Study of Aripiprazole in the Treatment of Children and Adolescents With Autistic Disorder (AD): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study of Aripiprazole in the Treatment of Children and Adolescents With Autistic Disorder (AD) (NCT00337571), a randomized phase 3 study. The primary outcome is Mean Change (Week 8 - Baseline) in the Autistic Behavior Checklist (ABC) Irritability Subscale Score.

2. Clinical Question

The posted results address differences in mean score change and include secondary measures of global impressions, response, compulsive symptoms, other ABC subscales, severity, safety, and body weight. The primary comparison is the clearest direct answer to the stated question, while the secondary outcomes provide additional, distinct descriptions of the study. A registry record alone does not establish whether the observed differences are clinically important or apply beyond the studied participants and follow-up.

3. Trial Design

The registry identifies this as a completed Phase 3 study with randomized allocation, double masking, and a parallel-group model. Its official title describes a multicenter, double-blind, randomized, placebo-controlled study of three fixed aripiprazole doses. The listed interventions are aripiprazole and placebo, and the active groups are identified as 5 mg, 10 mg, and 15 mg. Enrollment is recorded as 218. That is the registry enrollment figure; outcome-specific counts differ across measures.

The study record lists a start date of 2006-06-01 and primary completion on 2008-06-01. Its stated purpose was to compare aripiprazole with placebo for reducing serious behavioral problems in children and adolescents diagnosed with autistic disorder. The primary outcome was mean change from baseline in ABC Irritability score at Week 8. A negative change represents a decrease, and the outcome description says a decrease indicates improvement.

The primary endpoint population is described as all randomized participants minus specified exclusions: three placebo participants, one participant in the 5-mg group, and one participant in the 15-mg group. The record gives reasons for these exclusions, including loss to follow-up, withdrawal of consent, no longer meeting study criteria, and elevated potassium levels. It labels the data set as L. The record does not explain the meaning of that label or provide a fuller account of the analysis population.

Details such as the trial's original sample-size justification, planned interim analyses, and the complete statistical analysis plan are not reported in the registry record. The posted group summaries and recalculated contrasts should therefore be read as the available registry-level evidence, not as a substitute for a full protocol or clinical study report.

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:

Analysis population for the primary outcome, as defined in the registry: Efficacy population=all randomized participants minus 3 patients in the placebo group (1 lost to follow-up, 1 withdrew consent, 1 no longer met study criteria), 1 participant in the 5-mg group who withdrew consent, and 1 participant in the 15-mg group who had elevated potassium levels. Data set is LOCF.

5. Baseline Characteristics

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

6. Primary Endpoint

The registry lists the primary outcome as Mean Change (Week 8 - Baseline) in the Autistic Behavior Checklist (ABC) Irritability Subscale Score, measured over Week 8.

The primary endpoint is mean change from baseline to Week 8 in the Autistic Behavior Checklist Irritability subscale score. The ABC is described in the registry as a 58-item informant-based assessment of problem behaviors in children and adolescents with intellectual disability. Items use a 4-point rating scale from 0, meaning no problem, to 3, meaning severe problem, and contribute to five domain subscales. The irritability subscale focuses the primary question on one behavioral domain rather than the full range of symptoms or circumstances relevant to autistic disorder.

The registry explicitly states that a decrease in score indicates improvement. Accordingly, a more negative mean change describes a larger average decline from baseline; it does not by itself establish how individual participants changed or whether the change was important to families or clinicians. The outcome is reported as a group mean with a standard error. The standard error describes uncertainty around an estimated group mean, not the spread of individual scores.

This type of continuous endpoint can use information across the score range rather than reducing each participant to a response or nonresponse category. Its interpretation still depends on the quality and completeness of the informant ratings, the analysis population, and the study's follow-up period. A threshold for a clinically important change is not reported in the registry record. The endpoint therefore supports a comparison of average score changes, but the registry alone does not define what magnitude should be considered meaningful for an individual participant.

