Utility of a Smart Phone Application in Assessing Radial Artery Patency - the CAPITAL iRADIAL Study: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Utility of a Smart Phone Application in Assessing Radial Artery Patency - the CAPITAL iRADIAL Study (NCT02519491), a randomized na study. The primary outcome is Diagnostic Accuracy.

2. Clinical Question

The clinical motivation is the assessment of blood supply before procedures that use wrist artery access. The registry describes ultrasound as an accurate but time-consuming and expensive approach, and the Modified Allen Test as a commonly used clinical assessment. The smartphone application was proposed as a less subjective way to assess blood flow. These points explain the question; the trial results do not establish that either assessment changes patient outcomes or is suitable for every clinical setting.

3. Trial Design

CAPITAL iRADIAL, registered as NCT02519491, was a randomized, parallel-group trial with no masking. It was listed as not applicable for phase and is recorded as completed. The registry lists an enrollment of 438 and identifies Ottawa Heart Institute Research Corporation as sponsor. The start date is 2015-07 and primary completion is 2018-04.

The two arms were the Modified Allen's Test and the iPhone Assessment. Each arm has a posted group size of 219. The descriptions state that both assessments concerned radial and ulnar patency. Random allocation means that participants were assigned to study arms by chance, but details of sequence generation, allocation concealment, and implementation are not reported in the registry record. No masking was used, so the trial is recorded as open-label; who knew each assignment and whether outcome assessment was masked are not described further.

The primary outcome was diagnostic accuracy, expressed as a percentage, with a 95% confidence interval. Its stated timeframe was 1 year, and the primary population was intention to treat. The outcome description defines the target as the iRADIAL application’s accuracy for assessing competency of collateral blood supply to the hand, with Doppler ultrasound as the reference standard. The registry also posts accuracy results under both arms. Details of the trial visit schedule, the precise definition of an accuracy result, and how the reference standard was applied are not reported in the registry record.

The record lists one outcome measure and no posted analysis records. It does not report secondary outcomes. The absence of a posted analysis record means readers cannot assess a registry-posted inferential test or adjusted model for the arm comparison. It does not, by itself, establish what analyses may have been undertaken outside the registry.

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

Analysis population for the primary outcome, as defined in the registry: Intention to treat

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.

Population: Baseline analysis population is the same as the assignment in Participant Flow

6. Primary Endpoint

The registry lists the primary outcome as Diagnostic Accuracy, measured over 1 year.

The primary endpoint is diagnostic accuracy: the percentage of assessments that corresponded with the reference standard, according to the registry’s outcome description. The stated reference standard is Doppler ultrasound, and the target is competency of collateral blood supply to the hand. The registry labels the timeframe as 1 year and identifies the analysis population as intention to treat.

Diagnostic accuracy is used to describe how often a test classification agrees with a reference classification. In general, it provides an overall summary of agreement, but it does not show which types of errors occurred. A test can have the same overall accuracy as another while differing in its ability to identify people with or without the condition of interest. Sensitivity, specificity, predictive values, and the underlying classification counts are not reported in the registry record, so those features cannot be assessed from the posted outcome.

The posted accuracy is 81.7% for the Modified Allen's Test arm, with a 95% confidence interval from 76.0% to 86.6%. For the iPhone Assessment arm, it is 91.8%, with a 95% confidence interval from 87.3% to 95.1%. These estimates describe the recorded arm-level results. The registry does not provide a formal contrast between them, and the confidence intervals are not a substitute for a confidence interval or test of the difference between arms.

The record does not specify the exact denominator or operational rules used to calculate accuracy, how indeterminate assessments were handled, or whether all participants had evaluable results. Those details matter when interpreting a diagnostic endpoint, and are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The trial is recorded as randomized, parallel-group, and unmasked. In a randomized design, assignment by chance is intended to make comparison groups similar on average at the time of assignment. Randomization does not guarantee identical groups, and it does not remove uncertainty from the observed results. The registry does not report the sequence-generation procedure, how assignments were concealed before enrollment, or whether any individuals involved in assessment or analysis were masked. Those omissions limit assessment of how randomization was implemented and whether knowledge of assignment could have influenced measurements.

A parallel design assigns participants to one study arm rather than having each participant receive both assessment strategies in a crossover sequence. The registry lists the Modified Allen's Test and iPhone Assessment as the arms, with 219 participants in each. The primary population is described as intention to treat. In general, intention-to-treat analysis aims to retain participants in their assigned groups, regardless of what happens after assignment. The registry does not give enough detail to determine how that principle was applied to participants with missing or uninterpretable diagnostic results.

The endpoint is a percentage for overall diagnostic accuracy, accompanied by a 95% confidence interval. In general, a confidence interval describes the statistical uncertainty around an estimate under a particular analysis method. It is not a range containing a fixed proportion of individual participant outcomes, nor does it show the result for every person. The interval’s interpretation relies on assumptions about the sampling process and calculation method. The registry posts the interval bounds but does not report which method was used to calculate them.

