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 Exercise for Swallowing Problems After Stroke (NCT00722111), a randomized na study. The primary outcome is Maximum Isometric Tongue Pressure.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT00722111 |
| Phase | NA |
| Condition | Cerebrovascular Accident, Deglutition Disorders |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | None (open label) |
| Enrollment | 83 |
| Arms in the primary outcome | 2 |
| Lead sponsor | VA Office of Research and Development |
| Results posted | Yes |
2. Clinical Question
Population
People with swallowing problems after stroke were the intended population. The registry describes the condition as cerebrovascular accident with deglutition disorders. The total enrollment is listed as 83, but the posted outcome description says that data were available for only two arms and that data for the other treatment arms were unavailable because the principal investigator was not available.
Intervention
The study description focuses on oral exercise, including high-intensity lingual press and effortful swallowing. The registry also lists natural swallowing as a high-frequency, low-intensity swallowing regimen. These approaches represent different exercise tasks and intensities; the record does not provide enough outcome data to compare them all.
Comparator
A non-oral sham exercise is listed as a control approach. The posted primary outcome groups, however, are labelled “Lingual Strengthening” and “Lingual Pressure Norms.” The registry does not explain how the norms group relates to the randomized treatment arms, so it should not be assumed to be the sham group or a randomized comparator.
Primary question
The stated purpose was to determine whether stroke patients with swallowing problems would show greater swallowing improvement with intense oral exercise than with low-intensity oral exercise or sham exercise. The registered primary outcome was maximum isometric tongue pressure, measured in kPa at a timeframe of 16 months. Because results are posted for only two differently labelled groups and no analysis is posted, the registry record does not answer the stated comparative question.
Interpreting the result begins with separating the intended question from the information actually available. A randomized study can be designed to compare assigned regimens, but a comparative conclusion requires interpretable outcome data for the relevant groups and an analysis that accounts for the study design. The registry lists four intervention approaches, yet the primary outcome description says that only two arms had available data. The two posted group labels do not establish which randomized assignments those groups represent. The reported values therefore describe the entries in the registry, not a complete comparison of the intended exercise strategies.
3. Trial Design
The registry classifies the study as randomized, with a parallel design and no masking. Enrollment is listed as 83. The study status is completed, and its phase is recorded as not applicable. The sponsor is the VA Office of Research and Development. The recorded start is 2011-02 and primary completion is 2011-12.
The interventions include lingual press, effortful swallowing, natural swallowing, and a non-oral sham control exercise. The arm descriptions characterize lingual press as high-intensity oral exercise without swallowing; effortful swallowing as high-intensity swallowing; natural swallowing as high-frequency, low-intensity swallowing; and the sham as a non-oral control exercise. These descriptions show that the study was intended to examine several distinct exercise strategies rather than a single treatment against a single control.
The registry's posted primary outcome is maximum isometric tongue pressure. The outcome is described as peak isometric pressure at four sensors, with units of kPa and a timeframe of 16 months. The reported entries are for “Lingual Strengthening” and “Lingual Pressure Norms,” with sample sizes listed as 11.0 and 72.0. The registry does not clarify how those labels correspond to the randomized study arms. It explicitly notes that primary-investigator information was unavailable and that outcome data for the other treatment arms were unavailable.
Although the design is randomized, the record does not post a statistical comparison, analysis population, or analysis results. It also does not describe how allocation was implemented, whether allocation was concealed, or how the posted norms group was formed. Those details are not reported in the registry record. The design labels alone cannot resolve the mismatch between the intended multi-arm comparison and the two groups with posted outcome values.
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 | Lingual Strengthening | Lingual Pressure Norms |
|---|---|---|
| STARTED | 11 | 72 |
| COMPLETED | 11 | 72 |
| NOT COMPLETED | 0 | 0 |
Analysis population for the primary outcome, as defined in the registry: PI is not available, thus the numbers presented are what was initially entered by research team. For the other treatment arms, data is unavailable, as PI is not available. Data was only available for one arm. All efforts were made to retrieve data.
