Treatment Intensity - Apraxia of Speech: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Treatment Intensity - Apraxia of Speech (NCT02332915), a randomized na study. The primary outcome is Accuracy of Articulation of TREATED ITEMS Measured as Effect Size - Change From Baseline to End of Treatment Phase.

2. Clinical Question

3. Trial Design

NCT02332915, titled “Treatment Intensity - Apraxia of Speech,” was registered as a randomized, single-masked crossover study of behavioral Sound Production Treatment. Its official title was “Effect of Intensity of Treatment on Rehabilitation of Acquired Apraxia of Speech.” The registry lists the phase as not applicable and the status as completed. The sponsor was the VA Office of Research and Development.

The registry lists enrollment as 36. Participants were assigned to one of two treatment sequences. One sequence began with intense SPT and then received non-intense SPT; the other began with traditional, non-intense SPT and then received intense SPT. The registry summary describes comparing intensity while holding the total number of sessions constant. The detailed schedule and other treatment-delivery particulars are not reported in the registry record.

The primary outcome timeframe includes pretreatment assessment, assessment 2 weeks following the first treatment phase, and assessment 2 weeks following the second treatment phase. The registry reports outcome summaries with 24 observations for the intense condition and 24 for the traditional condition. Those condition-specific figures should not be added or treated as an explanation of the registry enrollment; the record does not explain the relationship between the enrollment figure and the analysis counts.

The listed secondary outcome concerns speech intelligibility, measured as percent intelligible words comparing baseline with the end of treatment phases. The registry lists three outcome measures and three analyses, but only one secondary outcome is posted in the record summarized for this analysis. Further details about sequence allocation, masking procedures, handling of period or order effects, and the analysis population are not reported in the registry record.

4. Participant Flow and Accounting

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

Overall Study

Reasons for not completing this period:

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

Population: Persons with aphasia and apraxia of speech who completed both phases of treatment.

6. Primary Endpoint

The registry lists the primary outcome as Accuracy of Articulation of TREATED ITEMS Measured as Effect Size - Change From Baseline to End of Treatment Phase, measured over Pre treatment, 2 weeks following the first treatment phase, 2 weeks following the second treatment phase.

The primary endpoint concerns the accuracy of articulation for sounds in words, phrases, and sentences designated for treatment. Participants’ productions were elicited through repetition in nontreatment probes. Percent accuracy was calculated for each probe, with the scale ranging from 0% to 100% correct. The outcome then expresses change from baseline to 2 weeks after a treatment phase as an effect size, intended to describe the magnitude of change while accounting for variability in performance.

This endpoint is used to focus on whether trained speech items changed after treatment, rather than relying only on a general impression of speech performance. The effect-size direction has a stated interpretation: positive values indicate increases in accuracy, and negative values indicate decreases. Larger positive effect sizes indicate greater improvement in the measured performance, according to the registry description. This measure is specific to treated items and the probe method described; it should not be read as a direct measure of every aspect of communication or daily functioning.

The registry reports a mean and standard deviation for each treatment condition. It does not give confidence limits for the individual condition means. The comparison between condition means, and its uncertainty, is therefore useful for interpreting the size and precision of the reported difference, while keeping in mind that the trial used a crossover design.

7. Results Posted in the Registry

• SPT - Traditional (Non-intense): All Participants: mean 28.90 versus 19.70 on control; difference 9.20 (95% CI 0.67 to 17.73), standardized effect d = 0.61; z-test p = 0.035.

8. Statistical Interpretation

9. Statistical Methods Explained

What does the mean summarize?

The primary outcome is continuous: it is an effect-size measure of change in articulation accuracy for treated items. A mean summarizes the average observed effect-size value within a treatment condition. The registry reports a mean of 19.7 for intense SPT and 28.9 for traditional, non-intense SPT. A mean describes the center of the observed values; it does not show whether every participant had a similar outcome or whether individual outcomes varied widely.

The standard deviation describes how dispersed the observed values are around their mean. The reported standard deviations are 12.6 for intense SPT and 17.2 for traditional, non-intense SPT. Standard deviation is a description of variability among observations, not a measure of uncertainty in the mean difference. The uncertainty in the comparison is instead conveyed by its confidence interval.

What does the mean difference represent?

The recalculated comparison subtracts the intense-condition mean from the traditional-condition mean. The resulting mean difference is 9.2 in the registry’s effect-size scale. A positive value under that direction means the traditional-condition mean is higher. This is an unadjusted recalculation based on the posted summary statistics; it is not a participant-level analysis and does not incorporate any design adjustment that may have been used by the sponsor.

