A Study for Patients With Attention-Deficit/Hyperactivity Disorder Treated With Atomoxetine: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study for Patients With Attention-Deficit/Hyperactivity Disorder Treated With Atomoxetine (NCT00471354), a na phase 4 study. The primary outcome is Correlation Between Change From Baseline and 24 Week Endpoint in Attention-Deficit/Hyperactivity Disorder Rating Scale-IV-Parent:Investigator-Administered and Scored (ADHDRS-IV-Parent:Inv) Total Score and School Grade Average (SGA).

2. Clinical Question

3. Trial Design

The registry describes a completed Phase 4 study with a single-group model and no applicable allocation. It lists atomoxetine as the intervention and enrollment of 228. No masking is listed. As there is no separate comparison group, this design can describe changes and associations observed among participants, but it cannot provide a randomized between-group estimate of treatment effect.

The study population is described in the official title as Asian children aged 8 to 11 years with ADHD. The intervention was atomoxetine taken by mouth once daily, starting at 0.5 mg/kg/day. The dose was titrated over 1 week to a target of 1.2 mg/kg/day for 23 weeks. The primary outcome timeframe is baseline and 24 weeks. The registry record does not report additional details about visit scheduling or the timing of titration beyond this description.

The stated purpose was to investigate the relationship between changes in academic performance and problem behaviors and changes in core ADHD symptoms in children treated with atomoxetine. The primary outcome focuses on the association between change in ADHD-RS total score and change in total SGA. The listed secondary outcomes include associations with grades in separate language, math, and science classes; changes in total and subject-specific grade averages; changes in ADHD-RS and Conners’ parent-rating scores; and Clinical Global Impressions measures.

The record lists a start date of 2007-04, primary completion of 2008-11, and status as completed. More detailed information about sites, recruitment procedures, eligibility criteria beyond the population description, and the protocol is not reported in the registry record.

4. Participant Flow and Accounting

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

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All patients with baseline and at least one non-missing post-baseline score for each of the variables, regardless of them having or not having received any ordinal grades (for the SGA).

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 Correlation Between Change From Baseline and 24 Week Endpoint in Attention-Deficit/Hyperactivity Disorder Rating Scale-IV-Parent:Investigator-Administered and Scored (ADHDRS-IV-Parent:Inv) Total Score and School Grade Average (SGA), measured over Baseline, 24 weeks.

The primary endpoint is a Spearman correlation coefficient between change in the ADHDRS-IV-Parent:Investigator-Administered and Scored total score and change in total School Grade Average (SGA) from baseline to the 24-week endpoint. It is designed to quantify how the two measured changes vary together across the analysis population. The correlation estimate itself is not reported in the registry record.

The ADHD rating scale measures 18 symptoms associated with ADHD. Each item ranges from 0, meaning none, never, or rarely, to 3, meaning severe or very often; total scores range from 0 to 54. SGA is calculated from grades in language, math, and science, with the total score ranging from 0 to 100. Higher SGA scores indicate better grades or aptitude. The record says ordinal grades were converted to numerical grades based on communication with relevant schools.

As a general statistical concept, correlation is used to describe the degree to which two quantities vary together. It does not by itself establish that change in one measure causes change in the other, and it is not a direct comparison of atomoxetine with an untreated or alternative-treatment group. The registry record does not report a confidence interval or a test result for this endpoint.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Spearman correlation is the stated method for the primary endpoint. As a general statistical concept, it assesses whether two variables tend to have a monotonic association: as one variable changes in rank, the other tends to change in a consistent direction. It does not require the relationship to follow a straight line. It is an association measure, not a causal estimate and not a comparison between treatment groups.

What are the two changes being related?

The first variable is change in the ADHDRS-IV-Parent:Investigator-Administered and Scored total score. The scale covers 18 ADHD symptoms, with individual item values from 0 to 3 and total values from 0 to 54. The second variable is change in total SGA. Grades from language, math, and science contribute to this measure, which ranges from 0 to 100; higher scores indicate better grades or aptitude. The endpoint timeframe is baseline to 24 weeks.

The numerical coding and direction of a change score matter when interpreting correlation direction. The scale description establishes that higher SGA scores indicate better grades, but the registry record does not specify the sign convention used to calculate each change score. The record also does not report the resulting correlation, so its direction cannot be interpreted from the endpoint description alone.

What does the analysis population mean?

The primary population is described as all patients with a baseline value and at least one non-missing post-baseline score for each variable, regardless of whether ordinal grades were received for SGA. This indicates that observations were required for both measures. The record does not report how many patients met the criterion, which post-baseline visit or visits contributed to the calculation, or how multiple available observations were handled. It does not describe imputation for missing values or sensitivity analyses for missing data.