7. Results Posted in the Registry

• Aripiprazole 5 mg: mean -12.40 versus -8.40 on control; difference -4.00 (95% CI -7.81 to -0.19), standardized effect d = -0.41; z-test p = 0.040.
• Aripiprazole 10 mg: mean -13.20 versus -8.40 on control; difference -4.80 (95% CI -8.46 to -1.14), standardized effect d = -0.50; z-test p = 0.010.
• Aripiprazole 15 mg: mean -14.40 versus -8.40 on control; difference -6.00 (95% CI -9.74 to -2.26), standardized effect d = -0.62; z-test p = 0.002.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Mean Clinical Global Impressions Improvement Scale (CGI-I) Score

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • Aripiprazole 5 mg: mean 2.60 versus 3.30 on control; difference -0.70 (95% CI -1.19 to -0.21), standardized effect d = -0.56; z-test p = 0.005.
  • Aripiprazole 10 mg: mean 2.50 versus 3.30 on control; difference -0.80 (95% CI -1.27 to -0.33), standardized effect d = -0.64; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Aripiprazole 5 mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Aripiprazole 10 mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Number of Participants With Response at Week 8

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Aripiprazole 5 mg: 55.8% (29/52) versus 34.7% (17/49) on control; risk difference 21.1 percentage points (95% CI 2.1 to 40.0), risk ratio 1.61 (1.02–2.53); unadjusted two-proportion z-test p = 0.034.
  • Aripiprazole 10 mg: 49.2% (29/59) versus 34.7% (17/49) on control; risk difference 14.5 percentage points (95% CI -4.0 to 32.9), risk ratio 1.42 (0.89–2.25); unadjusted two-proportion z-test p = 0.130.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Aripiprazole 5 mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Aripiprazole 10 mg and the control. The interval gives the range of effects that remain compatible with the data.

Mean Change (Week 8 - Baseline) in the Children's Yale-Brown Obsessive Compulsive Scale (CY-BOCS; Compulsion Scale Only)

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • Aripiprazole 5 mg: mean -2.60 versus -1.70 on control; difference -0.90 (95% CI -2.29 to 0.49), standardized effect d = -0.27; z-test p = 0.203.
  • Aripiprazole 10 mg: mean -2.40 versus -1.70 on control; difference -0.70 (95% CI -2.01 to 0.61), standardized effect d = -0.21; z-test p = 0.293.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Aripiprazole 5 mg and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Aripiprazole 10 mg and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes span clinician-rated global impressions, a participant response count, compulsive symptoms, other ABC subscales, severity, safety, and body weight. This range can help describe whether the primary behavioral-domain pattern appears alongside other kinds of measurement. The posted results do not provide an equally complete inferential account for every listed outcome, and the registry does not report analyses for all the secondary outcomes named in the record.

Overall, several reported secondary summaries show numerically different group means or response counts between aripiprazole and placebo. The pattern is not uniform in precision: some unadjusted comparisons have intervals that exclude zero, while others include zero. That variation means the secondary results should not be compressed into a single claim that every measure changed in the same way. The scales assess different constructs, and the registry does not establish a common clinical interpretation across them.

The secondary findings relate to the primary result by offering additional context, not independent confirmation of the same endpoint. CGI-I and CY-BOCS, for example, are distinct measures from ABC Irritability; response is a categorical outcome rather than an average score. Agreement across measures can be informative, but their different definitions, available observations, and uncertainty matter. No overall summary measure combining these endpoints is reported.