A confidence interval for a mean difference gives a range of values compatible with the data and the method used to construct the interval. The recalculated 95% confidence interval has positive endpoints, with the upper endpoint at 17.73. Since its lower endpoint is also above zero, the interval does not include a zero difference. The interval describes uncertainty about the contrast under the recalculation’s assumptions. It does not indicate the probability that the true difference lies in the particular interval.

How should the p-values be read?

The recalculated p-value is 0.035. In general, a p-value measures how unusual data at least as inconsistent with a null hypothesis would be if that null hypothesis and the test assumptions were correct. It is not the probability that the null hypothesis is true, and it does not quantify the importance of the observed difference. The posted analysis identifies the sponsor’s method as a two-sided t-test for superiority and reports a p-value that differs from the recalculated value. The record does not provide enough details to establish why the values differ.

A two-sided test allows for evidence in either direction, rather than testing only whether one named condition has a higher result. A t-test evaluates a difference in means using a reference distribution that accounts for estimated variability. In a crossover trial, however, paired observations and treatment order may matter. The registry record does not report whether the sponsor’s test incorporated the crossover structure, and the unadjusted recalculation does not model it.

Why does the crossover structure matter?

In a crossover design, participants receive more than one treatment condition in sequence. Each participant may therefore contribute information under both conditions, and comparisons can depend on how within-participant measurements are handled. Treatment sequence, period effects, and possible carryover are general considerations for such designs. The registry record gives the two sequences but does not report the details of any period, sequence, or carryover analysis. It would be inappropriate to assume a particular model or adjustment.

The recalculated comparison uses the posted means, standard deviations, and condition counts as an unadjusted summary-statistic comparison. It should not be interpreted as a paired crossover analysis. The sponsor’s own test is separately reported; details needed to reproduce or fully assess that analysis are not reported in the registry record.

What is the role of variability and over-dispersion?

General statistical explanation: dispersion matters because a mean difference is easier to estimate precisely when observations are less variable, all else being equal. Greater variability tends to widen uncertainty around an estimate. In count outcomes, over-dispersion means that variation exceeds what a simple count model expects; it can make standard errors too small if ignored. Over-dispersion is not the model issue identified for this continuous endpoint, and the registry does not report an over-dispersion analysis. Here, the relevant posted variability measures are the condition standard deviations.

General explanation: an effect-size measure may put change in relation to variability, but its interpretation depends on the definition used. The registry describes corrections for variability in performance and defines the direction of change, but it does not provide every computational detail of the effect-size calculation in the information available for this analysis. The reported effect-size means should therefore be interpreted on the registry’s stated scale, not converted into a percentage-point change or another scale.

10. Confidence Intervals and Precision

The recalculated mean difference is 9.2, comparing traditional, non-intense SPT with intense SPT. Its 95% confidence interval has a positive lower endpoint and an upper endpoint of 17.73. Because zero is outside this interval, the range of differences compatible with the recalculation’s assumptions is entirely in the direction of a higher mean for traditional treatment. The exact lower endpoint is not repeated; the important interpretation is that it is above zero.

The interval also communicates uncertainty about magnitude. Its span allows for a range of positive differences rather than identifying one exact value as certain. A confidence interval is not a guarantee that the true difference lies within its bounds, nor does it account for limitations that are not captured by the calculation, such as unreported analysis choices or the crossover structure. It quantifies uncertainty under the method used for the recalculation.

The condition-level standard deviations are 12.6 for intense SPT and 17.2 for traditional SPT. These describe variability in the outcome values and help explain why a mean comparison should not be read as a uniform result for every participant. The registry does not report confidence intervals for the condition means in the information, and no participant-level distribution is available in the record summarized for this analysis.

The sponsor’s two-sided t-test reports a different p-value from the recalculated test. Without fuller analysis details, the recalculated confidence interval should be treated as an unadjusted precision estimate, not as a confidence interval for the sponsor’s specific model. The registry record does not report the sponsor’s confidence interval for the treatment contrast.

11. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The planning figures are participants per arm, not total events. They are illustrations derived from the observed treatment contrast, rather than a report of the trial’s original design calculation. The planning assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. Each planning scenario considers detecting a stated fraction of the observed effect.

The 80% power column gives the illustrative number of participants per arm for a design with an 80% chance of detecting the specified effect if that effect were the true one and the planning assumptions held. The 90% power column gives the corresponding number per arm for a design with a 90% chance under those assumptions. These are prospective planning quantities, not summaries of what happened in this trial.

For planning at 100% of the observed effect, the figures are 43 participants per arm for 80% power and 57 participants per arm for 90% power. For planning at 75% of the observed effect, the figures are 75 and 101 participants per arm, respectively. For planning at 50% of the observed effect, the figures are 169 and 226 participants per arm, respectively. The pattern illustrates a general principle: if the effect to be detected is smaller, more participants are needed to achieve the same target power, all else being equal.

