Accuracy Assessment of an Automatic Blood Pressure Measurement Device in Adult Volunteers: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Accuracy Assessment of an Automatic Blood Pressure Measurement Device in Adult Volunteers (NCT02267577), a na na study. The primary outcome is Accuracy of Blood Pressure Measurement by Sphygmo Relative to the Clinical Standard..

2. Clinical Question

This is a device-measurement question rather than a comparison of treatment effects. Each participant's blood pressure was measured using both devices, with the participant seated and the arm at heart level. The study description says measurements were repeated during a single visit, and that the mean readings were used for analysis. The intended use of Sphygmo is monitoring and diagnosis of pre-eclampsia in pregnant women, particularly in low-resource settings. However, the registry describes the participants as adult volunteers and does not establish that the study assessed the device in pregnant women or people with pre-eclampsia. The record therefore describes an initial measurement-accuracy assessment, not evidence about clinical outcomes or performance in the intended patient population.

3. Trial Design

The registry describes a completed, single-group device study with enrollment of 90 adult volunteers. It lists no trial phase, no allocation, and no masking. Because the same participants were to have their blood pressure measured with Sphygmo and the GE Dinamap ProCare monitor, the intended comparison is within participants rather than between randomized treatment groups. The registry's allocation field is recorded as NA; this does not describe a randomized allocation process.

During a single visit, a participant sat in a comfortable chair with the arm at heart level. Researchers measured arm circumference and placed a cuff on the arm. The study description says blood pressure was measured by both devices, with up to 9 measurements and a waiting period of 45-60 seconds between measurements. The primary endpoint description says blood pressure was measured an average of 9 times per participant, and that the mean of the measurements was recorded for analysis. It describes the measurement period as approximately 30-45 minutes.

The registry gives a start date of 2015-01, a primary completion date of 2015-10, and a completed status. It says that 9 participants of the original 81 who completed the study were excluded because the observer reported difficulty hearing, movement, or uncertainty with the measurement. The record does not report a detailed participant flow or explain how those reasons were applied to each measurement.

The stated purpose was to use the results to optimize the Sphygmo blood-pressure detection algorithm and further develop the device. No outcome results for the primary measurement comparison are reported in the registry record. The record also does not provide a detailed protocol, masking procedures beyond NONE, or a more specific account of the clinical reference measurement.

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: Data from nine participants of the original 81 who completed the study was excluded on account of the observer reporting a difficulty in hearing, movement, or uncertainty with the measurement.

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 Accuracy of Blood Pressure Measurement by Sphygmo Relative to the Clinical Standard., measured over Blood pressure was measured an average of 9 times for each participant during their single visit and the mean of these measurements was recorded and used for analysis. The measurement period lasted approximately 30-45 minutes..

The primary endpoint is described as the mean difference, in mmHg, between systolic and diastolic blood-pressure readings from Sphygmo and the reference measurement. It is a continuous outcome: the result is intended to express the average difference in the same units as blood pressure. The registry says that measurements were repeated during a participant's single visit and that their mean was used in the analysis.

A mean difference can indicate average directional disagreement. If the device tends to read higher than the reference, the signed difference would tend in one direction; if it tends to read lower, it would tend in the other. A mean close to zero can still coexist with substantial variation among individual measurements, because positive and negative differences can offset each other. For that reason, average bias alone does not fully describe how close the readings are for individual users or measurements.

Standard deviation is listed as the measure of dispersion. In general, dispersion describes how spread out measurements or differences are around their average. The registry record does not report the endpoint's numerical mean difference or standard deviation, nor does it report confidence intervals or an agreement analysis. Without those results, the direction, size, and precision of the observed average difference cannot be assessed from the registry record. The endpoint is relevant to device comparison, but its numerical result and the amount of individual-level variation would both be needed for a fuller statistical interpretation.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry describes a continuous endpoint based on a mean difference in blood-pressure readings. Understanding that endpoint requires distinguishing average bias from agreement, repeated measurements from independent participants, and an estimate from its uncertainty. The record does not report the numerical outcome values or a detailed statistical analysis plan, so the explanations below describe general statistical principles rather than additional analyses from this study.

