Gut Response to Reviewers: Gastroenterology Revision Guide

Build one ledger that includes diagnosis and criteria, disease location and activity, prior surgery, medication and antibiotic exposure, diet or fasting context, sample compartment, collection timing, processing, endoscopy and histology, exclusions, missingness, follow-up, and analysis denominator.

Gut asks original research to explain what is known, what is new, and how findings might affect clinical practice. Treat that summary as a claim-control surface. A revision that weakens an endpoint, removes a classifier feature, or limits a mechanism must propagate into the abstract, significance box, visual abstract where required, Discussion, and conclusion.

Do not translate an observational association into a treatment recommendation. Name the patient group, timing, comparator, measurable decision, and validation still required.

In our pre-submission review work with Gut manuscripts, we inspect cohort definitions, disease phenotypes, stool and tissue provenance, immune and molecular assays, microbiome pipelines, endoscopy and histology, medication and diet exposures, clinical endpoints, missingness, statistics, figures, reporting checklists, abstracts, significance summaries, and visual artifacts. We map every reviewer comment to the compartment and inference it tests. These are qualitative Manusights patterns, not confidential BMJ editorial data.

Pattern 1: stool association becomes mucosal mechanism

In Gut revisions, a stool taxon or metabolite associates with disease activity and the paper describes a mucosal immune mechanism. Stool, mucosa, and blood represent different compartments. We require paired evidence, a perturbation model, or explicitly separated claims, and we trace collection and processing differences before interpreting biology.

Pattern 2: medication adjustment hides indication bias

Treatment exposure is added as one covariate, although treatment choice depends on severity, phenotype, prior response, and calendar time. For Gut manuscripts, we build an exposure timeline and ask whether restriction, propensity methods, target-trial framing, negative controls, or a narrower association claim is appropriate.

Pattern 3: endotypes are discovered and validated in the same information

A molecular or microbial cluster is tuned and evaluated in one cohort, or the split leaks patient, center, batch, or time information. We separate discovery from locked validation, preserve the deployment unit, and report instability instead of selecting only the favorable clustering resolution.

Pattern 4: clinical meaning outruns endpoint design

A biomarker association, surrogate change, or exploratory subgroup becomes a practice claim. We reconstruct the endpoint hierarchy, effect estimate, uncertainty, missingness, harms, and competing events, then revise the significance box and conclusion to the decision actually supported.

The distinctive Gut information gain is compartment and inference alignment: phenotype, sample, mechanism, exposure, endpoint, and patient claim must describe the same gastroenterology result.

Check the Gut response and revised evidence together.

One reviewer may ask for tissue mechanism while another prioritizes external clinical validation. These are not interchangeable. Tell the editor which evidence controls the primary conclusion, add the feasible discriminating test, and divide mechanistic, predictive, and treatment-effect statements. Do not let a paired biopsy subset silently stand in for an independent patient cohort.

Serious risks include disease-definition drift, compartment overreach, unhandled medication or diet confounding, compositional artifacts, discovery-validation leakage, inconsistent endoscopy or histology, endpoint switching, informative missingness, and practice language beyond the study design.

Most rejection-on-revision risk comes from a response that adds analyses while leaving the controlling compartment, confounding, validation, or endpoint problem unresolved.

Submit if: every comment is answered and located; phenotype and sampling ledgers reconcile; compartment claims match evidence; medication, diet, and procedure timing are explicit; microbiome methods respect compositionality; endpoints and denominators are stable; and practice language stays within the design.

Think twice if: more relative-abundance plots replace validation, stool is treated as mucosa, one adjustment variable is said to remove confounding, exploratory endpoints lead the abstract, or the response promises future mechanistic work.

We reviewed current Gut author and journal materials, BMJ reviewer and transfer guidance, reporting expectations, and response-writing evidence. Public sources establish the author framework. The decision letter and current ScholarOne task list control a live revision. The compartment-to-outcome audit is Manusights analysis.

1. Put editor priorities before reviewer sections.
2. Answer every comment with action, result, claim impact, and location.
3. Reconcile diagnosis, disease location, activity, endotype, and denominator.
4. Separate stool, mucosa, blood, liver, bile, tumor, and model claims.
5. Record medication, antibiotic, diet, procedure, and sampling timing.
6. Audit compositionality, batches, negative controls, leakage, and validation.
7. Reconcile endoscopy, histology, biomarker, clinical endpoints, and missingness.
8. Use the patient as the unit and model repeated measures appropriately.
9. Update the abstract, significance box, visual abstract, and Discussion.
10. Follow the decision letter and current reporting checklist.

Read final Search Console data after 14 complete days. At 21 days, keep, revise, consolidate, or stop based on exact-query ownership, impressions, clicks, query fit, and qualified starts. The source cluster had 3,404 impressions and one preview start; exact-query demand remains unproven.