Safety and Efficacy Study in Preschool Children Aged 4-5 Years With Attention-deficit/Hyperactivity Disorder (ADHD): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Safety and Efficacy Study in Preschool Children Aged 4-5 Years With Attention-deficit/Hyperactivity Disorder (ADHD) (NCT03260205), a randomized phase 3 study. The primary outcome is Change From Baseline in Clinician-Administered Attention-Deficit/Hyperactivity Disorder Rating Scale-IV (ADHD-RS-IV) Preschool Version Total Score at Week 6.

2. Clinical Question

3. Trial Design

The registry describes a phase 3, randomized, double-blind, multicenter, parallel-group, placebo-controlled, fixed-dose study of SPD489 in preschool children aged 4–5 years with ADHD. The study is listed as completed and enrolled 199 participants. The sponsor was Shire. The registry record gives a start date of 2017-09-06 and a primary completion date of 2018-10-23.

Participants were assigned to placebo or SPD489. The treatment description lists four SPD489 strengths: 5, 10, 20, and 30 mg, taken orally once daily for 6 weeks. The registry describes dose titration starting at 5 mg and continuing until the assigned fixed dose was reached. The placebo group received a matching capsule. The primary outcome analysis did not compare every dose separately: its pre-specified reporting pooled the 10, 20, and 30 mg groups and compared that pooled group with placebo. The 5 mg group therefore does not contribute to the posted primary pooled contrast.

Randomization and double masking are important design features because they can reduce systematic differences between treatment groups and limit the influence of treatment knowledge on participant care or outcome assessment. These general features do not, by themselves, establish that all possible sources of bias were avoided. Details such as the exact randomization procedure, allocation concealment, masking checks, and the number or location of study centers are not reported in the registry record.

The main outcome was change from baseline to Week 6 on the ADHD-RS-IV Preschool Version total score. Other listed outcomes included global improvement, dose-response assessment, adverse events, vital signs, height, weight, body mass index, and clinical laboratory values. The registry lists 12 outcome measures and posts results for 11 secondary outcomes. The original trial sample-size justification and the trial’s own statistical planning 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: A total of 284 participants were screened of them 85 participants were screen failures and did not receive any treatment. 199 participants (153 participants for SPD489 treatment and 46 participants for placebo treatment) were randomized, out of them 150 participants completed the study.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Full analysis set. Here, number of participants analyzed signifies participants who were evaluable for this outcome measure. This outcome was pre-specified to collect and analyze pooled groups of SPD489 doses. The data were planned, analyzed, and reported only for the placebo and the pooled 10, 20, 30 mg doses of SPD489 arms.

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: Safety analysis set consisted of all randomized set who had taken at least 1 dose of investigational product.

6. Primary Endpoint

The registry lists the primary outcome as Change From Baseline in Clinician-Administered Attention-Deficit/Hyperactivity Disorder Rating Scale-IV (ADHD-RS-IV) Preschool Version Total Score at Week 6, measured over Baseline, Week 6.

The primary endpoint was change from baseline to Week 6 in the total score on the clinician-administered ADHD-RS-IV Preschool Version. The registry describes this instrument as an 18-item questionnaire adapted from the ADHD Rating Scale-IV, with examples intended for the developmental level of preschool children. Each item rates the frequency of an ADHD symptom using a scale from 0, meaning never or rarely, to 3, meaning very often. The total score ranges from 0 to 54, and the items cover hyperactivity/impulsivity and inattentiveness.

Change from baseline is used to describe how a participant’s score differs at follow-up from their starting score. A negative change indicates a lower score at Week 6 than at baseline; in this scale, lower scores correspond to less frequent symptom ratings. Comparing average changes between randomized groups helps describe the relative pattern during the study period, while the placebo group provides a reference for changes that may occur apart from the active treatment assignment.

The endpoint is a continuous score, so the registry reports group means and standard errors as well as a model-based comparison. The primary analysis population was the full analysis set, defined for this outcome as participants in the safety analysis set with at least one post-dose assessment. The registry does not report a threshold for a clinically important change, so the statistical contrast alone does not establish how meaningful a given score difference is to children or families.

