External trigeminal nerve stimulation in youth with ADHD: a randomized, sham-controlled, phase 2b trial.

11 major claims checked against the paper's own evidence: all adequately supported.

Recomputed 2 tests: 2 consistent, 0 inconsistent, 2 via agent-written checks.

1 integrity concern flagged (0 high).

5 copyedit issues flagged: mostly consistency, clarity, punctuation.

Checked 63 references: 3 verified — 60 not checked.

Ready to submit after minor edits. The two most important fixes are: (1) correct the percentage error in Table 1 for left/mixed handedness (currently 50.6%, should be 25.3%), and (2) add a brief statement verifying the linear mixed model assumptions (e.g., residual plots or a reference justifying robustness). Addressing the other copyedit issues and reviewer suggestions will polish the manuscript for a clean submission.

1. 1.
   HIGHcopyedit

   Correct the percentage for 'Left/mixed handed' in the overall column of Table 1: 38 out of 150 participants is 25.3%, not 50.6%. Verify and correct the actual numbers or the percentage.

   This is a demonstrable arithmetic error that could lead to reviewer scrutiny about data quality.
2. 2.
   HIGHstatistics

   Add a brief statement in the Methods (Statistical analysis) that the assumptions of the linear mixed model (e.g., normality of residuals, homoscedasticity) were checked via residual plots or similar diagnostics, or cite a reference justifying their robustness for the sample size.

   Explicitly verifying model assumptions strengthens the statistical reporting and preempts reviewer concerns.
3. 3.
   MEDIUMreporting

   Include the pre-specified statistical analysis plan (SAP) as a supplementary file or explicitly reference the protocol supplement where it can be found.

   Improves reproducibility and transparency, as one reviewer suggested.
4. 4.
   MEDIUMreporting

   In the Discussion, explicitly note that the positive finding on the MEWS (commission errors) should be considered exploratory due to the lack of correction for multiple comparisons, even if pre-specified.

   Prevents over-interpretation of a single significant secondary outcome amidst null primary results.
5. 5.
   MEDIUMcopyedit

   Fix the formatting error in Table 3 for the commission errors row: change '0.95 1 (0.80, 1.13)' to '0.95 (0.80, 1.13)' or '0.951 (0.80, 1.13)' as appropriate.

   The extra space and digit are a typographical error that should be corrected for clarity.
6. 6.
   MEDIUMreporting

   Clarify whether the decision to not analyze teacher-rated secondary outcomes due to high missing data (80%) was a pre-specified criterion or a post-hoc decision, and state this in the Results.

   Transparency about deviations from the analysis plan is important for integrity.
7. 7.
   MEDIUMreporting

   Add exact p-values for all reported secondary outcomes in the text (currently only in Table 3 and for MEWS in text), as suggested by one reviewer.

   Consistency in reporting p-values improves completeness and reader trust.
8. 8.
   MEDIUMreporting

   Report the blinding assessment data for the 6-month follow-up if it was collected, to fully characterize blinding integrity over the longer term.

   Provides a more complete picture of blinding success for the entire trial duration.
9. 9.
   LOWreporting

   Consider adding a brief note in the Methods on how missing data for the primary outcome were handled in the mixed model (e.g., under missing-at-random assumption).

   Clarifies the missing data handling approach, which is good practice for clinical trials.
10. 10.
    LOWreporting

    Add a reference for the anticipated effect size or the formula used in the sample size calculation.

    Minor addition to improve completeness of the power analysis justification.