Sleep Environment Risks for Younger and Older Infants

Thank you for the opportunity to respond to the letter from Dr. Cutz. His letter stated that no distinction was made between different infant death categories and that the study did not take into account possible health problems in the infants who died. Although we provided the proportion of deaths that were attributed to specific causes of death, we analyzed the associated factors for the group as a whole.[1] Because the modifiable risk factors for the 3 most common causes of sleep-related infant death are remarkably similar, we felt that it was important to identify these risk factors across categories of sleep-related infant deaths, regardless of the specific "official" diagnosis. Second, we did adjust for known health problems, including genetic and metabolic disorders, in the cohort. Specifically, we adjusted for Complex Chronic Conditions, which are "any medical condition that can be reasonably expected to last at least 12 months (unless death intervenes) and involving either several different organ systems or one system severely enough to require specialty pediatric care and probably some period of hospitalization in a tertiary care center."[2] We also should reemphasize that 97.6% of the cohort had an autopsy performed; if a health problem was felt to contribute to the death was discovered by the medical examiner or coroner, this would have been indicated in the autopsy report, and this information was included in the database. We cannot speculate on whether additional unknown molecular or genetic abnormalities existed in the cohort, let alone attribute deaths to them.

Dr. Cutz was also concerned that the percentage of infants with the diagnosis of accidental suffocation and strangulation in bed (ASSB) was higher than that reported in other articles, specifically citing Shapiro- Mendoza et al.[3] The difference in our proportion of deaths attributed to ASSB (26.8%) compared to Shapiro-Mendoza (12.5% in the final year of that study) is likely due to different data sources. First, we acknowledge that the dataset does not contain all sleep-related deaths in the reporting states; it should therefore not be considered a population-based dataset. However, our data were derived from multidisciplinary state child death review teams that were able to comprehensively review all known aspects of a child's death, including past medical history, autopsy results, and death scene investigation. In contrast, Shapiro-Mendoza et al. relied on data derived from death certificates. Those authors noted that in many cases the information provided on the death certificate was limited, lacking some of the data from the autopsy reports and death scene investigation, and that cause of death determination was often pending when the death certificate was filed. Shapiro-Mendoza, in a more recent study using a combination of data similar to the data sources in our study, including medical examiner, law enforcement, and hospital records, found that 47.9% of deaths were attributed by the medical examiner to ASSB.[4] Even the article cited by Dr Cutz (Pasquale-Styles et al) found that "if the witnessed overlays, entrapments, strangulations and infants found with their noses and mouths blocked and/or their heads covered represent asphyxia deaths, then a minimum of 108 (51.7%) in this study died of asphyxia" and "this would represent a conservative estimate of asphyxia deaths, as it does not take into account infants who died in high -risk sleep situations."[5]

Dr. Cutz also found implausible the proposed mechanism of suffocation of the older age group, stating that "actual scene reconstruction studies are often inconclusive" and citing Pasquale-Styles et al. We disagree with this conclusion; again, Pasquale-Styles found that over half of the infant deaths in that study demonstrated a clear asphyxia mechanism.[5]

Finally, Dr Cutz expressed concern that our study and other recent studies "create the impression that the cause of SUID including SIDS has now been identified and a simple solution exists to prevent such death", thereby implying that further research into molecular and genetic etiologies to sleep-related infant deaths is unnecessary. That was certainly not our intention. We agree with Weese-Mayer et al. (cited by Dr Cutz) that "a number of genetically controlled pathways appear to be involved in at least some cases of SIDS."[6] Unfortunately, at this point in time, the technology to identify which infants have a genetic predisposition to SIDS is not available. Furthermore, as Dr. Cutz noted, there are likely both intrinsic and extrinsic risk factors for sleep- related infant deaths. The Triple Risk Model mentioned by Dr. Cutz posits that, if any of the 3 risks (vulnerable infant, critical period of development, and external stressors) is eliminated, infant death is unlikely to occur. Because we cannot yet identify which infants are vulnerable, and because the critical period of development is impossible to avoid, public health efforts have focused on eliminating the external stressors. Our analysis of risk factors was aimed at eliminating these external stressors; it was in no way intended to imply that the contribution of genetic causes is unimportant or to detract from the important work of investigators exploring genetic and molecular etiologies to sleep-related infant deaths.

