Our approach to evidence

Evidence in baby sleep research is complicated

Much of the research on baby sleep involves small sample sizes, parent-reported data, varying definitions of key terms (like "sleeping through the night"), and short follow-up periods. Randomised controlled trials are relatively rare in this field. Longitudinal studies are expensive and difficult to conduct.

This means that confident claims — "this works", "this is harmful", "all babies should do X by age Y" — are often more certain than the underlying evidence justifies. We try to reflect that honestly.

What we do

• Cite primary research where possible. We link to peer-reviewed studies rather than secondary summaries where we can. References are listed at the end of each article.
• Acknowledge the limits of evidence. When studies are small, poorly designed, or conflicting, we say so rather than cherry-picking the result we prefer.
• Distinguish between evidence and consensus. Some things are well-established; others are expert opinion or practical guidance without strong trial evidence. We try to signal which is which.
• Avoid catastrophising. We don't amplify scary findings beyond what the evidence actually supports, and we try not to create anxiety where the data doesn't warrant it.

What we don't do

• Present one study as definitive proof
• Cherry-pick research that supports a predetermined conclusion
• Cite low-quality or retracted research
• Use medical-sounding language to make opinion sound like fact
• Predict outcomes ("if you do X, your baby will Y")

On safe sleep

Safe sleep guidance — in particular around reducing the risk of SIDS — is based on epidemiological evidence and represents current clinical consensus. We follow and reflect guidance from organisations such as the Lullaby Trust (UK) and the American Academy of Pediatrics (AAP).

Safe sleep guidance does evolve as evidence develops. We aim to keep articles up to date, but always check current guidance from official health bodies for the most current recommendations.

Corrections and updates

If we get something wrong — or if evidence changes significantly — we update articles and note the revision date. Getting it right matters more than being consistent with what we originally said.