BOTOX® Treatment in Pediatric Lower Limb Spasticity: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for BOTOX® Treatment in Pediatric Lower Limb Spasticity (NCT01603628), a randomized phase 3 study. The primary outcome is Average Change From Baseline in Modified Ashworth Scale-Bohannon (MAS-B) Ankle Score With Knee Extended at Weeks 4 and 6.

2. Clinical Question

This is a question about a measured change in muscle tone, not a direct measure of every aspect of spasticity, function, or daily experience. The registry also lists physician-assessed Goal Attainment Score and change in ankle spasticity using the Modified Tardieu Scale among secondary outcomes. The primary comparison is between each dose and placebo; it does not establish how the two BOTOX® doses compare directly with one another. The record does not report a direct dose-to-dose analysis for this endpoint.

3. Trial Design

The registry describes a randomized, quadruple-masked, parallel-group phase 3 study sponsored by Allergan. Its listed enrollment was 384. The study is marked completed, with a start date of 2012-09-11 and primary completion date of 2017-06-28. The registry record does not report further details about who was masked or how the allocation sequence was generated.

Three groups are described: BOTOX® 8 U/kg, BOTOX® 4 U/kg, and normal saline placebo. Injections were given intramuscularly on Day 1 into specified muscles of the lower limb. Weekly physical therapy was part of each arm description, so the primary comparison concerns assignment to the injection treatment in the setting of that shared therapy, rather than BOTOX® versus no accompanying therapy.

The posted primary outcome is the average change from baseline in the Modified Ashworth Scale-Bohannon ankle score with the knee extended at Weeks 4 and 6. The registry reports an analysis population defined by baseline and post-baseline measurement availability, rather than simply describing every randomized participant as included in the endpoint analysis. The primary outcome was analyzed using a Mixed Model Repeated Measures model.

The listed secondary outcomes include physician-assessed Goal Attainment Score using a 6-point scale and change in ankle spasticity with the knee extended and flexed, calculated using the Modified Tardieu Scale. The record lists four outcome measures and 22 analyses. Those counts do not, by themselves, explain which findings were treated as confirmatory or how statistical errors across outcomes and comparisons were controlled. The trial's original sample-size rationale and other design assumptions 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.

Before assignment: Pediatric participants with lower limb spasticity were randomized 1:1:1 to one of three treatment groups: BOTOX® 4 or 8 U/kg (unit per kilogram) or placebo.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Participants from the mITT population, all randomized participants with a valid MAS-B baseline ankle score with knee extended and at least one post-baseline measurement at Weeks 2, 4, or 6 for the MAS-B of the ankle score with knee extended and the CGI by physician, with data available for analysis.

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: All randomized participants.

6. Primary Endpoint

The registry lists the primary outcome as Average Change From Baseline in Modified Ashworth Scale-Bohannon (MAS-B) Ankle Score With Knee Extended at Weeks 4 and 6, measured over Baseline (Day 1) to Weeks 4 and 6.

The primary endpoint is the average change from baseline in the Modified Ashworth Scale-Bohannon (MAS-B) ankle score with the knee extended at Weeks 4 and 6. It is a continuous measure of resistance encountered during passive movement of a limb. The registry describes grading on a 6-point scale, from 0 for no increase in muscle tone to 4 for the affected part being rigid in flexion or extension, and says scores were recoded for analysis. The recoding was 0=1, 1=1, 1+=2, 2=3, 3=4, and 4=5.

The endpoint uses a change from baseline rather than a single follow-up score, so it expresses how the score shifted over the specified follow-up period. The Week 4 and Week 6 scores were averaged. The registry states that a negative change indicates improvement. This direction matters when reading the group means and between-group estimates: a more negative change represents a greater decrease in the coded score.

The measure is used to quantify muscle tone through a standardized clinical rating of resistance to passive stretch. It does not directly capture every dimension of spasticity or functional ability. The registry's general description of the scale refers to grading resistance in the elbow and wrist, while the named primary endpoint concerns the ankle. That wording is a limitation in the endpoint description; the registry record does not clarify it further. The primary endpoint's listed unit is a score on a scale, and the group results are least-squares means with standard errors.

