Parent-Infant Co-sleeping and Its Relationship to Breastfeeding

Definitions of co-sleeping 

Difficulties arise when comparing the literature on co-sleeping because a universal definition for this term does not appear to exist. Numerous factors may be taken into account when defining this practice. Whereas some authors offer a very broad description, such as “sharing the bed with anyone” (Hauck et al., 2003), others do not even provide a working definition (Flick, White, Vemulapalli, Stulac, & Kemp, 2001).

Within the literature, one variance is the length of time the bed is shared each night. Some authors specify this length of time, such as a minimum of 5 hours per night on a regular basis (Hunsley & Thoman, 2002), whereas others claim that co-sleeping can occur for any portion of the night (Ball 2002, Jenni et al 2005, Weimer et al 2002). Furthermore, the number of nights per week that co-sleeping occurs also is inconsistent, ranging from at least once per week (Ball, 2003) to at least five nights per week (McKenna, et al., 1997). Most authors do not include length of time per night or number of nights per week in their definitions.

Discrepancies also exist regarding the number of family members or other individuals present in the bed while co-sleeping. Most definitions include one or both parents as co-sleepers with the infant (Ball 2002, Ball 2003, Brenner et al 2003, Jenni et al 2005, Latz et al 1999, Valentin 2005, Willinger et al 2003). Some vaguely refer to the co-sleeper as “another adult” (Abel et al 2001, Blair and Ball 2004, Weimer et al 2002) or “anyone” (Hauck et al 2003, Nelson and Taylor 2001). These definitions allow for other family members such as extended family or siblings to be considered co-sleepers. Furthermore, even though the association of co-sleeping and breastfeeding is a maternal-infant activity, only one author specified the term as such (McKenna et al., 1997). Thus, it is obvious that there is no universal term for co-sleeping, and variances undeniably exist between different families. Even though a standardized definition would help in research comparisons, such consistency is not useful in everyday practice because each family may define its own practices differently from another’s.

Methods of data collection 

The majority of the literature on co-sleeping utilizes qualitative methods for data collection. A common approach is to analyze parent-completed sleep logs (Ball 2003, Blair and Ball 2004, Buckley et al 2002, McKenna et al 1997). This method relies heavily on the parent’s ability to remember certain circumstances, yet parental recall may not always be reliable. Moreover, studies using parental recall of events occurring the night before may have more credibility than those asking parents to remember episodes from many nights or weeks earlier.

Another approach is to utilize only parental/caretaker interviews to gather data (Abel et al 2001, Latz et al 1999, Weimer et al 2002, Willinger et al 2003). Interviews allow for elaboration or clarification on certain aspects of the conversation, and they are excellent for presenting descriptive and narrative data such as in case studies. Finally, many authors combine interviews, sleep logs, and/or questionnaires for multidimensional qualitative data collection (Ball 2002, Ball 2003, Blair and Ball 2004, Brenner et al 2003, Clements et al 1997, Nelson and Taylor 2001, Valentin 2005). The use of multiple methods for data collection incorporates parental experiences and recall, as well as standardized questionnaires. No articles were found to have used questionnaires only.

Quantitative studies also have been performed on related aspects of co-sleeping, examining thermal sleep environments (Baddock et al., 2004), physical properties of bedding (Flick et al., 2001), arousability thresholds of breastfeeding infants (Horne et al., 2003), nocturnal breastfeeding behaviors using audiovisual recordings and polysomnographic measures (McKenna et al., 1997), and dangers associated with co-sleeping (Nakamura et al., 1999). Thus, quantitative studies tend to focus more on specific, measurable variables in the sleeping environments or physiological responses, rather than on parental experiences, practices, and/or rates of co-sleeping.

