Abstract
OBJECTIVES. We aimed to provide a quantitative analysis of the sleep arrangements and behaviors of bed-sharing families to further understand the risks and benefits as well as the effects of infant age and room temperature on bed-sharing behaviors.
METHODS. Forty infants who regularly bed shared with ≥1 parent ≥5 hours per night were recruited. Overnight video of the family and physiological monitoring of the infant was conducted in infants' homes. Infant sleep position, potential for exposure to expired air, head covering and uncovering, breastfeeding, movements, family sleep arrangements, responses to the infant, and interactions were logged.
RESULTS. All infants slept with their mother. Fathers were included in 18 studies and siblings in 4. Infants usually slept beside the mother, separated from the father/siblings (if present), facing the mother, with head at mothers’ breast level, touching, or with mother cradling. Median overnight breastfeeding duration was 40.5 minutes. Mothers commonly faced their infant, but infants were rarely in a position that potentially exposed them to maternal expired air. Fathers were seldom in contact with the infant during sleep. Of the 102 head-covering episodes observed in 22 infants, 80% were because of changes in adult sleep position. Sixty-eight percent of head uncovering was facilitated by the mother; half of these events were prompted by the infant. A 1°C increase in room temperature decreased infant head covering by 0.2 hours.
CONCLUSIONS. The mother-infant relationship is of prime importance during bed sharing, whether the father is present or not. The focus around breastfeeding often dictates the sleep position of the infant and mother, though room temperature may also influence this. In colder rooms infants tend to spend more time with their face covered by bedding. Frequent maternal interactions rely on the ability of the mother to arouse with little stimulation. Mothers, perhaps impaired by alcohol, smoking, or overtiredness, may not be able to respond appropriately.
Bed sharing has been investigated in many epidemiological studies in relation to the risk of sudden infant death syndrome (SIDS). This has lead to the identification of prenatal and postnatal factors that are associated with increased risk, such as maternal smoking in pregnancy,1–4 maternal alcohol consumption2,3 or overtiredness,2 excess bedding,2 infants <11 weeks,2–4 and infants sleeping with people other than parents.5 Observational studies have reported increased breastfeeding overnight,6–8 continuation of breastfeeding to an older age,9 increased overnight mother-infant interactions,8,10 decreased maternal tiredness,11 and emotional benefits to the mother.11 Evaluation of this evidence led to a recent recommendation by the American Academy of Pediatrics12 for parents to avoid bed sharing with their infant as practiced in the United States and other Western countries, rather than focusing on the specific situations to be avoided. The closed questions asked in case-control studies have very limited ability to place the risk factors in a broader context. Although these studies have identified many factors that increase the risk of SIDS, they are not able to describe the dynamic nature of sleep positions or adult-infant interactions through the night.
Perceptions about the advantages or risks of bed sharing are shaped by culture. It has been reported that maternal perception of infant bed sharing as normal or problematic depends on whether bed sharing is part of the cultural norm and whether interdependency or autonomy is valued.13 Bed sharing has been investigated in the context of a sleep problem in some studies,14,15 whereas night waking and frequent feeding is regarded as normal behavior in others.16–18
Whether traditional or not, bed sharing is often an integral part of an infant-centered parenting style and is perceived by parents to have substantial benefits.11,16,19 Studies in the United Kingdom and United States suggest that ≥50% of families spend some time bed sharing.20,21 Studies of infants who commonly sleep all night with a parent, from birth, report fewer sleep difficulties.13,22
Observational studies have been conducted with mother-infant pairs at low risk of SIDS in the controlled environment of the sleep laboratory. These studies have reported benefits such as increased breastfeeding, increased infant arousals, and synchronized mother-infant sleep patterns.6,23–25 Fathers have not participated in these studies; however, bed-sharing families report that fathers are often included in bed sharing at home.26
Observational studies measuring behavior and physiology have compared bed sharing and cot sleeping in the sleep laboratory19,27 and in the home,8,28 but this confines the investigation of bed sharing to factors that can also be described in cot sleep. This present study focuses solely on bed-sharing families to describe the variety of practices as observed in the home where usual practices are most likely to occur.
