Babies in boxes and the missing links on safe sleep: Human evolution and cultural revolution

Abstract

Concerns about bedsharing as a risk for sudden infant death syndrome and other forms of sleep‐associated infant death have gained prominence as a public health issue. Cardboard “baby boxes” are increasingly promoted to prevent infant death through separate sleep, despite no proof of efficacy. However, baby boxes disrupt “breastsleeping” (breastfeeding with co‐sleeping) and may undermine breastfeeding. Recommendations enforcing separate sleep are based on 20th century Euro‐American social norms for solitary infant sleep and scheduled feedings via bottles of cow's milk‐based formula, in contrast to breastsleeping, an evolutionary adaptation facilitating the survival of mammalian infants for millennia. Interventions that aim to prevent bedsharing, such as the cardboard baby box, fail to consider the implications of evolutionary biology or of ethnocentrism in sleep guidance. Moreover, the focus on bedsharing neglects more potent risks such as smoking, drugs, alcohol, formula feeding, and poverty. Distribution of baby boxes may divert resources and attention away from addressing these other risk factors and lead to a false sense of security wherein we overlook that sudden unexplained infant deaths also occur in solitary sleep environments. Recognizing breastsleeping as the evolutionary and cross‐cultural norm entails re‐evaluating our research and policy priorities, such as providing greater structural support for families, supporting breastfeeding and safe co‐sleeping, investigating ways to safely minimize separation for formula‐fed infants, and mitigating the potential harms of mother–infant separation when breastsleeping is disrupted. Resources would be better spent addressing such questions rather than on a feel‐good solution such as the baby box.

1. INTRODUCTION

Concerns about bedsharing as a risk for sudden infant death syndrome (SIDS) and other forms of sleep‐associated infant death have gained prominence as a public health issue. Most recently, interest in the cardboard “baby box” as a way to promote separate sleep has grown enormously, with baby box distribution programs now being instituted in several U.S. states. However, the baby box and guidelines around infant sleep must be re‐evaluated when placed in an evolutionary and sociohistorical context. The baby boxes and current infant sleep guidelines that emphasize the avoidance of bedsharing fail to consider the potential harmful consequences of mother–infant night‐time separation, including its impact on breastfeeding.

For generations, conventional wisdom has held that “sleeping like a baby” means that babies sleep long, deeply, and alone. These assumptions, and expert medical advice around infant sleep, were predicated upon and reinforced by nighttime separation of mothers and babies and scheduled infant feeding via carefully measured amounts of formula in bottles derived from the milk of the cow. These practices, however, are recent Euro‐American historical inventions (Tomori, 2014; Wolf, 2003). The species‐specific norm for infant feeding is breastfeeding, and breastfeeding comprises a sum total of human behaviour that is more than just nutrition. Indeed, anthropologists James McKenna and Lee Gettler (2015) have argued that “The mother's body provides the only environment to which the human neonate infant is adapted.” Breastfeeding cannot be separated from other infant activities such as sleeping or being held or carried. Drawing on data showing the mutually reinforcing relationship of breastfeeding and shared mother–infant sleep, McKenna and Gettler (2015) coined the term “breastsleeping” to connote that breastfeeding and (safe) co‐sleeping are part of the same process. By recognizing breastsleeping as an evolutionary adaptation that has contributed to the survival of our species and the ethnocentrism entailed in much of current U.S. and western infant sleep guidance, we must completely reframe how we examine current public health interventions and research questions on this topic.

The distribution of “baby boxes,” cardboard boxes filled with baby supplies which, when emptied, can be used as an infant sleeping environment, is an increasingly popular intervention introduced in North America and elsewhere to promote separate sleep and is generally paired with some form of education on safe sleep. Baby box programs are based on a Finnish government program initiated in the 1930s in which baby clothes and related items were given to mothers who attended prenatal appointments. The boxes also provided a safe place outside of parents' beds for infants to sleep, especially in homes that might have only rudimentary furniture (Rosenberg, 2016), as poverty was common.

