Comparing the Effectiveness of New Versus Older Treatments for Attention Deficit Hyperactivity Disorder (The NOTA Study): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Comparing the Effectiveness of New Versus Older Treatments for Attention Deficit Hyperactivity Disorder (The NOTA Study) (NCT00889915), a randomized phase 4 study. The primary outcome is Dichotomized Clinical Global Impression-Effectiveness (CGI-E) Scale.

2. Clinical Question

3. Trial Design

The NOTA Study is registered as a randomized, parallel-group, Phase 4 trial. Masking is listed as none, so the study is described in the registry as unmasked. The record lists an enrollment of 228 and a completed status. The study start is recorded as 2009-04 and primary completion as 2009-12.

Four medication arms are listed: Daytrana, a methylphenidate transdermal system; Vyvanse, lisdexamfetamine dimesylate; Concerta, osmotic-release oral system methylphenidate; and Adderall, mixed amphetamine salts extended release. The posted primary-endpoint groups contain 35 participants for Daytrana, 71 for Vyvanse, 54 for Concerta, and 40 for Adderall. These are the arm counts attached to the endpoint results; the record does not explain how they relate to enrollment or provide a participant-flow account.

The study summary describes a comparison of two newer psychostimulants with two older medications, with effectiveness, tolerability, and acceptability as its broad concerns. The primary endpoint combines therapeutic benefit and adverse impact into a dichotomized response classification. The two listed secondary outcomes are the Clinical Global Impressions-Improvement (CGI-I) Scale and Clinical Global Improvements-Acceptability (CGI-A) Scale.

The registry record does not report the randomization procedure, allocation concealment, treatment duration beyond the endpoint timing, or the details of medication administration. It also does not report a protocol-based sample-size justification or a statistical analysis plan. These omissions limit what can be established about the conduct and intended analysis from the record alone.

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

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 Dichotomized Clinical Global Impression-Effectiveness (CGI-E) Scale, measured over Measured at each participant's last visit, which can occur at or before Week 6.

The primary endpoint is the dichotomized Clinical Global Impression-Effectiveness (CGI-E) Scale. The registry describes CGI-E as combining each participant’s therapeutic benefit score with the adverse impact of tolerability and acceptability problems. A participant classified as much or very much improved therapeutically, with no or mild adverse impact, is described as a responder. The remaining specified categories are classified as non-responders when adverse impact is moderate or outweighs therapeutic effect.

This composite aims to represent both improvement and the burden of treatment rather than therapeutic change alone. That can make the measure relevant to a study asking about effectiveness alongside tolerability and acceptability. It also means that the endpoint’s interpretation depends on how the component ratings are combined and how the response categories are defined.

The registry says CGI scores are totaled for each participant and that a mean score is calculated, while also stating that a median total score is calculated for each treatment group. The posted primary results give a median and full range for each arm. Assessment is at each participant’s last visit, which can occur at or before Week 6. The record does not report the distribution of visit timing, component-level results, or an inferential comparison between arms.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a median summarize?

In general, the median is the central value after observations are ordered. It is useful when a measure is ordinal or when a distribution may not be well represented by its arithmetic mean. It can be less affected by unusually high or low observations than a mean. A median does not, by itself, convey how concentrated observations are near the center, how many observations fall in each response category, or how uncertain the estimate is.

For the primary CGI-E endpoint, the registry reports medians of 7 for Daytrana, 5 for Vyvanse, 6 for Concerta, and 6 for Adderall. These are descriptive arm summaries. The record also reports a full range of 1 to 16 in every arm. A range is the interval between the observed minimum and maximum. It communicates the endpoints of the observed values but gives no account of the number of observations at either endpoint or the pattern between them.

What does the endpoint’s dichotomization mean?

In general, dichotomization turns a measure or set of ratings into two categories. Here, the registry’s description calls those categories responders and non-responders, based on a combination of therapeutic improvement and adverse impact from tolerability or acceptability. The rule is intended to reflect both perceived benefit and impact from problems associated with treatment.

Reducing component ratings to two categories can make a response classification easier to interpret, but it also discards distinctions among people placed in the same category. It can make the result sensitive to the chosen classification rules. The record describes examples of how CGI-I and adverse-impact ratings map to responder status, but does not post component-level results or the number of participants in each response category. The posted median total scores therefore should not be mistaken for responder proportions.

Why is a direct treatment comparison not available?

In general, a comparative analysis estimates how outcomes differ between treatment groups and quantifies uncertainty in that estimate. Depending on the endpoint and analysis plan, this might involve a difference in means, a ratio of risks, an odds ratio, or another effect measure. A confidence interval can communicate the precision of such an estimate; a hypothesis test can assess compatibility with a specified null hypothesis under its assumptions.

The registry record gives arm-level medians and ranges but does not report a comparative effect estimate, confidence interval, or test for the primary endpoint. It lists zero analyses. Therefore, the posted descriptive values cannot be used to claim a formal statistical difference. No treatment contrast should be inferred simply from the relative ordering of the medians.

