Differences in Infant and Parent Behaviors During Routine Bed Sharing Compared With Cot Sleeping in the Home Setting

METHODS

Two groups of infants were studied: 40 bed-sharing infants and 40 cot-sleeping infants. The sleep practice criteria was that bed-sharing infants regularly slept in the parental bed for a minimum of 5 hours per night whereas cot-sleeping infants regularly slept in a cot or bassinette in the parental bedroom ≥5 hours per night. None of the infants reported prenatal or postnatal complications (questionnaire). Bed-sharing infants were recruited through local postnatal groups and media advertising. Cot-sleeping infants matched for age and season of study were recruited from the local maternity ward. Infants were aged 0–6 months with 13 infant pairs aged 0–12.9 weeks, 15 pairs aged 13–19.9 weeks, and 12 pairs aged 20–27 weeks. All of the infants were at >37 weeks' gestation (except 2 infants in each group who were 28 and 32 weeks' gestation). The age of the 4 premature infants was adjusted to be consistent with 40 weeks' gestation. There were 14 pairs of studies in the winter compared with 8 to 10 studies in each of the other seasons. The study was approved by the Southern Regional Health Authority Ethics Committee, New Zealand (protocol 97/04/036). Informed consent was obtained from the parent(s) of all of the infants studied.

RESULTS

As shown in Table 1, bed-sharing and cot-sleeping infants were comparable with regard to gestational age, birth weight, male:female ratio, age at study, and weight at study. All of the bed-sharing infants and 35 of 40 cot-sleeping infants were breastfed. The age of the mother and the proportion of mothers educated to tertiary level were similar between the groups. A small number in both groups were identified as Maori, indigenous New Zealanders. Maternal smoking was more common in the cot group (25%) compared with the bed-sharing group (8%). Maternal alcohol consumption was minimal in all, ranging from “rarely” to 3 glasses of wine or beer per week, with 17 of 40 bed sharers and 15 of 40 mothers of cot sleepers reporting no alcohol consumption during or after pregnancy. The practice of bed sharing was reported to be adopted by mothers because of factors such as the ease of breastfeeding, the provision of a close and secure environment for the infant, a more settled infant, and a natural environment.

DISCUSSION

This study clearly demonstrated different behaviors of both the infant and parents when comparing bed-sharing and cot-sleeping practices. Although the cot was usually immediately adjacent to the parents' bed, the presence of the infant in the adult bed for bed-share studies resulted in very different behaviors. The study showed that regular bed-share infants engaged in more feeding and more infant-mother interactions than cot-sleep infants, side sleeping position was more common during sleep and at final waking in bed-share infants, and prone sleeping position, although rare, occurred for short intervals in bed-share infants, whereas it lasted all night for 2 cot infants. Incidents where the bedding or clothing covered the infant's head were more common in the bed-sharing situation both during the night and on final waking.

Previous smaller studies comparing the same practices have been conducted mainly in the laboratory setting,^23,27 although attempts have been made to make the environment as home-like as possible. These studies used a crossover design that showed that, for many behaviors, the largest difference was recorded for regular bed sharers on their bed-share night compared with regular cot-sleepers on their cot-sleep night.^23,27 This emphasizes the importance of observing infants in their regular sleep arrangement, as in this study. The present study supports the observations from laboratory studies that bed-share infants engage in more feeding episodes and are checked by their mother more frequently than cot-sleeping infants.^23,28–30 Mothers often identified ease of breastfeeding as a reason for bed sharing. Population studies also support an association between bed sharing and breastfeeding³¹ and an association with breastfeeding persisting to an older infant age.³² Several large epidemiologic studies have shown a small but significant protective effect of breastfeeding against SIDS;^17,33–36 however, this has not been shown in some others.^37,38

Concern has been raised regarding the possibility of accidental asphyxiation from mothers falling asleep breastfeeding while lying down.³⁹ No instances were noted where the mother was in a position that might have resulted in mechanical obstruction of the airways, and no oxygen desaturation events <90% with head covered (data not given) or any increase in rectal temperature outside the reference range was observed.²⁵

The reason we videoed 2 nights was to establish whether there was any difference in behavior because of “first-night effect” or the presence of the sensors on the infant. Using Bland-Altman plots,⁴⁰ we found no significant difference on key behavioral indices, such as sleep time, number of infant movements per hour, feeds, and sleep position, suggesting that attachment of the sensors did not have a significant effect on sleep behavior.

Although there is no direct evidence that increased maternal checking reduces SIDS, mothers in this study and others^2,23,41,42 report an emotional benefit from bed sharing, because they can easily check their infant. Mothers have also been observed to actively check and modify infant temperature by rearranging bedding.⁴³ Room sharing compared with infant sleeping in a separate room is protective against SIDS^14,15,44 and may be related to increased maternal checks. It is likely that the dramatic reduction in the multivariate relative risk for infants not sharing the room and prone from 16.99 (95% CI: 10.43–27.69) to 3.28 (95% CI: 2.06–5.23) for infants sharing the room and prone⁴⁴ is explained by increased awareness and checking of infants while they are asleep. In our study, whereas bed-share parents checked their infant more often, many of these checks were brief, involving minimal disruption to the mother's sleep. These findings are confirmed by Mosko et al,⁴⁵ who reported that total sleep time of mothers was not decreased on bed-share nights compared with infant cot-sleep nights and that maternal awakenings were for shorter duration on the bed-share nights. It is likely that greater arousal and more disruption to maternal sleep would be needed to check an infant in a cot, even if it was nearby.