7. Results Posted in the Registry

• BOTOX® 8 U/kg: mean -1.06 versus -0.80 on control; difference -0.26 (95% CI -0.46 to -0.06), standardized effect d = -0.33; z-test p = 0.010.
• BOTOX® 4 U/kg: mean -1.01 versus -0.80 on control; difference -0.21 (95% CI -0.41 to -0.01), standardized effect d = -0.27; z-test p = 0.038.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary analysis used a Mixed Model Repeated Measures (MMRM) model. In general, this kind of model analyzes repeated measurements from participants over time while accounting for the fact that measurements from the same participant are related. It can estimate group means and between-group differences while using the available repeated outcome data under the model's specified structure. The registry record does not report the model's full specification, such as its covariance structure or all covariates, so those details cannot be assessed from the posted description.

What does a least-squares mean represent?

A least-squares mean is a model-based estimate of a group mean. It is not necessarily the same as the raw average calculated directly from all observed scores. The reported arm results are identified as least-squares means, and the comparisons are identified as least-squares mean differences. Interpretation of adjusted or model-based estimates depends on the model and its inputs. The registry identifies the MMRM approach but does not report enough detail to reproduce the model or determine all of the adjustments used.

Why use repeated-measures analysis?

Participants had assessments across follow-up, and the endpoint uses scores at Weeks 4 and 6. Repeated-measures analysis can use the pattern of outcome observations across visits instead of treating every measurement as unrelated. In general, accounting for within-participant correlation can make inference more appropriate for longitudinal data. A model can also incorporate observations from participants who have some, but not all, scheduled outcome measurements, depending on its specification and assumptions. The registry does not report the assumptions used to address missing measurements or the extent and reasons for missingness, so the performance of that aspect of the analysis cannot be evaluated from the record alone.

How should the score and direction be read?

The MAS-B is an ordinal clinical rating that the registry recoded for analysis. The record says that the score at Weeks 4 and 6 was averaged and that negative change from baseline indicates improvement. The posted results are therefore best read as model-based average changes on the recoded score, rather than as a percentage change or a count of participants who improved. A mean can summarize a group while hiding variation among individuals; it does not imply that every participant had the group-average experience.

What does the comparison estimate?

The contrast is the difference between the estimated change for one BOTOX® dose group and the estimated change for placebo. A negative difference means the estimated change is more negative in the BOTOX® group. The registry gives separate comparisons for 8 U/kg versus placebo and 4 U/kg versus placebo. It does not report a direct comparison between the two dose groups in the primary analysis. Nor does the endpoint comparison isolate the effect of injections from all other care: weekly physical therapy is described for all three arms.

What do the p-values tell us?

A p-value describes how compatible the observed data are with a specified null hypothesis under the analysis assumptions. It is not the probability that the null hypothesis is true, nor does it measure the size or clinical importance of an effect. The registry reports p = 0.010 for the 8 U/kg comparison and p = 0.033 for the 4 U/kg comparison. The record labels both analyses as superiority tests, but the registry record does not report whether the p-values were adjusted for multiple comparisons or outcomes. That omission matters when interpreting several statistical tests together.

In general, over-dispersion occurs when observed variability is greater than a model assumes. If present and not appropriately handled, it can make standard errors and inferential results unreliable. This endpoint is analyzed with an MMRM, and the registry provides standard errors for the group estimates; it does not report a specific over-dispersion assessment. There is no basis in the record for asserting that over-dispersion was present or absent.

There are also limits to what can be inferred from the method label alone. The record does not describe the prespecified analysis plan, the model's detailed terms, diagnostics, sensitivity analyses, or the handling of intercurrent events. These items are not reported in the registry record. General statistical interpretation can explain what MMRM estimates mean, but it cannot fill in those trial-specific methodological details.

