Clinical Validation for SpO2 Function of Masimo INVSENSOR00069: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Clinical Validation for SpO2 Function of Masimo INVSENSOR00069 (NCT06120777), a na na study. The primary outcome is SpO2 Accuracy of Masimo INVSENSOR00069.

2. Clinical Question

3. Trial Design

This was a completed, single-group device validation study sponsored by Masimo Corporation. The registry lists the condition as Healthy, enrollment as 89, and the intervention as Masimo INVSENSOR00069. The study started on 2023-11-01, with primary completion on 2023-12-12. The record gives no phase designation beyond NA.

The design is listed as single group, with allocation NA and masking NONE. The arm description says subjects enrolled into the test group and participating in data collection had the device. There is no randomized control arm or like-for-like control group described. Instead, the principal comparison is within the measurement process: device SpO2 readings are assessed against contemporaneous SaO2 reference values from ABG analysis.

The registry gives a primary outcome timeframe of 1-3 hours. That indicates the stated assessment window, but the record does not provide a more detailed schedule of measurement collection, the number of paired readings per subject, or how readings were selected for the endpoint. It also does not report the setting or additional eligibility criteria beyond the healthy condition.

The record identifies 10 subjects as not analyzed because of an incorrect sensor and 3 because they did not desaturate to the lowest bin, 70-80%. It does not say whether these categories overlap, how many observations were retained, or whether any other exclusions occurred. Enrollment alone should therefore not be treated as the analysis population for the primary endpoint.

Although the study is marked completed, the numerical primary accuracy result is not reported in the registry record. The study dates and completion status establish that the registry lists the evaluation as completed; they do not substitute for the missing endpoint estimate.

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: 10 subjects not analyzed due to incorrect sensor 3 subjects not analyzed due to not desaturating to the lowest bin (70-80%)

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 SpO2 Accuracy of Masimo INVSENSOR00069, measured over 1-3 hours.

The primary endpoint is described as the accuracy root mean-squared difference, abbreviated Arms, between SpO2 measured by INVSENSOR00069 and reference SaO2 values. It is intended to express the typical magnitude of disagreement between the device readings and the reference readings, with larger discrepancies contributing more heavily because the differences are squared before averaging. Taking a root returns the measure to the original saturation scale. This is a general explanation of the statistic.

For a device-validation question, comparing contemporaneous readings matters because the target is agreement with a reference measurement taken at the same time, rather than change over a longer period. The registry identifies arterial blood gas analysis as the source of reference values and says the measure was determined in accordance with ISO-80601-2-61. The registry record does not supply additional details about the implementation of that standard.

The endpoint is not a count of successful readings or a proportion of subjects within a specified tolerance. No tolerance threshold is stated in the endpoint description. Nor does the record provide an Arms estimate, confidence interval, or other numerical result. The listed unit is “% of oxygen saturated hemoglobin,” but without the endpoint value and its uncertainty, the magnitude and precision of agreement cannot be judged.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary measure is an accuracy root mean-squared difference between SpO2 device readings and SaO2 reference readings. In general, this type of measure starts with paired differences: each device reading is compared with its corresponding reference value. The differences are squared, averaged, and square-rooted. Squaring prevents positive and negative differences from cancelling and makes larger discrepancies contribute more to the final summary. The square root restores the measure to the scale of the original readings. This explanation describes the general statistic; the registry does not report the numerical result or the detailed computational steps used in the analysis.

Pairing is central to this endpoint. A device reading should be compared with the corresponding reference measurement, rather than with an unrelated reference value, because the question concerns agreement at the time of measurement. The registry says the reference was contemporaneous ABG analysis. It does not state how many paired observations were obtained, whether multiple readings per subject were included, or how repeated readings were handled statistically. Those details matter because observations from the same subject may be related rather than independent.

Root-mean-squared difference is a measure of agreement, not merely correlation. In general, two measurement methods can move together while still differing systematically in their values. A measure based on paired differences directly summarizes discrepancy, but it does not by itself describe direction: squaring removes the sign of each difference. It also does not show the full distribution of errors. A plot or additional summaries of signed differences could help describe direction, range, and possible patterns, but those results are not reported in the registry record.

The measure is described as being determined in accordance with ISO-80601-2-61. The record does not state the protocol details needed to assess that implementation, such as the reading-level inclusion rules or the exact aggregation procedure. The formula as printed in the record is not accompanied by a worked example or an endpoint value. It is therefore safest to interpret the named measure as the stated root-mean-squared accuracy difference, without reconstructing unreported analysis details.

Precision is a separate issue from the point estimate. In general, a confidence interval quantifies uncertainty around an estimated quantity under a statistical model and its assumptions. A narrow interval indicates greater precision than a wide interval, but precision does not establish that the underlying measurement process is unbiased or clinically adequate. The registry entry does not report a confidence interval for Arms, so the precision of the intended primary estimate is not available. It also does not report a standard error or another uncertainty measure.

