Tolerability and Efficacy of AM and PM Once Daily Dosing With Extended-release Guanfacine Hydrochloride in Children 6-12 With Attention-Deficit/Hyperactivity Disorder (ADHD) (The ADHD Tempo Study): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Tolerability and Efficacy of AM and PM Once Daily Dosing With Extended-release Guanfacine Hydrochloride in Children 6-12 With Attention-Deficit/Hyperactivity Disorder (ADHD) (The ADHD Tempo Study) (NCT00997984), a randomized phase 3 study. The primary outcome is Change From Baseline in Attention-Deficit/Hyperactivity Disorder-Rating Scale-IV (ADHD-RS-IV) Total Score at Week 8 - Last Observation Carried Forward (LOCF).

2. Clinical Question

The ADHD-RS-IV is a symptom-rating scale, and the registry describes its total score as ranging from 0 to 54, with higher item scores reflecting more severe symptoms. Consequently, a negative change indicates a decrease in the measured total score from baseline. The central statistical comparison is the difference in mean change between each dosing-time group and placebo, rather than a comparison of the within-group changes alone.

The registry also includes outcomes addressing global clinical ratings, daytime sleepiness, health utility, parent-rated behavior and functioning, and sleep-related measures. Those outcomes provide context beyond the primary symptom scale, but they do not replace the primary comparison. Details of the trial’s original statistical plan, including the rationale for choosing its primary endpoint, are not reported in the registry record.

3. Trial Design

This was a Phase 3 randomized, parallel-group, multicenter, placebo-controlled study with quadruple masking. The registry lists enrollment as 340. Its title describes the study as double-blind, while the design fields record quadruple masking; the registry does not explain how those descriptions relate to one another. The study status is completed, with a start date of 2009-11-17 and primary completion date of 2010-10-09.

Participants were assigned to placebo, SPD503 AM, or SPD503 PM. SPD503 was described as extended-release guanfacine hydrochloride, with once-daily dose optimization over 1, 2, 3 and 4mg/day. The registry’s arm descriptions contain no further details about titration, dose decisions, or treatment conduct, so those aspects of the protocol are not reported in the registry record.

The primary outcome was change from baseline in ADHD-RS-IV total score at Week 8, using LOCF. The primary population was the Full Analysis Set, defined as all subjects who had taken at least 1 dose of investigational product during the study. Reported group results are least-squares means with standard errors. The registry reports ANCOVA analyses for placebo versus the morning group, placebo versus the evening group, and placebo versus All-Active.

Enrollment is a trial-level figure and should not be confused with the endpoint-specific analysis counts. The primary outcome lists 110 participants for placebo, 104 for SPD503 AM, and 111 for SPD503 PM. Counts can differ across outcomes, as illustrated by the Week 8 CGI-I outcome, where the evening group has 112 participants. The registry record does not explain each difference in analysis counts.

Further details about the randomization procedure, allocation concealment, site structure, dose-optimization rules, original sample-size justification, or planned interim analyses are not reported in the registry record. The available design description supports a randomized comparison of dosing schedules against placebo, but it does not establish every operational or analytical detail of the trial.

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: Full Analysis Set (FAS) defined as all subjects who had taken at least 1 dose of investigational product during the study.

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 Change From Baseline in Attention-Deficit/Hyperactivity Disorder-Rating Scale-IV (ADHD-RS-IV) Total Score at Week 8 - Last Observation Carried Forward (LOCF), measured over Baseline and up to 8 weeks.

The primary endpoint was change from baseline in the ADHD-RS-IV total score at Week 8, with the last observation carried forward for participants without a Week 8 observation. The registry states that the scale has 18 items, each scored from 0 (no symptoms) to 3 (severe symptoms), and a total score ranging from 0 to 54. The endpoint therefore summarizes rated ADHD symptoms on a single total-score scale.

A change score expresses how the measured value at follow-up differs from the participant’s baseline value. A negative change in this measure indicates a lower score at the endpoint than at baseline. Comparing mean changes between randomized groups helps distinguish the pattern in each treatment group from changes also observed in the placebo group over the study period.

The registry reports least-squares means and standard errors for this outcome and identifies the primary time frame as baseline and up to 8 weeks. ANCOVA is the reported method for the sponsor’s primary comparisons. The record does not provide the model’s covariates, the precise handling of all missing observations, or a threshold defining a clinically important difference. Statistical evidence about a group contrast should therefore not be treated as a direct measure of how important that difference is to patients.