The aim of this study, therefore, was to provide a quantitative analysis of the sleep arrangements and behaviors of bed-sharing families in relation to infant sleep positions and potential exposure to expired air from the caregiver, infant head covering and uncovering, maternal responses to the infant, family interactions, and the presence of multiple family members. This was to increase understanding of the risks and benefits as identified by epidemiological studies. A further aim was to identify the effect of infant age and room temperature on behaviors.
METHODS
Forty infants who regularly slept in the parental bed with ≥1 parent for a minimum of 5 hours per night were studied. No prenatal or postnatal complications in any of the infants were reported (questionnaire). Families were recruited through local postnatal groups and media advertising. Infants were aged 0 to 6 months and were >37 weeks' gestation (except 2 infants who were 28 and 32 weeks' gestation). There were 13 infants aged 0 to 12.9 weeks, 15 aged 13 to 19.9 weeks, and 12 aged 20 to 27 weeks. The postnatal age of the 2 premature infants was adjusted to be consistent with 40 weeks' gestation. There were 14 studies in the winter compared with 8 to 10 studies in each of the 3 other seasons. The study was approved by the Southern Regional Health Authority Ethics Committee, New Zealand (protocol No. 97/04/036). Informed consent was obtained from the parent(s) of all of the infants studied.
Protocol
Infants were monitored over 2 consecutive nights in their own home. The first night involved video recording only and, on the second, both video and physiological recording. The physiological recordings involved placement of electrodes for recording raw electrocardiogram, oxygen saturation and heart rate, abdominal and chest movements of respiratory pattern, nasal airflow, shin and rectal temperature, and level of CO2 at the infants' face. The details of these recordings have been described previously.29 Sensors were attached to the infants and recordings started by the researchers. Families were left unattended for the night. Recordings were turned off in the morning when the researchers returned.
For the behavioral recordings, a small surveillance camera (CEC-C38; Panasonic, Osaka, Japan) was mounted on a stand above the bed so that the full width and the top third of the bed were in field of view to allow recordings of the infant's movements and positioning and any infant-parent interactions. A small, handheld portable television was used as a monitor to ensure correct positioning. An infrared light source (12 volt 880 Med 50; Dennard, Fleet, United Kingdom) was mounted on the stand to reflect light off the ceiling on to the recording area. The camera was connected to an analog video recorder (AG-TL700; Panasonic) set to “long play” that allowed 15 hours of recording on a 3-hour videotape. Mothers completed a questionnaire providing demographic information and answered open questions relating to reasons for bed sharing, history of bed sharing with other children, and frequency of bed sharing.
Video Analysis
Analysis of the video data for infant sleep position, potential for exposure to expired air, breastfeeding, movements, infant head covering and uncovering, family sleep arrangements, responses to the infant, and interactions was based on video from the second night to allow synchronization with physiological measures. Custom-developed computer software (C VideoTM; Envisionology, San Francisco, CA) was used to log all of the significant events into a database with time code for correlation with the physiological readings. The database (File Maker Pro 2.0; Claris Corporation, Santa Clara, CA) was customized to provide a file for each major behavior category and subcategory. The start and finish times and code for each event were logged in the database using computer key commands. The video counter was calibrated with the real time digitized on the recording tape. Although tapes were recorded as long play they were viewed at normal tape speed.
Offline logging of data started from when the infant was asleep. Sleep was defined from the video as starting after the infant was settled for 2 minutes. Start and stop times for behavioral categories were logged into the database from this start time until the final waking of the infant in the morning. Data were logged for the infant throughout the study and for any person adjacent to the infant. For example, if both parents were in the bed and the infant slept all night on the outside of the bed, next to the mother only, data were logged only for the mother and infant.
Statistical Analyses
Descriptive data are presented for infant and parent behaviors showing the median, interquartile range (IQR), and maximum value for each category investigated. Logistic or Poisson regression was used to identify any effect of infant age or room temperature on infant or parent behavior. For logistic regression analysis, the reference category was a hypothetical infant aged 12 weeks. The room temperature was centered on 15°C, the mean overnight temperature for family bedrooms in the study.
