Reviewer Guidelines

Guidance for fair and rigorous statistical review.

Reviewers assess methodological rigor, clarity, and relevance to health statistics practice.

Reviewers should assess methodological soundness, data quality, and clarity of conclusions.

Transparent reporting of data provenance and governance supports reproducibility and ethical compliance in health statistics.

Define statistical terminology clearly for multidisciplinary readers who apply methods in clinical settings.

Include brief rationale for study design choices to support reviewer understanding and methodological transparency.

Confidentiality and conflict of interest policies must be respected.

Well structured manuscripts accelerate peer review and help readers apply statistical insights to real world health decisions.

Summaries that connect statistical findings to health outcomes improve translation to policy and practice.

Use tables and figures to communicate effect sizes, uncertainty, and subgroup comparisons clearly.

Treat manuscripts as confidential and report conflicts of interest promptly.

Constructive feedback helps authors improve statistical reporting.

Provide uncertainty measures such as confidence intervals or credible intervals for key estimates and model outputs.

Report software versions and packages to support reproducibility across analytic environments.

If external validation is performed, describe population differences and implications for generalizability.

Provide constructive feedback and support comments with evidence where possible.

• Evaluate model assumptions and diagnostics
• Check data quality and transparency
• Assess interpretation and limitations

Clear statistical reporting improves the interpretability of health evidence for clinicians, policymakers, and research funders.

Explain how missing data were handled and why chosen strategies were appropriate for the study design.

When combining datasets, document linkage procedures and quality checks for matching accuracy.

Describe any model tuning or hyperparameter selection to support reproducibility in machine learning workflows.

We encourage authors to document assumptions and sensitivity analyses so conclusions remain robust across populations.

When presenting predictive models, report calibration, discrimination, and decision curve metrics where relevant.

Highlight ethical safeguards for patient privacy, especially when working with linked or sensitive datasets.