7. Results Posted in the Registry

• Pooled SPD489 Doses (10, 20, and 30 mg): mean -14.20 versus -9.10 on control; difference -5.10 (95% CI -10.66 to 0.46), standardized effect d = -0.35; z-test p = 0.072.

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.

Clinical Global Impressions Global Improvement (CGI-I) at Week 6

• 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:
  • Pooled SPD489 Doses (10, 20, and 30 mg): mean 2.80 versus 3.40 on control; difference -0.60 (95% CI -1.04 to -0.16), standardized effect d = -0.52; z-test p = 0.007.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Pooled SPD489 Doses (10, 20, and 30 mg) than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Dose Response Relationship for Change From Baseline in ADHD-RS-IV Preschool Version Total Score in Preschool Children at Week 6

• 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:
  • SPD489 30 mg: mean -15.30 versus -9.10 on control; difference -6.20 (95% CI -13.11 to 0.71), standardized effect d = -0.44; z-test p = 0.079.
  • SPD489 20 mg: mean -17.00 versus -9.10 on control; difference -7.90 (95% CI -15.23 to -0.57), standardized effect d = -0.52; z-test p = 0.035.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between SPD489 30 mg and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for SPD489 20 mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Change From Baseline in Height at Week 6

• 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 deviations.
• Result:
  • SPD489 5 mg: mean 0.70 versus 0.70 on control; difference 0.00 (95% CI -0.63 to 0.63), standardized effect d = 0.00; z-test p = 1.000.
  • SPD489 10 mg: mean 0.70 versus 0.70 on control; difference 0.00 (95% CI -0.53 to 0.53), standardized effect d = 0.00; z-test p = 1.000.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between SPD489 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 SPD489 10 mg and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes span global improvement, symptom-score dose-response, adverse events, vital signs, growth measures, body mass index, and laboratory values. The posted secondary analyses do not form a single consistent inferential set: some comparisons have intervals that exclude zero, while others include zero, and not every outcome has a sponsor-reported analysis. The registry lists 11 secondary outcomes with posted results, but it does not provide numerical results for every listed outcome in the information summarized by the record.

The secondary findings provide context around the primary endpoint rather than an independent confirmation of it. A global-improvement measure and a dose-specific symptom-score analysis address related but distinct questions; safety and growth measures concern different outcomes altogether. Agreement across related endpoints can be informative, but it does not erase the primary analysis discrepancy or establish that all findings were prespecified and analyzed with the same method.

Unadjusted secondary results carry less weight because many outcomes and dose comparisons create multiple opportunities for apparently notable findings. Unless an appropriate multiplicity strategy is used, nominal intervals and p values do not preserve the same overall false-positive protection as a single pre-specified test. The registry record does not report a multiplicity adjustment or a complete statistical hierarchy for secondary outcomes. These results are therefore best treated as supportive or exploratory, with emphasis on estimates and uncertainty rather than isolated thresholds.

Detailed sponsor analyses for several secondary outcomes, the handling of missing secondary measurements, and the intended role of each secondary endpoint are not reported in the registry record. The posted results can describe patterns, but the record does not establish that every secondary finding was confirmatory.

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

• Clinical Global Impressions Global Improvement (CGI-I) at Week 6
• Dose Response Relationship for Change From Baseline in ADHD-RS-IV Preschool Version Total Score in Preschool Children at Week 6
• Number of Participants With Treatment-Emergent Adverse Events (TEAEs)
• Number of Participants With Potentially Clinically Significant Changes in Vital Signs
• Change From Baseline in Height at Week 6
• Change From Baseline in Body Weight at Week 6
• Change From Baseline in Body Mass Index (BMI) at Week 6
• Number of Participants With Potentially Clinically Significant Changes in Clinical Laboratory Values

9. Statistical Interpretation

10. Statistical Methods Explained

The primary endpoint is continuous: each participant has a total score, and the analysis compares average change from baseline. A mean summarizes the group’s central value. The reported standard error describes the precision of an estimated mean, rather than the spread of individual scores. A standard deviation, by contrast, describes variation among individual observations. Keeping these quantities distinct is important when reading the posted group summaries and their uncertainty.