Unadjusted secondary results carry less weight because multiple outcomes and dose comparisons create more opportunities for chance findings. The record does not describe a correction or a prespecified hierarchy for these tests, and the reported registry analyses are absent. Consequently, secondary p-values should be treated as exploratory rather than as stand-alone confirmation. They can identify patterns worth evaluating in a full analysis, but do not settle the primary question or establish broader benefits.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Mean Clinical Global Impressions Improvement Scale (CGI-I) Score
• Number of Participants With Response at Week 8
• Mean Change (Week 8 - Baseline) in the Children's Yale-Brown Obsessive Compulsive Scale (CY-BOCS; Compulsion Scale Only)
• Mean Change (Week 8 - Baseline) in the Other ABC Subscale Scores
• Mean Change (Week 8 - Baseline) in CGI-Severity (CGI-S)
• Summary of Safety
• Change From Baseline in Body Weight

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is continuous: each participant has a score, and the registry reports a group mean change from baseline. A mean is a measure of the group's central value. It is useful for comparing average change, but it can conceal variation: participants within a group may have had quite different score changes. The registry reports standard errors alongside the means. A standard error quantifies the sampling uncertainty of an estimated mean; it is not the standard deviation of individual participants' scores.

What does a mean difference represent?

A mean difference is the active-group mean minus the placebo-group mean for the outcome change. For the primary endpoint, the recalculated differences are -4.0 for aripiprazole 5 mg, -4.8 for 10 mg, and -6.0 for 15 mg. The negative direction reflects a more negative mean change in the active group. Because lower ABC Irritability scores indicate improvement according to the registry, these estimates describe larger average decreases relative to placebo. The arithmetic direction is tied to this score's stated meaning; it is not a general rule for every outcome.

The differences are unadjusted recalculations from the posted group summaries. They do not incorporate covariates or other adjustments that might have been used in a sponsor analysis. Such an analysis is not reported in the registry record. The registry's discrepancy flag is false, so there is no identified conflict between these recalculations and a sponsor result in the record.

What do p-values say?

A p-value measures how incompatible the observed data are with a specified null model, under the assumptions of the test. It is not the probability that the null hypothesis is true, and it does not measure the size or clinical importance of an effect. The unadjusted primary p-values are 0.04, 0.01, and 0.002 for the 5-mg, 10-mg, and 15-mg comparisons, respectively. The registry does not report the sponsor's inferential analyses or the full testing procedure, so these p-values should be understood as recalculated, unadjusted results.

When several doses or outcomes are tested, the chance of at least one apparently unusual result under a broad null can increase. A multiplicity adjustment or a prespecified hierarchy can address that issue, but neither is described in the registry record. The primary outcome has a designated role, which makes it more central than the secondary measures, but the three dose comparisons still form a set of comparisons whose broader testing context is not specified.

Why are confidence intervals useful?

A confidence interval gives a range of estimates compatible with the data under the interval method. It communicates uncertainty and the precision of an estimate, not a guarantee that the true value lies in the particular interval. For the primary comparisons, the recalculated intervals are presented with the group results. An interval that does not include zero indicates that zero is not among the values covered by that unadjusted interval; it does not resolve clinical importance or multiplicity concerns.

The posted standard errors allow the uncertainty around group means to be described, while the comparison intervals address uncertainty in the difference between groups. Those are related but distinct quantities. A mean can have a relatively small standard error while the individual scores remain variable. Standard errors should not be read as ranges containing most participants.

How do the other outcome types differ?

The response outcome is binary at the participant level: a participant is counted as having a response or not. Its group summaries are numbers of participants, while the recalculated comparisons use risk differences. A risk difference expresses the difference between the observed proportions in two groups. Confidence intervals around a risk difference can include zero, indicating that the data are compatible with no difference under that unadjusted interval procedure. The registry also posts ratio measures for some comparisons, but these do not remove uncertainty or the effects of multiple testing.

Other secondary outcomes are continuous mean scores, including CGI-I and CY-BOCS compulsion scores. Their scales have different meanings and directions; the registry's endpoint description only explicitly states the direction of improvement for the ABC endpoint. Interpretations should not assume that a decrease has the same meaning on every scale. Outcomes can also have different participant counts, reflecting the data available for each measure.