The registry lists a single-group design, no masking, and no allocation designation. These design labels do not describe a randomized contrast between groups. The device is assessed against an external reference measurement, rather than against an assigned control intervention. That makes the endpoint a method-comparison question. It also means that group-comparison concepts such as a treatment effect between randomized arms do not describe the primary question in this record.

The record states that the outcome timeframe was 1-3 hours. A timeframe identifies when the endpoint was intended to be assessed, but it does not alone establish how often measurements were taken or whether all measurements within that window contributed. The collection schedule, subject-level aggregation, and any rules for handling measurements outside the target range are not reported in the registry record.

Sample-size planning and observed precision should not be conflated. The planning assumptions described for the statistical illustrations are a two-sided significance level of 0.05, equal allocation between two hypothetical arms, and a normal approximation; the target effect is expressed as a stated fraction of an observed effect. These are planning illustrations based on an observed effect, not the trial’s own design description. They are not observed power, and they do not establish that the single-group study was designed with those assumptions. No sample-size justification for this trial is reported in the registry record. Since the numerical endpoint result and the planning illustration values are not included in the record summarized by the trial details, no numerical planning estimate can be interpreted.

More generally, power is a design-stage probability calculated under assumptions about an effect, variability, significance level, and analysis. It is not a measure of whether a completed study has produced convincing evidence, nor does a retrospective calculation reveal the actual precision of an estimate. For a completed study, the endpoint estimate and its uncertainty are more informative for interpretation than a post hoc power calculation. Here, those endpoint results are not reported in the registry record.

Over-dispersion is a general concern when count outcomes vary more than a simple statistical model expects; it can make standard errors too small if ignored. The primary endpoint in this record is described as a measurement-difference summary, not as a count outcome. The registry does not identify a count model or report an over-dispersion assessment. The general concept therefore should not be mistaken for a method known to have been used in this study.

Likewise, the record does not describe an interim analysis, stopping rule, covariate adjustment, or a prespecified statistical analysis plan. Their absence from the registry entry does not prove that none occurred; each is not reported in the registry record. The same distinction applies to the handling of dependence across repeated measurements and to any sensitivity analyses.

10. Multiplicity, Missing Data and Analysis Populations

The record lists one primary outcome measure, the accuracy root mean-squared difference between device SpO2 and reference SaO2. It lists zero analyses, but this does not reveal the full analysis history or establish that no additional statistical comparisons were conducted. A multiplicity plan describing how multiple outcomes, subgroups, time points, or analyses were addressed is not reported in the registry record.

Missingness and exclusions are particularly relevant because the primary outcome description states that 10 subjects were not analyzed due to an incorrect sensor and 3 were not analyzed because they did not desaturate to the lowest bin, 70-80%. The record does not specify whether these categories overlap. It also does not explain whether the subjects were excluded before measurement, whether some readings were retained, or whether there were other missing or unusable observations. The final analysis population and the number of paired readings are not reported in the registry record.

These details affect what population the endpoint represents. If the analysis excludes subjects who do not reach a specified saturation range, the resulting estimate may describe readings in the analyzed range rather than performance across all enrolled subjects or all possible conditions. That is a general interpretive point; the record does not provide enough detail to characterize the actual data coverage or the consequences of the exclusions.

The registry does not report an intention-to-analyze approach, a missing-data method, imputation, or sensitivity analysis. It also does not describe a protocol for handling failed sensor readings beyond identifying the incorrect-sensor exclusion. Accordingly, neither the impact of missing data nor the robustness of the primary analysis can be assessed from 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.

Most frequent non-serious adverse event terms

The registry lists 0 serious adverse events among 86 subjects at risk in the Masimo INVSENSOR00069 group. It lists 9 other adverse events among 86 at risk. These are the posted counts; the record does not provide event descriptions, severity details for the other events, timing, or an assessment of whether the events were related to the device.

“At risk” is the denominator recorded for the adverse-event entries. It should not be assumed to be the same as the enrollment count or the primary endpoint analysis population. The record does not explain why the adverse-event denominator differs from the enrollment figure or provide further details about the safety population.

Zero reported serious adverse events means the registry entry records none in this category for the stated group and denominator. It does not establish that the device has no safety risks, and the other adverse-event count cannot be interpreted without information about event types, ascertainment, and attribution. The study is described as a single-group validation in healthy subjects, not as a comparative safety trial. No comparative risk estimate can be derived from these entries.

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

The record establishes that Masimo INVSENSOR00069 was evaluated in a completed, single-group study of healthy subjects, with SpO2 readings assessed against contemporaneous SaO2 values from ABG analysis. It identifies the intended primary statistic as an accuracy root mean-squared difference and gives an endpoint timeframe of 1-3 hours. These details explain the measurement question and the planned endpoint concept.