RESULTS
Characteristics of the infants and mothers have been described previously.29 Median gestational age for the group was 40.5 weeks (IQR: 39.0–41.0 weeks), median birth weight was 3615 g (IQR: 3190–4100 g), and the male/female ratio was 1.35:1. Median age and weight at study were 15.3 weeks (IQR: 9.9–20.4 weeks) and 6450 g (IQR: 5605–7755 g), respectively. All of the infants were breastfed. Median maternal age was 28.2 years (IQR: 25.05–32.45 years), and 88% of mothers were educated to tertiary level. The majority of mothers were New Zealand European, and 4 (10%) mothers identified as Maori (indigenous New Zealanders). Eight percent of mothers smoked during pregnancy, and maternal alcohol consumption was minimal with 17 of 40 mothers reporting no alcohol consumption during or after pregnancy. In 13% of studies, another household member was a smoker at the time of the study. The term “father” was used, because on almost all occasions the partner was the biological father of the infant, and this term implied a relationship with the infant. The 13 of 18 fathers who provided information on education level were all educated to tertiary level, and 4 fathers identified as Maori. The median room temperature 2 hours after sleep onset for families in this study was 16.0°C.29
Ninety percent of parents in this study bed shared with their infant 7 nights per week and had started bed sharing either from the birth of their infant or within 2 weeks of the birth. Most parents (26 of 31) stated that they intended to continue to bed share with their infant after the study. In 20 of 22 families where the parents had other children, the parent(s) had also bed shared with these children. (Data were not available for 9 families regarding intention to bed share and bed sharing with previous children.)
Reasons to Bed Share
Mothers gave multiple reasons as to why they chose to bed share with their infant. The most frequent responses were that they felt it provided comfort for the infant (14 mothers), it was natural (12), it facilitated breastfeeding (11), it was convenient (10), and it promoted bonding (8). Other reasons cited were that it provided enjoyment (6), reassurance (4), comfort for the mother (3), more sleep for mother (3), and helped keep the infant warm (1).
Who Slept in the Bed?
The infant slept with only the mother in 19 of 40 studies, with mother and father in 16 studies, with mother and same gender partner in 1 study, with mother and siblings in 3 studies, and with mother, father, and siblings in 1 study. In 4 mother-infant studies, the father was present in the house, but chose not to sleep with the mother and infant (2 slept in a separate room, 1 on a mattress on the floor of the parental bedroom, and 1 slept with a toddler on a single bed pushed against the king size bed of the mother/infant). In each case, this was the family's usual practice and was not a change of sleep arrangement because of the study. All of the families slept in a double (or larger) bed except for 2 mother-infant pairs who slept in a single bed, and 1 mother-father-infant group who slept in a three-quarter bed.
Sleep Arrangement of Multiple Family Members
When the father and/or siblings also shared the bed, sleep arrangements varied, but the most common position was for the father/siblings to be separated from the infant by the mother (Table 1 and Fig 1). In 2 studies, the infant slept all night in this location having no contact with the father/sibling in the bed. In the remaining 19 studies the majority of the time was still spent in this location, with the balance (median: 33%; IQR: 13%–51%; maximum: 84%) spent with the infant between the adults (Table 1). Other arrangements are given in Table 1.
Most common arrangement: infant on mother's side with mother and infant facing.
Sleep Arrangement for Studies Involving Mother, Infant, and Father/Siblings
Sleep Positions of Bed Sharers
Table 2 shows that mothers and infants slept facing each other for much of the night. Five infants spent time sleeping prone (range: 1.6–3.51 hours). These episodes all began with the infant sleeping on the mother's chest, sometimes after feeding (Fig 2). In 1 instance, the mother then placed the infant in the bed to sleep prone. In another, the infant was placed on his side and he rolled to prone when the mother moved.
Infant asleep prone on mother's chest.
Sleep Positions of Mothers and Infants for All Studies (N = 38)
Data for fathers was only recorded while they were next to the infant. This was almost exclusively while the infant was between the adults and occurred for a median of 3.1 hours per night. Table 3 shows that in this sleep arrangement very little time was spent with the infant facing the father or the father facing the infant. The most common position for the father was to sleep on his side, facing away from the infant and for the infant to face away from the father and toward the mother.