What does a mean difference represent?

In general, a difference between a test device and a reference is calculated for matched measurements, with the sign depending on which reading is subtracted from which. The average of those differences describes whether the test device tends to read higher or lower on average. A difference near zero would mean little average directional bias, but it would not prove that individual readings are close. Differences that are positive for some observations and negative for others can cancel in the average. The registry record does not specify the subtraction order, so the direction represented by a positive or negative result is not defined there.

For this study, the endpoint description refers to mean systolic and diastolic differences in mmHg, and says participant measurements were averaged before analysis. The record does not state whether the reported difference would be calculated from each participant's device-specific mean, or give further detail on the pairing of observations. This makes the intended endpoint understandable at a high level, but leaves the exact calculation procedure unclear.

Why does variability matter?

In general, a standard deviation describes the spread of observations around their mean. For paired device comparisons, variability in the differences helps show whether the average represents fairly consistent behavior or masks wide disagreement for individual measurements. If repeated readings are averaged, the spread of those participant-level averages may differ from the spread of single readings. The registry lists standard deviation as the dispersion measure, but does not identify the precise set of values to which it applies or report a numerical result.

A confidence interval is a general way to express uncertainty around an estimated quantity. It gives a range of values compatible with the data and the statistical method, rather than a guarantee that the true value lies in that range. A narrow interval indicates greater precision under the method's assumptions; a wide interval indicates less precision. No confidence interval for the primary endpoint is reported in the registry record. Without an estimate and its uncertainty, the observed magnitude cannot be assessed.

How should repeated readings be understood?

Repeated measurements within a participant are usually correlated: readings from the same person share features and should not automatically be treated as independent observations. Averaging repeated measurements can provide a more stable participant-level value, but it changes the unit of analysis from a single reading to a summary for each participant. The study description says that the mean of the measurements was recorded and used for analysis. It also says measurements were taken up to 9 times, while the endpoint description states an average of 9 measurements per participant. The record does not explain how this distinction was handled in practice.

The waiting period between readings and the seated position with the arm at heart level describe elements of the measurement procedure. They do not, by themselves, establish that measurement order was randomized, that observers were masked, or that all conditions were standardized. Those details are not reported in the registry record. The masking field is NONE, and the allocation field is NA; the record does not describe an additional masking or allocation procedure.

What can a sample-size illustration tell a reader?

The planning illustration is described as using the observed effect, a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It varies the effect to detect as a fraction of the observed effect. The registry's description explicitly makes this a planning illustration, not the trial's own design. Because this was a single-group device comparison and the illustrated assumptions include equal allocation between two arms, those planning assumptions should not be confused with the study's participant-level measurement procedure.

A sample size calculated from an observed effect is a hypothetical planning aid. It is not evidence that the original study had a particular probability of detecting an effect, and it should not be called observed power. The registry record does not report a prospective sample-size justification for the study. The numerical estimates in the planning illustration are not restated; they do not replace missing endpoint results or provide a retrospective assessment of the study's precision.

What does the study design support?

Because both devices were used with the same participants, a paired comparison is conceptually appropriate: each participant can serve as their own reference point, reducing some between-person variation. General statistical practice would analyze the within-participant differences and account for the dependence created by repeated measurements. The registry record does not describe the exact model, assumptions, handling of repeated readings, or any covariate adjustment. It reports no analysis entries. These omissions mean that the statistical method used for the primary result cannot be reconstructed from the registry record.

Finally, a device-comparison study aims to characterize measurement performance, not estimate a randomized treatment effect. A mean difference, even if reported, would apply to the specific participant group, measurement setup, reference method, and summary procedure used. The registry record's lack of outcome estimates and analytic detail limits how much can be concluded about accuracy or generalizability.