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: PI is not available, therefore, no data is available for reporting in this section.Efforts to retrieve data were unsuccessful.
| Characteristic | Unit | Lingual Strengthening | Lingual Pressure Norms | Total |
|---|---|---|---|---|
| Participants analysed | 0 | 0 | 0 | |
| Sex: Female, Male: Female | — | — | — | |
| Sex: Female, Male: Male | — | — | — | |
| Region of Enrollment: United States | participants | — | — | — |
6. Primary Endpoint
The registry lists the primary outcome as Maximum Isometric Tongue Pressure, measured over 16 months.
| Element | Registry entry |
|---|---|
| Result type | Mean |
| Measure of spread | Full Range |
| Unit | kPa |
| Endpoint classification | not reduced to a single comparison |
Maximum isometric tongue pressure measures the peak pressure a participant can produce while pressing the tongue against sensors without performing a swallow. The registry describes the outcome as peak isometric pressure at four sensors and reports the unit as kPa. As a measure of tongue strength, it captures a physical capability relevant to oral function. It is not itself a direct measure of swallowing safety, swallowing efficiency, or improvement in everyday eating and drinking.
The registry specifies a mean as the summary measure and lists a timeframe of 16 months. It describes the observed spread as a full range. For the “Lingual Strengthening” group, the posted mean is 49.0 kPa, with a range from 45.0 to 53.0 and a listed sample size of 11.0. For “Lingual Pressure Norms,” the posted mean is 78.0 kPa, with a range from 5.0 to 270.0 and a listed sample size of 72.0.
In general, a measure can be useful because it captures a specific biological or functional attribute, while still representing only one part of a broader clinical question. Tongue pressure may be informative about tongue strength, but the registry's stated purpose concerns swallowing improvement. The record does not establish that the posted pressure values are change from baseline, explain the relationship of the norms group to randomized assignment, or report another endpoint that would connect these values to swallowing function. The meaning of the measure should therefore remain specific to tongue pressure.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The posted primary outcome values are descriptive entries for two groups. “Lingual Strengthening” has a listed mean maximum isometric tongue pressure of 49.0 kPa, sample size 11.0, and full range 45.0 to 53.0. “Lingual Pressure Norms” has a listed mean of 78.0 kPa, sample size 72.0, and full range 5.0 to 270.0. These are the registry's reported summaries; they are not accompanied by a between-group estimate, confidence interval, or statistical test.
The difference in labels is important. The intervention descriptions identify exercise assignments, while the outcome entries use the names “Lingual Strengthening” and “Lingual Pressure Norms.” The registry does not say that the norms group was randomized, that it was the sham control, or that its data are directly comparable with the Lingual Strengthening group. Nor does it identify the posted values as changes from baseline. Without those clarifications, a numerical contrast between the two means would not be a valid estimate of the effect of an assigned exercise regimen.
The registry's outcome description states that data were available for only two arms and that data for the other treatment arms were unavailable because the principal investigator was not available. It also notes that the presented numbers were initially entered by the research team and that efforts were made to retrieve data. This disclosure limits what can be learned from the record: the posted values do not cover all of the described interventions, and the reasons for unavailable data do not establish what the unreported outcomes would have been.
No analysis is posted for the primary outcome. There is consequently no reported adjusted or unadjusted comparison, no p-value, and no confidence interval to characterize the uncertainty around a treatment contrast. In general, a mean and a range describe observed values but do not by themselves establish whether groups differ beyond what could arise from sampling variation. The range also does not show the distribution of observations within it. A range can be affected by extreme observations, and the much wider reported range in the norms group signals substantial spread in that group's recorded values, without identifying its cause.
The registry lists the trial as randomized, but randomization cannot repair absent outcome information or uncertainty about group labels. A trustworthy interpretation of a randomized comparison depends on knowing which participants were assigned to each regimen, which participants contributed outcome data, and how those data were analyzed. Those details are not reported in the registry record. The values are therefore best read as posted descriptive summaries, not as evidence that one exercise approach produced greater improvement than another.