International and United States literature 

Out of the 21 research studies analyzed, 12 (57%) were conducted in countries other than the United States. Of these 12 studies, most were conducted in the United Kingdom (five) or Australia and New Zealand (five). To compare and contrast international practices in countries outside of the United States, Nelson and Taylor published the International Child Care Practices Study in 2001. This study collected comparative information on child care practices, including co-sleeping, from 4656 families in 17 different countries. It was found that infants in Chongqing, China, had the highest rate of co-sleeping (88%), whereas those in Odessa, Ukraine, had the lowest rate (19%). Interestingly, the data were collected during the coldest 2 months of the year in each country. Because the Ukraine’s coldest months may be colder than China’s, the possibility exists that co-sleeping with the intent to keep an infant warmer may not be a significant factor. This finding echoes an analysis by (Blair & Ball 2004) suggesting that co-sleeping rates are not related to colder months.

Analyzing data from multiple countries across the globe is beneficial because it can reveal how diverse cultural practices may affect the selection of an infant’s sleeping environment. Just as different cultures view breastfeeding differently, so too may they view co-sleeping differently. In general, there appears to be a positive correlation between co-sleeping and breastfeeding reported in the international literature. (Ball 2003) found a significant relationship between co-sleeping and breastfeeding persistence. (Blair and Ball 2004) noted that 46% of co-sleeping mothers and infants in the United Kingdom were still breastfeeding at 3 to 4 months, twice as many as non–co-sleepers. A similar study in the United Kingdom found that a shortened duration of breastfeeding was associated with mothers who did not co-sleep (Clements et al., 1997).

In New Zealand, (Abel et al. 2001) revealed that co-sleeping is the norm of Pacific ethnic groups there (i.e., Maori, Tongan, Samoan, Cook Islands, Niuean, and Pakeha). These communities view co-sleeping as a safer, more family-connected way to sleep with many practical, psychological, and spiritual benefits for the infant. Furthermore, among these ethnic groups, breastfeeding was one of the parental topics on which there was highest consensus. Most women chose to breastfeed because of the perceived physical and emotional benefits for themselves and their infants.

Just as co-sleeping and breastfeeding practices vary worldwide, they also differ among geographic regions within the United States. Breastfeeding rates tend to be highest in western states like Washington and Oregon and lowest in southeastern states such as Mississippi and Louisiana (Centers for Disease Control and Prevention [CDC], 2004). Published rates of co-sleeping in the United States also vary depending on geographical region; however, data are limited to states in which specific studies have occurred. In the District of Columbia, co-sleeping was viewed as “normative behavior,” with almost 50% reporting they usually co-slept in the child’s first year of life (Brenner et al., 2003). A study of New Orleans parents found a co-sleeping rate of 88%, with 46% of caregivers co-sleeping at least 22 days in the last month (Weimer et al., 2002).

In a 7-year study of 8453 caregivers within all 48 contiguous states, (Willinger et al. 2003) reported that overall, 45% of infants had spent at least some time at night on an adult bed within the past 2 weeks. Between 1993 and 2000, the proportion of co-sleeping infants more than doubled. Thus, it appears that co-sleeping is growing within the United States, despite warnings against it from the AAP and CPSC. Factors reported to be associated with routine co-sleeping in the United States include mothers younger than 18 years of age, Black or Asian race, annual household income of less than $20,000, Southern states, and an infant less than 8 weeks of age.

In addition, much of the literature on co-sleeping focuses on urban settings, often concentrating on minority or socioeconomically disadvantaged populations. These communities tend to display consistently lower breastfeeding rates (CDC, 2004). In the aforementioned studies conducted by (Weimer et al. 2002) and (Brenner et al. 2003), the majority of subjects were inner-city, African-American mothers. Their results may suggest that this population has more of a tendency to co-sleep than other populations, yet additional research is needed to determine the validity of this notion. Furthermore, sleeping practices in rural communities also should be examined to present a more comprehensive picture of the rates of co-sleeping within the United States.

