This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for Preliminary Study 2 to Test the Effects of Ambulatory Voice Biofeedback (NCT03416868), a na na study. The primary outcome is Percent Compliance.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT03416868 |
| Phase | NA |
| Condition | Vocal Fold Nodules, Muscle Tension Dysphonia |
| Allocation | Not applicable (single group) |
| Design model | Single Group |
| Masking | None (open label) |
| Enrollment | 39 |
| Arms in the primary outcome | n/a |
| Lead sponsor | Massachusetts General Hospital |
| Results posted | Yes |
2. Clinical Question
Population
The study concerns patients with vocal hyperfunction, including the registry-listed conditions Vocal Fold Nodules and Muscle Tension Dysphonia. The primary-outcome description identifies a final group of 7 patients assessed under the final protocol. It also explains that data from the first 32 patients were used to refine and adapt study protocols, accounting for the difference from the total enrollment of 39. Further eligibility criteria, such as age, symptom duration, or other clinical characteristics, are not reported in the registry record.
Intervention
The intervention is ambulatory voice biofeedback, a behavioral intervention used alongside standard-of-care voice therapy. The study summary says ambulatory monitoring took place before therapy and after the first 3 voice therapy sessions, with biofeedback added to monitoring after the second session only. The arm description also refers to monitoring during the first 3 weeks of therapy. The registry does not give further details about the feedback display, the specific patient-level thresholds, or how patients were instructed to respond to feedback.
Comparator
There is no separate comparison group in the recorded design. The study is listed as single-group, with allocation and masking recorded as NA and NONE, respectively. The registry describes standard-of-care voice therapy in combination with ambulatory monitoring and biofeedback, but does not report a concurrent group receiving therapy without biofeedback. Consequently, the record does not set up a randomized, like-for-like comparison from which the effect of adding biofeedback can be isolated.
Primary question
The primary outcome is Percent Compliance, defined as the percentage of voicing within desired patient-specific numeric thresholds. The stated question is whether patients spend a measurable portion of their voicing within those thresholds across Week 1, Week 2, and Week 3, during the described therapy and monitoring sequence. The registry does not report numerical compliance results, a comparison of compliance across weeks, or a statistical analysis. It therefore documents the intended measurement, but does not establish from its posted results how compliance changed or how precisely it was estimated.
3. Trial Design
This is a completed, single-group behavioral study listed as NCT03416868, titled Preliminary Study 2 to Test the Effects of Ambulatory Voice Biofeedback. Massachusetts General Hospital is the recorded sponsor. The registry lists an enrollment of 39 and identifies Vocal Fold Nodules and Muscle Tension Dysphonia as conditions. The study start date is 2018-06-01, and the primary completion date is 2024-01-25. The phase is recorded as NA.
The study summary describes standard-of-care voice therapy with ambulatory voice monitoring before therapy and after the first 3 voice therapy sessions. Biofeedback was added to ambulatory monitoring after the second voice therapy session only. The primary outcome timeframe is Week 1, Week 2, and Week 3. The arm description also mentions monitoring during the first 3 weeks of therapy, but its text is incomplete in the registry record, so further monitoring details cannot be established from that description.
The primary-outcome description distinguishes the final 7 patients from the total enrollment of 39: data from the first 32 patients were used to refine and adapt study protocols, and the final 7 were acquired with the final protocol. The record does not report how the protocol changes affected endpoint measurement or whether data from the earlier patients were included in any outcome analysis.
Because this is a single-group study with no recorded control arm, it is not a randomized comparison. The registry records allocation as NA and masking as NONE. Further details of recruitment, eligibility, therapy delivery, adherence, and the rationale for the enrollment target are not reported in the registry record. The record lists one outcome measure and zero analyses. It does not report a statistical analysis plan or numerical primary-outcome findings.
