Clinical Performance of a Transcutaneous Bone Conduction Hearing Solution (Baha® Attract System): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Clinical Performance of a Transcutaneous Bone Conduction Hearing Solution (Baha® Attract System) (NCT01822119), a na na study. The primary outcome is Hearing Performance, PTA4 at 12 Weeks.

2. Clinical Question

The investigation concerns hearing performance with the Baha Attract System among people with hearing impairment who are candidates for Baha surgery. The registry lists deafness, hearing loss, and conductive hearing loss among the conditions studied. Its central question is how pure-tone hearing thresholds with the system compare with pre-operative unaided thresholds. The primary assessment was planned at 12 weeks after surgery.

The device is a transcutaneous bone-conduction solution using magnets on both sides of the skin: an implant magnet and an external sound-processor magnet to which a sound processor is attached. The registry’s stated aims also include describing hearing performance relative to a pre-test situation using a sound processor on a softband, selecting between sound processors, improving fitting, and selecting the sound-processor magnet at fitting and over time.

This is not a randomized comparison of separate treatment groups. The primary contrast is within the clinical pathway: pre-operative unaided hearing versus aided hearing with the Baha Attract System after surgery. Listed secondary outcomes include comparisons with a sound processor on a softband, speech performance in quiet and noise, individual-frequency hearing performance, and the Abbreviated Profile of Hearing Aid Benefit. Several of those outcomes are identified at 36 weeks.

3. Trial Design

The registry describes a multicentre, open, prospective clinical investigation. It is a single-group study, with allocation listed as not applicable and masking listed as none. Enrollment was 27. The record marks the study completed and lists its start as 2013-03 and primary completion as 2014-02. The official title describes a 3-month investigation with a 6-month follow-up.

Participants received the Baha Attract System. The device description identifies an implant magnet and an external magnet that holds the sound processor. Because the study has a single group rather than a randomized control group, comparisons in the outcomes are between hearing conditions or assessment times, not between separately assigned randomized arms. In particular, a pre-operative unaided assessment is a baseline condition, not an independent control group.

The primary outcome is hearing performance measured by PTA4 at 12 weeks, with baseline described as before surgery. The primary population is the intention-to-treat population, defined in the record as including all patients who received surgical intervention. That definition identifies the intended analysis population, but the record does not report the number contributing data to each outcome or whether every enrolled participant had an observed primary assessment.

The registry lists 20 outcome measures and 24 analyses. Secondary outcomes include individual-frequency hearing performance; PTA4 and individual-frequency comparisons between softband and Baha Attract at 36 weeks; speech in quiet and speech in noise under unaided, softband-aided, and Baha Attract conditions; and the Abbreviated Profile of Hearing Aid Benefit. The list indicates a range of hearing and patient-reported measures, but does not establish which comparisons were prespecified as confirmatory or how their results were intended to be interpreted together.

Details such as a formal sample-size justification, planned interim analyses, and a detailed statistical analysis plan are not reported in the registry record. The posted record therefore supports a description of the broad design and listed endpoints, but not a reconstruction of unreported design decisions.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

Analysis population for the primary outcome, as defined in the registry: Intention-to-Treat (ITT) population, includes all patients who received surgical intervention.

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, PTA4 at 12 Weeks, measured over Baseline before surgery and 12 weeks after surgery.

The primary endpoint is the change in pure-tone thresholds in free field, expressed as the difference in PTA4 between the pre-operative unaided condition and the aided condition with the Baha Attract System at week 12. PTA4 is defined as the mean of thresholds at 500, 1000, 2000, and 4000 Hz. The endpoint is a continuous measure in dB, and the planned summary is a mean with standard deviation.

Pure-tone thresholds quantify the level at which sounds at selected frequencies can be detected. Combining the listed frequencies into PTA4 gives a single summary measure rather than treating each frequency as a separate primary result. The free-field setting and aided condition specify how hearing performance is assessed for this endpoint.

General statistical context: a continuous endpoint can describe both the average observed change and the variability among participants. The mean summarizes the central tendency, while the standard deviation describes how dispersed individual measurements are around that mean. These are different features: a mean alone does not show whether participants’ responses were closely grouped or widely spread.

