Medical guidance often appears to lag behind research that has already been widely reported. The delay is a feature of how guidelines are constructed, and each stage in the process adds time deliberately.
A guideline rests on a body of evidence
Guidance is written from systematic reviews that gather all relevant studies on a question, assess their quality and combine their findings.
A single study, however striking, is one input among many, and results that do not reproduce are common enough that caution is warranted.
Waiting for replication is what prevents guidance from oscillating with each publication, at the cost of responding slowly to genuine advances.
Evidence is graded before it is used
Study designs are ranked by how well they control for confounding, with randomised trials generally weighted above observational work for questions about treatment effect.
Reviewers also assess consistency across studies, how directly the population studied matches the population being advised, and the precision of the estimates.
Recommendations carry a strength rating derived from this assessment, which is why some guidance is stated firmly and other guidance is expressed as an option.
Committees weigh harm alongside benefit
A change in guidance alters what happens to large numbers of people, including those who would have done well under the previous approach.
Committees therefore consider the harms of the new recommendation, the burden it places on patients, and what happens if it is applied imperfectly in practice.
Where benefit is modest and harm is plausible, the balance can favour leaving guidance unchanged until the evidence is stronger.
Thresholds and definitions have wide consequences
Much guidance turns on where a boundary is drawn, and moving that boundary reclassifies a large group of people at once.
Such a change alters demand for testing, follow-up and treatment across an entire health system, and those consequences are assessed before the change is made.
This is why definitional changes attract more scrutiny and take longer than adjustments that affect a narrower group.
Implementation adds further delay
A published guideline still has to reach clinicians, be built into local protocols and appear in the software that supports decisions at the point of care.
Different countries adapt international guidance to local resources and disease patterns, so the same evidence produces somewhat different recommendations in different systems.
Guidance is general by design and does not replace an assessment of an individual by a qualified clinician, which is where any specific decision belongs.