require careful reading. The analysis is labelled Wilcoxon (Mann-Whitney), while the reported parameter is an odds ratio. The registry does not explain the calculation linking that measure to the score-rate endpoint. This is a useful reminder that a method label, effect estimate, and endpoint definition need to be considered together, and that registry reporting may not contain enough detail for full reproduction.

Finally, the record’s stated enrollment does not align with its listed arm counts, and it does not document several features needed to evaluate the analysis completely, including missing-data procedures and multiplicity handling. These are not reasons to discard the reported statistics; they are reasons to keep conclusions appropriately bounded. Clear separation of objective, descriptive results, inferential comparison, and documentation limits is essential when reading registered trial results.

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.

Several trial-specific limitations affect interpretation of the registry result. The stated enrollment is 354, while the primary-outcome arm counts are 111 for SC True, 223 for SC Simulated, and 223 for DC True. The registry record does not explain how these counts relate to the overall enrollment. This discrepancy in reported group counts makes it difficult to understand the analysis population and should not be resolved by assuming a particular explanation.

The only reported inferential analysis compares SC True with DC True. SC Simulated is part of the listed design and has a primary outcome value, but no inferential comparison involving it is reported. The record does not provide arm-level uncertainty estimates for the score-rate values, nor does it report the distribution of individual scores. It also does not explain the calculation that connects the Wilcoxon (Mann-Whitney) method to the reported odds ratio.

The primary endpoint combines event scores and specially weighted outcomes. The record does not provide the component contributions to the score or the frequency with which death or premature authorized crossover determined the score. Although serious adverse-event counts and selected event terms are listed, the record does not explain their relationship to the primary endpoint or provide a formal safety comparison.

Other relevant details are not reported in the registry record: the randomization procedure, allocation concealment, missing-data handling, multiplicity adjustment, interim analyses, and a full statistical analysis plan. The record also does not report a sample-size justification. These limits prevent a complete assessment of the analysis choices and the robustness of the reported comparison. They do not establish that any specific procedure was absent; they define what cannot be evaluated from the registry record.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: DATAS: The Dual Chamber & Atrial Tachyarrhythmias Adverse Events Study — NCT00157820.
• Registry record last updated 2025-07-02; status last verified 2024-04.
• Results first posted 2012-08-21.
• 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

DATAS is a completed randomized, parallel study with no masking reported, sponsored by Medtronic Cardiac Rhythm and Heart Failure. The record describes three ICD programming groups: SC True, SC Simulated, and DC True. Its primary outcome is the clinical significant adverse-event score rate over a timeframe of 17 months, identified as an ITT outcome.

The listed score-rate values are 0.112 score/month for SC True, 0.128 score/month for SC Simulated, and 0.075 score/month for DC True. The single reported inferential analysis compares SC True with DC True using Wilcoxon (Mann-Whitney), reporting an odds ratio of 0.31, a 95% confidence interval of 0.14 to 0.67, and p = 0.003. No inferential comparison involving SC Simulated is reported.

The registry also lists serious and other adverse-event counts, but does not provide a statistical safety comparison. The stated enrollment and listed arm counts do not align, and details about missing data, multiplicity, and the full analysis plan are not reported in the registry record.