Grade the question, not the subject.
There is no single grade for sauna as a whole. A judgment concerns a defined outcome and the exposure being examined, in the populations actually represented. Evidence about a measured physiological response does not automatically receive the same interpretation as evidence about symptoms, daily functioning, or clinical events.
A grade also needs a stated claim. Evidence for an association can be reasonably consistent while evidence that assigning an intervention changes the outcome remains uncertain. The accompanying explanation must make that distinction explicit instead of allowing a letter to imply more than the sources establish.
The four editorial categories.
The categories are a reading aid. They are not a formal certification, a clinical guideline, or an interchangeable substitute for the assessment methods used by a particular systematic review. A page should state the reasons for its chosen category and the population to which its conclusion applies.
- A — Strong: consistent findings from multiple relevant randomized trials or a high-quality synthesis, with a conclusion supported by the quality and applicability of the evidence.
- B — Moderate: substantial prospective observational evidence with a coherent pattern, and/or smaller randomized studies that support a similar conclusion. The wording must still distinguish association from assigned-intervention effects.
- C — Emerging: limited trials, a single cohort, short-term physiology, uncertain consistency, or populations and exposures that differ materially from the reader's question.
- D — Insufficient or unsupported: the available material does not substantiate the claim, relies on anecdote or marketing, or includes evidence that weighs against the proposed conclusion.
Look beyond the number of papers.
The assessment considers design, precision, consistency, comparator choice, risk of bias, and relevance to the question. The result may be limited by a small sample, a brief follow-up, an outcome selected from many measurements, or missing information about methods. Those limitations should remain visible even when the direction of a result is favorable.
Several publications can describe the same people. A cohort paper, its subgroup analysis, and a later report using the same participants are useful contributions, but they are not independent replications. The evidence snapshot should distinguish publications from independent populations rather than presenting a larger count as stronger confirmation.
Exposure categories also matter. A synthesis combining different forms of heating may answer a broader question than a page about a particular room or device. Transferring a grade between them requires an explanation of what is shared, what differs, and what remains unknown.
A visible limit is part of the grade.
The current research records include abstract-only appraisals. These can identify reported findings and important restrictions, but they do not constitute complete risk-of-bias or funding assessments. A preliminary page judgment must disclose that scope and can change after fuller source examination or new evidence.
A grade is not a safety score, an individual prediction, or permission to start a routine. A person-specific decision may involve information outside the outcome being summarized. Read the safety context, source details, and medical disclaimer alongside the grade. Where the evidence cannot support a precise answer, the page should say so.