7. Results Posted in the Registry

• SPD503 AM: mean -20.00 versus -10.60 on control; difference -9.40 (95% CI -12.77 to -6.03), standardized effect d = -0.75; z-test p = <0.001.
• SPD503 PM: mean -20.40 versus -10.60 on control; difference -9.80 (95% CI -13.11 to -6.49), standardized effect d = -0.78; z-test p = <0.001.

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.

Improvement on Clinical Global Impression-Improvement (CGI-I) Scale at Week 8 - LOCF

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • SPD503 AM: 66.3% (69/104) versus 31.8% (35/110) on control; risk difference 34.5 percentage points (95% CI 21.9 to 47.1), risk ratio 2.09 (1.54–2.83); unadjusted two-proportion z-test p = <0.001.
  • SPD503 PM: 67.0% (75/112) versus 31.8% (35/110) on control; risk difference 35.1 percentage points (95% CI 22.8 to 47.5), risk ratio 2.10 (1.55–2.85); unadjusted two-proportion z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for SPD503 AM 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 proportion with the event is statistically higher for SPD503 PM than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Change From Baseline in Pediatric Daytime Sleepiness Scale (PDSS) Total Score at Week 8 - LOCF

• 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:
  • SPD503 AM: mean -2.70 versus -3.10 on control; difference 0.40 (95% CI -1.10 to 1.90), standardized effect d = 0.07; z-test p = 0.600.
  • SPD503 PM: mean -1.40 versus -3.10 on control; difference 1.70 (95% CI 0.23 to 3.17), standardized effect d = 0.31; z-test p = 0.023.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between SPD503 AM 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 higher for SPD503 PM than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Change From Baseline in Health Utilities Index-2/3 (HUI 2/3) Scores at Week 8 - LOCF

• 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:
  • SPD503 AM: mean 0.06 versus 0.04 on control; difference 0.01 (95% CI -0.02 to 0.05), standardized effect d = 0.11; z-test p = 0.434.
  • SPD503 PM: mean 0.06 versus 0.04 on control; difference 0.02 (95% CI -0.01 to 0.05), standardized effect d = 0.18; z-test p = 0.212.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between SPD503 AM 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 SPD503 PM and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes broaden the picture beyond ADHD-RS-IV symptom scores. They include global clinical ratings, daytime sleepiness, health utility, parent-rated behavior and functioning, and sleep-related measures. The registry lists outcomes such as CGI-S, CGI-I, PDSS, HUI 2/3, CPRS-R:S, bedtime resistance, sleep quality, and WFIRS-P global score. The posted results are not equally complete across these measures, and no single secondary pattern can be assumed to represent them all.

The CGI-I summaries show a higher percentage classified as improved in the active groups than in placebo, with sponsor-reported comparisons for All-Active and AM. In contrast, the PDSS analyses are not uniformly aligned with the primary result: the reported placebo comparisons for AM and All-Active have p-values of 0.596 and 0.099, while the simple unadjusted PM recalculation is statistically different from zero. HUI 2/3 group means are posted, but no sponsor analysis is listed. These findings illustrate variation across outcomes and analyses rather than a single, consistent treatment pattern.

Secondary outcomes are often interpreted more cautiously because multiple outcomes and comparisons create more opportunities for chance findings. An unadjusted p-value does not account for that broader testing context. The registry record does not report a multiplicity adjustment or a hierarchy specifying which secondary outcomes were confirmatory. Accordingly, individual unadjusted secondary results should be treated as exploratory evidence unless a more complete prespecified analysis plan is available.

The record does not provide enough information to judge whether all registered secondary endpoints were analyzed as planned or how missing observations affected each measure. Secondary outcomes can add useful context, but they do not overturn or independently validate the primary result. Their interpretation depends on the outcome-specific methods, completeness of reporting, and the unreported multiplicity strategy.