Although U.S. initiatives assume that sleeping in the boxes has resulted in lower infant mortality, no evidence to date supports this assumption. Experts in recent media reports have questioned the underlying reasons for the significant drops in infant mortality in Finland, which were observed over the period since the boxes were introduced (Cassin, 2017; Hafner, 2017) and note that less than half of Finnish babies currently sleep in the boxes (Hafner, 2017). Indeed, the nearby nations of Sweden, Norway, and Iceland never introduced baby boxes but have had similarly low infant mortality rates (Organisation of European Co‐operation and Development, 2016). They share with Finland universal health care systems, social safety nets, and paid maternity leave policies. Moreover, although many baby box initiatives claim that the Finnish program was designed to combat poverty‐associated infant mortality, it was actually designed to incentivize women to get screened and treated for prenatal syphilis, which was an epidemic at that time (Weeks, 2016). Furthermore, in most U.S. programs, it is assumed the distribution of baby boxes for infant sleep only confers advantages. Few questions have been raised about how these boxes may affect breastfeeding mothers and infants or introduce new hazards. The distribution of baby boxes also fails to consider the evolutionary context of normal nighttime infant behaviour, reinforces cultural historical norms about nighttime mother–infant separation and artificial feeding, and may inadvertently undermine breastfeeding.

Two related devices designed with lower sides for use in the adult bed in situations where high risk precludes bedsharing bear mentioning. The Pepi‐Pod from New Zealand and Australia (Bartholomew, 2017) and the Wahakura from New Zealand were both designed to address high sleep‐related infant death rates in indigenous populations associated with smoking and other risk factors (Baddock et al., 2017). The Wahakura, which is a woven flax bassinet‐like sleeper developed in the Maori community in 2006, was shown be associated with an increased in sustained breastfeeding compared to a traditional bassinet and a decrease in direct bedsharing (Baddock et al., 2017).

Baby boxes promote a separate sleep surface, following longstanding guidelines from the American Academy of Pediatrics (AAP) against bedsharing, even though AAP experts are not necessarily endorsing the boxes (Cassin, 2017). In its most recent infant sleep guidelines from October 2016, the AAP continues to recommend separate sleep surfaces for all mothers and babies, including those who are breastfeeding (Task Force On Sudden Infant Death, 2016). The AAP guidelines are historically predicated on the assumption that the normative culture is one where infants sleep alone and are fed artificially. These assumptions are reflected in the fact that the AAP issues separate guidelines for breastfeeding and for infant sleep. The AAP has recently acknowledged that proximity to mother matters for health (Feldman‐Winter, Goldsmith, Committee, & Newborn, & Task Force On Sudden Infant Death, 2016; Task Force On Sudden Infant Death S, 2016) and that breastfeeding matters for health (American Academy of Pediatrics and Section on Breastfeeding, 2012), but they are each discussed only as risk reduction strategies in guidance on reducing childhood morbidity and mortality. If breastsleeping were treated as the norm, these guidelines would be integrated, and instead, we would be asking about the risks of separation from mother, not solely the risks of sleeping with one's infant. The AAP is considering mother–infant behaviour in the context of only the last century or so of United States and Western European history, not in an evolutionary context where breastsleeping has been the norm and a survival strategy for not only humans but also primates and many other mammals. However, despite a cultural revolution, maternal–infant biology that was forged in our evolutionary past has not altered in the most recent 100 years.

Evidence shows that the bedsharing recommendations from AAP are not working. In the most recent study, led by one of the lead authors of the AAP recommendations, Rachel Moon, investigators conducted a randomized control trial using enhanced messaging to African American families to avoid bedsharing, and yet bedsharing actually increased in the intervention group (Moon et al., 2017). In another survey consisting of a survey of nearly 5,000 mothers, fully 25% of mothers reported falling asleep with their infants in hazardous locations such as sofas and recliners, many having done so in order to avoid bedsharing (Kendall‐Tackett, Cong, & Hale, 2010).

Much of the attention on preventing sleep‐associated infant deaths has focused on bedsharing, despite this being of debatable to no increased risk when other risk factors are absent (Bartick & Smith, 2014; Blabey & Gessner, 2009; Blair, Sidebotham, Pease, & Fleming, 2014). A recent New Zealand study showed that it is the combination of antenatal smoking with bedsharing that markedly increased the risk of sudden unexplained death in infants, whereas either by itself was associated with a non‐significant increased risk (Mitchell et al., 2017). The leading risk factors for sleep‐associated sudden and unexpected infant deaths include parental smoking, sleeping prone, falling asleep with an infant on a sofa or recliner, sharing a bed with an adult who is under the influence of drugs or alcohol, and formula feeding (Bartick & Smith, 2014). If it is an independent risk factor at all, the risk of bedsharing is tiny in comparison to the above mentioned risks (Bartick & Smith, 2014; National Institute of Health Care and Excellence, 2015). In an attempt to reduce sudden infant death syndrome, the U.S. medical establishment has come down hard on bedsharing, with multiple public health campaigns aimed at discouraging the practice. The “Alone. Back. Crib.” campaigns are especially popular: Infants should sleep alone, on their back, in a crib. Attention to the other risk factors is essentially neglected, even though they carry more substantial risks for infant death. In contrast, the latest UK guidance acknowledges that there is insufficient evidence to say that bedsharing causes SIDS and offers information on the elimination of bedsharing hazards, not of bedsharing itself (Ball, 2017).