The endpoint is suitable for a before-and-after description of hearing performance, but the registry record does not report its numerical result, the direction or size of the change, a confidence interval, or a p-value. It also does not report the detailed method used to form the analysis or handle incomplete measurements. The endpoint definition therefore clarifies what was intended to be measured, but does not by itself establish the observed result or its precision.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The statistical methods should be read in light of the endpoint definitions and the single-group design. The registry identifies a continuous primary measure, a mean summary, and standard deviation as its dispersion measure. It does not report the detailed analysis procedure, inferential model, or all numerical outcome results. The explanations below describe general statistical concepts rather than unreported choices in this investigation.

What does a mean summarize?

General: a mean is the arithmetic average of a set of measurements. For a continuous endpoint such as PTA4, the mean can summarize the group’s typical measured value or change. It is sensitive to the measurements included and can be influenced by unusually high or low observations. The registry identifies the mean as the primary endpoint parameter, but the primary mean result is not reported in the registry record.

A mean is not a description of every participant. Two groups of measurements could have the same mean but different patterns of individual responses. That is why a dispersion measure is useful alongside a mean. In this investigation, the stated dispersion measure is the standard deviation.

What does the standard deviation convey?

General: the standard deviation describes the spread of observations around their mean. A smaller standard deviation indicates that observations are more closely clustered around the mean; a larger one indicates greater spread. It does not, by itself, describe uncertainty in the estimated mean. The standard deviation and the standard error answer different questions: the former describes participant-level variability, while the latter concerns the precision of an estimated mean. The record specifies standard deviation, not a standard error for the primary endpoint.

For a before-and-after endpoint, the variability of individual changes may differ from the variability of measurements at either assessment alone. The registry defines the change being measured but does not report the detailed analysis formula or a standard deviation for the change in the registry record. The exact statistical handling of paired measurements is therefore not reported in the registry record and should not be assumed.

What would a confidence interval add?

General: a confidence interval gives a range of values compatible with the data and the method used to calculate it. It communicates uncertainty around an estimate; it is not a range containing a fixed proportion of individual participants’ results. A narrow interval indicates greater precision than a wide interval, under the assumptions of the method. The primary confidence interval is not reported in the registry record, so the precision of the primary estimate cannot be assessed from the registry information.

A confidence interval also helps distinguish the estimated magnitude from statistical evidence against a specified null value. Neither a p-value nor a confidence interval substitutes for assessing whether the endpoint is meaningful in context. The registry record does not report a primary p-value or an interpretation threshold for PTA4.

Why does the single-group design matter?

General: a comparison of two conditions measured in the same participants differs from a comparison of separate randomized groups. Within-participant measurements can account for stable differences between participants when an appropriate paired analysis is used. But such an analysis still does not create a randomized counterfactual. Changes between assessments may reflect more than the intervention itself, and the study design determines what causal claims can be supported.

The primary comparison is pre-operative unaided performance versus aided performance with the system at week 12. The record does not report the detailed analysis method used for that contrast. It is therefore appropriate to describe the endpoint as a before-and-after condition comparison, rather than naming an unreported statistical test or model.

How should multiple outcomes be considered?

General: when many outcomes or comparisons are examined, the chance of at least one apparently unusual result can increase if each is interpreted independently without a plan for multiplicity. Methods for addressing multiplicity include adjusting significance thresholds or defining a hierarchy of outcomes; these approaches are not interchangeable and depend on the study’s aims. The registry lists 20 outcome measures and 24 analyses, but a multiplicity strategy is not reported in the registry record. It is not possible to determine from the outcome list alone which results were intended to be confirmatory.

Outcome measures can also differ in scale and meaning. PTA4 summarizes pure-tone thresholds across specified frequencies, whereas speech measures assess speech performance under stated conditions and the Abbreviated Profile of Hearing Aid Benefit is a separate patient-reported measure. Evidence from one measure should not be treated as if it automatically establishes the result for the others.

What do planning calculations mean?

General: a sample-size calculation is a planning exercise that describes how many participants may be needed under specified assumptions about the effect to detect, variability, significance level, allocation, and statistical method. A calculation based on an observed effect is retrospective planning, not evidence that the original study had a particular probability of detecting an effect. It is not “observed power.”

