Arrhythmia Detection In Obstructive Sleep Apnea (ADIOS): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Arrhythmia Detection In Obstructive Sleep Apnea (ADIOS) (NCT02743520), a na na study. The primary outcome is Number of Participants With Definite Atrial Fibrillation (Defined as an Irregularly Irregular Rhythm With Absence of p Waves).

2. Clinical Question

The study’s stated purpose was to learn how often people with OSA have irregular heartbeats and to characterize which OSA patients are most likely to have them. The primary endpoint narrows that broad purpose to definite atrial fibrillation, while the secondary outcomes address other rhythm findings, including ventricular premature contractions, premature atrial contractions, and features of atrial fibrillation episodes.

This is a descriptive detection question, not a treatment comparison. The registry record does not describe an intervention intended to change health outcomes, nor does it identify a control condition against which the monitor’s findings could be compared. It also does not report a comparison of detection strategies. The record therefore supports describing the registered population, monitoring approach, and endpoint definition, but not estimating a relative effect between groups or attributing differences in outcomes to the monitor.

The endpoint result itself is not reported in the registry record. Consequently, the record does not establish how many participants met the endpoint or how precisely the frequency of definite atrial fibrillation was estimated.

3. Trial Design

NCT02743520, titled Arrhythmia Detection In Obstructive Sleep Apnea (ADIOS), is registered as a completed, single-group study sponsored by the University of Miami. Enrollment was 86. The registered study period began on 2017-03-06, and primary completion was recorded as 2019-04-03. The phase is recorded as NA.

Participants were asked to join because they had been diagnosed with obstructive sleep apnea. The arm description states that participants underwent evaluation with a two-week cardiac event monitor, and the named device was the Lifestar Act III. Masking is recorded as none. Allocation is recorded as NA, consistent with the registered single-group model rather than a randomized allocation between study arms.

The primary outcome is the number of participants with definite atrial fibrillation at Week 2. The endpoint description specifies new cases evaluated by the Lifestar Act III. The registry defines definite atrial fibrillation as an irregularly irregular rhythm with absence of p waves. Four secondary outcomes are listed: participants with frequent ventricular premature contractions, the longest duration of atrial fibrillation, atrial fibrillation episodes lasting more than six minutes, and participants with frequent premature atrial contractions. The registry records five outcome measures overall.

These design features frame the study as a monitoring and detection study in one enrolled group. There is no registered comparator arm, so the design does not support a randomized between-group estimate. The record does not report a detailed protocol, eligibility criteria beyond an OSA diagnosis, or an analysis plan. It also does not report participant characteristics or the endpoint result. Those omissions limit assessment of the population’s representativeness and of the statistical precision of the findings.

4. Participant Flow and Accounting

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

Overall Study

Reasons for not completing this period:

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 Number of Participants With Definite Atrial Fibrillation (Defined as an Irregularly Irregular Rhythm With Absence of p Waves), measured over Week 2.

The primary endpoint counts participants with definite atrial fibrillation at Week 2. Its registry definition requires an irregularly irregular rhythm with absence of p waves, and the description states that new cases are evaluated by the Lifestar Act III. This is a binary participant-level outcome: each participant either meets the stated definition during the endpoint assessment or does not. The registry reports the outcome as a count of participants rather than as a continuous measurement.

In general, a participant count can describe how frequently an event was identified in the enrolled group. If the number of participants meeting the definition and the relevant analysis population are reported, readers can understand the observed event frequency and assess the uncertainty around it. A confidence interval, when reported, describes a range of values compatible with the data under the specified method; it does not guarantee that the true value lies within that interval.

The endpoint is relevant to the study’s stated detection purpose because it records a defined rhythm finding among participants monitored for two weeks. It is not, by itself, a comparison of the monitor with another detection method, nor does it measure a clinical outcome beyond the rhythm finding. The registry record does not report the number meeting the primary endpoint, an uncertainty interval, or a detailed analysis population. The endpoint definition is available, but its observed result and precision are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registered primary outcome is a count of participants meeting a stated rhythm definition. Statistical methods for such an endpoint generally begin by identifying the analysis population, assigning each participant an outcome status, and reporting the number who meet the definition. The registry record gives the endpoint definition and enrollment, but it does not report the endpoint count or the number included in its analysis.

What does a participant count represent?

For a binary outcome, each participant is classified according to whether the event occurred under the endpoint definition. A count describes how many participants had the event; a proportion would relate that count to the number of participants assessed. In general, a proportion is easier to interpret when its denominator is explicit and matches the population and observation period of interest. No new proportion can be recovered from the registry record because the primary event count and analyzed denominator are not reported.

The distinction between enrollment and analysis population matters. Enrollment describes how many participants entered the study, while an endpoint analysis may include only participants with adequate monitoring and interpretable results. The registry record reports enrollment of 86 but does not state whether all enrolled participants contributed to the Week 2 outcome. It does not describe exclusions, incomplete recordings, or the handling of missing endpoint data.

How should uncertainty be understood?

