Evidence-Based Infant Sleep Recommendations

Dear Editor:

In “Speaking Out on Safe Sleep: Evidence-Based Infant Sleep Recommendations,” Bartick and Smith ¹ assert that the sleep recommendations of the American Academy of Pediatrics (AAP) overemphasize the risks of bedsharing, while minimizing other important risks, which may have “serious unintended consequences.” They suggest that the risk for death while bedsharing is because of additional risk factors, that routine bedsharing does not increase sudden infant death syndrome (SIDS) risk, and that the recommendation to avoid bedsharing interferes with breastfeeding. They conclude that “the only evidence-based universal advice is that sofas are hazardous places for adults to sleep with infants, that exposure to smoke, both prenatal and postnatal, increases the risk of death, and that sleeping next to an impaired caregiver increases the risk of death.” ¹

As members of the AAP Task Force on SIDS, we respectfully disagree on several points. The authors list several “leading” risk factors for SIDS and accidental suffocation (e.g., smoking, prone sleeping, formula feeding, infant sleeping unattended, poverty, sofa sleeping, and parental use of alcohol/drugs). These are important risk factors. However, it is unclear why the authors omitted soft bedding (adjusted odds ratio [aOR]=3.1–6.7), which is highly associated with bedsharing and increases the risk more than smoking (aOR=1.6–3.8) and bedsharing with someone who has consumed alcohol (aOR=1.7).

Blabey and Gessner ² found that 99% of bedsharing deaths were associated with other risk factors, implying that as long as those risk factors are not present, it is safe to bedshare. However, they also found that 40% of bedsharing mothers reported the presence of one or more risk factors. Parents assume that what they are doing is safe if nothing bad happens. It is only when an infant dies that parents realize that it was not in fact safe—and the 40% suddenly become the 99% with “additional risk factors” contributing to the death. We have, in our policy statement, emphasized factors that increase the risk of bedsharing, strongly recommending against these practices. However, as demonstrated in this study, there continues to be a great deal of high-risk bedsharing.

It is yet unclear if routine bedsharing increases SIDS risk. In the meta-analysis of Vennemann et al., ³ which Bartick and Smith ¹ cite, routine bedsharing did not increase risk (aOR=1.42; 95% confidence interval [CI], 0.85–2.38), but only two studies in the analysis had data on that variable. In contrast, the odds ratio in this same study for bedsharing with an infant <12 weeks of age (with nonsmoking parents) was 10.37 (95% CI, 4.44–24.21) (three studies provided data). Unfortunately, no studies have been able to analyze the risk for routine bedsharing in infants <12 weeks of age. However, there is strong evidence that bedsharing, compared with roomsharing without bedsharing, is a risk factor for SIDS among breastfed infants <12 weeks of age without other known risk factors (aOR=5.1; 95% CI, 2.3–11.4). ⁴

It is also unclear whether recommendations to roomshare (i.e., infant in a crib, bassinet, or bedside sleeper immediately adjacent to the mother) interfere with breastfeeding duration. To our knowledge, no large, quantitative studies have compared breastfeeding duration while roomsharing versus bedsharing. The studies cited as “compelling data” compared bedsharing versus not bedsharing (the infant in a separate room or the same room as the mother) or are qualitative. Data comparing roomsharing and bedsharing will be important to obtain. Until then, it is erroneous to assume that current recommendations interfere with breastfeeding. On the contrary, between 2000 and 2010, the same period in which recommendations against bedsharing were made, rates for breastfeeding initiation, at 6 months, and at 12 months in the United States have increased from 70%, 35%, and 16%, respectively, to 79%, 49%, and 27%, respectively. ⁵

The authors imply that more infant deaths occur on sofas than on adult beds. Although the risk of a sleep-related death is higher on a sofa, many more infants die on adult beds. A recently published analysis of U.S. infant deaths on sofas found that 50.5% and 12.9% occurred on adult beds and sofas, respectively. ⁶

We agree that “the issues around safe sleep are complex. No infant sleep environment is completely safe.” However, our goal is to ensure that parents have the most recent and unbiased evidence to inform their choices. Our policy statement strongly encourages breastfeeding and addresses all the issues and risk factors that Bartick and Smith ¹ discuss. Further research is needed to understand why parents continue to bedshare, particularly in high-risk ways. Although it is essential to acknowledge the benefits of bedsharing, we must also help parents understand that these potential benefits may be outweighed by the increased risk of infant death during bedsharing. Finding ways to communicate these messages effectively is our current challenge.