10. Confidence Intervals and Precision

The registry reports 95% confidence intervals for both primary comparisons. For BOTOX® 8 U/kg versus placebo, the estimated least-squares mean difference is -0.26 and the interval extends from -0.453 to -0.063. For BOTOX® 4 U/kg versus placebo, the estimate is -0.21 and the interval extends from -0.405 to -0.018. Each interval is compatible with a range of effect sizes under the model and analysis used, rather than identifying one exact effect as certain.

In general, a confidence interval conveys the precision of an estimate: a narrower interval indicates less statistical uncertainty than a wider interval, all else being equal. It does not capture every possible source of uncertainty, including limitations in measurement, model specification, missing-data assumptions, or selective reporting. The intervals describe uncertainty under the registry's MMRM analysis; the complete model specification is not reported in the registry record.

Both posted intervals lie below zero. This is consistent with the negative estimated differences and the registry's definition that a negative change from baseline indicates improvement on the score. The interval for each comparison also gives a sense of the range of differences compatible with the analysis. It does not state the probability that the true effect lies within that particular realized interval, and it does not establish whether the difference is important in clinical terms.

The record does not report a clinically important difference threshold for the MAS-B endpoint. Without such a threshold, statistical precision and clinical importance must remain distinct questions. The posted intervals also do not resolve uncertainty about multiplicity, because the registry record does not report the adjustment strategy, if any, for the set of analyses.

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 illustrations based on the observed effect, not the trial's original design calculations. For these continuous-outcome comparisons, the figures are participants per arm, not total events. The planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, a normal approximation, and an effect to detect set to the stated fraction of the observed effect. The trial's own original design assumptions are not reported in the registry record.

The 80% power column gives the estimated number of participants per arm for a design that would have an 80% chance of detecting the specified effect if that effect and the planning assumptions were correct. The 90% power column gives the corresponding planning number for a higher chance of detection. These are design calculations under assumptions, not guarantees about a future study's result.

For the BOTOX® 8 U/kg comparison, planning for 100% of the observed effect gives 146 participants per arm for 80% power and 195 for 90% power. Planning for 75% of the observed effect gives 259 and 346 participants per arm, respectively; planning for 50% gives 581 and 778. For BOTOX® 4 U/kg, the corresponding figures are 223 and 298 at 100% of the observed effect, 395 and 529 at 75%, and 889 and 1,189 at 50%.

Smaller effects require more participants to detect with the same target power, which is why the planning numbers rise as the assumed fraction of the observed effect falls. These figures are planning illustrations, not observed power. They do not tell us the probability that the completed trial's result is true or the probability that a repeat study will succeed. The sample-size assumptions do not account for every practical feature of a trial, and they should not be read as a reconstruction of the original design.

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 = -0.260; pooled standard deviation σ = 0.791.
2. n = 2 × (1.96 + 0.842)² × 0.791² / -0.260² = 145.2, rounded up to 146 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 <- -0.26                      # observed difference in means
sigma <- 0.790648457748345                     # 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(1603628)
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 = 146 (80% power), 195 (90% power)
 75% of observed effect: n per arm = 259 (80% power), 346 (90% power)
 50% of observed effect: n per arm = 581 (80% power), 778 (90% power)
power.t.test: 146.1 
Simulated power at n = 146 per arm: 0.803 (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 MMRM analyses as least-squares mean differences. For BOTOX® 8 U/kg versus placebo, the posted estimate is -0.26, with a 95% confidence interval from -0.453 to -0.063 and p = 0.010. For BOTOX® 4 U/kg versus placebo, it is -0.21, with a 95% confidence interval from -0.405 to -0.018 and p = 0.033.

The unadjusted recalculations from the posted arm-level summaries also give estimates of -0.26 and -0.21 for the two respective contrasts. The recalculated confidence limits use an approximation based on the reported standard errors, whereas the sponsor's analyses use the stated MMRM. The registry does not flag a discrepancy between the recalculated and sponsor-reported results. The sponsor's model-based estimates and inferential results therefore remain the primary reported analysis, rather than being replaced by the simpler summary-statistic recalculation.

A recalculation from arm summaries cannot reproduce all features of a repeated-measures model. It is useful as a check on the broad scale and direction of the posted contrast, but it is not an independent confirmatory analysis. The sponsor's analysis is the analysis reported for the trial, and the registry record does not provide enough model detail to independently reconstruct it.

