Clinical Performance of a New Implant System for Bone Conduction Hearing: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Clinical Performance of a New Implant System for Bone Conduction Hearing (NCT03086135), a na na study. The primary outcome is Hearing Performance: Threshold Audiometry, Pure Tone Average of 4 Frequencies, PTA4, Unaided Versus OSIA System at 3 Months.

2. Clinical Question

3. Trial Design

The registry describes a completed, single-group study of the Osia System, a bone-conduction hearing device. Enrollment was 51. The study’s stated purpose was to examine initial experience with implanting and fitting the system in patients with conductive, mixed, or single-sided sensorineural deafness. The registry lists the study as completed and records its start as 2017-03-09 and primary completion as 2018-10-30.

The design is not a randomized, parallel-group comparison. Allocation is recorded as NA, masking as none, and the study model as single group. The primary hearing assessment therefore compares measurements within the study’s unaided and aided conditions: baseline before surgery and the assessment with the Osia System at 3 months. That is a different question from whether the system performs better than a concurrently assigned control under the same study conditions. The record does not describe such a control comparison for the primary endpoint.

The primary endpoint is free-field threshold audiometry, expressed as the Pure Tone Average of four frequencies (PTA4). The registry specifies the mean of thresholds at 500, 1000, 2000 and 4000 Hz, with results expressed in dB. It describes the endpoint as the difference between unaided PTA4 and aided PTA4. The registry also lists secondary hearing assessments at other time points and measures of speech performance, hearing-related questionnaires, health utility, and comparison with a reference device. The primary endpoint’s observed mean, standard deviation, and analysis details are not reported in the registry record.

Because the primary results are not reported, the record does not allow a numerical account of the observed change or its precision. Its description of the design and planned assessments can clarify what the study sought to measure, but cannot substitute for the missing outcome estimates.

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 Hearing Performance: Threshold Audiometry, Pure Tone Average of 4 Frequencies, PTA4, Unaided Versus OSIA System at 3 Months, measured over Baseline before surgery, 3 months after surgery.

The primary endpoint is the change in hearing performance from the unaided condition before surgery to the aided condition with the Osia System at 3 months. It uses free-field threshold audiometry to measure PTA4, the mean threshold across 500, 1000, 2000 and 4000 Hz, in dB. A threshold represents the sound level at which a person can detect a sound under the test conditions; the registry describes a lower or more negative difference as more desirable because it reflects the ability to hear softer sounds.

PTA4 is used as a summary of hearing thresholds across specified frequencies. As a general statistical point, combining measurements into a prespecified average can make a continuous outcome easier to summarize than presenting each frequency separately. That summary does not show every aspect of hearing performance, and it should be interpreted alongside the test conditions and the other outcomes the study planned to assess.

The registry states that the endpoint is a mean and that its dispersion is to be described by a standard deviation. The standard deviation describes variability among observations; it is not a measure of uncertainty in the estimated mean. The registry record does not report the primary mean, standard deviation, participant-level analysis population, or an uncertainty interval. Consequently, the magnitude and precision of the primary endpoint result cannot be evaluated from the recorded outcome information.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry identifies the primary outcome as a continuous measure summarized by a mean and standard deviation. These statistics answer different descriptive questions. The mean represents the average value of the endpoint in the analyzed observations. The standard deviation describes how dispersed individual observations are around that average. Neither statistic is reported for the primary endpoint in the registry record, so the observed average and participant-level variability cannot be assessed.

What does the endpoint contrast represent?

The outcome is defined as the difference between unaided PTA4 and aided PTA4 with the Osia System. The measurements refer to baseline before surgery and the aided assessment at 3 months. The registry specifies that lower or more negative scores are more desirable. This direction should be kept in mind when reading any estimate: the sign has a stated interpretation, and it should not be reversed or treated as an absolute magnitude without reference to the registry’s definition.