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

• Assessment of Clinical Global Impression-Severity of Illness (CGI-S) at Week 8 - LOCF
• Improvement on Clinical Global Impression-Improvement (CGI-I) Scale at Week 8 - LOCF
• Change From Baseline in Pediatric Daytime Sleepiness Scale (PDSS) Total Score at Week 8 - LOCF
• Change From Baseline in Health Utilities Index-2/3 (HUI 2/3) Scores at Week 8 - LOCF
• Change From Baseline in Conner's Parent Rating Scale - Revised Short Version (CPRS-R:S) Score at Week 8 - LOCF
• Change From Baseline in the Bedtime Resistance Subscale of Child's Sleep Habits Questionnaire (CSHQ) at Week 8 - LOCF
• Post Sleep Questionnaire (PSQ) Quality of Sleep at Week 8 - LOCF
• Change From Baseline in Mean Weiss Functional Impairment Rating Scale - Parent Report (WFIRS-P) Global Score at Week 8 - LOCF

9. Statistical Interpretation

10. Statistical Methods Explained

What does the primary comparison estimate?

The primary outcome is a continuous score, so the reported group summaries are least-squares means of change from baseline. The main comparison asks how much the mean change in an active-treatment group differs from the mean change in placebo. The registry’s reported method is ANCOVA. In general, ANCOVA compares groups while accounting for specified covariates, but the covariates used in this trial are not reported in the registry record. The sponsor’s reported ANCOVA p-values are the available inferential results for these primary contrasts; adjusted effect estimates and confidence intervals are not listed.

A simple unadjusted recalculation can be made from the posted group summaries. For the primary endpoint, it subtracts the placebo mean change from the active-group mean change. The resulting estimates are -9.4 for AM versus placebo and -9.8 for PM versus placebo. Their confidence intervals use an estimated pooled standard deviation derived from the posted standard errors and group sizes; the record identifies this standard deviation approximation as approximate. These calculations do not reproduce the sponsor’s ANCOVA model and should be read as supplementary, unadjusted comparisons.

How should p-values be read?

A p-value describes how compatible the observed data are with a specified null comparison under the assumptions of the analysis. 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 registry gives p-values of <0.001 for each primary placebo comparison. The sponsor’s analysis method is ANCOVA. The recalculated p-values are also unadjusted and are not estimates from the sponsor’s model.

Some secondary results have larger p-values, and some recalculated secondary comparisons differ from the sponsor’s listed analyses. Such differences can arise because an adjusted model and a simple comparison of summary statistics answer related but not identical statistical questions. The registry record does not provide enough model detail to reconstruct the sponsor’s ANCOVA or Cochran-Mantel-Haenszel analyses.

What do confidence intervals add?

A confidence interval describes a range of effect estimates compatible with the data and the method used to calculate it. For a mean difference, the interval is expressed in the score’s units. Its width communicates precision: a narrower interval indicates less statistical uncertainty under the calculation’s assumptions. The intervals accompanying the simple recalculations are not the sponsor’s adjusted confidence intervals, which are not reported in the listed analyses.

An interval should be interpreted together with the endpoint definition, group summaries, analysis method, and study limitations. It does not account for unreported sources of uncertainty such as departures from modeling assumptions or selective reporting. General statistical interpretation also distinguishes the estimated group average from individual-level outcomes; an interval for an average does not imply that all participants had changes close to that average.

What does LOCF mean in this record?

LOCF means last observation carried forward: a participant’s most recent available measurement is used as the endpoint value when the scheduled endpoint measurement is missing. The registry explicitly names LOCF for the primary outcome and several secondary outcomes. This identifies one element of missing-data handling, but it does not report how often observations were missing, why they were missing, or whether other missing-data procedures were used in sensitivity analyses.

In general, carrying a previous observation forward can make an analysis depend on the assumption that the carried value is an appropriate stand-in for the unobserved endpoint. That assumption may not hold for every participant. The registry record does not report analyses that assess how sensitive the results are to alternative missing-data assumptions.

Why do analyses differ across outcomes?

For the primary continuous endpoint, the registry reports ANCOVA. For CGI-I, a binary outcome expressed as a percentage of participants, the listed sponsor method is Cochran-Mantel-Haenszel. In general, this method can compare categorical outcomes while accounting for strata, but the registry record does not identify the strata or other implementation details. HUI 2/3 is reported as a mean with standard deviation, and no sponsor analysis is listed for that outcome.

These methods are suited to different outcome formats, but method labels alone do not provide a complete analysis plan. Details such as covariates, stratification variables, model diagnostics, missing-data rules beyond LOCF, and adjustments for multiple comparisons are not reported in the registry record. Consequently, the posted analyses can be described, but their full specification cannot be independently reconstructed from the record.