Accuracy condenses classifications into the share that matches a reference classification. General diagnostic-methods considerations are that an overall accuracy measure can conceal different error patterns. Sensitivity describes identification among those with the target condition, while specificity describes correct classification among those without it. These measures, along with the underlying classification counts and the threshold defining a positive result, are not reported in the registry record. Therefore, the reported overall accuracy cannot establish how the assessments performed in each diagnostic category.

Doppler ultrasound is identified as the gold standard for the primary diagnostic question. In diagnostic research, a reference standard supplies the classification against which an index test is evaluated. The usefulness of an accuracy estimate depends on appropriate application and interpretation of that standard. The registry does not state who performed or interpreted the ultrasound, whether assessors were masked to the other test, what criteria defined adequate collateral supply, or how discrepant or indeterminate results were resolved. These details are relevant to understanding potential measurement and classification bias.

No inferential comparison between arms is posted in the registry record. The record contains arm-level estimates but no analysis records. A formal comparison would ordinarily report an effect measure for the difference or another prespecified contrast, its uncertainty, and any associated test. The separate confidence intervals do not provide that contrast directly. It would be inappropriate to treat the visual relationship between the intervals as a substitute for the missing comparison. No adjusted analysis or covariate model is described in the registry record.

Planning illustrations based on an observed effect are not evidence of the trial’s original design calculations. The stated illustration assumptions are a two-sided significance level of 0.05, equal allocation between the arms, and a normal approximation, with the effect to detect expressed as a fraction of the observed effect. Such illustrations describe hypothetical planning under those assumptions. They are not observed power and do not establish that the trial was designed around those assumptions. The protocol’s sample-size justification and original statistical plan are not reported in the registry record.

Over-dispersion is a general concern when count outcomes vary more than a chosen statistical model expects. It can make standard errors too small and confidence intervals too narrow if not accounted for. The primary endpoint is recorded as a percentage for diagnostic accuracy, and no count-model analysis is posted; over-dispersion is therefore not identified as a reported feature of this analysis. More generally, diagnostic classification data require methods suited to their structure, but the registry does not report sufficient analytic detail to assess modeling choices.

The posted results are 81.7% accuracy for the Modified Allen's Test arm, with a 95% confidence interval from 76.0% to 86.6%, and 91.8% for the iPhone Assessment arm, with a 95% confidence interval from 87.3% to 95.1%. The registry does not describe whether these estimates were adjusted, whether any clustering or site effects were considered, or how the confidence intervals were constructed. Those methods should not be inferred from the reported values alone.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists one outcome measure and no posted secondary outcomes. It does not report a multiplicity plan, a hierarchy of outcomes, or adjustments for multiple comparisons. The registry record also has no posted analysis records, so it does not provide an analysis plan specifying which contrasts or tests were intended. Whether the trial protocol prespecified additional outcomes or analyses is not reported in the registry record.

Missing-data handling is not described. The primary population is identified as intention to treat, but the record does not report how participants with missing assessments, indeterminate test results, or unavailable reference-standard results were handled. It also does not report the number of participants with evaluable results for the accuracy calculation or any imputation method. The arm sizes of 219 each should not be read as proof that every participant contributed to the endpoint.

Because diagnostic accuracy depends on classifying index-test and reference-standard results, absent or unclear results can affect the denominator and the interpretation of the percentage. The registry does not state whether missing results were excluded, classified as errors, or addressed by another approach. It also does not provide the underlying classification counts needed to examine these possibilities.

The posted data therefore support interpretation of the reported arm-specific accuracy estimates and their confidence intervals, but not an assessment of multiplicity control or the consequences of missing endpoint data. The statistical analysis plan, missing-data assumptions, and any sensitivity analyses are not reported in the registry record.

11. 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 serious adverse event fields report zero serious adverse events in the Modified Allen's Test arm and zero in the iPhone Assessment arm. The record also reports zero deaths in each arm. These are the posted counts; they do not establish that no adverse events of any kind occurred, or that no events were missed or unrecorded.

The registry lists 219 participants at risk for serious adverse events in each arm, and 219 at risk for other adverse events in each arm. The number affected by other adverse events is also recorded as zero for each arm. No serious adverse event terms are listed. These safety figures should be read as the registry’s recorded event counts and denominators, not as estimates of comparative risk with quantified uncertainty.

No formal between-arm safety analysis or confidence interval for adverse events is posted. With zero recorded events, the registry counts alone cannot establish that the true risk is zero, nor can they precisely quantify rare-event risks. The record does not describe the method or observation procedures for adverse-event collection. These figures provide limited descriptive safety information rather than a complete assessment of safety.

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

The registry reports overall diagnostic accuracy of 81.7% for the Modified Allen's Test arm and 91.8% for the iPhone Assessment arm, each with a 95% confidence interval. The iPhone arm’s point estimate is higher. The posted results do not include a formal statistical comparison between these arms, so the size and uncertainty of an arm-to-arm contrast are not reported in the registry record.