The planning assumptions described for the associated illustrations are a two-sided significance level of 0.05, equal allocation between two arms, a normal approximation, and effects expressed as a stated fraction of the observed effect. The record notes that these are planning illustrations computed from the observed effect, not the trial’s own design. The study itself is single-group, so a two-arm allocation assumption in a planning illustration does not describe the trial’s actual allocation. A formal sample-size justification for this investigation is not reported in the registry record.

These concepts help explain what different statistical summaries can and cannot say. They do not fill in the missing primary estimate, uncertainty interval, analysis method, or multiplicity plan. Those details remain not reported in the registry record.

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("Baha Attract System")
n   <- c(27)
m   <- c(-18.4)
sp  <- c(6.7)
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  Baha Attract System  27  -18.400  6.700  -21.050  -15.750

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 20 outcome measures and 24 analyses. The outcomes cover several dimensions, including PTA4, individual-frequency hearing performance, speech in quiet and noise, softband comparisons, and the Abbreviated Profile of Hearing Aid Benefit. Examining multiple outcomes can be informative, but it also raises the question of how results were prioritized and how the possibility of chance findings across many analyses was handled.

A prespecified hierarchy, a multiplicity adjustment, or a clear distinction between confirmatory and exploratory analyses can help readers interpret a collection of results. Such a strategy is not reported in the registry record. The count of outcomes and analyses alone does not establish that every analysis had equal status or that a particular adjustment was used.

The primary population is described as intention-to-treat and includes all patients who received surgical intervention. Enrollment was 27. The registry information does not give endpoint-specific sample sizes, numbers missing at each assessment, reasons for missingness, or the method used to handle missing values. It is therefore not possible to determine how closely each analysis population matched the stated primary population.

General statistical context: missing measurements can affect estimates when the probability of missingness is related to participant characteristics or outcomes. Methods for dealing with missing data rely on assumptions, so the method and its assumptions matter. No such method is reported in the registry record. Interim analyses and a detailed statistical analysis plan are also not reported in the registry record.

12. 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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: postoperative wound infection (7.4%), post procedural heamatoma (3.7%).

The registry reports serious adverse events for 2 of 27 participants in the Baha Attract System group, and other adverse events for 8 of 27 participants in that group. These are participant counts, not a comparison between randomized groups. The record lists postoperative wound infection at 7.4% and post procedural heamatoma at 3.7% among the top serious terms.

These figures describe what the registry reports, but the available details do not establish whether every event was caused by the device, surgery, or another factor. They also do not provide event timing, severity details beyond the serious-event classification, recurrence, or participant-level overlap between event categories. The percentages listed for serious terms should not be treated as counts unless the record separately provides those counts.

Deaths are not reported in the registry record. That absence of a reported value should not be read as evidence of zero deaths. Likewise, the reported participant counts do not by themselves show comparative safety, because this is a single-group investigation without a like-for-like control group.

General: adverse-event counts are most informative when interpreted with clear definitions, follow-up time, event-level detail, and a suitable comparison. The registry entries support a limited descriptive reading, not a causal or comparative conclusion.

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

The registry record establishes that the investigation was designed to describe hearing performance with the Baha Attract System and lists a primary measure based on PTA4 before surgery and at week 12. It also identifies additional hearing, speech, and patient-reported outcomes. These definitions clarify the questions the study set out to examine.

They do not establish the numerical primary result. The mean change, its standard deviation, confidence interval, and p-value are not reported in the registry record. Without those results, the size and precision of the primary outcome cannot be assessed. The listed secondary outcomes also do not include numerical results in the registry record, so their findings cannot be characterized.

The single-group, open design permits description of participants’ outcomes across assessment conditions and time points. It does not provide a randomized comparison against a separate control group. A pre-operative unaided assessment is a comparator condition for the primary endpoint, but it is not the same as a concurrent control group assigned to a different intervention. The study’s design therefore limits what can be concluded about whether observed changes were caused by the system alone.