The “Back to Sleep” campaign (to place babies in the supine position for sleep) has been associated with a reduction in sleep‐related infant deaths, a decline that began even prior to the 1992 campaign (Pelligra, Doman, & Leisman, 2005). Yet it is important to note that sleeping prone is not in the behavioural repertoire of normal human breastsleeping infants (McKenna & Gettler, 2015; Richard, Mosko, McKenna, & Drummond, 1996), and this important recommendation arose out of the recent cultural context of solitary sleep and artificial feeding, in which infants were frequently placed prone alone in their cribs. In the only video study done comparing bedsharing formula feeding and breastfeeding infants, the formula feeding infants were more likely to have their heads placed level with their mother's face, whereas the breastfeeding infant's head was placed at breast‐level, and the breastfeeding mothers spent more time turned towards their infants, who also faced their mothers, whereas the formula‐fed infants spent more time sleeping on their backs (Ball, 2006). Despite what appears to be potential risks for suffocation from pillows in this small study among the formula‐fed bedsharing infants, another recent study found that bedsharing (even with formula feeding infants included) was not associated with increased risk of death in the absence of other risk factors (alcohol, drugs, and sofa‐sleeping) (Blair et al., 2014). More research is needed to know if the different bedsharing positioning contributes to any increased risk of sleep‐associated infant death in formula feeding infants, and, if so, if bedsharing with formula feeding infants could be done in such a way that could minimize such risk.

In addition, because separate sleep can undermine breastfeeding, and baby boxes promote separate sleep, the boxes could hinder contact between breastfeeding mothers and infants and lead to early weaning. Research shows that bedsharing breastfeeding mothers nurse their infants 5.75 times during the night (often without realizing it), compared to 2.5 times a night for moms and babies who do not share a bed (McKenna, Mosko, & Richard, 1997). This increased breastfeeding is especially important for a mother to maintain a robust milk supply (Hartmann, Sherriff, & Mitoulas, 1998) and remain anovulatory so that her children are widely spaced (Labbok et al., 1997). The perception of low milk supply is one of the most important reasons women give for stopping breastfeeding (Ball et al., 2016). Persuading mothers not to breastsleep through separate sleep and baby boxes may thus undermine milk supply and result in difficulty attaining breastfeeding goals (Ball, 2003). Early weaning puts both the mother's and the infant's health at risk (Bartick et al., 2016; Chowdhury et al., 2015; Victora et al., 2016).

It is important to look at current infant sleep recommendations in their historical, physiological, and cultural context. Human milk is digested very quickly, and the rapidly growing infant needs to eat every 2 to 3 hr (Casiday, Wright, Panter‐Brick, & Parkinson, 2004; De Carvalho, Roberson, Friedman, et al., 1983). Such a feeding pattern would be difficult if the infant was not in constant contact with his breastfeeding mother, day and night. Indeed, ethnographic studies have shown that in traditional cultures all over the world, mothers and babies are in prolonged contact, being carried by day, sleeping together at night, and nursing at will for the first several months (Barry & Paxson, 1971). After that, infants remain in contact with other caregivers. Even with the return to breastfeeding over the last few decades in the United States and the growing emphasis on breastfeeding in the public health literature, we have often grown focused on how formula differs from the components of breast milk and on the delivery of expressed milk. In doing so, we miss the connection that breastfeeding is about physical and emotional contact as much as it is about the milk itself.