The record also includes adverse-event entries with zero counts for serious and other events in both listed groups. Death counts are not given. These entries should not be treated as a complete safety analysis: the outcome information is limited to the listed groups, and the record does not explain the reporting or ascertainment process. The entries describe what was posted, not a guarantee that no adverse experiences occurred.
Any planning illustration based on an observed effect is distinct from an observed treatment result. The stated planning assumptions use 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 fraction of the observed effect. Such illustrations do not establish that the original trial was designed with those assumptions, and they are not observed power. In this case, the ambiguity of the posted groups and the absence of a comparative analysis further constrain what an observed-effect-based planning exercise can represent.
9. Statistical Methods Explained
The registry record identifies the study as randomized and parallel, but does not report a statistical analysis for the primary outcome. The entries for maximum isometric tongue pressure are summary statistics rather than a documented inferential analysis. Understanding that distinction helps prevent descriptive values from being read as a treatment effect.
What does a mean describe?
In general, the mean is a summary of the observed measurements in a group. It gives a single location value, but it does not show how individual measurements are distributed around that value. The registry identifies the primary outcome parameter as a mean and reports a mean for each posted group. The “Lingual Strengthening” entry has a mean of 49.0 kPa, while the “Lingual Pressure Norms” entry has a mean of 78.0 kPa. These means summarize the posted groups; they do not, on their own, estimate the effect of randomized assignment.
For an effect estimate, the analysis needs a defined comparison that corresponds to the research question. It must be clear which groups are being compared and what quantity is being estimated, such as a difference in endpoint values or a difference in change from baseline. The registry does not specify that the posted values are changes, and it does not explain whether the norms group is a randomized study arm. No treatment contrast is reported. It would therefore be inappropriate to create a new contrast from these summaries and present it as the trial's result.
What do the ranges add?
In general, a full range reports the smallest and largest observed values. The “Lingual Strengthening” range is 45.0 to 53.0, and the “Lingual Pressure Norms” range is 5.0 to 270.0. A range gives a sense of the span of recorded observations, but it does not indicate how densely values are distributed within that span. It is also sensitive to extreme observations. The registry does not provide a standard deviation, standard error, or other measure of dispersion for these groups. The listed full range should not be mistaken for a confidence interval: a confidence interval concerns uncertainty about an estimated quantity, whereas a range describes observed values.
In general, over-dispersion means that observations vary more than a statistical model expects. It matters because a model that assumes less variability than is present can report uncertainty too narrowly. The wide recorded range for the norms group motivates caution about variability, but the range alone is not enough to diagnose over-dispersion or determine an appropriate model. The registry does not report a distributional assessment or a model for the primary outcome.
Why is a confidence interval useful?
In general, a confidence interval expresses the precision of an estimated effect under a specified analysis. It provides a range of effect values compatible with the data and method, rather than only a single point estimate. A narrow interval indicates greater precision than a wide interval, while an interval that includes values representing little or no difference may leave substantial uncertainty about the direction or size of an effect. These interpretations depend on having a defined estimand, an appropriate model, and data for the groups of interest. No confidence interval is posted for the primary outcome, so precision for a comparative effect cannot be assessed from the registry record.
A p-value, in general, measures how unusual data at least as incompatible with a null hypothesis would be under a specified statistical model, assuming that hypothesis and model. It does not measure the size or clinical importance of an effect, nor the probability that a treatment is effective. The record does not report a p-value or any hypothesis test for the primary outcome. That absence means there is no posted inferential test to interpret; it is not itself evidence for or against a difference.
What role would randomization play?
In general, random assignment aims to make treatment groups comparable in expectation with respect to factors that could affect outcomes. A parallel design assigns participants to a group and compares outcomes across groups. This rationale supports causal interpretation when assignment, follow-up, outcome measurement, and analysis are handled appropriately. Here, randomization and a parallel model are listed, but the registry does not explain allocation implementation or identify how the two outcome labels map to randomized groups. Randomization as a design label cannot establish that the posted norms group is a valid comparator.
An analysis could also account for baseline values or other prespecified factors, but the record does not report whether such adjustment was planned or performed. No statistical analysis plan is reported in the registry record. It would be speculation to describe a particular model, covariate adjustment, hypothesis, or decision rule as the study's method.
