Effects of SGLT2 inhibition on incident heart failure in carriers of cardiomyopathy-associated genetic variants.

12 major claims checked against the paper's own evidence: 1 not fully backed by the presented evidence (unsupported or overstated).

Recomputed 17 tests: 17 consistent, 0 inconsistent, 17 via agent-written checks.

2 integrity concerns flagged (0 high).

8 copyedit issues flagged: mostly consistency, typo, grammar.

Checked 30 references: 5 verified — 25 not checked.

4 data/code links checked; 4 live.

Ready after minor-to-moderate revision: the study design and data/code sharing gaps must be addressed (especially the data availability statement and code sharing), and the claims should be tempered to not overstate the clinical recommendation. The copyedit issues and statistical reporting gaps are minor. The paper's core findings are plausible and well-supported.

1. 1.
   HIGHdata code

   Replace the blanket data denial in the Data Availability section with a description of a managed-access pathway (e.g., contact TIMI Study Group, access via Vivli or YODA platform).

   A data-driven paper with no data access mechanism will be flagged by many journals as non-compliant with reproducibility standards.
2. 2.
   HIGHreporting

   Temper the abstract claim that 'SGLT2 inhibitor treatment should be started early to prevent HF' — revise to state that the findings suggest a potential benefit that requires prospective confirmation, consistent with the Discussion's call for a dedicated trial.

   The claim audit flagged this as overstated; a direct clinical recommendation from a post-hoc analysis without prospective trial support is a common reason for reviewer concern and potential editorial rejection.
3. 3.
   HIGHother

   Add explicit inclusion and exclusion criteria for the genetic substudy in the Methods, including details on why 499 of 13,184 sequenced samples were excluded after QC.

   The vague 'consented and had WES available' description leaves uncertainty about selection bias; QC exclusion details are needed for reproducibility.
4. 4.
   HIGHdata code

   Deposit custom analysis code (e.g., variant curation, statistical models) in a public repository (e.g., GitHub, Zenodo) with a persistent DOI, and reference it in the Data Availability section.

   Code sharing is increasingly expected for computational reproducibility; its absence was flagged by all three reviewers.
5. 5.
   HIGHreporting

   Reference a reporting guideline (e.g., STROBE for observational genetic studies, or STREGA) in the Methods or Reporting Summary.

   A formal guideline reference is best practice and was noted as missing by all three reviewers, though it is a minor gap.
6. 6.
   HIGHother

   Add a CONSORT-like flow diagram for the genetic substudy showing the derivation of the analytical cohort (e.g., 17,160 → 12,685).

   Clarifying how many participants were included and at which stage they were lost improves transparency and addresses reviewer concerns about inclusion criteria.
7. 7.
   HIGHother

   Specify the manufacturer, formulation, and dose of dapagliflozin used in DECLARE-TIMI 58 in the Methods (e.g., 'dapagliflozin 10 mg once daily, AstraZeneca').

   Investigational product specification is a standard reproducibility element; its omission was flagged by two of three reviewers.
8. 8.
   MEDIUMreporting

   Acknowledge the post-hoc, non-prespecified nature of this analysis explicitly in the Abstract (e.g., 'In this exploratory, non-prespecified analysis...').

   Failing to flag the exploratory nature at the highest level of the paper can mislead readers and was noted by two reviewers.
9. 9.
   MEDIUMstatistics

   Report exact p-values for the secondary endpoints currently described only as 'not statistically different' (e.g., CV death, all-cause mortality).

   Providing exact p-values (even non-significant ones) follows best reporting practices and addresses Reviewer 1's concern.
10. 10.
    MEDIUMother

    Add a statement that no independent replication was attempted or note that independent validation is needed, and consider a post-hoc power calculation for the interaction test.

    Acknowledging the lack of replication and quantifying the power of the interaction test would strengthen the design section and manage reader expectations.
11. 11.
    LOWcopyedit

    Fix subject-verb agreement in the Abstract: change 'These results needs to be confirmed' to 'These results need to be confirmed'.

    Grammar error flagged by copyedit pass.
12. 12.
    LOWcopyedit

    Define 'tv' (truncating variant) at first use in the Discussion; add a parentheses definition.

    Acronym not defined earlier in the paper, flagged by copyedit pass.
13. 13.
    LOWcopyedit

    Add a missing period at the end of the sentence about genetic testing in Discussion paragraph 2.

    Minor punctuation error flagged by copyedit pass.
14. 14.
    LOWcopyedit

    Ensure figure reference in Results is complete (replace 'Fig. ' with 'Fig. 1' or the appropriate figure label).

    Incomplete reference flagged by copyedit pass.
15. 15.
    LOWcopyedit

    Standardize 's.d.' vs 'SD' across tables (Table 1 vs Extended Data Table 1) and hyphenate 'cardiomyopathy-associated' consistently in the Abstract.

    Style inconsistencies flagged by copyedit pass; minor but improves professional presentation.