We use cookies to ensure that we give you the best experience on our website. You can change your cookie settings at any time. Otherwise, we'll assume you're OK to continue.

Durham University

Research & business

View Research Directory

Parent-infant bed-sharing: Who, why and how?

A research project of the Department of Anthropology.


This study investigated parental strategies for the night-time care of infants, with particular focus upon parent-infant bed-sharing. Using sleep logs, face-to-face interviews and small-group discussions we generated baseline information on the general night-time practises of a random sample of 253 families with infants in the North Tees Health District. We found that parent-infant bed-sharing is commonly practised. In 70% of the families studied, the infant bed-shared with one or both of its parents at least once by 3 months of age. Sixty-seven percent of babies slept in the same bed as their parents at least once in their 1st month (mean age = 17 days), and typically they slept with both parents (triadic bed-sharing). Very few neonates slept in a separate room from their parents, but by the age of 3 months an increasing number of non-bed-sharing babies were being moved out of their parents' room.

Bed-sharing mothers in our sample were on average 2 years older than non-bed-sharing mothers, and a greater proportion of them had post-secondary or graduate educational qualifications. We found no associations between bed-sharing and parity, marital status, socio-economic class, income level, paternal education, home ownership, housing tenure, ethnicity, or ward of residence. Feeding method was the primary variable distinguishing families who bed-share from those who don't. Two-thirds of all breast-feeders were bed-sharers, while almost three-quarters (73%) of formula fed babies never bed-shared with their parents between birth and 4 months of age.

For bed-sharers as a whole, there were no associations with maternal variables such as mother's return to work, smoking, alcohol consumption, and parental experience. Nor were there associations between bed-sharing and care-giving variables such as infant sleeping position, and number of layers of clothing and covers. Families considered to be 'high-risk' for SIDS were slightly less likely to bed-share than low-risk families (not significant), reflecting the lower proportion of breast-feeders in the high-risk sub-group. High-risk group parents, and particularly lone mothers, engaged in more practises designed to minimize their infant's disruption during the night (e.g. dummy use, different room) than the low risk group. Approximately half of all smoking parents, regardless of SIDS risk group, bed-shared with their infants. Our data do not suggest an association between bed-sharing and variables indicative of socio-economic deprivation or chaotic parenting in the UK (in contrast to some US studies).

The most frequent factor introducing risk into bed-sharing arrangements in our sample was paternal alcohol consumption. Of 99 regularly bed-sharing fathers 16 consumed 5 or more units of alcohol on at least one night when sleep logs were recorded, and more than half of the remainder consumed lesser amounts several times a week. The heaviest drinking bed-sharing fathers are generally characterised as being of middle income, socio-economic classes III and IV, with little post-16 education, whose partners are breastfeeding their first infant (primarily families in the 'low-risk' sub-group). Targeting bed-sharing safety information specifically to these fathers could be one constructive course of action.

Among those families where bed-sharing was irregular and unrelated to breast-feeding, settling a baby who was having trouble sleeping was a frequent reason for bringing an infant into the parental bed. We have evidence from our sleep logs that some of these unsettled babies were experiencing the prodromal phase of an illness at this time, which may have further implications for our understanding of some infant deaths that occur in adult beds.

The results of this qualitative and quantitative study confirm that bed-sharing is a complex phenomenon, and infants cannot simply be categorised as bed-sharers or solitary sleepers. The intersection of a range of variables causes each bed-sharing infants' sleep environment to be unique. Factors that are intrinsic to the infant (such as incipient illness) intersect with care-giving decisions (such as feeding method) and parental experience and expectations (such as uninterrupted sleep) to determine whether or not a baby will sleep in its parents' bed on any given night. An infant's bed-sharing safety is affected by parental acknowledgement of bed-sharing, understanding of bed-sharing safety, and willingness to modify their own behaviour in deference to their infant's presence in the bed.