Effects of co-sleeping on breastfeeding 

Out of the 11 articles specifically investigating the relationship between co-sleeping and breastfeeding, eight showed a positive correlation and three revealed a negative correlation or no correlation. Among the positive correlations, (McKenna et al. 1997) found that infants who routinely co-slept at home breastfed three times longer during the night than did infants who routinely slept separately, when tested in their routine conditions. (Blair and Ball 2004) demonstrated that almost half of co-sleeping mothers and infants were still breastfeeding at 3 to 4 months of age, twice as many as solo sleepers. In addition, (Clements et al. 1997) found that exclusive breastfeeding at discharge from the hospital was associated with later co-sleeping, suggesting that breastfeeding is the inducement for co-sleeping. Finally, in terms of benefits to the mother, Glenn & Quillin (2003) reported that breastfeeding mothers had more overall sleep than did bottle-feeding mothers if their infants co-slept for any part of the night.

Conversely, in the District of Columbia, Brenner et al. reported that breastfeeding was not significantly associated with co-sleeping at either of two follow-up interviews; therefore, no association existed between co-sleeping and either initiation of or concurrent breastfeeding. Interestingly, the authors report that 50% of the subjects claimed they usually co-slept during the first year of the infant’s life (2003). This study did not cite breastfeeding initiation rates of its subjects, but the CDC reports an initiation rate of only 65% among residents in the District of Columbia. Similarly, (Flick et al. 2001) found that at 8 weeks of age, breastfeeding was no more common among African-American infants who co-slept than among those who did not co-sleep. All subjects were known to be African-American women in St. Louis, yet other demographic factors (i.e., maternal age, socioeconomic status, education, employment, and previous breastfeeding experience) were not accounted for.

Finally, an Australian article reported that similar proportions of infants were breastfeeding, regardless of their sleeping environment: 76% of co-sleeping infants were breastfeeding, compared with 75% of solo-sleeping infants (Buckley et al., 2002). This finding may be attributed to the fact that breastfeeding is the cultural norm in Australia, where breastfeeding initiation rates are between 83% and 87% (Australian Bureau of Statistics, 2001). Furthermore, this research did not provide any definition of co-sleeping, so parental subjects could have differed on whether their practices were truly considered co-sleeping.

Implications for pediatric nurse practitioners 

Some parents may not be aware of the recommendations from the AAP or CPSC, whereas others may be overwhelmed by all the contradictory information on co-sleeping with their infant(s). Breastfeeding mothers, in particular, may be frustrated with the AAP’s conflicting recommendations; although a mother may desire to exclusively breastfeed for 6 months, she may find it difficult to be successful when she is advised against nighttime breastfeeding in bed. (Eidelman and Gartner 2006) argue that mothers wary of breastfeeding in bed may choose the even riskier practice of breastfeeding infants in an upholstered sofa or chair. This practice may carry an even greater risk of smothering or dropping the infant. Finally, some parents may not think to inquire about safe sleeping practices at their infant visits, especially if they believe their chosen method is “wrong.” Because client populations may display a wide range of knowledge on this topic, it is the responsibility of the pediatric nurse practitioner (PNP) to incorporate this discussion into everyday practice.

A full assessment of the infant’s sleep practices should occur not only at the initial infant visit but also at every visit thereafter, because infant sleep patterns, practices, and environments can change over time. This assessment should impart a nonjudgmental approach so as not to suggest there is a right or a wrong practice. An open-ended question such as, “Where does your infant sleep at night?” allows for a more detailed response. Probing questions include the location of the infant (e.g., adult bed or separate crib), hard or soft sleeping surface, presence of blankets or pillows, and proximity of the infant to the parents (e.g., in the same room or in a separate nursery). By encouraging the parent to provide more details on the infant’s sleeping environment, PNPs can better assess the safety of the chosen setting. For instance, an infant co-sleeping in an adult bed without covers or blankets may be less likely to experience hyperthermia than a solo-sleeping infant swaddled or covered in multiple blankets.