General statistical interpretation separates an observed estimate from its uncertainty. A confidence interval expresses the range of values compatible with the observed data under a particular statistical method and its assumptions. A narrower interval generally indicates greater precision, while a wider interval indicates less precision. An interval is not a guarantee about the true value, and it does not replace consideration of design limitations or the population studied. No confidence interval for the primary endpoint is reported in the registry record, so its precision cannot be assessed from the posted results.

A significance test, when used, addresses compatibility between the data and a specified null hypothesis under the assumptions of the test. A p-value is not the probability that a hypothesis is true, nor does it measure the size or importance of an effect. No primary analysis or p-value is reported in the registry record. The registered study is single-group, so a test framed as a comparison between randomized arms would not correspond to the design described.

Why does the study design affect the analysis?

In a single-group study, the main descriptive target is the outcome observed in that group during the defined assessment period. A second group is not available for a concurrent contrast. This means the study can describe findings among monitored participants, but it cannot supply a randomized estimate of a difference between monitoring strategies or treatment conditions. Differences from external populations or other studies would require additional design and comparability assumptions not reported in the registry record.

The primary timeframe is listed as Week 2, and the arm description says participants underwent a two-week monitor evaluation. In general, a clearly specified observation window helps determine which findings qualify as events and how each participant’s follow-up contributes. The record does not explain procedures for monitor interruptions, the timing of adjudication, or how episodes were classified when recordings were uncertain. Those details are not reported in the registry record.

What do the planning assumptions mean?

The sample-size assumptions describe a planning illustration computed from an observed effect, not the study’s own design. They specify a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the effect to detect is described as a stated fraction of the observed effect. The registry’s actual study is single-group, so the equal-allocation, two-arm assumption is not a description of its registered design. The planning illustration is not observed power, and it does not establish that the study had a particular probability of detecting an effect. The calculated planning values are not reported in the registry record.

In general, sample-size planning depends on the chosen outcome, target effect, variability or event frequency, significance level, desired power, allocation, and analysis method. If those assumptions differ from the actual study design or the outcome data, the planning result may not describe the precision achieved. The registry record does not report a protocol-based sample-size justification, the target effect for the study’s own design, or a detailed statistical analysis plan.

Do count-model considerations apply?

The primary outcome is a participant count for a binary event, rather than a repeated-event count for each person. General concerns about over-dispersion arise when count data vary more than a chosen count model allows. Whether such a model is appropriate depends on the outcome structure and analysis plan. The registry record does not report a count model or an over-dispersion assessment, and the endpoint description alone does not establish that one was used. No model-specific interpretation should be attributed to the study without a reported analysis method.

In general, transparent reporting would connect the endpoint definition to the assessed population, the event count, an uncertainty estimate, and the handling of missing observations. The registry record supplies the endpoint definition and study framework, but not those analysis details or primary results. That limits statistical interpretation to the design and measurement question rather than a quantified estimate of the observed frequency.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("Cardiac Event Monitor")
n      <- c(85)
events <- c(2)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  Cardiac Event Monitor  85  2  2.4  0.3  8.2  0.6  8.2  0.7  8.1

11. Multiplicity, Missing Data and Analysis Populations

The registry records five outcome measures: the primary atrial fibrillation count and four listed secondary outcomes. When a study examines several outcomes, the chance of finding at least one apparently notable result can depend on how many analyses are conducted and how they are interpreted. General approaches include identifying a primary outcome in advance, distinguishing confirmatory from exploratory analyses, and specifying any adjustment for multiple testing. The registry record identifies a primary outcome, but multiplicity procedures, a hierarchy among secondary outcomes, and adjusted analyses are not reported in the registry record.

No analyses are listed, and results for the primary and secondary outcomes are not reported in the registry record. It is therefore not possible to determine which outcomes were analyzed, whether additional analyses were performed, or how any multiplicity was addressed.

Missing-data handling is also not described. The record does not report how many participants had evaluable monitor recordings for the primary endpoint, whether any participants withdrew or had incomplete recordings, or how missing outcome information was treated. In general, missingness can affect an event estimate if participants without evaluable data differ from those with complete assessments. Without the analyzed denominator and a stated missing-data approach, that possibility cannot be assessed.

The analysis population is not reported in the registry record. In general, a report should clarify whether analyses include all enrolled participants, only those with adequate monitoring, or another defined group. Enrollment of 86 does not by itself establish the denominator used for the Week 2 endpoint. These reporting gaps prevent assessment of the completeness and robustness of the outcome analyses.

12. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

The registry reports safety figures for the Cardiac Event Monitor group. Serious adverse events were reported as affecting 0 participants among 86 at risk; other adverse events were reported as affecting 0 participants among 86 at risk; deaths were reported as 0. No top serious adverse-event terms are listed.

These are the posted counts for the categories recorded in the registry. They should be read as reported event figures, not as proof that no adverse events of any kind occurred or that the monitoring approach is risk-free. The record does not provide further details about event definitions, collection procedures, observation duration for safety, or whether all participants completed safety follow-up. Such details are not reported in the registry record.

The safety figures do not answer the primary rhythm-detection question. They describe the adverse-event categories recorded for the group, whereas the primary endpoint concerns definite atrial fibrillation detected during the monitoring assessment. The primary endpoint result is not reported in the registry record. The safety counts should therefore be interpreted separately from the unreported rhythm outcomes and without drawing conclusions beyond the recorded figures.

