Clinical Performance, Safety and Patient Reported Outcomes of an Active Osseointegrated Steady-State Implant System: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Clinical Performance, Safety and Patient Reported Outcomes of an Active Osseointegrated Steady-State Implant System (NCT04041700), a na na study. The primary outcome is Change in Hearing Performance With the Osia 2 System at 3 Months Post-surgery Compared to Preoperative Unaided Condition Assessed Via Audiometric Thresholds.

2. Clinical Question

This is a within-participant question about change over time, not a randomized comparison of the device with another treatment. Such a comparison can describe the difference between the preoperative unaided assessment and the postoperative assessment, but by itself it cannot separate the device’s contribution from other influences that may change over time. The investigation also planned to collect objective and subjective hearing-performance information, quality-of-life measures and safety information. The record lists additional assessments of speech recognition, self-reported hearing outcomes, health-related quality of life and surgical information. However, it does not report their numerical findings in the primary-endpoint entry. Without the primary endpoint estimate, its uncertainty and the associated analysis details, the registry record cannot establish the size or precision of the change in audiometric thresholds.

3. Trial Design

The registry describes a prospective, multicentre, open clinical investigation of the Osia 2 system in adults with conductive hearing loss, mixed hearing loss or single-sided sensorineural deafness. Its model is single group. Allocation is listed as NA, and the record does not identify a randomized allocation or a parallel control arm. Masking is listed as none. The study phase is listed as NA.

The record reports enrollment of 29 participants, a start date of 2019-08-16, primary completion on 2020-06-19 and a completed status. It identifies Cochlear as the sponsor. These registry entries describe the investigation’s structure and timing; they do not indicate that all enrolled participants contributed data to every outcome. For the primary endpoint, the record specifically says data were received for 27 participants. It explains that one participant did not receive the sound processor due to the Covid-19 pandemic and one participant did not attend either the 4 weeks visit or the 3 months visit.

The primary comparison is between baseline before surgery and the assessment 3 months after surgery. The outcome is the mean free-field Pure Tone Average 4, defined in the record using 0.5, 1, 2 and 4 kHz. The planned unit is dB HL (decibel Hearing Level), with standard deviation as the dispersion measure. The record also lists outcomes at 4 weeks, 3 months and 6 months, as well as questionnaires and other hearing assessments. It does not report the detailed protocol for analysis, the handling of repeated assessments, or a statistical analysis plan. Because the investigation has one group and no concurrent control comparison, its design supports description of observed change over time but does not provide a randomized estimate against an alternative intervention.

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:

Analysis population for the primary outcome, as defined in the registry: Data was only received for 27 participants. One participant did not receive the sound processor due to the Covid-19 pandemic and one participant did not attend neither the 4 weeks visit nor the 3 months visit.

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 Change in Hearing Performance With the Osia 2 System at 3 Months Post-surgery Compared to Preoperative Unaided Condition Assessed Via Audiometric Thresholds, measured over Baseline before surgery, 3 months after surgery.

The primary endpoint is change in hearing performance at 3 months after surgery compared with the preoperative unaided condition, assessed using free-field audiometric thresholds. The measure is Pure Tone Average 4, described as the mean of thresholds at 0.5, 1, 2 and 4 kHz. Its unit is dB HL (decibel Hearing Level), and the registry identifies the endpoint summary as a mean with standard deviation.

Audiometric thresholds characterize the sound level at which a person detects tones under the assessment conditions. In general, combining thresholds across specified frequencies into an average can provide a concise summary of hearing sensitivity across those frequencies. The registry’s choice of a defined frequency average makes the endpoint a quantitative measure that can be compared between the preoperative unaided assessment and the postoperative assessment. The record does not specify a threshold for a clinically important change, nor does it state how the direction or size of change should be interpreted for this endpoint.

The registry identifies the two assessment periods but does not report the mean change, standard deviation of the change, confidence interval, or a statistical test for this endpoint. It also does not report the individual threshold data. Therefore, the endpoint definition alone describes what was intended to be measured; it does not show whether or how much the measured thresholds changed.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Statistical methods can clarify what an endpoint and its summaries would mean, but the registry record does not report the analysis procedures used for the primary outcome. The explanations below are general statistical context, not claims about an unreported analysis.

