Provision of knee bracing for knee osteoarthritis (PROP OA): multicentre, parallel group, superiority, statistician blinded, randomised controlled trial.

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

Recomputed 5 tests: 5 consistent, 0 inconsistent, 5 via agent-written checks.

4 copyedit issues flagged: mostly consistency, other, clarity.

Checked 35 references: 30 verified — 5 not checked.

2 data/code links checked; 2 live.

The manuscript is ready for submission after addressing a short list of pre-submission actions: moving the analysis code to a persistent repository, adding a CONSORT reporting guideline reference, adding a dedicated limitations section, and making minor copyedit fixes. No critical methodological problems were identified.

1. 1.
   HIGHdata code

   Upload the analysis code to a version-controlled public repository (e.g., GitHub, Zenodo, or the same institutional repository that hosts the data) and update the Data Availability Statement to include the repository link and persistent identifier.

   Code in supplementary files is not independently citable or version-controlled; a persistent repository ensures long-term access and reproducibility.
2. 2.
   HIGHreporting

   Add an explicit statement in the Methods section that the trial report follows CONSORT guidelines, and provide the completed CONSORT checklist as a supplementary file.

   Journals and readers expect reference to the applicable reporting guideline; its absence may trigger a reviewer request and delays.
3. 3.
   HIGHreporting

   Add a dedicated 'Strengths and Limitations' subsection to the Discussion that transparently addresses potential biases (e.g., open-label design, contamination in the AIE group who received motivational interviewing), generalisability, and the impact of COVID-19 adaptations.

   A dedicated limitations section is a standard element of a complete clinical trial report and helps readers interpret the findings appropriately.
4. 4.
   MEDIUMreporting

   Report exact p-values (e.g., p=0.006) alongside the 95% confidence intervals for all primary and key secondary outcomes in the Results and Tables 4–5.

   While confidence intervals are sufficient for inference, some meta-analysts and funders require exact p-values; their absence may be noted during peer review.
5. 5.
   MEDIUMreporting

   Add the Bristol Online Tool or another validated adherence measurement instrument reference to complement the description of the adherence intervention (motivational interviewing).

   This would strengthen the methodology description and allow others to replicate the adherence component.
6. 6.
   MEDIUMreporting

   In the Abstract, consider adding a brief statement of the study's limitations to balance the conclusions.

   Abstracts are frequently read in isolation; a limitations sentence helps prevent overinterpretation.
7. 7.
   MEDIUMcopyedit

   In the Methods: Randomisation and masking section, rephrase 'Although masking participants or physiotherapists to treatment allocation was not possible ...' to 'Masking of participants and physiotherapists was not feasible due to the nature of the intervention; however, the trial statistician was blinded to treatment allocation.'

   Improves clarity and flow of a key methodological statement.
8. 8.
   LOWcopyedit

   Standardise the formatting of confidence intervals throughout: use en dash or 'to' consistently (e.g., '0.96 to 5.82' in text, ensure tables match).

   Minor formatting consistency improves readability and avoids copyeditor queries.
9. 9.
   LOWcopyedit

   In Table 2 and Table 4, standardise the display of subscale names (e.g., remove extra colon in 'KOOS-4: (0-100):').

   Minor formatting issue that can be caught with a final proofread.
10. 10.
    LOWcopyedit

    In Table 1 footnote, clarify whether the reported intraclass correlation coefficient (0.94, 95% CI 0.90 to 0.96) refers to intra-rater or inter-rater reliability, as the model described ('two-way random effects model with absolute agreement') is typically used for inter-rater reliability.

    Resolves a potential confusion for readers about the type of reliability assessed.
11. 11.
    LOWreporting

    Consider providing the brace adherence data as a supplementary dataset with a DOI to enhance the reproducibility of the complier average causal effect (CACE) analysis.

    Would strengthen the transparency and reusability of the trial data.