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
| Stage | Standard of Care Voice Therapy With Ambulatory Voice Biofeedback. |
|---|---|
| STARTED | 39 |
| COMPLETED | 36 |
| NOT COMPLETED | 3 |
Reasons for not completing this period:
| Reason | Standard of Care Voice Therapy With Ambulatory Voice Biofeedback. |
|---|---|
| Withdrawal by Subject | 2 |
| Lost to Follow-up | 1 |
Analysis population for the primary outcome, as defined in the registry: There is a discrepancy in the numbers of enrolled patients here (total = 7) and compared to total enrolled (total = 39). This is because the first 32 patients' data were used to refine and adapt study protocols. The final 7 patients were acquired with this final protocol.
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.
Population: There is a discrepancy in the numbers of enrolled patients here (total = 7) and compared to total enrolled (total = 39). This is because the first 32 patients' data were used to refine and adapt study protocols. The final 7 patients were acquired with this final protocol.
| Characteristic | Unit | Standard of Care Voice Therapy With Ambulatory Voice Biofeedback. |
|---|---|---|
| Participants analysed | 7 | |
| Age, Continuous | years | 35 (18 to 57) |
| Sex: Female, Male: Female | Participants | 6 |
| Sex: Female, Male: Male | Participants | 1 |
| Ethnicity (NIH/OMB): Hispanic or Latino | Participants | 1 |
| Ethnicity (NIH/OMB): Not Hispanic or Latino | Participants | 6 |
| Ethnicity (NIH/OMB): Unknown or Not Reported | Participants | 0 |
| Race (NIH/OMB): American Indian or Alaska Native | Participants | 0 |
| Race (NIH/OMB): Asian | Participants | 0 |
| Race (NIH/OMB): Native Hawaiian or Other Pacific Islander | Participants | 0 |
| Race (NIH/OMB): Black or African American | Participants | 1 |
| Race (NIH/OMB): White | Participants | 6 |
| Race (NIH/OMB): More than one race | Participants | 0 |
| Race (NIH/OMB): Unknown or Not Reported | Participants | 0 |
| Percent compliance | Percentage of voicing below threshold | 89.46 (78.23 to 98.67) |
6. Primary Endpoint
The registry lists the primary outcome as Percent Compliance, measured over Week 1, Week 2, Week 3.
| Element | Registry entry |
|---|---|
| Result type | Mean |
| Measure of spread | Full Range |
| Unit | percentage of voicing crossing threshold |
| Endpoint classification | not reduced to a single comparison |
The primary endpoint is Percent Compliance, expressed as the percentage of voicing crossing a patient-specific numeric threshold, with the desired state described as voicing within the thresholds. In practical terms, the endpoint is intended to quantify how much of a patient’s monitored voicing falls within the target range rather than how a clinician rates voice quality or how a patient reports symptoms. The measurement is specified for Week 1, Week 2, and Week 3.
Patient-specific thresholds make the measure tailored to each participant’s voice target. This can be useful when the relevant target is not identical across patients. It also means that the exact thresholds and the rule for combining individual measurements matter to interpretation; those details are not reported in the registry record. The recorded endpoint parameter is a mean, and its dispersion is described as the full range. The registry does not give the numerical mean or range at any of the stated timepoints.
As a general statistical point, a percentage-based measure can translate repeated voice monitoring into a quantity that is easier to compare across observation periods. Its meaning still depends on the monitoring procedure, the threshold definition, and the amount of usable monitoring data. Those operational details, along with the numerical endpoint results, are not reported in the registry record. The endpoint description alone therefore specifies what the study intended to measure, not the level of compliance observed.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The registry record documents a study question and an endpoint, but its posted results do not include numerical Percent Compliance findings. There is no reported mean or full range for Week 1, Week 2, or Week 3, and the record lists zero analyses. Without those results, the direction, size, and precision of any observed change cannot be described. It is not possible to determine from this record whether compliance increased, decreased, or remained similar across the stated weeks.