14. Multiplicity, Missing Data and Analysis Populations

The registry lists four outcome measures and 22 analyses, including the primary endpoint and secondary outcomes. Multiple statistical tests can increase the chance of obtaining at least one apparently unusual result when all null hypotheses are true. In general, a multiplicity strategy may control that risk by specifying which tests are primary, arranging tests hierarchically, or adjusting inference. The registry record does not report whether such a strategy was prespecified or used, and it does not state whether the posted p-values account for multiple comparisons.

The primary result includes two dose-versus-placebo comparisons. The record reports a p-value and confidence interval for each, but does not explain how inference across these comparisons was handled. The results should therefore be understood as the posted analyses, without assuming a particular family-wise error control or adjustment. The presence of several posted analyses is not, by itself, evidence that an adjustment was or was not performed.

The primary analysis population is described as mITT: randomized participants with a valid baseline ankle score with the knee extended and at least one post-baseline measurement at Weeks 2, 4, or 6 for that score and the physician CGI, with data available for analysis. This definition makes clear that inclusion in the analysis depended on outcome data availability. The registry does not report the number excluded from the primary analysis population or provide a full accounting of missing outcome data by group.

MMRM can, in general, use available repeated observations without requiring every participant to have every scheduled measurement, but validity depends on model specification and assumptions about missingness. The registry does not report the missing-data assumptions, reasons for missingness, or sensitivity analyses that might test those assumptions. It also does not report the full analysis plan, protocol deviations, or whether any analyses were changed after the data were observed. These items are not reported in the registry record, so the robustness of the primary inference to those issues cannot be determined from the posted information.

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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: seizure (1.6%), extrasystoles (0.8%), tachycardia (0.8%), tonsillar hypertrophy (0.8%), gastoeneteritis (0.8%).

The registry reports serious adverse events for groups at risk of 128 participants in the BOTOX® 8 U/kg arm, 126 in the BOTOX® 4 U/kg arm, and 128 in the placebo arm. Serious adverse events affected 0 participants in the 8 U/kg arm, 3 in the 4 U/kg arm, and 4 in the placebo arm. The record reports 0 deaths in each arm.

These are counts of participants affected, not a complete description of the type, timing, duration, or severity of every event. The registry lists seizure at 1.6%, and extrasystoles, tachycardia, tonsillar hypertrophy, Gastoeneteritis, and radicular pain at 0.8% each among the top serious adverse event terms. These terms and percentages do not establish that an event was caused by the assigned treatment.

Other adverse events affected 24 participants in the 8 U/kg arm, 26 in the 4 U/kg arm, and 35 in the placebo arm. The record's counts should be read with their at-risk group sizes and as registry-reported summaries. They do not by themselves establish comparative safety, and the registry record does not provide further detail about event-level follow-up or causal assessment.

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

The posted primary analysis found negative estimated changes in all three groups on the recoded MAS-B score. The registry defines negative change from baseline as improvement. The two BOTOX®-versus-placebo estimates were also negative, indicating a greater estimated decrease in the score for each BOTOX® group than for placebo over the endpoint's averaged Week 4 and Week 6 measurements.

The numerical findings describe group-level model estimates. They do not mean that every participant improved, and they do not reveal the distribution of individual responses. The record does not report a clinical-importance threshold for this score, so the statistical differences cannot be translated from the registry alone into a judgment about the importance of the change for participants.

The primary endpoint is a rating of resistance during passive movement, with the ankle assessed while the knee was extended. It is not a direct measure of all symptoms, mobility, participation, or quality of life. The record lists other outcome measures, but the primary result does not substitute for those outcomes. It also represents the specified follow-up window, not an estimate of how long any observed change persisted.

The findings do not compare BOTOX® directly against no treatment: all groups received weekly physical therapy, and the control group received saline injections. Nor do the two placebo comparisons establish which BOTOX® dose is preferable; a direct comparison between dose groups is not reported for this endpoint. The statistical results should be read as evidence about these specific comparisons, outcome definition, analysis population, and follow-up, rather than as a broad conclusion about all aspects of treatment or care.