The record does not report the numerical primary endpoint result. It therefore does not show the size of the observed discrepancy between device and reference readings. With no reported confidence interval or comparable uncertainty measure, it also does not show how precise an estimate might be. No conclusion about the device’s measured accuracy can be drawn from the posted endpoint information alone.

The endpoint, even if its estimate were available, would summarize a particular set of paired readings under the study conditions. It would not automatically describe every subject population, every clinical setting, or every range of oxygen saturation. The record identifies healthy subjects and explains that some subjects were not analyzed because of an incorrect sensor or failure to desaturate to the stated lowest bin. It does not provide enough information to judge how representative the analyzed readings were.

The reported adverse-event entries are also limited: the registry lists 0 serious adverse events and 9 other adverse events among 86 at risk, without event descriptions or attribution. These entries do not provide a comparative safety estimate.

In short, the registry supports a description of the study question, design, and certain enrollment and safety counts. It does not provide the primary numerical finding needed to quantify agreement, nor the statistical detail needed to assess its precision, distribution, or generalizability.

13. Why This Trial Matters Statistically

This study illustrates a distinction between a device-validation question and a randomized treatment comparison. The central task is to compare paired readings from a device with a reference method, not to estimate a difference between assigned intervention groups. Understanding that structure helps readers focus on agreement, measurement error, and the quality of pairing rather than on between-arm treatment effects.

The choice of an accuracy root mean-squared difference also highlights why the endpoint definition matters. A root-mean-squared summary weights larger errors more heavily and removes the direction of individual differences. It can be informative as an overall discrepancy measure, but it cannot by itself show systematic over-reading or under-reading, variation across the reading range, or the influence of individual outliers. Interpreting such a measure requires the estimate and ideally complementary information about uncertainty and error patterns.

The exclusions noted in the registry make analysis-population reporting important. The record identifies subjects not analyzed for specific reasons, including incorrect sensor use and failure to reach the lowest stated bin. Without the final analysis population, reading-level counts, or handling rules, readers cannot determine precisely which measurements contributed to the primary endpoint.

Finally, a completed study record is not the same as a complete statistical report. The registry supplies a clear account of the intended question and some design information, but not the primary endpoint value or its precision. This demonstrates why endpoint definitions, analysis populations, numerical estimates, and uncertainty measures are all needed for a statistical interpretation.

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.

The main limitation for interpretation is that the numerical Arms result is not reported in the registry record. Without that estimate, the magnitude of device-reference disagreement cannot be evaluated. The record also does not report a confidence interval or other measure of precision, so uncertainty around the endpoint cannot be assessed.

The study is described as single group, with no masking and no randomized allocation. Its reference comparison is contemporaneous ABG analysis rather than a randomized control arm. This design is appropriate to describe as a measurement-validation setup, but it does not support a randomized between-group contrast. The registry does not report further procedures that might clarify how measurement collection or reading assessment was standardized.

The population is listed as healthy. The record does not report demographic characteristics, additional eligibility criteria, setting, or details about the measurement range represented in the analyzed data. It identifies 10 subjects not analyzed because of an incorrect sensor and 3 because they did not desaturate to the lowest bin, 70-80%, but does not state whether those groups overlap or give the final analysis population. The number of paired measurements and the handling of repeated readings are also not reported.

Further statistical limitations arise from missing detail. The analysis plan, treatment of missing readings, multiplicity approach, and any sensitivity analyses are not reported in the registry record. The primary endpoint is said to follow ISO-80601-2-61, but the registry does not give implementation details beyond that statement. These gaps limit assessment of reproducibility and the scope of the findings.

The safety information is limited to event counts and the recorded at-risk denominator. Event types and attribution are not reported, so those entries cannot characterize the nature of the adverse events or establish a comparative safety profile.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Clinical Validation for SpO2 Function of Masimo INVSENSOR00069 — NCT06120777.
• Registry record last updated 2024-11-27; status last verified 2024-11.
• Results first posted 2024-11-27.
• 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

NCT06120777 is a completed, single-group study sponsored by Masimo Corporation that evaluated the SpO2 function of Masimo INVSENSOR00069 in healthy subjects. The record lists enrollment of 89, no masking, and an endpoint timeframe of 1-3 hours. The intended primary measure was the accuracy root mean-squared difference between device SpO2 readings and contemporaneous SaO2 reference values from ABG analysis.

The record identifies 10 subjects not analyzed because of an incorrect sensor and 3 not analyzed because they did not desaturate to the lowest bin, 70-80%. It does not report the final primary analysis population or numerical Arms result. The safety entries list 0 serious adverse events and 9 other adverse events among 86 at risk. Without the endpoint estimate and its uncertainty, the degree and precision of measurement agreement cannot be assessed from the registry record.