As a general statistical explanation, a change score compares measurements from the same study condition across time or circumstances. It summarizes the observed contrast for the people included in the analysis. A change score does not, on its own, establish why the measurements differ. In a single-group study, there is no concurrent randomized group in the primary comparison to represent what would have happened under another condition over the same period. The record does not report a method for accounting for other influences on the contrast.

How should variability and uncertainty be distinguished?

General statistical interpretation separates variation among observations from uncertainty about an estimated mean. Standard deviation describes the former. A confidence interval is one common way to express the latter, conditional on the model and analysis used. A confidence interval’s width depends on the amount of information and the variability in the data, as well as the method used to calculate it. The registry record states that the endpoint is to be reported as a mean with standard deviation, but does not report a confidence interval, a standard error, or the primary endpoint values. It is therefore not possible to judge the precision of the primary estimate.

Statistical significance and practical importance are also different concepts. In general, a test of statistical significance assesses compatibility between the observed data and a specified null model. It does not, by itself, describe the size or practical relevance of an effect. The registry record does not report a primary hypothesis test, p-value, or decision threshold for the primary endpoint. No claim about statistical significance can be made from the outcome information recorded.

What does the single-group design imply?

The registry lists allocation as NA, masking as none, and the model as single group. The primary assessment compares unaided hearing before surgery with aided hearing at 3 months. This is not a randomized arm-to-arm contrast. In general, random assignment can help make groups comparable with respect to factors that influence outcomes, while a within-group contrast does not provide that same concurrent comparison. The registry record does not describe an adjustment strategy or other design feature that would support a causal interpretation of the observed contrast.

Masking is listed as none. The record does not explain whether or how outcome assessment was protected from knowledge of treatment condition. For a measured threshold outcome, interpretation depends on the test procedures and the analysis population as well as the stated endpoint. Details of the testing protocol beyond the free-field PTA4 description, and details of assessor procedures, are not reported in the registry record.

How should multiple assessments be understood?

The registry lists secondary outcomes at several time points, covering hearing thresholds, speech in quiet, speech recognition in noise, and questionnaire or health-utility measures. It also lists a threshold comparison with a reference device. These outcomes concern distinct measurements and assessment occasions. In general, testing many outcomes or time points can increase the chance of observing an apparently notable result by chance if analyses are interpreted without regard to multiplicity. Methods such as prespecifying a primary endpoint, adjusting for multiple testing, or treating secondary analyses as exploratory can address different aspects of that issue. The registry record does not report which multiplicity strategy, if any, was used.

Some measures may have distributions that do not fit a simple normal model. As a general point, over-dispersion means that observed variability is greater than a model assumes; ignoring it can make uncertainty appear too small. The registry does not report a distributional assessment or modeling approach for the primary outcome, so no claim can be made about whether such issues arose or how they were handled.

The record’s enrollment figure is not itself evidence about the number analyzed for the primary endpoint. Missing follow-up measurements, exclusions, or a change in analysis population could affect the observations used in an estimate. Those details are not reported in the registry record. The absence of these analysis details, together with the absence of endpoint estimates, limits statistical interpretation to the endpoint definition and study structure rather than a quantitative evaluation of results.

10. Confidence Intervals for Each Arm

No two-arm comparison is shown for this trial, but each arm's mean can be given with a t-based 95% confidence interval computed from the posted mean, spread and number of participants.

R code and output

arm <- c("Bone-conduction Hearing Device")
n   <- c(49)
m   <- c(-25.8)
sp  <- c(9.1)
sd  <- sp      # posted standard deviations
for (i in seq_along(n)) {
  se <- sd[i] / sqrt(n[i]); h <- qt(0.975, n[i] - 1) * se
  cat(sprintf("CARM|%s|%d|%.3f|%.3f|%.3f|%.3f\n", arm[i], n[i], m[i], sd[i], m[i] - h, m[i] + h))
}

CARM  Bone-conduction Hearing Device  49  -25.800  9.100  -28.414  -23.186

11. Multiplicity, Missing Data and Analysis Populations

The registry lists a primary hearing endpoint and multiple secondary outcomes. The secondary assessments include hearing thresholds at additional time points, speech in quiet, adaptive speech recognition in noise, hearing-related questionnaires, health utility, and a threshold comparison with a reference device. When many outcomes or time points are assessed, general statistical practice distinguishes a primary question from secondary or exploratory questions and considers whether any adjustment for multiple comparisons is appropriate. The registry record does not report a multiplicity strategy, a hierarchy among secondary outcomes, or how conclusions were adjusted for the number of analyses.

