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

Sleeping arrangements for twin infants

A research project of the Department of Anthropology, part of the Medical Anthropology research group.


The project is funded by the following grant.

  • Co-bedding In Twins (£70358.00 from Foundation for Study of Infant Deaths)


This study assessed sleeping arrangements used by parents for their twin infants to determine what information might be useful for parents regarding twin infant sleep.


The study was conducted in 3 parts: An assessment of sleeping arrangements for 60 pairs of twin infants from 0-5 months; Video observations of the behaviour of 10 pairs of twin infants under 3 months of age sleeping together in their home environment; Behavioural and physiological monitoring of 14 pairs of twin infants under 3 months of age sleeping together and apart in a sleep lab environment. Part 1: a convenience sample of 60 families recruited via TAMBA (Twin and Multiple Birth Association) maintained sleep logs and participated in telephone interviews when their twin infants were 1, 3 and in some cases 5 months of age. At 1 month of age the majority of twins were co-bedded (60%), with parents taking their cues on sleeping arrangements from those experienced on the post-natal ward (X2=9.5, df=1, p=0.002). Co-bedded infants predominantly slept together in a cot (92%) while infants who slept apart were predominantly in separate Moses baskets (86%) in the 1st month (X2=36.13, df=1, p<0.00001). Co-bedded twins were initially positioned side-by-side (68%), however at least 8 co-bedding configurations were recorded. The vast majority of infants slept supine (86% in 1st month, 92% in 3rd month) with no association found between sleeping arrangement and sleeping position. Potential disturbance of each other’s sleep was the primary parental reason for sleeping twins apart, but overheating and unspecified ‘cot-death fears’ were mentioned by a handful of families. ‘The babies prefer it’ was the most common explanation for co-bedding although several parents also mentioned that co-bedded babies were more synchronised in their feeding and waking patterns which facilitated night-time care-giving. Concerns about overheating or co-twin suffocation caused 4 families to fashion barriers to separate their twins within a cot. A third of the pairs co-bedded in the 1st month were separated for sleep by the 3rd month (overall proportions = 40% together, 60% apart in 3rd month). A greater proportion of twin infants slept in a separate room from their parents in the 1st month than did singleton infants in a previous study (27% cf 8%). Part 2: 10 twin pairs (9DZ, 1MZ) were videoed sleeping together at home, 4 in head-to-head and 6 in side-by-side configurations. Sleep duration was unaffected by co-bedding configuration. Babies sleeping side-by-side were observed to occasionally obstruct one another’s airways with arms across the other’s face, but also exhibited a greater degree of observed sleep state synchrony than those sleeping head-to-head. Part 3: 14 twin pairs (12DZ, 2MZ) participated in a 2-condition trial sleeping together and apart in the sleep lab. We found no consistent difference in temperature patterns between the two conditions. Those infants that spent the greatest proportion of the co-bedded night physically touching one another exhibited overlapping temperature patterns on the co-bedded and the separate night. Neither the nadir of the post-sleep temperature fall, nor the peak temp recorded during REM or Quiet sleep, differed between the two sleep conditions. Airway obstruction of one twin by the other’s arm during co-bedding was observed for 1 twin pair and was associated with lowered oxygen saturation, followed by arousal and movement by the affected twin. Head covering by bedding was not observed during the co-bedded night.


Conclusions: In this study co-bedding did not appear to be more risky than separate sleeping for twin infants under 3 months of age. Co-bedded infants did not wake more frequently in the night than those sleeping separately, in fact their arousal patters were more synchronous. Co-bedded infants sleeping in close bodily contact with one another did not exhibit increased core temperature, although parents should be given guidance on appropriate use of bedcovers. Although not discussed above, swaddling co-bedded infants may be problematic. Regarding SIDS reduction, the primary reason to encourage co-bedding of twins is to facilitate keeping twins in the parental bedroom by the use of a single cot. Parents of twins may benefit from specific information that it is the presence of an adult caregiver, not the presence of another infant that is associated with reduced SIDS risk.

Published Results

Journal Article

  • Ball, Helen L. (2007). Together or apart? A behavioural and physiological investigation of sleeping arrangements for twin babies. Midwifery 23(4): 404-412.
  • Ball, Helen L. (2006). Caring for twin infants: sleeping arrangements and their implications. Evidence Based Midwifery 4(1): 10-16.


From the Department of Anthropology

From other departments