Contrast the human physiologic pattern of frequent feeding with that of cows, the primary source of food upon which artificial feeding is based. On some farms, nursing calves are separated from their mothers and are routinely allowed to suckle only 2 or 3 times a day (Alvarez‐Rodriguez, Palacio, Casasus, Revilla, & Sanz, 2009; Bar‐Peled et al., 1997; Conneely et al., 2014). The higher protein in the cows' milk allows these calves to grow normally whereas nursing far less frequently than a human infant would require. Having one fourth the protein content of cows' milk (Hernell, 2011), human milk is digested very quickly, and the rapidly growing infant needs to eat every 2 to 3 hr and will awaken to do so. Differences in milk composition may be why we see that baby humans fed cows' milk products are less arousable from sleep than babies who nurse from their mothers (Tikotzky et al., 2010). This difference in arousal levels may partly explain the higher risk of sudden infant death syndrome in infants who are fed formula (Horne, Parslow, Ferens, Watts, & Adamson, 2004). Bedsharing among breastfeeding dyads, in particular, is associated with more infant arousability compared to breastfeeding solitary sleepers, both by increasing arousability during Stages 3–4 sleep and by decreasing the amount of Stage 3–4 sleep, both of which are particularly important considering that normal infant arousability is greatly diminished in Stages 3–4 sleep (Mosko, Richard, & McKenna, 1997a). The differences between breastmilk and the composition of cow's milk‐based formula may also explain why parents of formula‐fed infants report more consolidated sleep (Ramamurthy et al., 2012).

Concerns about infant sleep were virtually absent until the late 19th and especially the early 20th centuries. Historians Stearns and colleagues demonstrated that “Manuals for parents, produced in abundance [in the mid 19^th century], simply did not deal with children's sleep in marked contrast to their counterparts by the 1920s. Medical researchers did not study sleep extensively ‐ another point noted when research began to accelerate a half‐century later” (Stearns, Rowland, & Giarnella, 1996, pp. 345–46). They seem to have arisen as a result of the development of solitary infant sleep in the late 19th and especially the early 20th centuries as an ideal among certain middle class cultural groups, facilitated by medical experts (Stearns et al., 1996). Medical experts also played a crucial role in the normalization of scheduled artificial feeding, cemented by the growing number of mothers giving birth in hospitals and the industrial production and marketing of cows' milk‐based breast milk substitutes. In the 1917 edition of a popular manual (Holt, 1917), which became the basis of the infant care pamphlet distributed by the government to millions of parents, Dr. Emmett Holt recommended that babies sleep in nurseries separate from their mothers and that they be fed only once or twice during the night through the first 4 months and then once between 4 and 7 months (Tomori, 2017b (in press)‐b). Thereafter, they were not to be fed at all during the nighttime. If infants awoke during the night and were not scheduled for a feeding, they were to “cry it out” for up to 2 to 3 hr. Experts like John Watson and Benjamin Spock further developed these ideas about “training” infants to sleep alone by “crying it out.” As late as 1976, Dr. Benjamin Spock (first published in 1946) wrote that a healthy 1‐month old infant should be able to sleep through the night and should be left to cry for up to half an hour if he woke; and as late as 1992, wrote about the “tyranny” of children who would not fall asleep when put down alone in the crib (Spock & Rothenberg, 1992). Dr. Richard Ferber further popularized sleep training starting in 1986, purposely leaving infants alone for progressively longer periods to “cry it out” (Ferber, 2006; Tomori, 2014). Because of their nutritional and emotional needs, very young infants are neither physiologically nor psychologically adapted to either sleep training nor “crying it out.”

“Crying it out” occurs in a context in which western parents have come to seeing crying infants as normal, in a society where infants are routinely separated from direct physical contact with their caregivers both at night and in daytime. Yet anyone who has spent time in the developing world, particularly Africa, where infants are carried and strollers are not a part of life, will have witnessed that it is rare to see a baby crying in public (Bleah & Ellett, 2010). Even in the United States, as hospitals become Baby‐Friendly and infants are kept in proximity and skin‐to‐skin, one of us (MB) frequently hears staff at many hospitals make remarks such as, “we never hear crying any more. Our unit is so much more quiet now.” Hospital staff frequently note that the unusual sound of baby crying on units that have eliminated maternal–infant separation will trigger their immediate concern, whereas before such crying was often disregarded and thought of as normal. Such observations illustrate the cultural context in which acceptability of crying and infant distress occurs where separation of mothers and infants is also considered normal; it is not until mothers and babies are routinely together that one realizes that crying appears unusual and people become more sensitized to the sound of a distressed, crying infant.