The registry lists 24 outcome measures and 156 analyses. Those counts indicate that the record encompasses multiple analyses, but do not specify which results were available, which analyses were prespecified, or how multiplicity was handled. The primary endpoint results themselves are not reported in the registry record. No unadjusted or adjusted test result for the primary outcome is recorded in the available outcome information.

Missing-data handling and analysis populations are also not reported in the registry record. The enrollment figure does not establish how many participants had usable baseline and follow-up measurements, whether all enrolled participants were included in the primary analysis, or how missing observations were treated. In general, missing follow-up data can affect both the estimate and its uncertainty, particularly when the chance of missingness is related to participant characteristics or outcomes. Without the analysis population and missing-data approach, the impact cannot be assessed.

The registry record also does not report a detailed statistical analysis plan, interim analyses, or a sample-size justification. No assumptions should be inferred about those features. Any planning illustration based on an observed effect is a hypothetical design calculation, not observed power and not a substitute for the trial’s own design rationale. The record does not provide the primary observed effect needed to interpret such a planning exercise quantitatively.

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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: wound infection (2.0%), infection and infestations (2.0%), atrial fibrilation (2.0%).

The registry’s safety figures list 2 serious adverse events affected among 51 at risk, 42 affected for other adverse events among 51 at risk, and 0 deaths. The listed serious-event terms include Wound infection, Infection and infestations, and Atrial fibrilation, each recorded as 2.0%. These are registry entries and should be read as reported rather than as a complete account of event timing, severity, or relationship to the device or procedure.

The record does not clarify whether the serious-event terms refer to overlapping categories, whether a person could contribute to more than one category, or how the term-level percentages were calculated. It also does not report event narratives, duration, resolution, or an attribution assessment. The broader other-adverse-event figure is not defined in greater detail in the registry record. Therefore, the serious and other adverse-event figures should not be combined, compared as if they were mutually exclusive, or interpreted as counts of distinct participants.

In general, adverse-event counts describe events recorded under the study’s monitoring and reporting procedures; they do not by themselves establish causation. The single-group design also means that these figures do not provide a randomized safety comparison. The registry record does not report enough detail to assess comparative risks or the completeness of event ascertainment.

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

The registry record establishes that the study examined the Osia System in a single-group setting and defined a primary hearing outcome based on free-field PTA4. The endpoint contrasts unaided hearing before surgery with aided hearing at 3 months. Its units, frequency components, and preferred direction are stated. However, the primary outcome estimate and its dispersion are not reported in the registry record. The observed change, and how precisely it was estimated, therefore cannot be described from the recorded primary results.

This distinction is important: a well-defined outcome is not the same as a reported outcome result. A study can specify an endpoint and assessment schedule without the registry record containing the corresponding numerical findings. The secondary outcome list likewise indicates assessments that were planned or registered, not the results observed for those measures.

The within-study comparison also has a limited interpretation. It compares unaided and aided conditions in a single-group study; it does not constitute a randomized comparison with a concurrent control group. As a general statistical principle, an observed before-and-after contrast can describe a change across conditions, but does not by itself identify the cause of that change. The record does not report analyses that would establish a causal effect or address alternative explanations.

The safety entries offer a separate, limited description: the registry records serious adverse events, other adverse events, and deaths. The available information does not establish whether particular events were caused by the device or procedure, and it does not describe overlap among categories. Safety counts should therefore be interpreted as registry reporting, not as proof of causation or a comparison of risks.