What does a mean describe?

For a continuous measurement such as an audiometric threshold, a mean is a summary of the observed values. In a change analysis, the mean change would summarize the participant-level differences between the postoperative assessment and the baseline assessment, if that is how the analysis were defined. The registry labels the primary parameter as a mean, but does not give the mean values or describe the exact calculation of change. It is therefore not possible to reconstruct or verify an estimate from the record.

A mean can be sensitive to unusually high or low observations. It also compresses variation across participants into a single value, so it does not show the full distribution of individual changes. A group mean does not imply that every participant changed by that amount, or even that individual changes all went in the same direction. Individual-level data are not reported in the registry information for this endpoint.

What does standard deviation describe?

In general, a standard deviation describes how spread out observations are around their mean. For an endpoint expressed as a mean, it can help readers understand variability among the measurements summarized. The registry identifies standard deviation as the dispersion measure, but does not provide its value or clearly state whether it refers to measurements at each time point or to participant-level changes. Those summaries answer different questions and should not be treated as interchangeable.

A standard deviation is not itself a measure of how precisely the group mean is estimated. Precision of an estimated mean is typically conveyed with a standard error or a confidence interval. A confidence interval describes a range of values compatible with the data and analysis assumptions; it helps communicate uncertainty rather than only the point estimate. No confidence interval or standard error for the primary endpoint is reported in the registry record.

Why does the comparison structure matter?

In a single-group before-and-after investigation, the same participants may be measured at multiple times. A within-participant analysis can account for the fact that repeated measurements on the same person are related. A comparison that instead treats observations at different times as independent may represent uncertainty differently. The registry does not say which approach was used, how repeated measurements were modeled, or whether a model adjusted for any other factors.

More broadly, an observed before-and-after difference includes any changes that occur over the interval, not only changes attributable to the intervention. A concurrent comparison group can help distinguish intervention-related change from influences shared over time, although the design and analysis still matter. This investigation is registered as single group, with no randomized comparison. Its structure therefore supports reporting change over time but does not identify a randomized contrast against a control intervention.

What would a confidence interval add?

As a general matter, a confidence interval around a mean change would show the range of estimates reasonably compatible with the data under the specified statistical procedure. A narrow interval indicates greater statistical precision than a wide interval, but precision does not establish clinical importance. Interpretation also depends on the scale, the direction of favorable change, and a meaningful-change threshold. The registry record does not report a threshold for clinically important change or a confidence interval for the primary endpoint.

A p-value, if reported, would address compatibility between the observed data and a particular null hypothesis under specified assumptions. It would not measure the magnitude or clinical importance of a change, and it would not establish that the device caused the change. The registry record does not report a primary-endpoint p-value or the test used. Without the estimate and its uncertainty, statistical significance cannot be assessed from this record.

How should planning calculations be understood?

In general, sample-size planning calculations estimate how many participants may be needed under stated assumptions about an effect, variability, allocation and statistical procedure. The registry’s planning illustration is described as using a two-sided significance level of 0.05, equal allocation between two arms and a normal approximation, with the effect to detect expressed as a stated fraction of the observed effect. It specifies participants per arm for non-time-to-event outcomes. These are planning assumptions, not evidence about the investigation’s observed precision or power.

The planning illustration should be distinguished from the registered design, which is a single-group investigation rather than an equal-allocation comparison between two arms. The record does not provide the resulting planning values in the information describing this endpoint, nor does it report the observed effect needed to interpret the illustration. Such calculations cannot substitute for a reported endpoint estimate, confidence interval or analysis.

Over-dispersion is a general concern for count outcomes when variability exceeds what a simple count model assumes. It is not a description of the primary endpoint, which is continuous, and the registry does not report an over-dispersion analysis. This distinction illustrates why the statistical method should match the outcome type and data structure rather than be inferred from a general methods description.