10. Multiplicity, Missing Data and Analysis Populations

The registry record lists 2 outcome measures and 0 analyses, but does not describe a multiplicity plan. It identifies a primary endpoint concerning systolic and diastolic blood-pressure differences, yet does not clarify how these components were treated statistically, whether one was designated as primary over the other, or whether additional comparisons were planned. When several outcomes or analyses are examined, multiplicity can increase the chance that at least one result appears unusual by chance. In general, a prespecified analysis plan helps readers distinguish confirmatory questions from exploratory ones. Such a plan is not reported in the registry record.

The primary endpoint description says data from 9 participants of the original 81 who completed the study were excluded because the observer reported difficulty hearing, movement, or uncertainty with the measurement. The record does not give a detailed account of which readings or participants were affected, whether exclusions were decided before reviewing results, or how the exclusions affected the endpoint analysis. It also does not report an imputation method, a missing-data sensitivity analysis, or a complete analysis population definition.

Because this is a device-measurement study rather than a randomized treatment trial, conventional intention-to-treat terminology may not describe its analysis population well. The registry record does not state a named analysis population or explain whether all eligible measurements were included. The reported exclusions and the absence of a detailed missing-data strategy mean that the potential effect of incomplete or uncertain readings cannot be evaluated from the registry record. The record also does not provide enough information to determine whether exclusions could have made the analyzed measurements more favorable or less representative.

11. 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 0% reporting threshold.

Most frequent serious adverse event terms

Serious events with the highest arm-level rate: high blood pressure (1.2%).

The registry's adverse-event information lists the group as Adults Volunteers. It reports 1 serious event among 85 participants at risk for serious events, and 0 other, non-serious events among 85 participants at risk for those events. The listed serious-event term is High Blood Pressure, with a reported value of 1.2%. These are the figures recorded for safety; they do not describe whether the event was judged related to either device.

The death field is null. A null value does not establish that there were no deaths; the death count is not reported in the registry record. The record also does not provide event narratives, timing, severity details beyond the serious classification, or an explanation of how the adverse-event at-risk groups relate to the study's measurement analysis population.

These safety figures should be read as descriptive registry entries, not as a comparison of devices. The study is listed as single-group, and the adverse-event information does not assign events to a randomized device group. The counts therefore do not estimate a relative safety effect between Sphygmo and the GE Dinamap ProCare monitor. They also do not establish causation. For this device-accuracy study, the main endpoint is measurement difference; the safety entries are a separate record of reported events.

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

The study addresses an important technical question: how Sphygmo blood-pressure readings compare with a commercially available monitor under a specified measurement procedure. Its single-group design, with both devices used for the same adult volunteers, is suited in principle to a within-participant comparison. Repeated measurements and participant-level averages are part of the described procedure, and the planned endpoint is a mean difference in mmHg.

The registry record does not report the numerical primary endpoint results. It therefore cannot establish the size or direction of the average difference, how variable differences were, or how precise an estimate might be. No conclusion that Sphygmo was accurate or inaccurate follows from the study objective or measurement protocol alone. The safety entries likewise do not provide a device-to-device comparison and do not establish whether an event was caused by a device.

The stated intended use is monitoring and diagnosis of pre-eclampsia in pregnant women, especially in low-resource settings. The participants are described as healthy adult volunteers, and the registry does not report whether they were pregnant or had pre-eclampsia. The entry consequently does not establish performance in the intended-use population or clinical benefit in that setting. A measurement-device assessment is also not, by itself, evidence about diagnosis, treatment decisions, or health outcomes.

More generally, the interpretation of a device-comparison study depends on the reference method, the way readings are paired, the spread of individual differences, and the population and setting in which measurements are made. The registry record leaves some of these details unclear and does not report the primary results. The appropriate reading is therefore limited: the record documents the study's purpose and procedure, but does not provide enough outcome information to judge measured accuracy.