For a binary outcome, the risk difference and its interval are not interchangeable with a mean difference and its interval. Each is defined for its own data type. The registry's secondary response comparisons include unadjusted estimates and p-values, but does not provide reported sponsor analyses. All such estimates are best understood as descriptive or exploratory unless the study's prespecified analysis plan establishes otherwise.

What is not specified?

The registry does not report the original sample-size justification, full missing-data methods, interim analyses, or a complete statistical analysis plan. It also does not provide a multiplicity procedure. General statistical explanations of these methods cannot fill gaps in the trial record. Without those details, the posted summaries support a careful account of observed group patterns and their unadjusted precision, but not a reconstruction of every analysis decision.

11. Confidence Intervals and Precision

Precision concerns how narrowly the data estimate a group mean or a difference between groups. The primary outcome reports a standard error for each mean: 1.39 for placebo, 1.36 for aripiprazole 5 mg, 1.25 for 10 mg, and 1.31 for 15 mg. These standard errors describe uncertainty in the group means. They are not measures of how widely individual score changes varied.

The recalculated primary comparison intervals show uncertainty in the mean differences. For the 10-mg group, the difference was -4.8, with a 95% confidence interval from -8.464 to -1.136. For the 15-mg group, it was -6.0, with a 95% confidence interval from -9.744 to -2.256. Each interval is below zero, while the posted interval for the 5-mg comparison also lies below zero. The intervals represent uncertainty under the unadjusted calculation, not a probability statement about a particular effect being true.

Confidence intervals help distinguish an estimate's direction from its precision. A narrow interval is more precise than a wide interval, all else equal. An interval's width is affected by the variability in the data and the amount of information available. The outcome-specific sample counts are reported with each result, and these are not identical across endpoints. A confidence interval does not capture every source of uncertainty, including possible bias from missing data, measurement limitations, or an analysis choice that is not documented.

Intervals and p-values are connected: an unadjusted interval that excludes zero corresponds to evidence against a zero difference under the corresponding testing framework. That relationship does not address whether a difference matters clinically, nor whether the interpretation should change after accounting for the several doses and endpoints. Multiplicity-adjusted intervals are not reported in the registry record. The posted intervals should therefore be read as unadjusted estimates of precision, not definitive bounds on a treatment effect across all analyses.

12. 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 for the continuous-outcome comparisons; they are not total study enrollment or total events. The planning illustration assumes a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. It uses an effect equal to the stated fraction of the observed effect. The registry does not report whether the original trial's design used these assumptions or how its original sample size was justified.

The 80% power column describes the number of participants per arm that the illustration estimates would be needed for an 80% chance of detecting the specified effect under its assumptions. The 90% power column uses a higher target chance of detecting that effect, which generally requires more participants under the same assumptions. These are planning illustrations based on the observed effect, not observed power.

Rows that plan for 75% or 50% of the observed effect represent smaller effects to detect. Smaller effects are harder to distinguish from random variation, so the illustrated sample sizes increase as the target effect becomes smaller. The calculations are not evidence that the trial was adequately powered for its original objectives, and they do not account for every possible design feature, missing observations, or multiplicity. They are conditional planning examples, not a retrospective verdict on the study's success.

13. 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 = -4.000; pooled standard deviation σ = 9.770.
2. n = 2 × (1.96 + 0.842)² × 9.770² / -4.000² = 93.6, rounded up to 94 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 <- -4                      # observed difference in means
sigma <- 9.76979389379698                     # 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(337571)
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 = 94 (80% power), 126 (90% power)
 75% of observed effect: n per arm = 167 (80% power), 223 (90% power)
 50% of observed effect: n per arm = 375 (80% power), 502 (90% power)
power.t.test: 94.6 
Simulated power at n = 94 per arm: 0.793 (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.