Sleep Positions of Infant and Father When Infant Was Between the Adults (N = 19)
Studies With Siblings
In 2 of the 4 studies that involved siblings, the infant was separated from the sibling by the mother or father for the entire night, so no behavioral data were recorded for these siblings. In the remaining 2 studies, siblings spent some of the night next to the infant (12 minutes and 6 hours, respectively). When the behavioral data set for all of the infants was reviewed, outlying values from these 2 studies were found in 1 category. The time that the mother spent facing away from the infant was 1.8 and 2.4 hours for the studies with siblings compared with a median of 0.7 hours for all of the studies. The siblings did not spend any time touching or cradling the infant despite sleeping on their side facing the infant for 0.5 and 0.2 hours, respectively. Because there seemed to be no measured differences in the responses of siblings compared with fathers, they were included in the father data set.
Infant-Adult Contact
It was most common for infants to sleep in some form of contact with their mother: touching only (median: 3.1 hours; IQR: 1.6–4.6 hours; maximum: 10.3 hours) or being cradled in the mother's arm (median: 1.2; IQR: 0–4.9; maximum: 8.4), but sleeping without parental contact also occurred (median: 1.7; IQR: 0–3.6; maximum: 8.2). The maximum values indicate, however, that individuals did sleep for long periods of time in each of the contact positions. When the infant was between the parents there was negligible contact with the father, although 1 infant spent 2.5 hours cradled by the father.
Position of Infant's Head Relative to the Mother
Infants spent most time sleeping with their head at the level of the mother's breast (median: 6.1 hours; IQR: 2.4–8.4; maximum: 10.6) and some time (median: 1.2; IQR: 0–3.2; maximum: 9.9) with their head at the same level as the mother's. Infants slept <10 cm from the mother's face for a median of 0.2 hours (IQR: 0–1.0; maximum: 7.4), illustrated in Fig 3, and from 10 to 20 cm for a median of 0.7 hours (IQR: 0.2–1.8; maximum: 5.4). Only 1 infant slept with the head higher up the bed than the mother's head (6.9 hours), and another slept with head at a level below the mother's breast (2.6 hours).
Infant potentially exposed to mother's expired air. Infant separated from toddler by the mother.
Infant Feeding
The median duration of overnight feeding for infants was 40.5 minutes. The median number of feeding sessions per night was 3 (IQR: 2–4).8 Feeding times of 2 to 3 hours were recorded for some bed share infants. This reflected time that the infant was attached to the breast but not necessarily with continuous active sucking. Infants frequently fell asleep at the breast, sometimes arousing later and resuming sucking. Mothers were observed to spend almost all of the feeding time in a drowsy, half asleep state (median: 37.8 minutes; IQR: 12–97.8; maximum: 184.8) and sometimes seemed to attach the infant to the breast while half asleep. They almost always fed lying down.
Infant Head Covering
Table 4 shows that 80% of infant head-covering episodes, to the eyes or above, occurred as a result of inadvertent movement of 1 or both adults as they changed position during sleep; 15% of these were because of adult repositioning after feeding. The mother deliberately positioned the blankets up to the infant's eyes on 2 occasions, and the remaining 17% of episodes were because of infant movements: wriggles or head turning. Infants cleared their own face in 32% of episodes, and infant movements prompted clearing by the mother in a further 36%. Whether prompted by the infant or not, the mother deliberately cleared the infant's face in 35% of the episodes. Infant arousal during head covering led to immediate feeding 16% of the time. Of the 22 infants with head covering episodes, the infant's head remained covered on final waking for 5 infants and was uncovered between 3 and 30 minutes before final waking for 5 infants and >30 minutes before final waking for 12 infants.
Behavior Leading to Covering and Uncovering of the Infant Head (Head Covering to at Least the Level of the Eyes)
Effect of Infant Age and Room Temperature on Infant-Mother Behavior
Mothers spent a mean of 10.7 minutes (range: 0–74.7 minutes) checking or responding to their infant by touch (excluding touching during breastfeeding). Increasing infant age (age range: 0–27 weeks) was associated with less maternal time checking or responding to infants (relative risk: 0.94; 95% CI: 0.90–0.98; P = .007) but did not change the number of parental checks. An increase in room temperature did not affect any category of parent or infant movement or the number of infant feeds.