What does the MMRM result represent?

The sponsor’s reported method is a mixed model for repeated measures (MMRM), and the reported parameter is the difference in least square means. In general, an MMRM is used when outcomes are measured repeatedly over time. It estimates group contrasts while accounting for the structure of repeated observations. A least square mean is a model-based group estimate, rather than simply the arithmetic average of the available observations. The registry identifies the method and the result, but the model’s covariates, covariance structure, handling of missing observations, and other specification details are not reported in the registry record.

The primary outcome timeframe is listed as baseline and Week 6. Although MMRM is a repeated-measures method in general, the registry information available for this endpoint does not describe additional post-baseline assessment times or give the model formula. It would therefore be inappropriate to infer which visits, covariates, or assumptions were included beyond what the record states.

How should the mean difference be read?

The primary reported difference in least square means is -5.9 for pooled SPD489 relative to placebo. The negative sign indicates a lower model-estimated change score for the pooled SPD489 group than for placebo, in keeping with the more negative group mean changes. The confidence interval expresses uncertainty about the model-based contrast. It is not a range containing a fixed proportion of individual participant scores, nor does it give the probability that the true effect lies within that particular interval.

The simple recalculation is different in kind. It uses the posted group means and standard errors to estimate an unadjusted difference and interval. It does not include the sponsor’s repeated-measures model or any model adjustments that may have been used. Its unadjusted p value is not interchangeable with the sponsor’s model-based p value. Since the registry analyses differ, the sponsor’s reported analysis takes precedence for interpretation of the trial’s primary result.

Why does over-dispersion matter?

Over-dispersion is a general issue in analyses of counts or binary outcomes: observed variation can exceed what a simple probability model expects. It can make standard errors too small if the extra variation is ignored. The primary endpoint is a continuous rating-scale score, not a count outcome, so over-dispersion is not the central issue for its mean comparison. For secondary outcomes that count participants with events, interpretation depends on the outcome definition and analysis method; detailed model methods are not reported in the registry record.

What do unadjusted secondary comparisons mean?

The posted secondary summaries include continuous outcomes and safety counts. Some recalculated comparisons use simple, unadjusted contrasts between a treatment group and placebo. Such calculations do not account for the trial’s full analysis model and should be read as descriptive statistical comparisons, not replacements for sponsor analyses. The registry does not provide sponsor-reported analyses for every posted secondary contrast. The calculated comparisons also involve multiple outcomes and, in some cases, multiple dose groups. Without a documented multiplicity procedure, nominal p values and intervals should be interpreted cautiously.

For a continuous endpoint, a mean difference expresses the separation between group averages in the endpoint’s units. For a binary or event-count endpoint, the underlying count and denominator describe how many participants were affected among those at risk; that is not the same as a continuous mean contrast. The registry lists several safety and growth outcomes, but detailed methods for each are not reported in the registry record. A reader should not assume that all outcomes used the same model or estimand.

What does statistical significance not tell us?

A p value evaluates how compatible the observed data are with a specified null model under the analysis assumptions. It does not measure the probability that the null hypothesis is true, the probability that a result will recur, or the size of a treatment effect. Confidence intervals add information about the range of estimates compatible with the model and data, but they do not resolve questions about study conduct, outcome relevance, or missing-data assumptions.

The registry identifies the primary outcome population as the full analysis set: participants in the safety analysis set with at least one post-dose assessment of the scale. That definition indicates who could contribute to this outcome analysis, but it does not state how missing measurements were handled within the analysis. The detailed missing-data assumptions and sensitivity analyses are not reported in the registry record.

11. Confidence Intervals and Precision

The sponsor-reported primary contrast is -5.9, with a 95% confidence interval from -11.01 to -0.78. A confidence interval gives a range of values that are compatible with the observed data under the analysis model and its assumptions. It communicates uncertainty in the estimated group difference; it is not a range for individual children’s score changes and does not mean that 95% of individual outcomes fall inside it.