In general, a multi-arm trial may use planned pairwise contrasts, an overall test across arms, or a hierarchical strategy. The choice affects the interpretation of results and can affect how multiplicity is addressed. The registry record does not report which, if any, strategy was planned or performed. It also does not report a statistical analysis plan, the method used to analyze the endpoint, or whether analyses were adjusted for other variables.

What can the group counts tell us?

The posted primary endpoint groups contain 35 participants for Daytrana, 71 for Vyvanse, 54 for Concerta, and 40 for Adderall. In general, the amount of information available for an estimate depends partly on the number of observations and on the variability and structure of the outcome. Group counts alone do not determine precision, and they cannot establish whether a study was adequately powered for a particular contrast.

The registry lists total enrollment as 228, but does not explain participant flow from enrollment to the posted endpoint groups. It does not state whether all randomized participants were analyzed, how missing endpoint values were handled, or which analysis population was used. Thus, the posted group counts should be treated as counts accompanying these endpoint summaries, not as a complete account of trial participation.

How should planning calculations be read?

In general, a sample-size calculation asks how many participants might be needed under specified assumptions to detect an effect with a chosen error rate and power. A planning illustration based on an observed effect is conditional on that effect estimate and the assumptions used in the calculation. It is not a description of the original trial’s design and is not observed power.

The registry record does not report the trial’s own sample-size justification. It also does not report a protocol-based rationale for the group sizes. Any planning illustration associated with the posted results should therefore be read as a hypothetical calculation under stated assumptions, not as evidence that the original study was designed around those assumptions. Because no formal primary-endpoint comparison is reported, the record does not establish a comparative effect estimate from which to make a definitive planning claim.

Why can over-dispersion matter?

In general, over-dispersion occurs when the variability in observed data exceeds what a statistical model assumes. If a model fails to account for that extra variability, standard errors may be too small and confidence intervals may be too narrow. Appropriate methods depend on the outcome type and study design; model choice should be based on the analysis plan and data structure.

The registry record does not describe a model for the primary endpoint or report a dispersion assessment. This general issue is not evidence that over-dispersion occurred in this trial. It illustrates why the endpoint’s data structure, analysis method, and uncertainty estimates matter for interpreting comparative results.

10. Multiplicity, Missing Data and Analysis Populations

The study lists one primary outcome and two secondary outcomes: the Clinical Global Impressions-Improvement (CGI-I) Scale and the Clinical Global Improvements-Acceptability (CGI-A) Scale. With multiple outcomes and four treatment arms, a complete analysis plan would clarify which comparisons were planned and how the chance of false-positive findings across multiple tests would be addressed. The registry record does not report a multiplicity strategy, a hierarchy of comparisons, or formal comparative analyses.

Missing-data methods and analysis populations are also not reported in the registry record. The posted primary endpoint groups contain 35 participants for Daytrana, 71 for Vyvanse, 54 for Concerta, and 40 for Adderall, while enrollment is listed as 228. The record does not explain participant disposition, missing endpoint values, exclusions, or whether the analysis followed an intention-to-treat or another population definition. It would be inappropriate to infer these details from the group counts alone.

The primary endpoint is measured at each participant’s last visit, which can occur at or before Week 6. The registry record does not describe how differences in visit timing or missing last-visit assessments were addressed. It also does not report sensitivity analyses that might show whether conclusions depend on assumptions about missing outcomes. These reporting gaps limit assessment of the robustness of the endpoint summaries and prevent a clear account of how all randomized participants contributed to the results.

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 2% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: tachycardia / arrhythmia (3.0%), hypersomnia (3.0%), assaultive/aggressive behavior (1.9%).

The registry reports serious adverse events assessed as related to study treatment for each safety group. The counts are 2 among 33 at risk for Daytrana, 0 among 67 for Vyvanse, 1 among 53 for Concerta, and 0 among 39 for Adderall. These are event counts and at-risk group sizes as listed in the registry; they do not establish comparative safety, particularly without a reported statistical analysis or additional context about exposure and follow-up.

The record also lists other adverse events affecting 16 among 33 at risk for Daytrana, 33 among 67 for Vyvanse, 18 among 53 for Concerta, and 15 among 39 for Adderall. The registry does not give further details in these summary counts about event severity, duration, or how events were distributed across participants.

The three serious-event terms with the highest listed percentages are tachycardia / arrhythmia at 3.0%, hypersomnia at 3.0%, and assaultive/aggressive behavior at 1.9%. The registry record does not specify the denominator or further calculation details for these percentages. Death counts are not reported in the registry record. These summaries describe recorded events; they do not support a treatment recommendation or a causal conclusion beyond the registry’s classification of related events.