The risks and/or benefits associated with increased waking have been debated. In adults and children, sleep fragmentation is associated with many negative effects, such as increasing the frequency and duration of obstructive sleep apnea⁴⁶ and increasing the arousal threshold.^47,48 However, none of these studies have been conducted on breastfeeding women. It is not known whether the multiple, brief, drowsy awakenings through the night during bed sharing would have more or less impact on the mother than the few, full awakenings required to attend to an infant in the cot. Interestingly, mothers in this study, as in others, report “increased sleep” as a reason to bed share, along with “having a more settled infant.” Studies investigating the effect of mild sleep deprivation on infants report a possible increased propensity to upper airway obstruction^49,50 and changes in autonomic control of cardiac function.⁵¹ However, it is likely that experimentally induced sleep deprivation has different physiologic manifestations from infant-initiated awakenings through the night associated with breastfeeding.

The finding in this study that the side sleep position was the most common sleep position for bed-share infants is in agreement with Ball's findings.³⁰ However, whereas bed-share infants had significant periods of side sleep in laboratory-based studies,^23,52 the supine position was predominant. The increased instrumentation for recording electroencephalogram, electro-oculogram, and electromyogram in these 2 studies might have affected the position mothers placed their infant to sleep (Helen Ball, PhD, written communication, 2002). The side sleep position has been identified as increasing the risk of SIDS,^15,53 reportedly associated with the tendency of side-sleep infants to roll prone.¹⁵ However, the evidence for this has been established from infants sleeping in a cot. There is no data to establish the risk of this position during bed-share sleep. In our study, 12 cot-sleep infants spent some time sleeping on their side, but none were observed to roll to the prone position. All 38 of the bed-share infants slept some time on their side, and 1 infant, aged 23 weeks, was observed to roll to the prone position when the mother moved away from the infant. A characteristic sleep position of mother and breastfed infant that seems to prevent rolling has been described: mother sleeping in a lateral position, facing the infant, with her knees drawn up under the infant feet and the mother's arm positioned above the infant's head.^54,55 This was observed in our study, but it was not universal, despite all of our mothers being breastfeeders.

Head covering by blankets occurred more often in the bed-share group, a finding also observed by Ball (Helen Ball, PhD, written communication, 2002) but not reported by others. Young²³ found no instances of head and body completely covered by bedding. This may reflect a difference in home monitoring compared with the sleep laboratory, where arguably parents are more relaxed and more likely to engage in usual practices. Whether this behavior places these infants at risk is another question. Being found with head covered has been reported in several studies as increasing the risk of SIDS (odds ratio: 12.5; 95% confidence interval: 6.47–24.1).¹⁵ Although head covering was common among the bed-share infants in this study, only a quarter of infants with head-covering episodes during the night ended up with head covered at the end of sleep. Bedding tended to be moved on and off infants more often during the natural course of sleep through the night. This may help explain why bed-share infants are found with the head covered at the end of sleep less often than cot-sleep infants.⁵⁶ Infants in the present study often stayed (without significant movement) in the head-covered position for long periods of time (eg, 3.5 hours by 1 infant), suggesting they were not uncomfortable. Our previous studies suggest that the risk of significant rebreathing into bedding depends on the type and thickness of covering,^57,58 as well as the ability of the infant to mount both a respiratory and arousal response. Because infants of smoking mothers may well be the infants least likely to respond to this stress,^59–61 the large interaction between smoking and bed sharing noted in epidemiologic studies may be explained by poor responsiveness to this particular occurrence during bed-sharing sleep.

Bed-share infants in this study had a different pattern of movements than the cot infants. Although there were the same numbers of posture change episodes in both groups, episodes were shorter in the bed-share group, resulting in a highly significant, reduced total posture change time. The presence of the mother, often touching or cradling the infant during sleep, may also have brought rapid reassurance to the bed-share infant and reduced nonawake movement episodes and, consequently, reduced stress experienced by infants.

It would seem that the bed-share infants in this study were at low risk of SIDS, because there were very few maternal smokers, all of the infants in the bed-sharing group were breastfed, most mothers had some form of tertiary education, and families actively chose to bed share because of perceived advantages to themselves and their infant. The findings may be quite different in bed-sharing families where many SIDS risk factors are prevalent^13,17,62 and breastfeeding is not common.

Although this study has identified potential hazards that may be encountered during bed sharing, for example, head covering, it has also identified many potential benefits, for example, increased parental checks. This was not a surprising finding, because in many societies around the world, bed sharing is the preferred sleep arrangement. It is only relatively recently that white societies have moved to a solitary sleep arrangement, where conditioning infants to sleep through the night without waking is a goal valued by society.⁶³ However, there is a growing trend among whites to choose to bed share as a parenting style.^7–9

This study has highlighted many factors that seem to be common to both bed-sharing and cot-sleep infants but in fact vary in important ways because of the different physical environments and the presence of adults. Thus, risk factors identified for infants sleeping in a cot, for example, side sleep, may not be directly applicable to bed-sharing infants and require investigation by epidemiologic studies using cases and controls in the bed-share environment. Secondly, the benefits of bed sharing, for example, increased maternal checking, breastfeeding, and faster and more frequent maternal responses, rely on the mother's ability to arouse, at least partially, and respond to the infant through the night. Mothers impaired, for example, by alcohol or extreme overtiredness, may not be able to respond appropriately, thus stressing the importance of a healthy, nonimpaired mother in the bed-share partnership.