Finally, the planning figures are illustrations based on observed effects, not evidence about the trial’s original planning or its achieved power. General power calculations describe how often a test would reject a null hypothesis under stated assumptions if a particular effect were present. The registry record does not report the original sample-size justification or its assumptions. The planning illustrations use their own stated assumptions and should not be interpreted as observed power or as confirmation that the trial was adequately designed for every secondary outcome.

11. Confidence Intervals and Precision

The primary simple unadjusted recalculation for SPD503 AM versus placebo is a mean difference of -9.4, with a 95% confidence interval from -12.77 to -6.032. For SPD503 PM versus placebo, it is -9.8, with a 95% confidence interval from -13.11 to -6.488. Both intervals lie below zero, consistent with the more negative mean changes in the active groups than in placebo. The recalculated p-values are each <0.001.

These intervals convey precision for unadjusted estimates reconstructed from the posted summary statistics. The calculation uses approximate pooled standard deviations derived from standard errors and group sizes. It does not reproduce the sponsor’s ANCOVA adjustment. The registry reports 95% confidence levels for the sponsor analyses, but their adjusted contrast limits and estimates are not listed. Thus, the recalculated intervals should not be presented as the sponsor’s confidence intervals.

For a continuous outcome, an interval gives a range of plausible values for the group mean contrast under the calculation’s assumptions. A confidence interval is not a range expected to contain individual participant responses, and it does not show whether a difference meets a clinically meaningful threshold. No such threshold is reported in the registry record.

Precision also depends on the outcome and analysis population. The HUI 2/3 comparisons have intervals that include zero, while the recalculated PM comparison for PDSS has an interval that does not. These are secondary, unadjusted comparisons, and the record does not report a multiplicity procedure. The intervals therefore describe statistical uncertainty for their individual calculations, not certainty across the collection of outcomes.

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 80% power and participants per arm for 90% power. They are planning illustrations based on the observed effect, not the trial’s original design assumptions. The stated calculation assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is the stated fraction of the observed effect. For continuous outcomes, the figures are participants per arm; the general convention for time-to-event outcomes would be total events, but no time-to-event planning figures are described for these outcomes.

The 80% power column describes a planning scenario with an 80% chance of detecting the specified effect under the calculation assumptions, if that effect were the true effect. The 90% power column describes the corresponding scenario with a 90% chance. These are not guarantees for a future study: the actual result depends on the true effect, variability, analysis, and other conditions.

Rows for 100% of the observed effect use the observed estimate as the target effect. Rows for 75% and 50% of the observed effect use smaller target effects. Smaller effects are harder to distinguish from random variation, so the planning figures increase as the target effect becomes smaller. For example, for the primary AM comparison, the illustration gives 29 participants per arm for 80% power at 100% of the observed effect and 113 participants per arm at 50% of the observed effect. For PM, the corresponding figures are 26 and 104 participants per arm.

These calculations are planning illustrations and are not observed power. They do not show the probability that the completed trial’s findings are correct, and they do not establish the sample-size rationale used by the trial team. The original design assumptions and sample-size justification are not reported in the registry record.

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 = -9.400; pooled standard deviation σ = 12.565.
2. n = 2 × (1.96 + 0.842)² × 12.565² / -9.400² = 28.0, rounded up to 29 participants per arm for 80% power.

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

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
delta <- -9.4                      # observed difference in means
sigma <- 12.5652607357942                     # 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(997984)
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 = 29 (80% power), 38 (90% power)
 75% of observed effect: n per arm = 50 (80% power), 67 (90% power)
 50% of observed effect: n per arm = 113 (80% power), 151 (90% power)
power.t.test: 29 
Simulated power at n = 29 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 registry reports ANCOVA for the primary comparisons of placebo with SPD503 AM, SPD503 PM, and All-Active. Each listed primary p-value is <0.001, and the reported confidence level is 95%. The sponsor’s adjusted contrast estimates and confidence limits are not included in the listed results.

The simple recalculations compare the posted mean changes without reproducing the sponsor’s ANCOVA model. They estimate -9.4 for AM versus placebo and -9.8 for PM versus placebo, with 95% confidence intervals from -12.77 to -6.032 and from -13.11 to -6.488, respectively. Their p-values are each <0.001. These results are directionally consistent with the reported primary analyses, but they are not the same estimand or model-based result.