These calculations are not observed power. The original trial’s sample-size justification and design assumptions are not reported in the registry record. The planning exercise also uses a two-arm equal-allocation framework, whereas the trial itself is registered as a crossover study, so the figures should not be read as a reconstruction of the trial’s design requirements.

12. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = 9.200; pooled standard deviation σ = 15.076.
2. n = 2 × (1.96 + 0.842)² × 15.076² / 9.200² = 42.2, rounded up to 43 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
delta <- 9.2                      # observed difference in means
sigma <- 15.0764717357875                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(2332915)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 43 (80% power), 57 (90% power)
 75% of observed effect: n per arm = 75 (80% power), 101 (90% power)
 50% of observed effect: n per arm = 169 (80% power), 226 (90% power)
power.t.test: 43.1 
Simulated power at n = 43 per arm: 0.798 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

13. Analyses Reported by the Sponsor

The registry reports the sponsor’s own analysis as a two-sided t-test for superiority comparing the two SPT conditions. It gives a p-value, but does not report a test statistic, confidence interval, or further details of the analysis in the record summarized for this analysis. That reported test is the sponsor’s trial analysis and should be distinguished from the unadjusted recalculation based on the posted condition means and standard deviations.

The recalculation gives a mean difference of 9.2, a 95% confidence interval with a positive lower endpoint and upper endpoint of 17.73, and an unadjusted p-value of 0.035. The sponsor’s reported p-value differs from that recalculated value. Since the underlying analysis details are not reported, the difference cannot be attributed to a particular modeling choice from the registry record alone. The recalculation does not replace the sponsor’s test, and the sponsor’s analysis is the primary account of the trial’s own inferential result.

The trial is described as a crossover study, but the record does not clarify whether the sponsor’s t-test accounted for within-participant pairing or treatment sequence. The recalculation is explicitly unadjusted and should not be mistaken for an analysis that resolves those design questions.

14. Multiplicity, Missing Data and Analysis Populations

The registry lists three outcome measures and three analyses, while only one secondary outcome is posted in the record summarized for this analysis. The primary outcome concerns articulation accuracy for treated items; the listed secondary outcome concerns percent of intelligible words from baseline to the end of treatment phases. The other outcome and analysis details are not reported in the registry record.

Multiplicity arises when several outcomes, comparisons, or analyses are tested. In general, conducting multiple tests increases the chance that at least one result will appear statistically notable by chance, if no adjustment or other control is used. A multiplicity strategy may specify which outcome is primary, how secondary outcomes are interpreted, or whether p-values are adjusted. The registry record does not report whether any multiplicity adjustment was planned or applied, nor does it give a complete account of the three analyses.

The crossover design adds further analysis questions. The record identifies treatment sequences and assessment after each phase, but does not report how the analysis treated period, sequence, within-participant correlation, or possible carryover. These considerations may affect standard errors and interpretation in a crossover study. No particular model should be inferred from the reported t-test label alone.

The record also does not report missing-data amounts, reasons for missingness, imputation methods, or sensitivity analyses. It does not identify an intention-to-treat, per-protocol, or other named analysis population. The condition summaries each report 24 observations, but the registry’s enrollment figure is 36, and the record does not explain the relationship between those counts. Without those details, it is not possible to assess whether missingness or analysis-population decisions could have affected the estimates.

These omissions do not establish that particular methods were or were not used. They mean that the registry record does not support a detailed evaluation of multiplicity control, missing-data handling, or the exact analysis population.

15. 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 0% reporting threshold.

The serious adverse event listings report 0 affected in the intense SPT condition and 0 affected in the traditional, non-intense SPT condition. The registry also lists 0 deaths for each condition. The at-risk counts are reported as 24 for each condition. These are the adverse-event figures recorded for the listed treatment conditions.

A zero in the registry’s serious adverse event count means that no serious events are recorded for that category in the posted results. It does not establish that no less serious symptoms or other experiences occurred, and it does not show that every possible safety outcome was assessed or captured. The registry lists no serious adverse event terms. Other adverse events are also listed as 0 affected in each condition, but the record does not provide detailed event descriptions in the registry record for this analysis.

These figures should be interpreted as registry reporting, not as proof that the intervention has no risks. The record does not report additional safety-monitoring details or specify how adverse events were collected. The counts are descriptive and do not provide a precise estimate of the likelihood of uncommon events.

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

The posted outcome summaries show a higher mean effect-size value for traditional, non-intense SPT than for intense SPT: 28.9 compared with 19.7. The unadjusted recalculation gives a mean difference of 9.2 in the direction of the traditional condition. Its 95% confidence interval remains above zero, and its unadjusted p-value is 0.035. The sponsor also reports a two-sided superiority test, with a p-value that differs from the recalculation.