Ordinal grades could be converted to numerical grades based on communication with relevant schools. The registry does not describe the conversion rules, whether those rules differed between schools, or how often such conversion was needed. These details could affect the comparability of academic scores, but their impact cannot be assessed from the registry record.

How should precision and statistical significance be assessed?

A confidence interval is a general statistical way to express uncertainty around an estimate: it gives a range of values compatible with the data under the interval’s assumptions and procedure. The registry record does not report a confidence interval for the primary correlation. Without it, the precision of the estimate cannot be evaluated. Statistical significance is also not reported; no p-value for the primary endpoint appears in the registry record. The record does not report the statistical analysis plan, including whether the correlation test was one-sided or two-sided.

General statistical interpretation should distinguish the size of an estimate from its uncertainty. A small p-value, if reported, would not by itself indicate a strong association or establish causation; an estimate and its precision are also important. Here, neither the estimate nor its uncertainty is reported, so those aspects cannot be judged from the registry record.

What do the planning assumptions represent?

The planning assumptions describe an illustration using a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with the effect to detect expressed as a fraction of the observed effect. Such sample-size calculations are planning illustrations, not observed power and not evidence about the study’s actual design or statistical plan. The assumptions refer to a two-arm illustration, whereas this registry record describes a single-group study. The calculation does not establish that the trial used those assumptions, and the sample-size justification is not reported in the registry record.

The registry lists 7 outcome measures and 11 analyses. Those counts do not, by themselves, identify which statistical tests were used for every outcome, whether all analyses were prespecified, or whether results were adjusted for multiple testing. Those details are not reported in the registry record. The main methodological information available is the named Spearman correlation, the defined variables, the endpoint timeframe, and the stated analysis-population criterion.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists 7 outcome measures and 11 analyses. The primary endpoint is one association between changes in ADHD-RS total score and total SGA. Secondary outcomes include related associations for separate school subjects, changes in academic scores, changes in ADHD symptom ratings, and Clinical Global Impressions measures. Testing or estimating several outcomes can raise multiplicity concerns: as a general statistical issue, interpreting many results without an appropriate plan may increase the chance that some findings appear notable by chance.

The registry record does not report whether the analyses were prespecified in a statistical analysis plan, whether any multiplicity adjustment was used, or how the primary result was distinguished from secondary or exploratory results. It also does not report results for the secondary measures. The count of outcomes and analyses alone cannot establish which hypotheses were tested or how their results should be prioritized.

For the primary analysis, participants needed a baseline and at least one non-missing post-baseline score for each variable. The number included under this criterion is not reported in the registry record. The record does not explain how missing data were handled beyond the stated inclusion criterion, whether observations were excluded for other reasons, or whether sensitivity analyses assessed the effect of missingness. Consequently, completeness of the paired measurements and the robustness of the primary result cannot be assessed from this information.

The record notes conversion of ordinal school grades to numerical values based on communication with relevant schools. It does not describe the conversion process in detail or how missing grades were handled. No further analysis-population details or missing-data methods are reported in the registry record.

11. 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: hyperthyroidism (0.4%).

The registry lists 1 serious adverse event among 228 at risk in the atomoxetine group. It also lists 175 affected among 228 at risk for other adverse events. These are separate categories in the record and should not be combined into a total. The listed top serious-event term is hyperthyroidism, reported as 0.4%.

These figures describe recorded events in a single-group study. Without a separate comparison group, they do not establish how the event frequency compares with another treatment or with no treatment. The counts also do not establish that atomoxetine caused an event. The record does not provide additional detail on event timing, severity beyond the serious-event classification, or individual clinical circumstances.

Deaths are not reported in the registry record. The listed adverse-event figures should be read as registry summaries, not as a complete account of every possible safety outcome or as a basis for estimating comparative risk. The information available does not support a treatment recommendation.

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

The record establishes that the study asked whether changes in ADHD symptoms were associated with changes in academic grades among Asian children aged 8 to 11 years treated with atomoxetine. It specifies a single-group design, the dose schedule, the baseline-to-24-week timeframe, the two measures, and the use of Spearman correlation. These details clarify what the primary analysis was intended to examine.

The record does not report the primary correlation estimate, its confidence interval, or a p-value. It therefore does not show whether the measured changes were associated, in what direction they were associated, or how precisely the association was estimated. The outcome definition cannot substitute for the missing result.