The reported interval lies below zero, so the sponsor’s model-based estimate and the interval are consistent with a negative difference in change scores for pooled SPD489 relative to placebo. The interval also shows that the estimate is not exact: the range includes values of different magnitudes. Precision is limited by the amount of information available, the variability of the outcome, and the analysis model. The registry does not report a clinically important difference threshold for interpreting the scale’s units.

The recalculated unadjusted interval is -10.66 to 0.456, which extends across zero. It does not match the sponsor’s MMRM interval. The discrepancy is consequential for inferential interpretation, so the sponsor’s analysis takes precedence. The unadjusted interval is useful only as a transparent calculation from the posted group summaries, not as a replacement for the model-based result.

For secondary outcomes, confidence intervals and nominal p values also need context. Some posted comparisons have intervals that include zero and others do not, but these contrasts are not all supported by a reported sponsor model. A narrow interval can indicate greater statistical precision without establishing clinical importance; a wider interval indicates less precision and leaves a broader range of compatible effects. Multiplicity and the absence of fully documented analysis details further limit how confidently secondary intervals can be treated as confirmatory.

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 sample-size illustrations are planning calculations based on the observed effect, not the trial’s original sample-size design. For the continuous comparisons listed, the figures are participants per arm, not total participants. No time-to-event comparison is being interpreted in these calculations. The stated assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. Each planning scenario uses the stated fraction of the observed effect as the effect to detect.

The 80% power column gives the number of participants per arm that the illustration estimates would be needed for an 80% chance of detecting the assumed effect at the specified significance level, if the assumptions hold. The 90% power column gives the corresponding per-arm planning figure for a 90% chance. These are prospective planning quantities: they describe hypothetical designs, not what the completed study achieved.

Rows using 100%, 75%, and 50% of the observed effect show how the planning illustration changes when the assumed target effect is reduced. Smaller effects are harder to distinguish from random variation, so they require more participants under the same assumptions. The calculations are based on the observed effect, which is uncertain and can be an optimistic basis for planning. The figures are not observed power and do not tell whether the completed trial had adequate power under its original design. That original design justification and its assumptions are not reported in the registry record.

The sample-size curve should be read in the same planning context: it represents how required per-arm sample size changes with the assumed effect size, under the listed assumptions. It is not a record of enrollment or a forecast of the actual results. The registry’s total enrollment was 199, but the primary outcome analysis involved 37 placebo participants and 84 pooled SPD489 participants; these outcome counts should not be conflated with planning illustrations.

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 = -5.100; pooled standard deviation σ = 14.379.
2. n = 2 × (1.96 + 0.842)² × 14.379² / -5.100² = 124.8, rounded up to 125 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 <- -5.1                      # observed difference in means
sigma <- 14.3790200809163                     # 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(3260205)
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 = 125 (80% power), 168 (90% power)
 75% of observed effect: n per arm = 222 (80% power), 297 (90% power)
 50% of observed effect: n per arm = 500 (80% power), 669 (90% power)
power.t.test: 125.8 
Simulated power at n = 125 per arm: 0.800 (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. Analyses Reported by the Sponsor

The sponsor reported an MMRM analysis of the pooled SPD489 doses against placebo. Its estimated difference in least square means was -5.9, with a 95% confidence interval from -11.01 to -0.78. The reported p value was 0.024. This is the registry’s own model-based analysis for the primary contrast.

The unadjusted recalculation uses the posted group means and standard errors instead. It gives a mean difference of -5.1, a confidence interval from -10.66 to 0.456, and an unadjusted p value of 0.072. These results do not match the sponsor’s analysis. The recalculation is not a reproduction of the MMRM and cannot establish why the estimates and intervals differ; the necessary model specification and underlying participant-level data are not reported in the registry record.