These statistics describe change in articulation accuracy for treated items, measured through repetition probes and expressed as an effect size. They do not directly describe every speech or communication outcome. The listed secondary outcome concerns intelligibility, but detailed results for that outcome are not included in the record summarized for this analysis. The primary outcome result should not be generalized to outcomes that are not reported.

The comparison also does not establish a universal ranking of treatment intensities. It pertains to the trial population, treatment procedures, measurement approach, and analysis information reported in the registry. The record describes a crossover design, but does not provide enough detail to determine how the sponsor handled sequence, period, pairing, or carryover in its analysis. The recalculation does not account for those design elements.

A statistically notable difference is not the same as proof of clinical importance. The registry record does not state a threshold for a clinically meaningful effect-size difference, nor does it report how the observed contrast relates to participants’ everyday communication. The data therefore support describing the measured contrast and its uncertainty, not claiming a level of practical importance that the record does not establish.

The registry’s adverse-event figures report no serious events or deaths in either condition. Those counts are useful as a description of the posted safety record, but they do not prove the absence of all harms or quantify rare risks. Overall, the result is best read as a comparison of the reported treated-item outcome under the registry’s conditions, with important analysis and generalizability details unreported.

17. Why This Trial Matters Statistically

This trial raises a statistical question that is common in rehabilitation research: how should treatment intensity be compared when participants receive multiple conditions in sequence? A crossover design can make within-participant information relevant, but interpretation depends on how treatment order, period, and repeated observations are handled. The registry names the design and sequences, while leaving key analysis details unreported. That gap matters because a simple comparison of condition summaries is not necessarily equivalent to a design-aware crossover analysis.

The outcome also illustrates the importance of defining the scale before interpreting a mean difference. The endpoint is an effect size for change in articulation accuracy of treated items, rather than a raw percentage-point contrast. The registry explains the direction of the scale and its connection to variability, but the result should remain on that stated scale. Clear separation between a measure’s units and a more familiar clinical quantity helps avoid overstatement.

Finally, the record distinguishes a sponsor-reported analysis from an unadjusted recalculation. Both can be informative, but they answer questions under potentially different analytic assumptions. Reporting the estimate, uncertainty, test, and missing analytic details separately helps readers avoid treating one p-value as a complete account of the evidence. The planning illustration likewise demonstrates how assumed effect size and desired power affect prospective sample-size requirements without implying that the trial’s actual design was based on those calculations.

18. 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 registry record leaves important details unavailable for evaluating the result. It does not report how the sponsor’s t-test handled the crossover structure, including paired measurements, treatment sequence, period effects, or possible carryover. The unadjusted recalculation does not account for those features. The record also does not provide a sponsor confidence interval for the treatment contrast or enough information to reproduce the sponsor’s test.

The enrollment figure is 36, while each condition summary reports 24 observations. The registry record does not explain the relationship between these figures, so they should not be combined or interpreted as a reconciled participant count. Details about missing observations, participant flow, analysis populations, and handling of missing data are not reported.

Three outcome measures and three analyses are listed, but only one secondary outcome is posted in the record summarized for this analysis. The record does not describe multiplicity procedures or give results for the other measures. It also does not report a threshold for a clinically meaningful difference, participant-level distributions, or fuller demographic and eligibility details. These limits constrain assessment of clinical relevance, robustness, and generalizability.

The posted safety summaries record no serious adverse events or deaths in either condition, but the registry information does not describe the full ascertainment process or provide detailed event narratives. The registry’s planned sample-size assumptions are also not reported; the available sample-size illustrations are based on the observed contrast and should not be treated as the original design justification.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Treatment Intensity - Apraxia of Speech — NCT02332915.
• Registry record last updated 2022-07-12; status last verified 2022-03.
• Results first posted 2020-11-17.
• 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.

22. Record Summary

NCT02332915 was a completed randomized, single-masked crossover study of Sound Production Treatment intensity for acquired apraxia of speech. The registry lists enrollment as 36 and identifies two sequences: intense treatment first or traditional, non-intense treatment first. The primary outcome measured effect-size change in articulation accuracy for treated items.

The reported mean effect sizes were 19.7 for intense treatment and 28.9 for traditional treatment. An unadjusted recalculation gives a mean difference of 9.2, with a 95% confidence interval above zero and an unadjusted p-value of 0.035. The sponsor reports a two-sided superiority t-test with a different p-value. The record does not report key details needed to assess crossover adjustments, missing data, analysis populations, or multiplicity. Serious adverse events and deaths are reported as 0 in both conditions.