Although “cry it out” is hotly contested among parents, various forms of “sleep training” and the emphasis on “self‐soothing” and “sleeping through the night” remain prominent in parenting advice in 2017. SIDS is now added to the list of reasons why infants should not share a bed with their parents, and other reasons have taken more of a back seat. At the same time, cultural worries also linger in many parents' discomfort with sharing a bed with their babies and their concerns about needing to get their babies to sleep through the night in their own room (Tomori, 2014). In the 2016 AAP guidelines, room‐sharing has been emphasized as a risk reduction for SIDS, but even this recommendation has received some backlash (Fallon, 2016) in a context where “sleeping through the night” in the baby's own room is considered necessary to achieve “independence” (Tomori, 2014).

Finally, in addition to safety concerns about the need for separate sleep, most public health guidelines, including the AAP guidelines, ignore the role of poverty in sleep‐related infant death. The risk of such death is higher in socioeconomically disadvantaged families in the UK, although they are less likely to share a bed with their infants (Blair, Heron, & Fleming, 2010). In the United States, rates of SIDS are higher among Black and Hispanic infants than in Whites, groups who have lower income levels on average, and the higher SIDS rate is partially explainable due to lower rates of any and exclusive breastfeeding in both groups after the post‐partum period (Bartick et al., 2017).

Thus, given that breastsleeping is the evolutionary and cross‐cultural norm, we must ask another fundamental question: What, if any, are the consequences of separating parents from infants? Does it harm children and/or parents when we assume that babies can and should sleep apart from their parents? Could a separate sleep surface cause other harms besides the undermining of breastfeeding? Could sleeping in a baby box cause harm just from mother–infant separation itself?

We know that separation from maternal skin‐to‐skin contact for even an hour can have profound physiological stress on 2‐day‐old infants (Morgan, Horn, & Bergman, 2011). Research in a small study of 4‐ to10‐month‐old infants also shows that after separation into different rooms for sleep, infant and maternal levels of the stress hormone cortisol are high, and the mothers respond to the infant's cries and signs of distress. However, after a few nights of separation for sleep, the infant cortisol levels remained high, indicating physiologic stress, even though they no were no longer crying and appeared to “self‐settle.” In contrast, the mothers' cortisol levels decreased and were no longer correlated with those of their infants, reflecting that they were unaware of their infants' stress and were out of sync with them (Middlemiss, Granger, Goldberg, & Nathans, 2012). Prolonged childhood stress can create long‐term changes in brain architecture and behaviours that could even be passed on to the next generation, in a phenomenon known as “toxic stress” (Shonkoff et al., 2012). We do not know if the stress caused by separation, such as that seen in newborns, would abate over time, or how much stress would need to occur in an infant to result in brain changes associated with “toxic stress.” At the very least, we cannot assume that enforced separate sleep is without harm to the infant through the repeated stress of being separated from the only environment he has ever known and to which he is uniquely adapted. Finally, separation may also make it difficult for parents to address other potential threats. For instance, research shows anecdotal reports of parents saving their children from acute life threatening events that would have gone undiscovered that the children been sleeping alone (McKenna & Volpe, 2007).

In addition to safety concerns about bedsharing, modern questions have arisen around the values surrounding parenting that affect assumptions about sleep practices: Should the parents' needs be subsumed by those of their children? When we look cross‐culturally, not every culture sees keeping the child in contact with the parent as a conflict. Breastsleeping does not inherently constitute a greater burden on parents, nor does it have to correspond with a parenting philosophy where mother's needs are subsumed to those of the child. Parents just get more rest (Doan, Gardiner, Gay, & Lee, 2007; Montgomery‐Downs, Clawges, & Santy, 2010), even though co‐sleeping mothers may experience more sleep fragmentation (Volkovich, Ben‐Zion, Karny, Meiri, & Tikotzky, 2015). Many breastsleeping mothers are simply not aware of how many times they nursed their babies throughout the night, because they were not fully awake (Gottlieb, 2004; Morelli, Rogoff, Oppenhein, & Goldsmith, 1992; Tomori, 2014). Bedsharing breastfeeding mothers spend more time in Stages 1 and 2 sleep (lighter sleep) and less time in Stages 3 and 4 sleep (deeper sleep) than solitary sleeping breastfeeding mothers (Mosko, Richard, & McKenna, 1997b), which may facilitate responsiveness to the infant. When some babies awaken frequently to nurse, mothers may find it unpleasant, but do not perceive it as problematic (Gottlieb, 2004, Tomori, 2017a (in press)‐a). Mothers are also not usually left to care for their infants alone and to be completely responsible for all other tasks right after birth. There may be periods of mandated rest for the first 40 days or so, and often there are others to help support mothers and care for infants and young children (Eberhard‐Gran, Garthus‐Niegel, Garthus‐Niegel, & Eskild, 2017).