14. Multiplicity, Missing Data and Analysis Populations

The registry identifies one primary outcome and several secondary outcomes. The primary endpoint is mean change in ABC Irritability score at Week 8. Naming a primary outcome helps distinguish the central analysis question from additional measures, but it does not by itself specify how multiple dose comparisons were handled. The primary comparisons include each aripiprazole dose versus placebo. A multiplicity adjustment, a testing hierarchy, or another familywise error strategy is not reported in the registry record.

The posted p-values are unadjusted recalculations. They do not account for testing the three doses, the secondary outcomes, or any additional analyses. This matters because interpreting many unadjusted tests as if they were a single isolated test can overstate the strength of evidence. The p-values remain descriptive of their individual comparisons, but should not be treated as a multiplicity-controlled finding.

For the primary efficacy population, the record says all randomized participants were included except three placebo participants, one participant in the 5-mg group, and one participant in the 15-mg group. It lists reasons including loss to follow-up, withdrawal of consent, no longer meeting study criteria, and elevated potassium levels. The record labels the data set as L, but does not define the label or report enough detail to infer the exact missing-data procedure. The handling of missing outcome values among remaining participants is not reported in the registry record.

Outcome counts differ, and the record does not describe a common imputation or sensitivity-analysis strategy for missing observations. It also does not report the original analysis population definitions beyond the primary outcome description. This leaves uncertainty about how robust the estimates are to assumptions about unavailable measurements. The number of reported analyses is listed as zero, and the secondary outcomes do not include reported analyses in the record.

These gaps do not erase the posted descriptive summaries, but they limit claims about confirmatory testing and robustness. A protocol or full clinical report might supply important analysis details; those details are not reported in the registry record and should not be inferred.

15. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: presyncope (1.9%), aggression (1.7%).

The registry lists serious adverse events by group and gives a number at risk for each group. It records one serious adverse event for aripiprazole 10 mg, zero for aripiprazole 15 mg, one for aripiprazole 5 mg, and zero for placebo. The associated at-risk counts are 59, 54, 52, and 51, respectively. These figures describe the registry's recorded serious-event counts and group denominators; they are not a comparison adjusted for differences in follow-up or other factors.

The record lists PRESYNCOPE at 1.9% and AGGRESSION at 1.7% as the top serious terms. It does not identify the group assignment for those listed terms in the information, so they should not be attributed to a particular arm. Death counts are not reported in the registry record; a blank value should not be read as a confirmed zero.

Serious adverse event counts are distinct from the count of participants with other adverse events, also reported by group. Event counts alone do not establish whether an event was caused by an intervention, and small counts do not rule out uncommon harms. These registry entries describe recorded events rather than providing a complete safety assessment.

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

The primary registry results show mean decreases in ABC Irritability score in the placebo group and all three aripiprazole groups by Week 8. The mean decreases were numerically larger in each aripiprazole group than in the placebo group. The unadjusted recalculated mean differences and intervals describe the size and uncertainty of those group contrasts. The registry states that a decrease indicates improvement on this particular scale.

That pattern does not mean every participant improved, and group means cannot show how many participants had a substantial change unless an individual-level response definition is applied. A change in one subscale also does not represent every aspect of autistic disorder, behavior, or daily functioning. The record does not supply a threshold for interpreting a clinically important ABC change, so statistical evidence and clinical importance remain distinct questions.

The result is limited to the recorded outcome at Week 8 and the defined efficacy population. It does not establish whether differences persisted after that timeframe, whether they apply to people outside the study population, or how the doses compare directly with one another. The registry's fixed-dose arms permit comparisons with placebo, but no direct active-dose comparison is reported.

Unadjusted comparisons do not account for the broader set of doses and secondary outcomes. Registry entries also do not provide the full analysis plan, missing-data methods, or sponsor inferential analyses. The estimates are therefore best understood as a statistical description of the posted summaries, not a complete account of all evidence or a basis for individual treatment decisions.