The study’s recorded enrollment is 39, while the primary-outcome description identifies 7 patients acquired with the final protocol and says data from the first 32 patients were used to refine and adapt protocols. This distinction is important when interpreting any outcome data: the record does not state that all enrolled patients contributed to the final-protocol outcome assessment. It also does not report an analysis population, participant-level completeness, or how data from the protocol-refinement group were handled in outcome analyses.
The design is single-group. It contains no concurrent group receiving standard-of-care voice therapy without biofeedback, and the registry does not describe a randomized allocation. Therefore, even if outcome values were available, a change over time in this group would not by itself isolate the contribution of biofeedback from other influences, including the therapy course, time, or changes in monitoring and protocol. These are general interpretive limitations of an uncontrolled before-and-after or repeated-measure design, not evidence that any particular influence occurred in this study.
The registry does not report confidence intervals, hypothesis tests, or an analysis model for the primary endpoint. As a general statistical point, a confidence interval communicates the range of estimates compatible with the data under the method used; without an estimate and interval, precision cannot be assessed. Likewise, the absence of a reported test means that no statistical significance claim can be drawn from the registry entry. The posted record also does not provide sufficient endpoint results for a meaningful assessment of the spread of patient outcomes, despite specifying full range as the dispersion measure.
The adverse-event entry lists 7 participants at risk for other events and 0 affected, with 0 deaths. For serious events, the registry records 0 at risk and 0 affected. The serious-event denominator is therefore not shown as the study cohort, and the entry cannot establish the frequency of serious events among all enrolled or final-protocol participants. Overall, the record supports a description of the planned measurement and the study structure, but not a conclusion about the observed level of compliance or an effect attributable to biofeedback.
9. Statistical Methods Explained
What does a single-group design allow?
The registry identifies the study model as single-group. This means the recorded design does not include a distinct control group against which outcomes can be compared at the same time. The primary outcome is measured across Week 1, Week 2, and Week 3, but the registry does not report the statistical method, if any, used to compare those timepoints. As a general statistical point, repeated measurements can describe how an outcome varies over time within the participants observed. They do not, on their own, distinguish an intervention effect from other changes that occur over the same period.
The record also lists allocation as NA and masking as NONE. These entries do not describe a random assignment or a masked assessment. Further details about who collected the monitoring data, how compliance was calculated, and whether endpoint assessment was independent of treatment are not reported in the registry record. Those procedural details can affect the risk of measurement differences, but their absence does not establish that a particular bias occurred.
What does the endpoint summary describe?
The outcome is named Percent Compliance, with a mean as the parameter and full range as the dispersion. In general, a mean describes the average across the observations included in an analysis, while a full range describes the lowest and highest observed values. These summaries answer different questions: the mean gives a central summary, whereas the range conveys the span of observed values and can be influenced by extreme observations. The registry gives no numerical mean or range for any of the listed weeks, so these summaries cannot be interpreted from the posted results.
The endpoint is based on patient-specific numeric thresholds and the amount of voicing within the desired thresholds. General methodological considerations for this kind of endpoint include how each threshold is chosen, whether the same rule is used at each timepoint, and how monitoring gaps or unusable recordings are handled. The registry does not report those details. It also does not describe whether the percentage is first calculated within each participant and then summarized, or calculated through another approach. That distinction can affect what an overall mean represents.
Can the analysis method be identified?
The registry lists one outcome measure and zero analyses. It does not report a model, an inferential procedure, an adjustment strategy, a confidence interval, or a hypothesis test for Percent Compliance. Accordingly, no particular statistical method should be attributed to the study on the basis of this record. General options for repeated measurements can account for the relationship between observations from the same participant, but the record does not state whether such an approach was used.
Over-dispersion is a general concern when outcome variability is greater than assumed by a statistical model. If a model assumes less variability than is present, uncertainty may be understated and intervals may be too narrow. Whether over-dispersion is relevant depends on the measurement and analysis model; neither is described sufficiently in the registry record to assess it for this study. This is a general explanation, not a finding about the posted data.
What information would support precision assessment?