When the sponsor’s reported result and an unadjusted recalculation disagree, the sponsor’s own analysis takes precedence for describing the trial’s primary inferential finding. The recalculation remains a check based on summary statistics, but it should not be presented as though it were the registry’s primary analysis or used to overrule the reported model. The mismatch is a reason to be explicit about which estimate is being discussed and what method generated it.

15. Multiplicity, Missing Data and Analysis Populations

The registry lists 12 outcome measures, including the primary endpoint and multiple secondary outcomes. The secondary outcomes include several kinds of measures, and the dose-response assessment includes separate dose groups. Testing or estimating many outcomes can increase the chance of at least one apparently unusual result arising by chance. A multiplicity strategy can address this problem by adjusting inference, organizing outcomes hierarchically, or distinguishing confirmatory from exploratory analyses. The statistical plan for multiplicity, including any adjustment or hierarchy, is not reported in the registry record.

As a result, nominal secondary p values and confidence intervals should not be read as if they were the sole planned test. This is especially relevant when considering multiple dose-specific comparisons alongside multiple outcomes. It does not mean the secondary observations are uninformative; it means they require more cautious interpretation and should not be treated as equivalent to a single prespecified primary analysis.

The primary endpoint population is described as the full analysis set, consisting of participants in the safety analysis set who had at least one post-dose ADHD-RS-IV Preschool Version total-score assessment. The registry does not provide a complete account of missing outcome values, the reasons for missingness, the assumptions used to handle missing measurements, or sensitivity analyses assessing those assumptions. The MMRM method is identified, but its detailed model specification and missing-data approach are not reported in the registry record.

The original sample-size rationale, anticipated effect, and design assumptions are also not reported in the registry record. The planning illustrations based on the observed effect are not a substitute for that information. Interim analyses, stopping rules, and any associated adjustment of the primary inference are not reported in the registry record either. These gaps limit independent assessment of the analysis plan and of how fully the posted results reflect the trial’s intended inferential framework.

These limitations concern what can be established from the registry record. They do not demonstrate that particular procedures were or were not followed; they mean that the relevant details cannot be evaluated from the record.

16. 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 non-serious adverse event terms

The registry reports serious adverse events by treatment group. The serious-event count was 0 among 45 participants at risk in placebo, 0 among 39 at risk in SPD489 5 mg, 0 among 35 at risk in SPD489 10 mg, 0 among 34 at risk in SPD489 20 mg, and 0 among 38 at risk in SPD489 30 mg. No deaths were reported in these groups.

These figures mean that no serious adverse events were recorded in the listed groups during the study period as reported in the registry. They do not establish that serious events are impossible, nor do they describe risks beyond the observed study participants and follow-up. The registry’s listed timeframe for treatment was 6 weeks. Counts at risk are group denominators, not evidence that every possible safety outcome was assessed with equal completeness.

The registry also reports participants affected by other, non-serious adverse events, but those counts should not be conflated with serious adverse events. The serious-event summary does not provide detailed event narratives or a full account of severity, duration, causality assessment, or timing. Such details are not reported in the registry record.

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

The posted results describe a short-term randomized comparison in preschool children aged 4–5 years with ADHD. For the primary symptom-rating outcome, both groups had negative mean changes by Week 6, and the pooled SPD489 group had a more negative posted mean change than placebo. The sponsor’s MMRM analysis estimated a negative between-group difference, with a 95% confidence interval below zero. The unadjusted recalculation does not agree with that inferential result, so the sponsor’s analysis takes precedence.

This does not establish how an individual child would respond. Group averages can conceal substantial variation, and the registry record does not report individual-level distributions or a clinically important difference threshold for the ADHD-RS-IV Preschool Version. A statistically supported group contrast is not the same as proof that every participant experienced a meaningful change.

The result also does not answer every question about treatment. The study’s listed treatment period was 6 weeks, so the primary result concerns the assessment at that follow-up point. Longer-term outcomes, durability, and effects beyond the studied population are not established by these data. The registry lists other outcomes, but their evidence varies and the record does not report a multiplicity strategy or complete analysis details for all of them.

