STROBE Checklist Checker
Review observational-study reporting with STROBE 2007. Choose cohort, case-control or cross-sectional and inspect evidence for each applicable item.
2007 · STROBE Statement combined checklist (2007) · Paraphrased summaries · Verified · 22 numbered items
AGREE II is disabled: appraisal instrument; commercial use restricted.
Manuscript
Up to 150,000 characters. Choose Excerpt for a partial manuscript; missing reporting then stays unclear.
Drag a file here, or
PDF with selectable text, TXT or Markdown; maximum 10 MB. DOCX not supported. Scanned PDFs need OCR first.
One free manuscript. Rechecking identical text against another guideline is free.
OpenAI only: manuscript text is sent to our backend and OpenAI; not stored by this application. OpenAI retention policies apply. Your manuscript and report stay in this tab until you close or reload it; copy or download first.
Continue from reporting review to research
Use AnswerThis to explore literature relevant to the gaps identified in your reporting review. Copy or download this reporting assessment first.
How to use STROBE Checklist Checker
- Select the guideline and edition that match your report; choose a study-design variant for STROBE.
- Paste the manuscript or upload a readable file. Choose Full manuscript only when the entire relevant report is supplied; otherwise select Excerpt.
- Check reporting, then inspect each requirement, verdict, quotation and location against the official checklist.
- Copy or download the text report or CSV. Resolve missing information with the author team before preparing a journal submission.
What to check
- Confirm the guideline edition and document stage.
- Check all applicable parts of an item, not just a matching phrase.
- Compare quotations and locations with tables, supplements and the original manuscript.
- Treat the findings as an AI-assisted draft requiring author confirmation, never certification.
Choose the observational design first
STROBE 2007 addresses reporting of cohort, case-control and cross-sectional studies. The original checklist has 22 numbered items, with lettered parts that differ by design. This page requires you to choose the appropriate variant before assessment. The choice changes participant-selection, statistical-method and outcome requirements; it is not just a label added to a generic result. If the manuscript combines designs, identify the part being assessed and review any additional reporting requirements with the author team.
A cohort report explains participant selection and follow-up, reports follow-up duration and presents events or outcome summaries over time. A case-control report explains case ascertainment and control selection, including the rationale, then describes exposure distributions. A cross-sectional report describes participant sampling and outcome summaries at the relevant observation period. Matched cohort and case-control studies have matching requirements that do not belong in the cross-sectional form. The tool preserves these distinctions rather than declaring every design-specific item missing.
STROBE is not a quality-assessment instrument
A transparent account of a weak design may report its methods well. Conversely, a carefully conducted study may be poorly described. STROBE helps readers locate and understand what was done; it does not turn reporting completeness into scientific validity. This checker therefore gives item verdicts and evidence, not a quality score, risk-of-bias classification or compliance percentage. Use a separate appropriate appraisal method if your task is to evaluate bias rather than reporting.
Bias, confounding and missing observations still matter to the reporting review. The manuscript should explain how relevant variables were defined and measured, which possible biases were addressed and what analytical decisions were made. Adjusted estimates need their chosen confounders and reasons, not merely the phrase “multivariable model.” When measurements differ between comparison groups, those differences should be visible. Review the quotations and missing-detail notes with someone familiar with the study, because the model cannot establish measurement validity simply by finding a methods sentence.
Use design-specific findings in your revision
Begin by checking participant numbers at each stage and the reasons for nonparticipation. Then examine descriptive data, missing-value counts and outcome or exposure summaries. The original guidance asks for group-specific information where appropriate, including cases and controls and relevant exposed and unexposed groups. A pooled total may be real evidence without satisfying that requirement. The checker should retain the quotation and identify what remains unresolved rather than invent a breakdown.
Interpretation should relate findings to the objectives, limitations, multiple analyses and other evidence. Discussion of limitations includes the likely direction and size of potential bias, not only a list of generic weaknesses. Generalisability requires context about where the results might apply. If you submit only results or discussion, use Excerpt so unavailable methods remain unclear. Check the loaded text for broken tables and misplaced headings before running. Keep the exported report alongside an author-reviewed official checklist; neither an automated badge nor a correctly located quote replaces that final review.
Choose a reporting checklist
MOOSE 2000 is unavailable while we verify its item list against an accessible authoritative source.
- Reporting Checklist Checker — Review manuscript reporting against an editioned guideline. Get item-by-item findings, verified quotations, locations and missing details for author review.
Frequently asked questions
Which STROBE design should I choose?
Choose cohort for follow-up from exposure to outcomes, case-control for sampling by outcome status, or cross-sectional for measurements at a defined time. The variant changes the relevant requirements.
Where should matching and follow-up appear?
For a matched case-control study describe matching criteria and controls per case. For a cohort describe follow-up duration and losses. These details cannot be supplied by choosing the variant alone.
Does reporting confounders establish low risk of bias?
No. Describe the variables and adjustment methods, then report both estimates and uncertainty. Use an appropriate appraisal instrument to judge whether confounding was adequately addressed.
Can I upload unpublished research?
Manuscript text is sent to our backend and OpenAI for assessment. This application does not store manuscript text; OpenAI retention policies apply. Upload only material you are authorized to share with those services. The current manuscript and report remain in this tab until you reload or close it.
Does this guarantee compliance, and why is AGREE II disabled?
No. This is an AI-assisted reporting review, not certification or a study-quality assessment. AGREE II is a guideline appraisal instrument with 23 items, six domains and 1–7 ratings, requiring at least two appraisers. Commercial use is restricted, so its dropdown option is disabled and no substitute or score is generated.
How many manuscripts can I check for free?
This tool includes 1 successful free run per browser. Failed requests and copying or downloading an existing result do not use another run. The unit is one manuscript across all five pages. Rechecking the same normalized text against another checklist is free, subject to the service’s daily request budget; edited manuscript text is a new manuscript.
What happens when I sign up?
Signup opens the AnswerThis app. It does not unlock more runs on this page or transfer your draft. Copy or download your result first.