13. Why This Trial Matters Statistically

Blood-pressure devices are judged not only by whether their average reading resembles a reference, but also by how consistently they perform across measurements and people. This study is statistically relevant because its stated endpoint focuses on the mean difference between Sphygmo and a clinical reference, while the described procedure includes repeated measurements and participant-level averaging. These design features illustrate why a device comparison requires attention to both paired differences and their variability.

The study also highlights a distinction between development-stage testing and evidence in an intended-use population. The brief summary describes Sphygmo as a device being developed for monitoring and diagnosis of pre-eclampsia in pregnant women, particularly in low-resource settings. The registry describes adult volunteers and does not establish that the participants were pregnant or had pre-eclampsia. The question of whether a device's measurement performance carries over to its intended context is therefore important, but cannot be answered by the information in this record.

The record is also a useful example of why a study objective and a registry result are not interchangeable. The endpoint is named and its general calculation is described, yet the numerical difference and dispersion are not reported. Without those estimates, readers cannot assess either average bias or the precision of the comparison. A planning illustration based on an observed effect can help frame future study planning, but it cannot substitute for outcome reporting or serve as observed power.

Finally, clarity about the reference device, repeated-reading procedure, exclusions, and analysis population is central to reproducibility. The record contains some procedural details and notes exclusions, but leaves other aspects unspecified. Transparent reporting of these elements would help readers understand what the estimated difference represents and to whom it may apply.

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 registry record does not report numerical results for the primary endpoint, so the main measurement comparison cannot be evaluated. It also does not provide confidence intervals, an endpoint standard deviation, an individual-level agreement summary, or detailed analysis methods. As a result, neither the magnitude and precision of average bias nor the spread of differences is available for interpretation.

The participant description limits generalizability. The study is described as involving healthy adult volunteers, while the stated intended use is monitoring and diagnosis of pre-eclampsia in pregnant women. The record does not establish pregnancy status or pre-eclampsia status for participants. The registry also lists Pre Eclampsia as a condition, but does not explain how that listing relates to the volunteer population.

The comparator is described inconsistently: the intervention listing names the GE Dinamap ProCare automatic monitor, while the primary endpoint description refers to a “gold standard sphygmomanometer.” The record does not clarify whether those labels refer to the same device or method. It also does not specify the subtraction order for the mean difference, the precise pairing of repeated readings, or whether measurement order was controlled.

The repeated-measurement description says measurements were taken up to 9 times, while the endpoint description says an average of 9 measurements was recorded for each participant. The registry does not explain whether this reflects a planned average, actual variation in the number of readings, or another distinction. It reports that 9 participants of the original 81 who completed the study were excluded for observer-reported difficulty hearing, movement, or uncertainty, but provides no detailed participant flow, missing-data analysis, or sensitivity analysis.

Finally, the registry does not report a prespecified statistical analysis plan, sample-size justification, or multiplicity approach. These limitations make it difficult to assess the robustness, reproducibility, and intended scope of any analysis beyond the basic study description.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Accuracy Assessment of an Automatic Blood Pressure Measurement Device in Adult Volunteers — NCT02267577.
• Registry record last updated 2020-03-06; status last verified 2020-03.
• Results first posted 2019-02-15.
• 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

This completed, single-group device study evaluated Sphygmo automatic blood-pressure measurements against a commercially available monitor in adult volunteers. The registry reports enrollment of 90, with repeated readings during a single visit and use of participant-level mean measurements. The primary endpoint is described as mean systolic and diastolic difference in mmHg, with standard deviation as the dispersion measure.

The registry record does not report numerical results for the primary endpoint or a detailed statistical analysis. It notes that 9 participants of the original 81 who completed the study were excluded for observer-reported measurement difficulties. Safety entries list 1 serious event, termed High Blood Pressure, and 0 other events; the death count is not reported in the registry record. The study's stated purpose was to support further development of Sphygmo, but the available record does not establish its measured accuracy or performance in pregnant women or people with pre-eclampsia.