Even if a correlation estimate were available, it would describe co-variation rather than prove that a change in ADHD symptoms produced a change in grades. The study has no separate randomized comparator group listed, so it does not provide a randomized estimate contrasting atomoxetine with another condition. Changes during follow-up could reflect influences beyond the intervention, and the registry record does not describe analyses that separate those influences.

The enrollment figure is not necessarily the number contributing to the primary analysis. The record requires a baseline observation and at least one non-missing post-baseline observation for both variables, but does not report how many participants met that requirement. It also does not report the missing-data methods or the detailed analysis plan.

The adverse-event summaries describe events recorded in the atomoxetine group, not comparative risks. Taken together, the record is informative about the question, measures, design, and some safety counts, but the numerical primary result and key analysis details are not reported in the registry record. No conclusion about treatment benefit or lack of benefit follows from those missing results.

13. Why This Trial Matters Statistically

This study’s primary endpoint connects two kinds of outcomes that clinical researchers may want to examine together: parent-rated ADHD symptoms and school performance. Its definition makes the measurement choices explicit. ADHD symptoms were assessed with the ADHDRS-IV-Parent:Investigator-Administered and Scored total score, while academic performance was represented by grade averages in language, math, and science.

The endpoint is statistically distinct from a treatment-versus-control comparison. A correlation can describe whether measured changes tend to vary together, but it cannot by itself identify a causal pathway or quantify an intervention’s effect relative to a comparator. That distinction is particularly important for this single-group design, which lists no randomized comparison group.

The registry also illustrates why an endpoint definition is not enough to interpret a result. Assessment requires the estimate, its uncertainty, the analysis population, and details such as handling of missing observations and multiple analyses. The record gives the primary analysis population criterion, but does not report the number included, the correlation estimate, or its precision. As a result, the statistical question is clear while the empirical answer is not available in the registry record.

The academic measure includes a conversion of ordinal grades to numerical grades based on communication with schools. This makes the construction of the outcome relevant to interpretation: readers would need to know the conversion approach and how consistent it was across settings to assess measurement comparability. Those details are not reported in the registry record.

For biostatistical reading, the central lesson is to separate the study’s design, the association it set out to measure, and the numerical evidence needed to evaluate that association. The study description supports the first two; the reported record does not provide the primary result needed for the third.

14. 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 single-group design is a central limitation for interpretation. The registry lists no comparator and no applicable allocation, so observed changes over time cannot be contrasted with randomized outcomes under another treatment or control condition. A correlation between symptom and grade changes would remain an association within the study, not an isolated estimate of treatment effect.

The primary numerical result is not reported in the registry record. The correlation estimate, its confidence interval, and its p-value are not reported, preventing assessment of direction, magnitude, and precision. The number of participants meeting the primary analysis criterion is also not reported, despite enrollment of 228.

The analysis-population description requires baseline and at least one non-missing post-baseline score for both variables, but the record does not state how missing observations were handled beyond this criterion. It does not report sensitivity analyses or the detailed statistical analysis plan. The registry also lists 7 outcome measures and 11 analyses without reporting whether multiplicity was addressed.

Academic grades were converted from ordinal values to numerical values based on communication with relevant schools. The conversion rules, consistency across schools, and handling of missing grades are not reported in the registry record. The record describes higher SGA scores as better grades or aptitude, but does not specify the change-score sign convention used in the correlation.

The intervention schedule is described, but adherence and participant-level exposure are not reported. Further details about sites, recruitment, and eligibility beyond the stated population are not reported in the registry record. These limitations leave important uncertainties about measurement, analysis, and generalizability.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: A Study for Patients With Attention-Deficit/Hyperactivity Disorder Treated With Atomoxetine — NCT00471354.
• Registry record last updated 2010-01-12; status last verified 2010-01.
• Results first posted 2009-12-31.
• 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.

18. Record Summary

NCT00471354 is a completed Phase 4, single-group study of atomoxetine in Asian children aged 8 to 11 years with ADHD. Enrollment was 228. The listed regimen began at 0.5 mg/kg/day by mouth once daily, was titrated over 1 week to 1.2 mg/kg/day, and continued for 23 weeks. The primary endpoint was the Spearman correlation between change in ADHD-RS total score and change in total SGA from baseline to the 24-week endpoint.

The registry defines the symptom and academic measures and describes the primary analysis population as patients with baseline and at least one non-missing post-baseline value for each variable. The correlation estimate, confidence interval, p-value, number in the primary analysis, and detailed missing-data and multiplicity methods are not reported in the registry record. The record lists 7 outcome measures and 11 analyses. It reports 1 serious adverse event among 228 at risk, 175 affected among 228 at risk for other adverse events, and hyperthyroidism at 0.4% among the top serious-event terms.