The safety figures likewise describe recorded serious adverse events, not causal effects. Low event counts in a registry summary cannot establish that serious harms are absent. A balanced interpretation keeps the primary score pattern, uncertainty, study scope, missing analysis detail, and safety reporting limitations in view without treating any single statistic as a complete measure of benefit or risk.

17. Why This Trial Matters Statistically

This trial is statistically instructive because it compares multiple fixed doses with a placebo group while measuring a continuous primary outcome. Mean changes preserve more information than a simple response classification, and the ABC Irritability endpoint has a stated direction: a decrease indicates improvement. These features make it possible to interpret the sign of the observed mean differences in relation to the scale.

The record also illustrates why estimates and uncertainty should be read together. The mean differences describe the observed size and direction of group contrasts, while confidence intervals show their precision under unadjusted calculations. P-values alone cannot communicate magnitude, clinical importance, or the consequences of testing multiple doses and outcomes.

Another statistical lesson is that a named primary outcome does not answer every question about analysis conduct. The registry does not report a multiplicity procedure, the sponsor's inferential analyses, or the full missing-data approach. Registry summaries can support transparent description, but incomplete analysis documentation limits how far confirmatory conclusions can be taken.

Finally, the outcome set demonstrates that different measures answer different questions. Average irritability change, global impressions, a categorical response, compulsive symptoms, and safety events should not be collapsed into a single judgment. The study offers a focused comparison at Week 8, while interpretation depends on scale meaning, precision, outcome availability, and the limits of the record.

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.

Several trial-specific limitations affect interpretation of the posted results. The primary endpoint is an informant-based rating of one ABC domain, not a direct measure of every behavioral concern or of overall functioning. The registry's scale description explains how to interpret a decrease, but a clinically important difference threshold is not reported. Statistical differences therefore cannot be translated from the registry alone into a judgment about meaningful change for an individual.

The primary outcome is assessed at Week 8. The record does not establish the persistence of changes beyond that endpoint timeframe. The study uses fixed-dose groups, but the posted comparisons are each against placebo; a direct statistical comparison among active doses is not reported. The numerical ordering of group means should not be taken as a tested dose-response relationship.

Participant counts vary by outcome. The primary efficacy population excludes specified participants, and the registry does not fully describe missing-data handling or sensitivity analyses. That leaves uncertainty about the effect of unavailable observations on the estimates. A complete protocol, analysis plan, or clinical study report may contain details not reported in the registry record.

Unadjusted recalculations are available, while sponsor analyses are not reported. No multiplicity strategy is reported despite multiple dose comparisons and secondary outcomes. Secondary findings consequently have less confirmatory weight. The registry also lists serious-event figures without enough detail in the summarized record to attribute the top serious-event terms to specific treatment groups or make causal claims.

These limitations do not negate the observed group summaries. They define what those summaries can support: a description of average changes and unadjusted comparisons for this study population and timeframe, rather than a complete account of long-term outcomes, individual response, or all safety considerations.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Study of Aripiprazole in the Treatment of Children and Adolescents With Autistic Disorder (AD) — NCT00337571.
• Registry record last updated 2026-05-13; status last verified 2026-04.
• Results first posted 2009-07-23.
• 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

NCT00337571 is a completed Phase 3 randomized, double-blind, parallel-group, placebo-controlled study of aripiprazole in children and adolescents with autistic disorder. The registry lists enrollment of 218 and fixed-dose aripiprazole arms of 5 mg, 10 mg, and 15 mg. The primary outcome was mean change from baseline in ABC Irritability score at Week 8; a decrease indicates improvement.

Posted mean changes were -8.4 for placebo, -12.4 for aripiprazole 5 mg, -13.2 for 10 mg, and -14.4 for 15 mg. Recalculated unadjusted comparisons were more negative for each active dose than for placebo. The registry record does not report sponsor analyses, a multiplicity procedure, full missing-data methods, or the trial's original sample-size justification. Serious adverse-event counts were low in the posted groups, but the record does not establish causation or support interpreting unreported death values as zero.