The registry’s reported analyses take precedence for describing the sponsor’s own statistical conclusions. The record does not flag a discrepancy between those results and the recalculations. The recalculations are useful as a transparent check on the posted summaries, but they cannot replace adjusted results or reveal the model details that are not reported in the registry record.

For secondary outcomes, reported analyses and simple recalculations are also distinct. For example, the registry lists Cochran-Mantel-Haenszel results for CGI-I and ANCOVA results for selected PDSS comparisons. Recalculated comparisons from arm summaries may not match these analyses because they do not incorporate the sponsor’s adjustment or modeling choices.

15. Multiplicity, Missing Data and Analysis Populations

The registry lists 15 outcome measures and 24 analyses. Testing several outcomes or comparisons increases the chance that at least one result will appear statistically notable even when there is no underlying effect. General statistical approaches to this issue include adjusting significance thresholds, controlling an overall error rate, or defining a hierarchy of primary and secondary tests. The registry record does not report whether any such procedure was used, nor does it describe a testing hierarchy beyond identifying the primary outcome.

This missing information matters especially for secondary findings. A nominal p-value describes an individual comparison under its own analysis assumptions; without the multiplicity strategy, it cannot establish how the result should be judged across the full set of outcomes. The primary outcome is clearly designated, and its reported comparisons can be described as primary analyses. The record does not establish that every listed secondary comparison had the same confirmatory status.

LOCF is named for the primary endpoint and several secondary outcomes. The record does not report the amount or pattern of missing data, reasons for missingness, or sensitivity analyses using alternative assumptions. In general, an analysis that carries a last observed value forward depends on that carried value being an appropriate proxy for the endpoint that was not observed. The available record does not show whether this assumption was examined.

The primary population is the Full Analysis Set, defined as subjects who took at least 1 dose. Endpoint-specific counts vary, and the record does not explain all differences. Further details about exclusions, analysis populations for each secondary outcome, covariate adjustment, and treatment of missing values beyond the stated LOCF approach are not reported in the registry record. Those omissions limit assessment of robustness and prevent full reconstruction of the statistical plan.

16. 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: syncope (0.9%), self-injurious ideation (0.9%), suicidal ideation (0.9%).

The registry lists serious adverse event counts and the number at risk for placebo, SPD503 AM, SPD503 PM, and All-Active. The serious adverse event affected counts are 0 of 112 for placebo, 1 of 107 for SPD503 AM, 2 of 114 for SPD503 PM, and 3 of 221 for All-Active. These are the recorded group counts; the All-Active entry is a separate registry summary and should not be added to the individual active-group counts.

The listed serious adverse event terms are syncope, self-injurious ideation, and suicidal ideation, each reported as 0.9%. The record does not provide a term-by-arm breakdown alongside these percentages. Death counts are not reported in the registry record.

These figures should be read as descriptive safety reporting, not as a comparison adjusted for exposure or other factors. The record does not provide duration-adjusted event rates or enough detail to infer why events occurred. The listed counts alone do not establish a causal relationship, and the absence of a recorded event in a group does not establish absence of risk.

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

The posted primary results show larger decreases in mean ADHD-RS-IV total score in the SPD503 AM and SPD503 PM groups than in placebo, and the registry reports p-values of <0.001 for the primary placebo comparisons. The unadjusted recalculations produce mean differences of -9.4 and -9.8, with confidence intervals that lie below zero. This describes a statistical difference in group averages over the study’s baseline-to-Week 8 endpoint period.

That finding does not describe how much any particular child’s score changed, nor does it indicate that every participant had a decrease. It also does not, on its own, determine the clinical importance of the difference. The registry record does not report a threshold for a clinically important ADHD-RS-IV change or adjusted effect estimates for the sponsor’s ANCOVA analyses.

The primary result also does not establish that morning and evening dosing are equivalent, or that one schedule is better than the other. The listed primary analyses compare each dosing group with placebo; a direct AM-versus-PM analysis is not reported in the registry record. Similar group means are not, by themselves, evidence of equivalence or noninferiority.