In practical statistical terms, the record supports understanding the question the study set out to examine and the structure of its primary endpoint. It does not support a quantitative conclusion about the size or precision of the primary hearing result, nor a controlled estimate of comparative benefit or risk. Those limitations arise from information not reported in the registry record, not from the endpoint definition itself.

14. Why This Trial Matters Statistically

This study is statistically informative as an example of how the design and outcome definition shape the question a result can answer. Its primary endpoint is a continuous hearing measure, PTA4, assessed in unaided and aided conditions. The registry describes the relevant frequencies, unit, timing, and direction of the difference. Those details matter because a numerical estimate is only interpretable when its scale and sign are clear.

The single-group design is equally important. The primary contrast is within the study rather than between randomized groups. As a general design principle, a within-group change can characterize how measured outcomes differ across conditions, while a concurrent comparison can help assess what would have occurred under an alternative condition. These designs answer different questions, so the registry’s structure should guide how any result is framed.

The record also highlights the distinction between registering an outcome and reporting its results. A clear endpoint definition does not supply an estimate, measure its uncertainty, or show how many observations contributed to the analysis. Here, the primary result and its variability are not reported in the registry record. This makes it important to avoid treating an outcome description as evidence of a particular magnitude or precision.

Finally, the range of secondary outcomes underscores the need for transparent analysis choices. Multiple measures and time points can serve different scientific purposes, but interpretation depends on knowing which questions were primary, how missing data were handled, and whether multiplicity was considered. Those analysis details are not reported in the registry record. The statistical significance of this study’s record is therefore in the careful separation of design, measurement, and available evidence: each can be described, but the absent primary estimate prevents an assessment of the result itself.

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.

The primary limitation for interpreting this study is that the registry record does not report the observed primary endpoint result. It defines PTA4 and the baseline-to-3-month contrast, but provides no primary mean, standard deviation, or confidence interval. The size and precision of the hearing change therefore cannot be evaluated from the registry’s outcome information.

The single-group design limits the comparison. The primary outcome contrasts unaided hearing before surgery with aided hearing at 3 months, rather than comparing randomized groups. This structure can describe a within-study difference, but does not by itself distinguish the effect of the system from other influences over time. The registry record does not report an analysis that addresses this limitation.

Important reporting details are absent. The analysis population, missing-data approach, statistical test or model, and any adjustments are not reported in the registry record. The record also does not explain whether the multiple outcomes and time points were handled under a prespecified multiplicity strategy, or provide a detailed statistical analysis plan. Without these details, the robustness of an endpoint estimate could not be assessed even if its value were available.

The safety record is also limited in detail. Although serious-event and other-event counts are listed, event narratives, attribution, timing, and overlap among categories are not reported. The registry does not describe the completeness of safety ascertainment or provide a concurrent control group for comparison.

These are limitations of what can be interpreted from the registry record. They do not establish that the underlying study lacked additional analyses or procedures; those details are simply not reported in the registry record.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Clinical Performance of a New Implant System for Bone Conduction Hearing — NCT03086135.
• Registry record last updated 2021-10-14; status last verified 2020-01.
• Results first posted 2020-02-05.
• 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

The registry describes a completed, single-group study of the Osia bone-conduction hearing system, with enrollment of 51 and no masking. The study’s primary endpoint is the difference in free-field PTA4 between unaided baseline before surgery and the aided assessment at 3 months. PTA4 is the mean threshold at 500, 1000, 2000 and 4000 Hz, expressed in dB; the registry describes a lower or more negative result as more desirable.

The registry record does not report the primary endpoint estimate, its standard deviation, or an uncertainty interval. It lists 24 outcome measures and 156 analyses, as well as serious and other adverse-event figures and 0 deaths. The design does not include a randomized control comparison for the primary endpoint. Missing-data methods, analysis populations, and multiplicity handling are not reported in the registry record, limiting interpretation of the planned outcomes and recorded safety information.