A confidence interval is a general statistical summary of uncertainty around an estimate under a stated method and its assumptions. It helps readers assess precision as well as the estimated value. The registry does not report a confidence interval for the endpoint. It also does not provide numerical endpoint estimates or the underlying participant-level data. As a result, precision cannot be evaluated from the posted results, and no interval can be reconstructed without making assumptions or performing calculations not documented in the registry.
The record’s explanation that the first 32 patients contributed to protocol refinement and the final 7 were acquired under the final protocol raises a further analysis question: which participants and observations were included in any endpoint summary? The registry does not provide an analysis population or a participant flow accounting for endpoint availability. The distinction between enrollment and the described final-protocol group should therefore be retained when reading any summary of the study.
The sample-size assumptions listed for planning illustrations describe a two-sided significance level of 0.05, equal allocation between two arms, a normal approximation, and an effect stated as a fraction of an observed effect. They are planning assumptions, not the trial’s own design justification. The study itself is recorded as single-group, so the two-arm planning setup does not describe its observed design. A planning illustration is not observed power and does not provide evidence about the result. The registry does not report the trial’s sample-size justification, and no calculated planning figures are reported in this record.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists one primary outcome measure, Percent Compliance, with three stated timepoints: Week 1, Week 2, and Week 3. It does not report whether the timepoints were analyzed separately, combined, or treated as repeated observations in a single model. It also does not report a multiplicity plan or specify whether additional outcomes or subgroup analyses were prespecified. A single listed outcome does not by itself resolve multiplicity concerns if several timepoint comparisons or analyses were conducted; whether that occurred is not reported in the registry record.
Missing-data handling is not reported in the registry record. The entry does not state how incomplete monitoring, missed therapy sessions, unavailable recordings, or participants without usable endpoint measurements were handled. These details matter because a percentage calculated from limited monitoring may not represent the same observation period or amount of voice use as a percentage based on more complete monitoring. This is a general consideration; the registry does not report that monitoring was incomplete.
The record reports total enrollment of 39 and explains that the first 32 patients’ data were used to refine and adapt study protocols, while the final 7 were acquired with the final protocol. It does not describe a formal analysis population, the number with endpoint data at each timepoint, or how any protocol-related changes were reflected in analysis. The registry lists zero analyses, so no analysis population can be inferred from a reported statistical result.
The primary-outcome description gives a reason for the difference between total enrollment and the final-protocol group, but does not report a missing-data procedure or an intention-to-treat approach. These analysis choices and their rationale are not reported in the registry record. Without them and without numerical results, readers cannot assess how incomplete observations or the protocol-refinement phase influenced endpoint estimates.
11. What the Result Does, and Does Not, Mean
The registry record establishes that the study evaluated ambulatory voice biofeedback alongside standard-of-care voice therapy and defined Percent Compliance as the primary endpoint. It identifies the intended assessment period as Week 1, Week 2, and Week 3, and it explains that the final 7 patients were enrolled under the final protocol after the first 32 patients’ data had been used for protocol refinement. These registry entries describe the study’s scope and measurement plan.
The posted results do not give numerical compliance estimates, ranges, or statistical analyses. They therefore do not show the observed level of compliance, any change over time, or the precision of an estimate. No conclusion about whether compliance improved or worsened can be drawn from the registry record. The record also does not report an analysis that would quantify an association between biofeedback and compliance.
Even if changes across the stated weeks were reported, the single-group design would require cautious interpretation. As a general point, changes in an uncontrolled group cannot be attributed solely to the intervention because there is no concurrent control group to indicate what might have happened without biofeedback. The record does not support a treatment-effect comparison or a claim that biofeedback caused an outcome change.
The endpoint is a behavioral measure of time spent within individually defined voice thresholds. It is not, by itself, a report of a clinical outcome beyond that measurement. The registry does not provide additional patient-reported outcomes or other clinical results in its listed outcome entry. It also does not report enough detail to determine how representative the final-protocol group is of a broader patient population.