Table 5 shows that infant age had no significant effect on blanket height, but a 1°C increase in room temperature increased the time that the blankets were below the chin by 0.3 hours and decreased the time that the blankets were partially over the infants face (to the eyes) by 0.2 hours. No effect of infant age or room temperature was found for total feed time, infant contact time (separate, touching mother, or cradled by mother), level of infant's head relative to mother, mother's sleep position, or infant sleep position.
Effect of Infant Age and Room Temperature on Blanket Height Relative to the Infant's Face
DISCUSSION
The families in this study made a deliberate choice to bed share with their infant. They were mainly of European descent without risk factors identified as contributing to SIDS and were not from a culture in which bed sharing is the norm. Smoking was unusual, alcohol consumption minimal, and all were breastfeeding. Their sleep arrangements were, however, diverse, reinforcing the importance of home studies for identifying normal behavior rather than the controlled environment of the sleep laboratory. Some sleep positions promoted mother-infant interactions, facilitating breastfeeding and frequent contact, including uncovering the infant's head from bedding. That the mother-infant partnership was of great importance confirms earlier studies of mother-infant pairs10,25 and now also shows this to be the case when the father shares the bed.
The finding that fathers frequently participate in bed sharing is consistent with parental reports.26,30 Their presence, however, does not seem to alter the mother-infant relationship but may be important with regard to bed size and space for the infant. The effect of the presence of older siblings is even less clear unless the mother's attention is distracted from the infant. Siblings sharing the bed have been shown to increase the risk of SIDS in a high-risk population,5 but the risk of siblings sleeping with infants at low risk of SIDS is unknown.
It is apparent that the success and satisfaction of bed sharing is related to the mutual orientation of infant and mother: side-sleeping at breast level and in close contact allowing minimal disturbance during breastfeeding and frequent checks on the infant. Other authors confirm this as a typical orientation in breastfeeding bed sharers25,28,31 but not in nonbreastfeeding pairs,28 suggesting that breastfeeding is an important component of “low-risk bed sharing.” Breast-level sleeping, although diminishing the chance of the infant's face pressing into an adult's pillow,28 does increase the likelihood of head covering.8 Therefore, it is important to examine how uncovering occurs, because it seems more likely that the mother's actions, often triggered by the infant, rather than the infant's own movements, are effective. Further evidence from cot-sleeping infants shows that infants themselves (2.5 and 5 months old) are often unable to clear blankets from their face when sleeping prone or supine.32
The increased risk of SIDS associated with bed sharing with a mother who has consumed alcohol or is overtired2,3 may be because of the inability of the mother to act on infant cues to attend to her infant. Similar findings about methods of uncovering the head have been observed previously (Helen Ball, PhD, written communication, 2002). Because 85% of infants who experienced head covering during the night were found with their head clear from bedding at final waking, it is likely that head covering during bed sharing is underreported in case-control studies.
Prone sleeping is a well-known risk factor for SIDS, and in most SIDS prevention programs, the side sleep position is discouraged because of the risk of rolling to prone. In the bed-share context, the possibility of rolling to prone may be reduced by the positioning of the mother relative to the infant.33 In 1 case-control study,34 there were more deaths in the prone position among the cot-sleeping infants compared with the bed sharers, although similar numbers in each group were placed prone to sleep.
The observations of bed-share infants in the prone position were almost all brief, with the infant lying on the mother's chest, usually after a feed. This may not constitute the same risk to infants as sleeping prone on soft bedding where rebreathing or hyperthermia may occur.35,36 This contrasts strongly with previous results from this study, where some age-matched cot-sleep infants slept prone all night.8 Other authors report bed-share infants as having either no prone sleep at all31 or briefly on the mother's chest.25 It is not known whether parents report the latter as prone sleep.
