The conversation around co-sleeping in the US is unusually fraught. The American Academy of Pediatrics recommends against bed-sharing under any circumstances; the international evidence (UNICEF UK, Australia, much of Europe) is considerably more nuanced. Most families end up bed-sharing at least occasionally — somewhere between 40–60% in US surveys — and the absolutism of the official US position has, paradoxically, contributed to unsafe bed-sharing because it discourages parents from asking how to do it safely.
Here’s the version most pediatric sleep experts actually agree on in private practice, drawn from the international consensus.
When bed-sharing is genuinely unsafe
The following are absolute contraindications. Do not bed-share under any of these conditions:
- The baby was born prematurely (under 37 weeks).
- The baby is under 4 months old.
- Either parent smokes (including during pregnancy).
- Either parent has consumed alcohol or sedating medications.
- The baby will be on a sofa, recliner, or armchair (these are dramatically higher risk than a bed).
- The bed has soft bedding the baby could become entangled in (duvets, pillows near the baby, sheepskins).
- The baby is bottle-fed rather than breastfed (the breastfeeding mother’s positioning behavior — the protective “C” curve — appears to be load-bearing in the safety data).
If any of those apply, bed-sharing carries a meaningfully elevated SIDS/SUID risk and should not happen.
When bed-sharing appears to be safe
Under all of the following conditions:
- Baby is 4+ months old, term, healthy.
- Mother is breastfeeding.
- Neither parent smokes, drinks, or uses sedating medications.
- The bed is firm, the bedding is minimal, and there are no gaps between mattress and frame.
- The baby sleeps on its back, beside (not between) the parents.
- No other children or pets in the bed.
Under these conditions, the bed-sharing SIDS/SUID risk is approximately equivalent to room-sharing in a separate cot — which is the safest configuration.
The room-sharing middle ground
For families who want closeness without bed-sharing, the AAP’s first-line recommendation (and the strongest evidence) is room-sharing with the baby in a separate sleep surface for the first 6–12 months. This reduces SIDS risk by roughly 50% compared to separate-room sleep, without the bed-sharing risk profile.
A bedside bassinet that attaches to the parent’s bed (Halo, Snoo, etc.) is the practical implementation. The baby is at arm’s length but on a separate sleep surface.
A note on the cultural context
Bed-sharing is the global default. Most of human history and most contemporary non-Western cultures bed-share. The Western nuclear-bedroom is unusual. The point isn’t that bed-sharing is risk-free — it isn’t — but that the US-specific moral framing around it isn’t a useful guide to safety.
The honest summary: room-sharing in a separate cot is the safest documented configuration through ~6 months. Bed-sharing under the strict conditions above is approximately as safe. Bed-sharing under any of the contraindicated conditions is genuinely dangerous. The official US position collapses all three into “never,” which is why the conversation is so unhelpful.
— Sources: Ball, H., Durham Infancy & Sleep Centre research; UNICEF UK Baby Friendly Initiative co-sleeping guidance (2019); McKenna, J., Mother-Baby Behavioral Sleep Lab (Notre Dame).