The safety summary reports no serious adverse events or deaths in the listed groups, but a zero count in this study does not rule out events outside the observed sample or timeframe. The record also lists non-serious adverse events and additional safety measures. Taken together, the registry results are a bounded statistical account of outcomes in this trial, not a complete assessment of all possible benefits, harms, or individual circumstances.

18. Why This Trial Matters Statistically

This trial raises several statistical issues that are useful in clinical research. First, the distinction between a posted descriptive contrast and a model-based primary analysis matters. The group means and standard errors support a transparent unadjusted calculation, but that calculation differs from the sponsor’s MMRM result. Readers need to keep the estimate, interval, and method attached to the analysis that produced them rather than treating all contrasts as interchangeable.

Second, the primary outcome population is defined by having at least one post-dose assessment. That definition is more specific than simply counting everyone enrolled, but it does not explain how missing follow-up observations were handled. Evaluating the implications of missing data requires information about the amount and pattern of missingness, the analysis assumptions, and sensitivity analyses; those details are not reported in the registry record.

Third, the primary analysis pools multiple SPD489 doses. Pooling can address a broad treatment comparison, while dose-specific summaries address different questions. The pooled result should not be mistaken for evidence that every dose had the same effect. The registry separately lists dose-response and individual dose outcomes, but their inferential role and multiplicity strategy are not fully documented.

Finally, statistical significance, estimate precision, clinical importance, and safety are separate dimensions. The confidence interval communicates uncertainty about the model-based contrast; the scale measures symptom ratings; and the safety counts describe recorded events in the study groups. Considering these dimensions separately makes the registry results easier to interpret without claiming more than the record supports.

19. 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 limitations are specific to the information available for this study. The primary outcome comparison pools the 10, 20, and 30 mg groups, while the trial also included 5 mg. The pooled estimate therefore does not describe each dose separately. The posted outcome counts are 37 for placebo and 84 for pooled SPD489, smaller than the total enrollment of 199; the registry defines the primary analysis population but does not give a full accounting of the difference between enrollment and outcome counts.

The reported primary MMRM analysis and the unadjusted recalculation disagree in their estimates and in whether their confidence intervals include zero. The sponsor’s result takes precedence, but the registry record does not provide enough model detail or participant-level data to reproduce the MMRM or explain the discrepancy. This limits independent checking of the primary inferential result.

The registry record does not report the original sample-size justification, model specification, multiplicity strategy, missing-data assumptions, sensitivity analyses, interim analyses, or detailed protocol deviations. It also does not give a clinically important difference threshold for the primary scale. These omissions restrict assessment of the prespecified analysis framework and the practical meaning of the estimated score difference.

Follow-up for the listed treatment was 6 weeks. The registry therefore does not establish longer-term patterns from this trial. The serious adverse event summary reports zero events in the listed groups, but event detail and broader safety context are limited in the record. Secondary outcome evidence should be interpreted cautiously because the registry does not report a complete multiplicity plan or sponsor analyses for every posted contrast.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Safety and Efficacy Study in Preschool Children Aged 4-5 Years With Attention-deficit/Hyperactivity Disorder (ADHD) — NCT03260205.
• Registry record last updated 2021-06-08; status last verified 2021-05.
• Results first posted 2020-01-18.
• 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.

23. Record Summary

The registry record describes a completed phase 3, randomized, double-blind, multicenter, parallel-group, placebo-controlled study of SPD489 in preschool children aged 4–5 years with ADHD. It enrolled 199 participants and compared placebo with SPD489 doses of 5, 10, 20, and 30 mg over 6 weeks.

The primary endpoint was change from baseline to Week 6 in the ADHD-RS-IV Preschool Version total score. The sponsor’s MMRM analysis reported a difference in least square means of -5.9, with a 95% confidence interval from -11.01 to -0.78. An unadjusted recalculation differs, so the sponsor’s analysis takes precedence. The registry reports no serious adverse events or deaths in the listed treatment groups. Important details about the original sample-size rationale, multiplicity, missing-data handling, and full model specification are not reported in the registry record.