How should planning illustrations be read?
The registry's stated planning assumptions describe an illustration based on the observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The target effect is expressed as a fraction of the observed effect. These assumptions describe a planning calculation, not the trial's own design justification. The registry does not report the original sample-size justification. Planning sample sizes are not observed power and do not establish whether a completed study did or did not have adequate power. An illustration based on an uncertain or ambiguous observed effect inherits that uncertainty.
In general, sample-size planning depends on the effect considered important, the variability of the outcome, the desired operating characteristics, and the planned analysis. If the effect is based on a small or selected set of observations, it may not represent the effect expected in a new study. No additional planning assumptions or design rationale are reported in the registry record. The most defensible statistical reading is therefore limited: the record supplies descriptive outcome summaries and general planning assumptions, but does not supply an inferential analysis of the randomized comparison.
10. Multiplicity, Missing Data and Analysis Populations
The registry describes several exercise approaches and lists two outcome measures, but it posts no analyses. It does not report a primary or secondary analysis set, a hierarchy of hypotheses, or a plan for addressing multiplicity. In general, multiplicity arises when several outcomes, groups, time points, or statistical tests are examined; testing many hypotheses can increase the chance of a result that appears persuasive by chance. A prespecified analysis plan can clarify which comparisons are primary and how additional analyses should be interpreted. Such a plan is not reported in the registry record.
The primary outcome description states that data were available for two arms, while data for the other treatment arms were unavailable because the principal investigator was not available. The record says that the posted values were initially entered by the research team and that efforts were made to retrieve data. It does not give participant-level reasons for missing outcome measurements, the number of missing observations within each group, or the timing and handling of missingness. Therefore, the available information does not distinguish between unavailable arm-level results and individual participants lacking measurements.
In general, missing outcome data can affect both the representativeness of the observed sample and the uncertainty of an estimate. The consequences depend on why data are absent and whether missingness relates to assignment, outcome, or other characteristics. No missing-data method or sensitivity analysis is reported. The registry also does not identify an intention-to-treat population, a per-protocol population, or another analysis population. These omissions prevent assessment of whether the posted groups represent all randomized participants or a selected subset.
The trial's completed status and total enrollment do not resolve these analysis questions. The posted outcome sample sizes and group labels are not enough to reconstruct participant flow or determine how all randomized participants were handled. Any conclusion about selective outcome availability, the direction of bias, or the effect of missing data would go beyond what is reported.
11. What the Result Does, and Does Not, Mean
The registry shows that maximum isometric tongue pressure values were posted for groups labelled “Lingual Strengthening” and “Lingual Pressure Norms.” The listed means are 49.0 kPa and 78.0 kPa, respectively, with sample sizes 11.0 and 72.0. Their full ranges are 45.0 to 53.0 and 5.0 to 270.0. These summaries describe the recorded entries; they do not establish a causal effect of exercise or a difference between randomized assignments.
The stated research question concerns whether intense oral exercise produces greater swallowing improvement than low-intensity oral exercise or sham exercise. The registry does not provide outcome results for all of these interventions, and it does not explain how the two posted group labels correspond to the randomized arms. It also does not identify the reported tongue-pressure values as changes from baseline. As a result, the stated comparative question remains unanswered by the posted outcome information.
The entries should not be interpreted as showing that any approach works or does not work. They do not establish how much participants changed, whether the groups were comparable at baseline, or whether an observed difference would be precise enough to distinguish from sampling variation. No comparative estimate, confidence interval, or statistical test is posted. Statistical uncertainty about the treatment contrast is therefore not quantified in the record.
Maximum isometric tongue pressure is also narrower than the broad concept of swallowing improvement. It measures peak pressure at sensors, as described in the registry, rather than directly summarizing all aspects of swallowing. The outcome can describe tongue strength while leaving other functional outcomes unaddressed. The registry lists another outcome measure but posts no analyses. The available record therefore does not support a broader conclusion about swallowing function.