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("Osia 2 System")
n   <- c(27)
m   <- c(-27.3)
sp  <- c(8.7)
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  Osia 2 System  27  -27.300  8.700  -30.742  -23.858

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 32 outcome measures, including objective hearing assessments, speech-recognition measures, patient-reported questionnaires, health-related quality of life and surgical information. When a study examines many outcomes or time points, the chance of finding at least one apparently unusual result can increase if each is tested separately. General approaches to this issue include defining a primary outcome in advance, controlling the overall error rate, or clearly labeling analyses as exploratory. The registry record does not report a multiplicity strategy, a hierarchy among outcomes beyond identifying the primary endpoint, or which outcomes were formally tested.

The record lists 0 analyses. It therefore does not provide analysis-level estimates, confidence intervals or p-values for assessing the listed outcomes. This absence does not show that a particular outcome had no change; it means the registry record does not report the corresponding analysis. The planned primary endpoint is clearly described, but the result itself and statistical analysis plan are not reported in the registry record.

Missing data are relevant to the primary endpoint. The record says data were received for 27 participants and identifies one participant who did not receive the sound processor due to the Covid-19 pandemic and one participant who did not attend either the 4 weeks visit or the 3 months visit. The record does not describe imputation, a missing-data model, sensitivity analyses, or the analysis population used. It also does not specify how missing observations for other outcomes were handled.

In general, conclusions from a complete-case analysis can be affected if participants with missing assessments differ from those with observed data. Methods for addressing missingness rely on assumptions, which should be stated and examined where possible. Because the registry record does not explain the analysis approach, the impact of missing primary-endpoint observations on the estimate cannot be evaluated from the posted information.

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: gastroenteritis (3.4%), infected skin/ wound with subcutaneous collection (3.4%), infected wound/possible flap compromise (3.4%), preoperative bradycardia (3.4%).

For the Osia 2 System group, the registry reports 29 participants at risk for serious adverse events and 3 participants affected. For other adverse events, it reports 29 at risk and 21 affected. The reported death count is 0. These are the safety counts recorded for the study group; they are not a comparison with an untreated or alternative-treatment group.

The record lists four top serious adverse-event terms, each with a reported frequency of 3.4%: Gastroenteritis; Infected Skin/ Wound With Subcutaneous Collection; Infected Wound/Possible Flap Compromise; and Preoperative Bradycardia. These terms describe events recorded in the registry. Their listing does not establish that the device caused them. The record does not provide event timing, duration, outcome, relatedness assessment or participant-level detail for these terms.

Adverse-event counts are not a measure of efficacy, and a group-level safety count alone cannot determine comparative risk. The record does not report a control group or an exposure-adjusted analysis. The denominator and time at risk should therefore be read in the context of the registered single-group investigation, rather than as evidence of a rate relative to another treatment.

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

The registry record establishes that a prospective, multicentre, open investigation was conducted in a single group and identifies a preoperative-to-postoperative audiometric endpoint. It also reports safety counts and lists additional planned hearing and patient-reported outcomes. Those details describe the scope and structure of the investigation. They do not, on their own, demonstrate a particular amount or direction of hearing change.

The primary endpoint is defined as a mean free-field Pure Tone Average 4 at baseline and 3 months after surgery, in dB HL. However, no endpoint mean, change estimate, confidence interval, standard deviation value or p-value is reported in the registry record. Without these results, the magnitude, precision and statistical compatibility of any change cannot be judged from the posted information. No conclusion about whether the change met a clinically meaningful threshold is possible; such a threshold is also not reported.

Even if a before-and-after difference were available, the single-group design would not isolate the device’s effect from other factors that could change over time. The investigation has no concurrent control arm and masking is listed as none. As a general matter, these features limit causal interpretation and may also matter for subjective assessments. The registry record does not describe additional procedures that might address these concerns.

The reported safety counts document what the registry lists for this group, including 3 participants with serious adverse events, 21 with other adverse events and 0 deaths. They do not establish whether the events were caused by the device, nor do they allow a comparative safety conclusion. The record does not include a concurrent comparison group for that purpose.