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

The posted results establish that the registry contains descriptive primary-endpoint summaries for four medication arms. The median CGI-E values are 7 for Daytrana, 5 for Vyvanse, 6 for Concerta, and 6 for Adderall, with a full range from 1 to 16 in every arm. The endpoint combines therapeutic benefit with adverse impact and is assessed at each participant’s last visit, at or before Week 6.

These summaries do not establish that one treatment is more effective, more tolerable, or more acceptable than another. No formal comparative estimates, confidence intervals, or tests are reported in the registry record. The ordering of medians is not a substitute for a statistical comparison, and the arm-level scores do not show the number or proportion of responders.

The study’s stated aim was to compare newer and older psychostimulant medications across effectiveness, tolerability, and acceptability. The posted results offer descriptive information about the primary composite endpoint and adverse-event counts, but do not answer the comparative question with an inferential analysis. The registry record also does not report the analysis plan or explain the difference between enrollment and endpoint group counts.

In general, a nonsignificant result, if one were reported, would not prove equivalence; likewise, descriptive summaries alone cannot establish that treatments are equivalent or different. Equivalence or non-inferiority conclusions require a defined margin and an analysis designed for that purpose. Such a margin and analysis are not reported in the registry record.

The record is therefore most informative as a description of the trial design and the posted arm-level outcomes. It should not be read as a definitive ranking of the four medications or as evidence for a clinical decision. No medical advice follows from these registry summaries.

13. Why This Trial Matters Statistically

This trial illustrates a common interpretive challenge in clinical research: a randomized design can generate arm-level results without the registry record presenting the comparative analyses needed to answer the main treatment question. Randomization is an important design feature, but it does not by itself provide a treatment-effect estimate or quantify its precision. Those depend on the analysis and results that are reported.

The endpoint also illustrates how clinical outcomes can combine several dimensions. CGI-E is described as joining therapeutic benefit with adverse impact related to tolerability and acceptability. Such a composite may be intended to reflect an overall clinical judgment, but its meaning depends on the classification rules and on how results are summarized. Median scores do not reveal the responder counts or the separate contribution of benefit and adverse impact.

The four-arm design highlights the importance of stating which comparisons matter. A question framed as newer versus older treatments can involve multiple possible pairings or an overall comparison. Without a reported comparison plan, effect estimates, and multiplicity approach, readers cannot determine which contrasts were tested or how strongly the posted summaries address the study’s stated aim.

Finally, the registry record demonstrates why transparency about participant flow, missing data, analysis populations, and uncertainty is valuable. These details help readers understand how posted outcome groups relate to enrollment and how precisely results are estimated. Their absence does not negate the descriptive information that is reported, but it limits the conclusions that can be drawn from it.

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.

A key limitation of the registry record is the absence of formal comparative results for the primary endpoint. It reports medians and ranges by arm, but no comparative effect estimates, confidence intervals, or hypothesis tests. The registry also lists zero analyses. As a result, the record does not quantify uncertainty in between-arm differences or establish a statistical comparison.

The endpoint is assessed at each participant’s last visit, which can occur at or before Week 6. The record does not report the distribution of visit timing or how timing differences were handled. It also does not report component-level therapeutic benefit and adverse-impact results or the number of participants classified as responders.

Although enrollment is listed as 228, the primary endpoint counts are 35 for Daytrana, 71 for Vyvanse, 54 for Concerta, and 40 for Adderall. Participant flow, reasons endpoint counts differ from enrollment, missing-data handling, and analysis-population definitions are not reported in the registry record. The record also does not provide a sample-size justification, the statistical analysis plan, or a multiplicity strategy.

The trial is listed as unmasked. The record does not provide further detail about procedures that might limit the influence of unmasked treatment knowledge on ratings. It also does not report treatment exposure details or the context needed to interpret adverse-event follow-up. These are limits on what can be concluded from the registry entry, rather than evidence that a particular bias or problem occurred.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Comparing the Effectiveness of New Versus Older Treatments for Attention Deficit Hyperactivity Disorder (The NOTA Study) — NCT00889915.
• Registry record last updated 2013-09-30; status last verified 2013-07.
• Results first posted 2013-09-30.
• 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

The NOTA Study, registered as NCT00889915, is a completed, randomized, parallel-group, Phase 4 trial with no masking listed. It enrolled children and adolescents with ADHD and included four medication arms: Daytrana, Vyvanse, Concerta, and Adderall. Enrollment is listed as 228.

The primary endpoint is the dichotomized CGI-E score at each participant’s last visit, at or before Week 6. Posted arm-level medians are 7 for Daytrana, 5 for Vyvanse, 6 for Concerta, and 6 for Adderall; the full range is 1 to 16 for each arm. The registry also lists CGI-I and CGI-A as secondary outcomes.

Formal comparative analyses, uncertainty estimates, missing-data methods, and a statistical analysis plan are not reported in the registry record. The results support descriptive reading of the posted arm summaries, not a formal conclusion about differences among treatments.