Breastfeeding helps mothers and infants both quickly fall asleep due to hormones released in the mother's brain (oxytocin) and hormones in the milk itself, yet both mother and baby are easily aroused, which is not the case if the pair are not breastfeeding (Blyton, Sullivan, & Edwards, 2002; Horne et al., 2004). In breastsleeping dyads, their sleep cycles are synchronized, and the infants' airway is naturally protected from blankets and pillows by the infant's position with his head across from mother's breasts, her arm, and shoulder forming a natural barricade from a potentially smothering pillow (Ball, 2006).

We see that breastsleeping is an elegant dance between mother and infant. Both mother and child benefit from the close physical contact and increased breastfeeding, physical warmth, and emotional connection. This process, honed through millennia, cannot happen with a baby in a box. Although it may seem like a simple, if costly, solution to give out baby boxes, we should not expect this to solve sleep‐related infant mortality problems, when the key underlying problems are access to health care, poverty, and lack of support for breastfeeding or for treatment for tobacco and substance use. Spending proportional resources on the most important risk factors for infant death, such as smoking, substance use, and formula feeding are likely to have greater impact than a feel‐good solution like a box. If resources are diverted from these efforts to baby boxes and if emphasis on sleeping in the boxes comes at the expense of breastfeeding, the boxes may have the potential to increase both maternal and child morbidity and mortality if they result in early weaning (Bartick et al., 2016; Chowdhury et al., 2015; Victora et al., 2016). Further, discouraging breastsleeping may have other developmental harms to the infant from stress that have not yet been studied.

Moving forward, our frame of reference in determining risk and public policy to manage risk must be normative human physiology, not an artificial intervention based on the physiology of solitary sleeping infants being fed the milk of another species in a bottle. We should address how we can better support safe breastsleeping and investigate the potential harms of disrupting breastsleeping and mother–infant separation. We should also develop advice specific for parents of formula feeding infants and support them in safely achieving closer human contact. In addition to lowering risky behaviours themselves, interventions in high‐risk families, such as the Pepi‐Pod and Wahakura should be continued. Moreover, we need to be able to offer flexible guidance for families who combine breastfeeding and formula feeding and who are transitioning from breastfeeding to formula feeding. Finally, we must address the role of poverty and lack of paid family leave in supporting new parents and the roles these may play in infant and maternal mortality related to breastfeeding and sleep‐related infant death. Instead of getting more babies into boxes, nighttime infant care guidance informed by evolutionary theory and cross‐cultural practices should foster greater opportunities for safe connection for all infants and their families.

Key Messages.

• Cardboard ”baby‐boxes” are increasingly promoted to prevent infant death through solitary sleep, despite no proof of efficacy.

• Such interventions are based on recent cultural innovations of solitary infant sleep and scheduled bottle‐feeding with cow's milk‐based formula.

• Boxes disrupt the evolutionary adaptation of breastfeeding with co‐sleeping—”breastsleeping”, may undermine breastfeeding, and divert resources away from addressing more potent risk factors for infant death.

• Instead of distributing boxes, we should consider sleep and breastfeeding as one integrated evolutionary process, develop support for safe breastsleeping, and examine the consequences of mother–infant separation.

CONFLICTS OF INTEREST

Dr. Bartick has no competing conflicts of interest regarding breastfeeding and infant sleep. Her current funding is for breastfeeding and economics research and comes from the WK Kellogg Foundation. Dr. Tomori has no current sleep‐related funding. Her past funding came from The Alfred P. Sloan Center for the Ethnography of Everyday Life; the Eunice Kennedy Shriver National Institute of Child Health and Development; the Rackham Graduate School, the Department of Anthropology, and the Center for the Education of Women at the University of Michigan. Dr. Ball's recent research was funded by the Lullaby Trust, the Scottish Government, the UK Economic and Social Research Council (ERSC), and the University of Durham. Previously research has been funded by the ESRC, the UK National Institute for Health Research, and UK health and medical charities.

CONTRIBUTIONS

MB conceived of the manuscript with CT, and MB was primarily responsible for constructing the drafts. HB and CT were responsible for shaping and editing the manuscript and contributing their expertise as anthropologists and experts in breastfeeding and infant sleep.

Bartick M, Tomori C, Ball HL. Babies in boxes and the missing links on safe sleep: Human evolution and cultural revolution. Matern Child Nutr. 2018;14:e12544 10.1111/mcn.12544