These findings concern agreement with Doppler ultrasound as the reference standard for assessment of collateral blood supply to the hand. They do not show that the assessments predict clinical complications, alter treatment decisions, or improve patient outcomes. Nor do the accuracy estimates establish performance in populations or settings beyond the trial’s recorded population. Eligibility criteria and the detailed recruitment setting are not reported in the registry record.

Overall accuracy does not reveal the balance of missed cases and incorrect positive classifications. Without sensitivity, specificity, predictive values, and classification counts, readers cannot determine how errors were distributed or how the assessments performed across diagnostic categories. The definition of a positive result and the denominator used for accuracy are also not reported.

The separate confidence intervals communicate uncertainty in the arm-specific estimates, not a direct interval for their difference. The record does not state a threshold for clinically meaningful accuracy or for a meaningful contrast between the strategies. Accordingly, the numerical pattern should be interpreted descriptively rather than as evidence of a confirmed advantage or equivalence.

The safety record contains zero reported serious adverse events and zero deaths in both arms. This is a description of the registry entries, not proof of zero risk. Taken together, the record offers arm-specific diagnostic accuracy estimates and limited safety counts, while leaving important details about classification, missing data, and comparative inference unreported.

13. Why This Trial Matters Statistically

This trial addresses a practical measurement question: whether a smartphone application can assess wrist artery patency and collateral blood supply in comparison with an established clinical test and Doppler ultrasound. Its statistical interest lies in distinguishing a diagnostic-accuracy estimate from a treatment-effect estimate. The primary result concerns classification agreement with a reference standard, rather than a change in health outcomes.

The registry presents accuracy with confidence intervals for both randomized arms. That format allows readers to consider each estimate’s precision, but it does not answer whether the difference between strategies is statistically compatible with chance variation. A direct contrast and its uncertainty are needed for that inference. The registry has no posted analysis records, making the distinction between descriptive arm results and a formal comparative conclusion especially important.

The endpoint also illustrates why a single summary measure can be incomplete. Overall accuracy aggregates correct classifications, but does not show which errors occurred. Diagnostic studies are more fully interpretable when they report classification counts and measures that distinguish performance among those with and without the target condition. Those details are not reported in the registry record.

Finally, this record demonstrates that a randomized design does not by itself answer every methodological question. Information about allocation implementation, reference-standard procedures, missing results, and statistical methods is needed to evaluate potential bias and precision. The posted numbers remain useful as registry-reported estimates, but their scope is bounded by the outcome and analysis details available in the record.

14. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

Several trial-specific details needed to assess the diagnostic results are not reported in the registry record. The operational definition of competency of collateral blood supply, thresholds used to classify tests, and exact calculation of overall accuracy are unspecified. The underlying counts for correct and incorrect classifications are also absent. Consequently, the reported percentages cannot be independently reconstructed from the registry entries, and the patterns of false-positive and false-negative results cannot be assessed.

The record identifies Doppler ultrasound as the gold standard but does not describe how it was performed or interpreted, whether the ultrasound assessor was masked to the arm or index-test result, or how discordant assessments were resolved. The procedures and timing for the Modified Allen Test and smartphone assessment are not detailed enough in the registry record to evaluate consistency of test administration. Although the timeframe is listed as 1 year, the relation between this timeframe and the diagnostic measurement schedule is not explained.

The registry reports no analysis records and no formal between-arm contrast. It does not specify the confidence-interval method, adjustment strategy, or missing-data approach. The intention-to-treat population label does not resolve these gaps because the treatment of missing, unusable, or indeterminate diagnostic results is not described. The record also does not report a protocol-based sample-size justification or statistical analysis plan.

Finally, the trial is recorded as unmasked, and the registry does not clarify whether outcome assessors or reference-standard readers were masked. This is relevant in diagnostic assessment because interpretation may be influenced by knowledge of other results. The enrollment and arm sizes alone do not establish how many participants had complete paired diagnostic information. These limitations restrict interpretation of the reported accuracy estimates and the strength of any comparative conclusion.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Utility of a Smart Phone Application in Assessing Radial Artery Patency - the CAPITAL iRADIAL Study — NCT02519491.
• Registry record last updated 2023-02-08; status last verified 2022-10.
• Results first posted 2021-05-04.
• 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.

18. Record Summary

NCT02519491, the CAPITAL iRADIAL Study, was a completed randomized, parallel-group, unmasked trial sponsored by Ottawa Heart Institute Research Corporation. The registry lists coronary heart disease as the condition, enrollment of 438, and study dates from 2015-07 to primary completion in 2018-04. The arms were the Modified Allen's Test and iPhone Assessment, with 219 participants recorded in each.

The primary outcome was diagnostic accuracy against Doppler ultrasound for assessing competency of collateral blood supply to the hand. The Modified Allen's Test arm had reported accuracy of 81.7% (95% confidence interval 76.0% to 86.6%); the iPhone Assessment arm had 91.8% (95% confidence interval 87.3% to 95.1%). No formal between-arm analysis or secondary outcomes are posted. Serious adverse events and deaths are recorded as zero in both arms. Important details about diagnostic classification, missing data, analysis methods, and the statistical plan are not reported in the registry record.