The registry’s adverse-event counts are also descriptive. They do not establish that the listed events were caused by the system, and there is no randomized control group for a comparative safety estimate. The record does not report deaths as a numerical result.

General: a study’s findings are best interpreted by considering the endpoint, analysis population, uncertainty, missing data, design, and multiplicity together. Here, the endpoint and broad study structure are described, but important numerical and analytic details are not reported in the registry record. The record therefore supports a careful account of what was planned and which selected event counts were posted, but not a conclusion about the magnitude, precision, or causal attribution of hearing-performance outcomes.

14. Why This Trial Matters Statistically

This investigation is statistically useful as an example of how clinical performance can be framed around a continuous within-participant outcome. The primary measure compares pre-operative unaided PTA4 with aided PTA4 at week 12, and the registry defines the frequency components used to form that summary. Clear endpoint definition is important because it distinguishes the target measurement from broader claims about hearing performance.

The study also illustrates why design labels matter. It is open and single-group, with allocation not applicable. The primary contrast is between conditions and assessment times rather than randomized arms. That distinction affects how a change should be interpreted: a measured before-and-after difference is descriptive of the participants and assessments, but it is not by itself a randomized estimate of the effect relative to an alternative treatment or no intervention.

The breadth of listed outcomes raises familiar statistical questions about prioritization and multiplicity. Hearing thresholds, speech outcomes, and patient-reported benefit capture different aspects of performance. A prespecified analysis plan can clarify which outcomes are central and how secondary findings should be interpreted, but that plan is not reported in the registry record.

The record also shows why enrollment alone is not enough to assess precision or completeness. Although enrollment was 27, endpoint-specific analyzed counts, missing-data handling, and numerical primary results are not reported in the registry record. Without those details, readers cannot assess the uncertainty of the main estimate or the influence of incomplete follow-up.

Finally, the reported adverse-event counts demonstrate the difference between a descriptive safety summary and a comparative safety analysis. Counts can be reported transparently while still leaving causality and relative risk unresolved. These are general lessons in interpreting clinical trial evidence, not conclusions about an unreported outcome estimate.

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 design is open and single-group, with no randomized control group. The primary comparison is between pre-operative unaided hearing and aided hearing with the Baha Attract System at week 12. This design can describe performance across those conditions, but it does not isolate a causal effect against a concurrent randomized comparator.

The registry does not report the numerical primary endpoint result, its uncertainty, or the detailed method used to analyze it. Enrollment was 27, but the number contributing data to each outcome is not reported in the registry record. The approach to missing measurements and the reasons for any missingness are also not reported.

The registry lists 20 outcome measures and 24 analyses, but a multiplicity plan and a detailed statistical analysis plan are not reported in the registry record. It is therefore unclear how the outcome family was prioritized or how secondary analyses were intended to inform conclusions. A formal sample-size justification and interim-analysis plan are not reported either.

The primary population is defined as including all patients who received surgical intervention. The record does not give further detail about eligibility, participant flow, or the extent to which each outcome’s analyzed population followed that definition. The study title describes a multicentre investigation, but the number and locations of centers are not reported in the registry record.

The safety information is limited to reported participant counts and listed serious-event terms. Event timing, detailed descriptions, causal assessments, and comparative data are not reported in the registry record. These limitations restrict interpretation of both the clinical-performance outcomes and the safety profile.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Clinical Performance of a Transcutaneous Bone Conduction Hearing Solution (Baha® Attract System) — NCT01822119.
• Registry record last updated 2017-02-06; status last verified 2016-12.
• Results first posted 2017-02-06.
• 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

This completed, multicentre, open, prospective, single-group clinical investigation enrolled 27 participants and evaluated the Baha Attract System. The primary outcome was the change in free-field PTA4 from the pre-operative unaided condition to the aided condition at week 12, in an intention-to-treat population described as including patients who received surgical intervention.

The registry lists 20 outcome measures and 24 analyses, including hearing, speech, and patient-reported outcomes. It reports serious adverse events for 2 of 27 participants and other adverse events for 8 of 27. The primary numerical outcome result, endpoint-specific analyzed counts, missing-data methods, multiplicity strategy, and detailed statistical analysis plan are not reported in the registry record.