The adverse-event entry records no other events among the 7 listed as at risk and no deaths, but records 0 participants at risk for serious events. Since that serious-event denominator is not the cohort size, the entry should not be read as establishing a serious-event rate for the study participants. More broadly, the registry record supports describing what was intended to be measured and which participants the protocol description distinguishes; it does not support a conclusion about efficacy, clinical benefit, or the magnitude of any intervention effect.
12. Why This Trial Matters Statistically
This study is statistically notable for the distinction between total enrollment and the group described as having completed the final protocol. The registry reports 39 enrolled, but states that data from the first 32 patients were used to refine and adapt study protocols and that the final 7 were acquired with the final protocol. That distinction affects the population to which a final-protocol endpoint summary would apply and makes clear why enrollment alone cannot be treated as the endpoint sample size.
The endpoint also illustrates a common measurement challenge in behavioral research: translating continuous real-world monitoring into a percentage tied to patient-specific thresholds. Such a measure can summarize whether monitored voicing falls within a target, but interpretation depends on how thresholds and monitoring periods are defined and how incomplete observations are treated. Those operational details are not reported in the registry record.
From a design perspective, a single-group study can document feasibility or describe observations across a therapy sequence, but it does not provide a concurrent comparison that separates the contribution of biofeedback from other influences. The registry does not report numerical results or analyses, so the record cannot establish what pattern was observed. The value of the entry is therefore chiefly in clarifying the study structure, endpoint concept, and protocol-refinement history, while also identifying information needed for stronger statistical interpretation.
13. Limitations
- The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
- Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.
The main trial-specific limitation is that the registry record lists zero analyses and provides no numerical Percent Compliance results. The stated mean and full-range summaries cannot be evaluated for any of the three timepoints. There are no reported estimates, confidence intervals, or tests with which to assess effect direction, magnitude, or precision.
The study is recorded as single-group, without a randomized comparator. The registry itself notes that the design is not a like-for-like control comparison. Any within-group pattern, if reported elsewhere, would not alone establish that biofeedback caused the pattern. The registry does not provide a separate standard-of-care-only group for comparison.
The difference between total enrollment of 39 and the final-protocol group of 7 is explained by the use of the first 32 patients’ data for protocol refinement. However, the record does not detail the protocol adaptations, the endpoint data available for each participant, or the analysis population. That leaves uncertainty about how the refinement phase and any changes in measurement were treated in analyses.
The endpoint relies on patient-specific thresholds, but the registry does not provide the threshold-setting procedure, monitoring-data processing rules, or missing-data approach. The arm description is incomplete, limiting interpretation of the monitoring schedule beyond the details stated in the study summary. Further limitations include the absence of a reported statistical analysis plan, sample-size justification, and multiplicity strategy. These are limitations of the registry information; they do not establish that the underlying study lacked such procedures.
14. Related Tutorials
15. Related Statistical Calculators
16. Sources
- ClinicalTrials.gov: Preliminary Study 2 to Test the Effects of Ambulatory Voice Biofeedback — NCT03416868.
- Registry record last updated 2024-04-17; status last verified 2024-04.
- Results first posted 2024-04-17.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
17. Record Summary
NCT03416868 is a completed single-group behavioral study sponsored by Massachusetts General Hospital. It addresses ambulatory voice biofeedback alongside standard-of-care voice therapy for patients with vocal hyperfunction, with Vocal Fold Nodules and Muscle Tension Dysphonia listed as conditions. The registry records total enrollment of 39 and identifies a final-protocol group of 7 after data from the first 32 patients were used for protocol refinement.
The primary endpoint is Percent Compliance, the percentage of voicing within patient-specific numeric thresholds, with assessments specified for Week 1, Week 2, and Week 3. The registry lists one outcome measure and zero analyses. Numerical endpoint findings, an analysis population, missing-data procedures, and a statistical analysis plan are not reported in the registry record. The single-group design has no recorded concurrent control arm, so the record does not establish a comparative effect attributable to biofeedback.