Nor do primary findings imply that all secondary outcomes changed consistently. The registry’s secondary results include a different pattern for daytime sleepiness and incomplete reported analyses for other domains. The record does not provide multiplicity details, a complete account of missing-data sensitivity analyses, or all model specifications. Those gaps limit how broadly the statistical results can be interpreted.

Finally, the trial’s age range, diagnosis, treatment schedule, endpoint, and study period define the context of these observations. The registry record alone does not establish whether the findings apply to people outside the studied population or beyond the recorded endpoint period. The statistics describe the registered trial comparisons; they are not individualized predictions or medical advice.

18. Why This Trial Matters Statistically

This trial is statistically instructive because it compares two dosing times with a shared placebo group and reports both a continuous symptom-score endpoint and several secondary measures with different data types. The primary outcome uses mean changes and ANCOVA, while CGI-I is categorical and analyzed with a Cochran-Mantel-Haenszel method. The contrast between methods illustrates why an analysis should be matched to the outcome rather than treating every endpoint as the same kind of measurement.

The distinction between sponsor-reported analyses and simple recalculations is also important. The posted arm summaries permit a straightforward unadjusted comparison, but that calculation does not reproduce the reported ANCOVA. Clear labeling preserves the value of a transparent numerical check without implying that it overrides the trial’s own analysis.

The record also demonstrates why statistical significance, effect size, and precision are separate ideas. The primary p-values are small, and the unadjusted intervals are relatively far from zero, yet the registry does not report a clinical-importance threshold or adjusted contrast estimates. A statistical result is not a complete account of its practical meaning.

Finally, multiple secondary outcomes and incomplete reporting of multiplicity and missing-data procedures make the interpretation of secondary findings more complicated. Recognizing those limits is part of reading trial results responsibly. The study’s statistical record is most informative when the primary result, outcome-specific analyses, confidence intervals, and reporting gaps are considered together.

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.

The registry record does not report the original sample-size justification, the assumptions used to plan enrollment, or any interim analyses. Planning illustrations based on observed effects cannot fill those gaps. The full statistical analysis plan, including covariates and model specifications for ANCOVA, is also not reported.

The sponsor’s primary ANCOVA results include p-values and a 95% confidence level, but the adjusted contrast estimates and confidence limits are not listed. The available numerical confidence intervals are simple unadjusted recalculations using approximate pooled standard deviations. They are informative supplementary estimates, not replacements for the sponsor’s adjusted results.

LOCF is specified for the primary endpoint, but the registry does not report how many observations were carried forward, the reasons for missingness, or sensitivity analyses under alternative missing-data assumptions. Endpoint counts vary, and the record does not explain each difference. Those details affect assessment of the analyzed populations and the robustness of the estimates.

The record also does not describe multiplicity control across the outcome measures and analyses. This limits the strength of inferences from secondary p-values. Some outcomes have group summaries without a listed sponsor analysis, while other outcomes are named without detailed numerical results in the available record.

Although the study is described as quadruple-masked in the design fields and double-blind in its title, the registry does not explain the masking procedures. Details of randomization implementation, dose optimization, treatment adherence, and model diagnostics are not reported. Finally, the record does not report a clinically important difference threshold for ADHD-RS-IV, so the practical importance of the statistical contrast cannot be judged from the registry results alone.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Tolerability and Efficacy of AM and PM Once Daily Dosing With Extended-release Guanfacine Hydrochloride in Children 6-12 With Attention-Deficit/Hyperactivity Disorder (ADHD) (The ADHD Tempo Study) — NCT00997984.
• Registry record last updated 2021-06-08; status last verified 2021-05.
• Results first posted 2011-10-26.
• 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 describes a completed Phase 3 randomized, parallel-group, multicenter, placebo-controlled study of once-daily extended-release guanfacine hydrochloride in children aged 6-12 with ADHD. It compared morning and evening dose-optimized SPD503 with placebo. The primary endpoint was ADHD-RS-IV total-score change from baseline to Week 8 using LOCF in the Full Analysis Set.

The reported primary ANCOVA p-values for placebo comparisons with AM, PM, and All-Active are each <0.001. Simple unadjusted recalculations from the posted summary statistics estimate mean differences of -9.4 for AM and -9.8 for PM versus placebo. These are distinct from the sponsor’s adjusted analyses. Secondary outcomes show a less uniform pattern, and the registry record does not report multiplicity procedures, detailed model specifications, or sensitivity analyses for missing data.