The research does not provide one clear answer
We have brought together 121 studies from 1984 to 2026 examining how co-sleeping is associated with children’s wellbeing, sleep and development. The largest longitudinal study, which followed the same children over many years, found no association between sleeping in their parents’ bed as babies and their wellbeing later in childhood. However, a new British study from 2026 found that children who first began sleeping in their parents’ bed between the ages of 1½ and 3½ were more likely to have an anxiety disorder at age 8. Studies can show whether two things are associated, but not whether one causes the other.
What this means
- Taken together, the research does not clearly indicate either harm or benefit for children who sleep with their parents.
- The latest study looked at older children: those who first began sleeping in their parents’ bed between the ages of 1½ and 3½ were more likely to have an anxiety disorder at age 8. The relationship may also run in the opposite direction: anxious children may be more likely to come into their parents’ bed.
- There are studies on both sides. The overview below includes them all, so you can see how the findings are distributed.
- Baby safety is a separate issue: research shows a higher risk of SIDS when babies under 3 months sleep in their parents’ bed. This is covered in the box further down.
Of the 121 studies in the research overview, 39 are from USA. They are marked with a dark ring in the chapter overview.
All studies at a glance
Each dot represents one study, positioned by year and by whether its findings support or oppose co-sleeping. “Neutral” means the study found nothing either for or against it, while “mixed” means it found some of both. The larger the dot, the more children took part. The darker the dot, the more robust the study. Select a dot to see what the study found.
Showing 121 studies: 34 support co-sleeping, 23 are neutral, 35 are mixed, and 29 oppose it. On a computer, you can also move between studies using the arrow keys.
Bilgin et al. 2024
There was no association between bed-sharing at 9 months and children's trajectories of internalising symptoms (anxiety, sadness) or externalising symptoms (restlessness, behaviour) from ages 3 to 11. The study
Source: Research databases PubMed, Europe PMC and OpenAlex; all studies rechecked on 28/10/2026, with new studies added and checked between 2/11 and 4/11/2026.
You may use the graphic under these terms. The studies the figures are based on belong to the researchers.
Selected studies from both sides
The largest and most widely discussed studies: those that find no harm and those that find a link to problems.
scientific papers reviewed: no convincing evidence either for or against.
From the study, in English: “We found general design limitations and a lack of convincing evidence in the literature, which preclude making strong generalizations.”Mileva-Seitz et al. 2017 · The Netherlands · robust study (A) · PubMed
Finds no harm
British children: sleeping in their parents' bed at 9 months was not associated with anxiety, sadness or restlessness between the ages of 3 and 11.
From the study, in English: “This finding suggests that bed-sharing at 9 months has no positive or negative influence on the development of internalizing and externalizing symptoms across childhood.”Bilgin et al. 2024 · United Kingdom · robust study (A) · PubMed
families: no difference in behaviour or cognitive ability at age 5 after accounting for family background.
From the study, in English: “There seem to be no negative associations between bed-sharing in toddlerhood and children's behavior and cognition at age 5 years.”Barajas et al. 2011 · USA · robust study (A) · PubMed
205 families followed: sleeping in their parents' bed as a young child was linked to neither problems nor benefits at age 18.
From the study, in English: “We suggest that such fears are without warrant if bedsharing is practiced safely as part of a complex of valued and related family practices.”Okami et al. 2002 · USA · moderately robust study (B) · PubMed
Finds a link to problems
OR (odds ratio) measures how much more common something is in one group than another. 1 means no difference, while a value above 1 means it is more common. An OR of 1.7 is not the same as 1.7 times as many children.
odds of an anxiety disorder at age 8 among children who first began sleeping in their parents' bed between the ages of 1½ and 3½: almost three times the odds. Sleeping there only as a baby, at 6 months, was not associated with anxiety. This says nothing about infant safety. 7,077 British children.
From the study, in English: “Findings suggest that not all bedsharing is associated with later adverse outcomes.”Bilgin et al. 2026 · United Kingdom · robust study (A) · PubMed
odds of a psychiatric diagnosis at age 6 among children who had slept in their parents' bed over a long period. 3,583 Brazilian children.
From the study, in English: “Bed-sharing is a common practice in our setting and is associated with impaired child mental health at the age of six years.”Santos et al. 2017 · Brazil · robust study (A) · PubMed
odds of behavioural problems at ages 10-13 among children who slept with parents, grandparents or others at ages 3-5. 1,274 Chinese children.
From the study, in English: “Early childhood co-sleeping predicts preadolescent internalizing and externalizing behavior after controlling for baseline behavior problems.”Chen et al. 2021 · China · moderately robust study (B) · PubMed
odds of asthma at age 6 among children who slept in their parents' bed at age 2. 6,160 Dutch children.
From the study, in English: “This study suggests that bed-sharing in toddlerhood is associated with an increased risk of asthma at later ages, and not vice versa.”Luijk et al. 2015 · The Netherlands · moderately robust study (B) · PubMed
You may use the graphic under these terms. The studies the figures are based on belong to the researchers.
Babies are a separate issue. Here, the research points more clearly in one direction: for breastfed babies under 3 months whose parents did not smoke, the odds of cot death were 5.1 times as high in their parents' bed as in their own bed in their parents' room (Carpenter et al. 2013). For babies under 12 weeks, a combined analysis of several studies found that the odds of cot death were 10.37 times as high in their parents' bed (Vennemann et al. 2012). As this is a separate issue, the 50 studies on baby safety and breastfeeding are not included in the overview above. See official guidance.
How we calculated it
Selection: Our collection includes 175 sources relating to children. The overview covers the 121 studies on child development, attachment, sleep, long-term outcomes, sleep training, culture and school-age children. It excludes 50 studies on baby safety and breastfeeding and 4 sources that are not studies.
For, against, mixed or neutral: FamBed's assessment of which side the study's main finding supports. A, B and C reflect our assessment of how robust the study is and how many participants it included, with A being the most robust.
OR (odds ratio): A measure of how much more common something is in one group than another. 1 means no difference. An odds ratio of 1.7 does not mean 1.7 times as many children.
Verification: Every study was opened and checked again in PubMed, Europe PMC or OpenAlex on 28/10/2026, and 56 descriptions were corrected, mostly because they had been worded too strongly. Studies added in November 2026 were opened and checked as they were added, between 2/11 and 4/11/2026. For some of them, we have only read the abstract.
The 2026 study: Bilgin and Parsons followed 7.077 British children and asked where they slept at 6 months, 1½ years and 3½ years old. We have only been able to read the abstract, so we have not checked the group sizes or participant attrition. An odds ratio of 2,83 does not mean 2,83 times as many children.