Safety counts are separate from the primary efficacy analysis and should not be inferred from the primary score. Serious and other adverse event summaries are descriptive registry figures, and the posted information does not by itself establish causality. Overall, the record provides estimates and uncertainty for a particular outcome; it does not support medical advice or conclusions beyond what those analyses measure.

17. Why This Trial Matters Statistically

This trial illustrates several statistical issues that arise in clinical research. First, the endpoint is a repeated clinical score, and the primary analysis uses a repeated-measures model. The model-based estimate is designed to address repeated observations, but understanding its exact behavior requires more than the model name: covariance structure, covariates, missing-data assumptions, and diagnostics are not reported in the registry record.

Second, the endpoint is a recoded rating scale. Reporting a mean change can make group comparisons compact, but an average does not show the spread or pattern of participant-level changes. The clinical meaning of a score difference also depends on an interpretable threshold, which is not reported for this endpoint. Statistical significance and clinical importance are distinct questions.

Third, two dose groups are each compared with a shared placebo group, while several outcomes and analyses are listed. Multiplicity is therefore an important interpretive issue. The registry reports the individual comparisons but does not report a correction strategy. Readers can distinguish the posted estimates from any broader claim that all possible statistical error has been controlled.

Finally, planning calculations based on observed effects are sensitive to the effect size chosen. The illustrations show that targeting a smaller fraction of the observed effect entails larger participant requirements under the stated assumptions. Such calculations are not the same as the trial's original design rationale, which is not reported, and they are not observed power. This separation between observed results, inferential uncertainty, and forward-looking design planning is essential when interpreting clinical trial statistics.

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 does not report the full statistical analysis plan or detailed MMRM specification. In particular, the covariance structure, covariates, model diagnostics, and sensitivity analyses are not described. Without those details, the posted model-based results cannot be independently reconstructed from the record.

The record does not report the original sample-size justification, the assumptions used in the trial's design, or whether interim analyses were planned or conducted. It also does not describe a multiplicity strategy for the two dose comparisons or the broader set of listed analyses. The posted p-values and intervals can be read as reported, but the record does not establish how they relate to any prespecified family of tests.

The primary analysis population required a valid baseline score and post-baseline data availability. The record does not provide a complete accounting of missing endpoint data by group, reasons for missingness, or sensitivity analyses. It therefore does not permit a full assessment of whether missing data could affect the estimates.

The endpoint description has a wording mismatch: the primary outcome is specified as an ankle score, while the general description of the scale refers to the elbow and wrist. The record does not clarify this language. The scale is also a recoded clinical rating, and no threshold for a clinically important difference is reported.

Finally, the primary endpoint averages assessments at Weeks 4 and 6, limiting interpretation to that combined follow-up measure. The registry lists secondary outcomes but this account focuses on the primary result. Safety counts are summarized without further event-level detail or causal assessment in the posted information. These limitations do not negate the reported estimates, but they narrow what can be concluded from the registry record.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: BOTOX® Treatment in Pediatric Lower Limb Spasticity — NCT01603628.
• Registry record last updated 2018-08-14; status last verified 2018-07.
• Results first posted 2018-08-14.
• 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

NCT01603628 is a completed randomized, quadruple-masked, parallel-group phase 3 study sponsored by Allergan. The registry lists enrollment of 384 and compares intramuscular BOTOX® at 4 U/kg or 8 U/kg with normal saline placebo; weekly physical therapy was described for all groups.

The primary endpoint was average change from baseline in the MAS-B ankle score with the knee extended at Weeks 4 and 6, analyzed using MMRM in the defined mITT population. The registry reports differences versus placebo of -0.26 for BOTOX® 8 U/kg and -0.21 for BOTOX® 4 U/kg, with respective 95% confidence intervals of -0.453 to -0.063 and -0.405 to -0.018.

The record also lists secondary outcomes and adverse event summaries. It does not report the original sample-size rationale, full analysis-plan details, or a multiplicity strategy.