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

The registry record documents a study of cardiac rhythm monitoring among participants diagnosed with OSA. It specifies a single-group design, the Lifestar Act III device, a two-week monitoring assessment, and a primary endpoint defined as the number of participants with definite atrial fibrillation at Week 2. It also records enrollment of 86 and completion status.

Those details clarify what the study set out to measure, but they do not supply the primary outcome count. The number of participants with definite atrial fibrillation, the number assessed for that endpoint, and an estimate of uncertainty are not reported in the registry record. As a result, the record does not quantify how often the endpoint occurred in this enrolled group.

Even if a primary event count were available, a single-group detection study would describe findings in its monitored participants rather than estimate a randomized difference between monitoring strategies. It would not by itself establish that OSA caused a rhythm finding, that the monitor performs better than another method, or that identifying a rhythm finding changes later health outcomes. Those questions require evidence and comparisons not described in the registered design.

The secondary outcomes cover additional rhythm features, but their results are also not reported in the registry record. The posted safety figures record no serious adverse events, no other adverse events, and no deaths in the reported group; the record does not provide enough detail to extend those counts into a broader safety conclusion.

The most defensible reading is therefore about the study’s scope and reporting status: it was a completed, single-group monitoring study with a defined atrial fibrillation endpoint, but the outcome results needed to assess detection frequency and precision are absent from the registry record. No conclusion about the observed frequency or comparative performance can be drawn from the record alone.

14. Why This Trial Matters Statistically

This study is statistically informative as an example of how a clear endpoint and a limited design answer different kinds of questions. The registered question concerns detection of a defined rhythm finding among people with OSA who underwent monitoring. A single-group design can be suited to describing observed findings, but it does not create a concurrent control comparison. Recognizing that distinction prevents descriptive results from being interpreted as comparative effects.

The endpoint is stated in concrete terms: a participant count for definite atrial fibrillation, assessed at Week 2 and defined by an irregularly irregular rhythm with absence of p waves. A precise definition helps establish what qualifies as an event. However, a useful definition is only one part of a statistical result. Interpretation also depends on how many participants met the definition, how many were assessed, and how uncertain the estimate is. Those results are not reported in the registry record.

The record also illustrates why enrollment alone is not enough to evaluate precision. Enrollment is reported as 86, but the analysis population and endpoint denominator are not reported. Without them, readers cannot tell whether all enrolled participants contributed usable endpoint information. Missingness, outcome assessment, and the analysis method can all shape the interpretation of a count.

Finally, planning illustrations should be kept separate from trial results. The stated planning assumptions include a two-arm allocation framework, while the registered study is single-group. Such an illustration is not observed power and does not replace a reported analysis. The statistical value of the record lies in its description of the study question and design, while its absent outcome reporting limits conclusions about the findings themselves.

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 limitations specific to this record constrain interpretation. The primary outcome count and its analysis denominator are not reported in the registry record, so the observed frequency of definite atrial fibrillation cannot be quantified. The record also provides no confidence interval, p-value, or other analysis estimate from which to assess statistical precision.

The registered design is single-group, with no comparator. It therefore does not estimate a contrast between the Lifestar Act III and another device, a different monitoring duration, or an unmonitored group. The registry records masking as none and allocation as NA. The record does not describe independent outcome adjudication or additional safeguards against measurement or classification bias.

The population description is limited to participants diagnosed with OSA. Further eligibility criteria, participant characteristics, and details needed to assess how representative the enrolled group may be are not reported in the registry record. The record also does not explain monitor adherence, incomplete recordings, endpoint adjudication procedures, or handling of missing data.

Although secondary outcomes are listed, their results and the analysis plan for multiple outcomes are not reported in the registry record. The reported safety counts have no accompanying detail about collection procedures or safety follow-up. The planning assumptions describe an illustration using equal allocation between two arms, which is not the registered single-group structure; the illustration is not evidence of achieved power. Together, these reporting and design limits mean that the record describes the study framework more fully than its statistical findings.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Arrhythmia Detection In Obstructive Sleep Apnea (ADIOS) — NCT02743520.
• Registry record last updated 2020-11-19; status last verified 2020-10.
• Results first posted 2020-11-19.
• 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

NCT02743520, Arrhythmia Detection In Obstructive Sleep Apnea (ADIOS), is a completed single-group study sponsored by the University of Miami. It enrolled 86 participants diagnosed with OSA, who underwent a two-week evaluation with the Lifestar Act III cardiac event monitor. The primary outcome was the number of participants with definite atrial fibrillation at Week 2, defined as an irregularly irregular rhythm with absence of p waves.

The registry lists four secondary rhythm outcomes and records five outcome measures overall. It reports 0 serious adverse events among 86 at risk, 0 other adverse events among 86 at risk, and 0 deaths. The primary and secondary outcome results, analysis population, missing-data approach, and detailed statistical plan are not reported in the registry record. The single-group design supports a descriptive monitoring question, not a randomized comparison.