How we work

Editorial Standards

Readable health journalism with named human accountability at every gate.

This page is the editorial home for Rivium Journal’s sourcing, review, corrections and disclosure policies. The permanent framework is rendered by the dedicated theme template.

Our editorial principles

Evidence before claims

We distinguish established evidence, emerging research, lived experience and uncertainty.

Named human accountability

Every article names its writer. Clinical content also names the qualified professional who reviewed it.

Useful, not clinical

We explain research in plain language without diagnosing readers or replacing individual medical care.

Commerce disclosed

Rivium Labs funds this publication. Product interests never change the evidence standard or review outcome.

How an article is made

  1. Frame: an editor selects the article type, reader question, scope and safety boundaries.
  2. Research: the writer searches current guidelines, systematic reviews and primary studies. Long-form clinical explainers normally use at least 20 substantive sources.
  3. Draft: the writer turns that evidence into clear, useful prose and links material claims to their sources.
  4. Source check: every citation is opened, matched to the claim and checked for retractions, relevance and date.
  5. Editorial review: a separate editor challenges structure, unsupported certainty, readability and conflicts of interest.
  6. Medical review: an appropriately qualified reviewer checks clinical accuracy, risk, contraindications and the limits of the evidence.
  7. Publish and maintain: the named writer, reviewer and dates appear on the article. Material medical updates trigger another review.

Sources and freshness

We prioritize current clinical guidelines, systematic reviews, meta-analyses and original peer-reviewed research. Most research should come from the previous 15 years, while older landmark papers may be used when they define a still-current concept. A source count is a floor, not a substitute for quality.

AI-assisted work

Automation may help discover literature, organize notes, test links or identify inconsistencies. It is never credited as an author or reviewer. A named human opens every cited source, verifies that it supports the claim and remains accountable for the final text.

Corrections

We correct material errors promptly and add a dated note when a change affects the meaning, evidence or practical takeaway. Questions and correction requests can be sent to [email protected].