Read beyond the headline.
Five questions to ask before a research finding becomes part of your treatment conversation.
Explore the studiesStart with the people. End with the uncertainty.
Open a study in the library and use this guide alongside it. Keep the original paper in reach: the summary helps you navigate the evidence, while the source supplies the details.
Who was studied?
Look for the condition, symptom duration, age range and inclusion criteria. Ask how closely this group resembles the person considering treatment.
Compared with what?
Find the comparison: sham treatment, exercise, another intervention or no comparison group. Improvement after treatment alone does not establish that the treatment caused it.
What changed—and when?
Read the measure and its units. Pain, function and imaging results answer different questions. A result immediately after treatment does not describe the longer term.
How uncertain is the result?
Look at the effect estimate, confidence interval and limitations together. Statistical significance alone does not tell you whether a difference matters to patients.
What about harms and effort?
Check adverse-event reporting, treatment sessions and follow-up. Missing harm information is a gap in the report, not proof that no harm occurred.
Try it with a real review
Open the review of shockwave therapy for chronic low back pain. Compare its outcomes with its limitations. The wiki notes analytical problems that reduce confidence; a favourable estimate should not erase those problems.
Read the review analysis →Keep the result in its own units
A pain-scale point, a percentage and a standardized mean difference are different kinds of numbers. Read the label before comparing results from different papers.
Further reading
Sources consulted on 22 October 2026.