Expired air from the mother may cause small increases in the infants inspired Pco2.37 Infants in this study spent little time with their face positioned close to and in line with mothers expired air. Indeed, it has been argued that mild hypercapnia would be protective, mildly stimulating ventilation and decreasing periodic breathing.37,38 There is no evidence that breathing mother's expirate could simulate the “repeated exposure to hypoxic conditions,” which animal studies have shown to blunt the arousal response to hypoxia.39 The close contact between mother and infant and exposure to mother's expired air may pose a hazard if the mother is a smoker. Nicotine has been shown to accumulate in the air, in dust, and on household surfaces even when parents smoke only outside,40 and if the mother breathes closely over the infant, even some time after smoking a cigarette, this may increase the exposure of the infant to harmful products of cigarette smoke.
Head covering is possibly of concern as a cause of overheating. It is more likely to occur in cooler rooms, which we have reported to be typical of bed sharers (16.0°C during bed sharing versus 17.5°C 2 hours after sleep onset in adult bedrooms with cot-sleeping infants).29 The increased shin temperature of bed-share infants compared with cot-sleep infants29 is evidence of thermolytic activity induced by the thicker bedding on adult beds,29 and head covering could be a problem in this situation. In New Zealand, central heating of homes is not common despite cool winter temperatures in the south. In this study, mean bedroom temperature dropped through the night to reach 14.4°C by 8 hours after sleep onset. It may be that parents are less likely to heat the bedroom overnight if the infant is sleeping in the adult bed. Heating adult bedrooms overnight could influence the amount of and the position of bedding, helping to reduce head covering.
The side sleep position was most common for infants in this study despite the widely publicized message to avoid this sleep position. At the time of data collection, New Zealand advice was that mothers who smoked should not bed share with their infant; however, many families, particularly those of European descent, perceived a general anti–bed-share culture and felt that they were sleeping in a way that did not have widespread support.16,17 This alienation from mainstream practice may have led families to disregard advice that was clearly aimed at cot-sleeping infants. There were no parenting pamphlets aimed specifically at bed-sharing families with advice about safe sleeping at this time. This emphasizes the importance of establishing bed share-specific guidelines and of using appropriate strategies for reaching the target group.
CONCLUSIONS
This study has reinforced the importance of the mother-infant relationship during bed sharing, whether the father or siblings are present or not. The focus around breastfeeding in many instances dictates the sleep position of the infant and mother. However, room temperature may also influence sleep position. In colder rooms, infants tend to spend more time with their face covered by bedding. Frequent maternal checking and response to infant cues are characteristics of bed sharing that rely on the ability of the mother to arouse with little stimulation. Mothers, perhaps impaired by alcohol, smoking, or overtiredness, may not be able to respond appropriately. Future studies aimed at targeting these practices associated with bed sharing may shed light on the harmful factors associated with bed sharing and SIDS.
Acknowledgments
This study was supported by a grant from the Health Research Council of New Zealand.
We thank the families who participated in the study and also Amanda Phillips and Charrissa Makowharemahihi for research assistance, Assoc Prof Sheila Williams for statistical advice, Christine Rimene for advice on cultural aspects, and Paul Bennington and Gordon Yau for assistance with customizing the database for video logging.
Footnotes
- Accepted July 24, 2006.
- Address correspondence to Sally Baddock, PhD, School of Midwifery, Otago Polytechnic, Private Bag 1910, Dunedin, New Zealand. E-mail: sbaddock{at}tekotago.ac.nz
The authors have indicated they have no financial relationships relevant to this article to disclose.
REFERENCES
- Copyright © 2007 by the American Academy of Pediatrics
In this issue
Jump to section
Related Articles
Cited By...
- SIDS and Other Sleep-Related Infant Deaths: Evidence Base for 2016 Updated Recommendations for a Safe Infant Sleeping Environment
- SIDS and Other Sleep-Related Infant Deaths: Evidence Base for 2016 Updated Recommendations for a Safe Infant Sleeping Environment
- Nocturnal Video Assessment of Infant Sleep Environments
- Is parent-child bed-sharing a risk for wheezing and asthma in early childhood?
- Hypoxic and Hypercapnic Events in Young Infants During Bed-sharing