Despite the risks and the recommendations by the AAP and CPSC, automatically condemning co-sleeping may not be appropriate in all families. Instead, the individual needs of the infant/child, the family context, and the cultural backgrounds need to be taken into consideration. Some children may need close parental proximity during the night as during the day depending on their developmental level and attachment behavior (Jenni et al., 2005). Furthermore, different parental ideologies, beliefs, and styles may emphasize certain aspects of infant sleep that are of greater importance. For example, if parents strongly desire the enhanced bonding effects of co-sleeping, they may believe that this benefit outweighs the associated risk for SIDS. On the other hand, if parents are incredibly fearful of accidentally overlying the infant while asleep, they may prefer to place the infant in a separate bed where this risk is eliminated. Thus, it is the responsibility of the PNP to inquire about and understand the reason(s) why parents choose their current sleeping environment and to provide information on how each method (co-sleeping or solo sleeping) can be performed safely.

Moreover, although no sleep environment is completely safe, many of the known risks associated with infant sleep environments are modifiable (Mesich, 2005). If a parent decides to co-sleep with his or her infant, guidelines can be provided to increase the safety of the experience (see Box). PNPs can utilize these guidelines to evaluate the safety of an established sleeping environment and/or to provide education on creating one. These guidelines also can be offered to parents in a written format. Furthermore, the PNP should always assess for frequency of parental drug or alcohol use on a routine basis, regardless of the sleeping environment.

Because research into the practice of co-sleeping is a relatively new endeavor, there are numerous implications for expanding the knowledge on this topic. The inclusion of a definition for co-sleeping is imperative for future research. This definition should include at least the length of time per night, the number of nights per week, and the number and type of persons (e.g., mother, father, and siblings) present in the sleeping environment.

Rates of co-sleeping among all geographic regions of the United States (especially in rural areas) also should be established. Furthermore, little is known about co-sleeping practices among Native American, Asian-American, and Hispanic populations. An awareness of this information may help PNPs and researchers better understand co-sleeping trends in the United States and be more aware of how families of diverse cultural backgrounds may perceive and practice co-sleeping. Ideally, having national data on co-sleeping would be advantageous so that the relationship between breastfeeding and co-sleeping could be better understood.

The relationship of co-sleeping and breastfeeding must be examined with use of multiple research modalities. If parental recall is utilized to gather data on co-sleeping practices, the data should be collected immediately after the event occurred. For example, if one is examining how many times an infant wakes to feed while co-sleeping, the parent should report these occurrences immediately the next morning so as to minimize errors in recall.

Finally, especially among low-income African-American families—a group found to have relatively high rates of co-sleeping but low rates of breastfeeding—there is a lack of knowledge as to why families choose to co-sleep or not. It is unclear whether the decision to co-sleep is linked to a lack of separate sleeping surfaces, to insufficient blankets or other materials for warmth, to a desire to be physically closer to the infant, or to other reasons. Focus groups and/or case studies may be appropriate for expanding on this question, yet the experiences of numerous subjects would need to be investigated in order to report accurate trends.

Conclusion 

Whether infants are safer in bed with their parent(s) or alone in a crib remains open for further exploration. Most likely, this multifaceted controversy will continue to be contested until sufficient research can support or undermine the risks and benefits associated with co-sleeping. At this point, the literature lacks ample data on co-sleeping rates both worldwide and in the United States, especially among distinct geographical and racial/ethnic groups. Furthermore, although the data on its relationship to breastfeeding is not conclusive, the literature appears to support the notion that co-sleeping promotes breastfeeding.

Because the opportunities to co-sleep with an infant may arise multiple times each day, it is essential that parents be informed of the risks and benefits, as well as how to maximize safety if they choose this practice. PNPs have the responsibility to question clients about their infant’s sleeping locations and to provide accurate, evidence-based information on this topic. In addition, as breastfeeding advocates, PNPs should be reminded of how co-sleeping could positively affect breastfeeding among their patient populations.