All readings

Learn

Reading a research claim

A short path from a broad promise to a question you can check.

Find the comparison

NCCIH’s reading guide explains that controlled trials compare outcomes between groups receiving different interventions or conditions. The specific comparison is part of the study design. 1

When you encounter “research proves,” look for a link to the cited work. You can note what happened in each group and which outcome was measured. If you cannot find that information, leave it as an open question.

Check the certainty

Cochrane considers factors including bias, inconsistent findings, indirect evidence, imprecision, and publication bias when assessing certainty across a body of evidence. It does not reduce certainty to the number of studies. 2

You can look for the review’s explanation of its limits and how it reached its conclusion. Note the words it uses to describe certainty without replacing them with a stronger promise.

Compare the claim and the paper

Try two columns: in one, describe the advertisement’s promise in your own words; in the other, summarize what you could verify in the publication. Leave room for differences and for anything you do not yet understand.

In a fictional example, an advertisement promises to “change your life,” while the paper describes a questionnaire completed one week later. Your note can preserve that difference without adding conclusions about other outcomes.

Keep a provisional reading

You can record the title, link, date, and a question to revisit. You do not need to evaluate an entire field in one reading. Keep clear which part you checked and which part remains a claim you encountered elsewhere.

Sources

  1. NCCIH · How to make sense of a scientific journal article: study design
  2. Cochrane Handbook · Grading the certainty of evidence

Sources checked November 18, 2026.

Open the free map