Finally, zero serious and other adverse-event counts are recorded for both listed groups, while death counts are not reported. These entries should be interpreted as registry entries rather than a complete assessment of safety. Overall, the result's meaning is limited to the posted descriptive values and the disclosed gaps in outcome availability and analysis.
12. Why This Trial Matters Statistically
This study illustrates an important distinction in clinical-trial interpretation: the question a trial was designed to address is not always the question its accessible results can answer. The registry describes randomization and multiple exercise strategies, and it states a comparative objective. Yet the posted primary outcome contains values for only two groups with labels that do not clearly identify randomized assignments. A familiar design label cannot substitute for a traceable link between assignment, participants, outcomes, and analysis.
It also shows why descriptive summaries and inferential results serve different purposes. Group means and ranges can report what was recorded, but they do not quantify a treatment effect or its precision. Without an identified comparison and a confidence interval or other inferential analysis, readers cannot judge how compatible the observed summaries are with a range of possible treatment effects. The large difference in the labels' sample sizes and the broad range recorded for the norms group further make it important not to treat the summaries as a straightforward randomized comparison.
Missing outcome information is not merely a technical inconvenience. When results for some arms are unavailable, the evidence may no longer reflect the comparisons that motivated the trial. A clear accounting of group assignment, participant flow, missing outcomes, and analysis populations helps readers determine what the posted estimates represent. Those details are not reported in the registry record.
The record also provides a useful reminder that planning calculations do not replace the trial's own design rationale or outcome analysis. A planning illustration based on an observed effect is not observed power, and it cannot clarify an ambiguous comparison. Statistical transparency depends on the estimand, analysis plan, and available data being interpretable together. Here, the principal statistical lesson is to keep the claim aligned with the evidence: the registry reports outcome summaries, but does not report a complete comparative analysis of the exercise strategies.
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 most important limitation is that the primary outcome data are available for only two arms, according to the registry's description. The study lists multiple exercise interventions, but the posted results do not cover all of them. The record attributes the unavailable arm data to the principal investigator not being available and says the displayed numbers were initially entered by the research team. The values are therefore incomplete relative to the stated study question.
The two outcome groups are labelled “Lingual Strengthening” and “Lingual Pressure Norms.” Their relationship to the randomized intervention arms is not explained. In particular, the registry does not identify the norms group as a randomized treatment or control group. Without that mapping, an arm-to-arm comparison cannot be verified from the record.
The outcome summaries are means and full ranges, with no reported standard deviations, confidence intervals, or inferential analysis. The record does not say whether these values are endpoint measurements or changes from baseline. The timeframe is listed as 16 months, but further detail about measurement timing and participant follow-up is not reported. The registry does not explain how tongue pressure measurements were aggregated across the four sensors.
Other trial details needed to assess validity are also absent from the record, including allocation implementation, analysis populations, missing-data methods, and a statistical analysis plan. The record does not report the trial's sample-size justification. Adverse-event entries are limited to the listed groups, and death counts are not reported. These limitations are specific to the information available in the registry and do not establish what may have been recorded elsewhere.
14. Related Tutorials
15. Related Statistical Calculators
16. Sources
- ClinicalTrials.gov: Exercise for Swallowing Problems After Stroke — NCT00722111.
- Registry record last updated 2018-03-21; status last verified 2018-03.
- Results first posted 2018-03-21.
- 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
NCT00722111 is a completed, randomized, parallel study of exercise approaches for swallowing problems after stroke. The registry lists enrollment of 83, no masking, and the sponsor as the VA Office of Research and Development. The recorded study start is 2011-02, and primary completion is 2011-12.
The stated objective was to compare intense oral exercise with low-intensity oral exercise or sham exercise. The primary outcome is maximum isometric tongue pressure in kPa at a timeframe of 16 months, described as peak pressure at four sensors. Posted summaries are available for “Lingual Strengthening” and “Lingual Pressure Norms,” but the registry says data for other treatment arms were unavailable and does not explain how the two labels map to randomized assignments.
No analysis is posted. The registry record therefore supplies descriptive values but does not report a comparative estimate for the intended treatment question. Missing-data methods, analysis populations, and the original sample-size justification are not reported in the registry record.