Accordingly, the strongest reading is descriptive: the record identifies the intended hearing-performance comparison, reports that data were received for 27 participants, and gives selected safety counts. The statistical results needed to quantify the primary outcome and its uncertainty are not reported in the registry record.

14. Why This Trial Matters Statistically

This investigation raises several statistical interpretation issues that apply broadly to clinical research. Its primary question concerns change within a single group, comparing a preoperative unaided assessment with a postoperative audiometric assessment. Such a design can document what was measured over time, but the lack of a concurrent control group limits how confidently observed change can be attributed to the intervention.

The endpoint is a continuous measure summarized by a mean and standard deviation. Understanding an average change requires more than the point estimate: readers also need information about participant-to-participant variability and the precision of the estimated change. The registry identifies a dispersion measure but does not report the endpoint result, its variability or an interval estimate. This makes the distinction between a defined outcome and a reported finding especially important.

The investigation also lists a range of secondary hearing, speech, questionnaire, quality-of-life and surgical outcomes. Multiple outcomes and assessment times can create interpretive complexity. A prespecified outcome hierarchy or multiplicity approach can help distinguish confirmatory findings from exploratory ones, but neither is reported in the registry record. The record lists 32 outcome measures and 0 analyses, leaving the numerical findings and formal inferential framework unavailable.

Finally, the primary endpoint has missing observations, with the record giving specific reasons for two participants. Because the handling of missing data is not reported, readers cannot evaluate whether the analysis accounts for incomplete follow-up or how sensitive results might be to assumptions about missingness. These design and reporting details are central to judging precision and causal interpretation, not merely technical additions to a result.

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 investigation affect interpretation of the registered information. The study is single group, with allocation listed as NA and no randomized comparison described. The endpoint compares postoperative hearing with the preoperative unaided condition, so any observed difference would not by itself distinguish the intervention’s effect from other changes over time. Masking is listed as none; the record does not clarify whether outcome assessors were masked.

For the primary endpoint, the record reports data for 27 participants and describes two reasons for missing data: one participant did not receive the sound processor due to the Covid-19 pandemic, and one participant did not attend either the 4 weeks visit or the 3 months visit. The analysis population and handling of these missing observations are not reported in the registry record. It also does not report whether missing data affected other outcomes.

The primary endpoint is specified as a mean Pure Tone Average 4 in dB HL, with standard deviation, at baseline and 3 months after surgery. Yet the record does not provide the numerical endpoint results, uncertainty estimates, analysis method, or a threshold for clinically important change. The listed secondary outcomes and the total of 32 outcome measures do not resolve this gap because numerical analyses are not reported.

The record identifies the study as multicentre but does not report the number of centres in the information available for this description. It also does not report detailed follow-up for safety events, causal relatedness, or a comparative safety analysis. These limitations constrain what can be concluded from the registry record, even though it provides the broad design, endpoint definition and selected safety counts.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Clinical Performance, Safety and Patient Reported Outcomes of an Active Osseointegrated Steady-State Implant System — NCT04041700.
• Registry record last updated 2021-08-18; status last verified 2021-07.
• Results first posted 2021-08-18.
• 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

NCT04041700 is a completed, prospective, multicentre, open, single-group clinical investigation sponsored by Cochlear. It evaluates the Osia 2 system in adults with conductive hearing loss, mixed hearing loss or single-sided sensorineural deafness. Enrollment was 29; the registry lists a start date of 2019-08-16 and primary completion on 2020-06-19.

The primary endpoint is change from the preoperative unaided condition to 3 months after surgery in free-field Pure Tone Average 4, measured in dB HL. The record says data were received for 27 participants and gives reasons for the two missing observations. It lists 32 outcome measures and 0 analyses. Numerical primary-endpoint results and the statistical analysis plan are not reported in the registry record.

For the Osia 2 System group, the registry reports 3 participants affected by serious adverse events, 21 affected by other adverse events and 0 deaths, with 29 at risk for each category. The record lists four serious adverse-event terms, each reported at 3.4%.