Jeffrey D. Colvin, MD, JD

Vicki Collie-Akers, PhD

Christy Schunn, MSW

Rachel Y. Moon, MD

1. Colvin JD, Collie-Akers V, Schunn C, Moon RY. Sleep Environment Risks for Younger and Older Infants. Pediatrics. 2014:1-7. doi:10.1542/peds.2014-0401.

2. Feudtner C, Feinstein JA, Satchell M, Zhao H, Kang TI. Shifting place of death among children with complex chronic conditions in the United States, 1989-2003. JAMA. 2007;297(24):2725-2732. doi:10.1001/jama.297.24.2725.

3. Shapiro-Mendoza CK, Kimball M, Tomashek KM, Anderson RN, Blanding S. US Infant Mortality Trends Attributable to Accidental Suffocation and Strangulation in Bed From 1984 Through 2004: Are Rates Increasing? Pediatrics. 2009;123(2):533-539. doi:10.1542/peds.2007-3746.

4. Sauber-Schatz EK, Sappenfield WM, Shapiro-Mendoza CK. Comprehensive Review of Sleep-Related Sudden Unexpected Infant Deaths and Their Investigations: Florida 2008. Matern Child Health J. 2014:1-10. doi:10.1007/s10995-014-1520-1.

5. Pasquale-Styles MA, Tackitt PL, Schmidt CJ. Infant Death Scene Investigation and the Assessment of Potential Risk Factors for Asphyxia: A Review of 209 Sudden Unexpected Infant Deaths. Journal of Forensic Sciences. 2007;52(4):924-929. doi:10.1111/j.1556-4029.2007.00477.x.

6. Weese-Mayer DE, Ackerman MJ, Marazita ML, Berry-Kravis EM. Sudden Infant Death Syndrome: Review of implicated genetic factors. Am J Med Genet. 2007;143A(8):771-788. doi:10.1002/ajmg.a.31722.

Conflict of Interest:

None declared

The paper by Colvin et al is yet another example of a study warning of the risks of sudden unexpected infant death (SUID) including sudden infant death syndrome (SIDS) by exposing infants to unsafe sleeping environment such as co-sleeping or the presence of toys and other objects in the sleep area. The study analysed 8207 cases of SUID categorised as SIDS (35% or 2873 cases) ,accidental suffocation and strangulation in bed (26% or 2196 cases) and as unknown /undetermined(38% or 3130 cases). The authors conclude that the predominant risk factor for younger infants is bed-sharing, whereas for older infants rolling into objects in the sleep area. However the study has several shortcomings. No distinction is made between different infant deaths categories , leaving an impression that all infants were perfectly healthy and died as a result of accidents that are preventable. The study does not take into consideration a possibility that these infants, especially SIDS harbor an intrinsic biological (genetic/molecular) defect that according to the triple risk model makes them vulnerable to extrinsic risk factors such as prone sleeping, mild upper respiratory infection, maternal smoking (other well documented SIDS risk factors) [1].The number of cases of accidental suffocation/strangulation in bed appears extraordinarily high considering that similar studies found that such cases accounted for between 2.8-12 % of SUID [2]. In addition, the mechanism of suffocation ie. "rolling over toys and objects " appear implausible since the actual scene reconstruction studies are often inconclusive in most such cases [3].

This and similar recent studies create an impression that the cause of SUID including SIDS has now been identified and a simple solution exists to prevent such death implying that further scientific research into biological underpinning of SIDS is unnecessary. This would be a grave mistake given recent advances in molecular studies suggesting that up to one third of SIDS cases could be based on demonstrable mutations affecting genes involved in neurotransmission, energy metabolism, autonomic control, response to infection and duration of cardiac action potential [4].Clearly a more balanced approach is required that on one hand provides sound advice to parents on safe sleep environment and on the other hand advocates molecular testing and supports robust scientific research into the true causes of SUID and SIDS. Pediatrics and American Academy of Pediatrics should be active participants in promoting these goals.

References.
[1] Kinney HC, Thach BT. The Sudden Infant Death Syndrome .New Engl J Med 2009;361: 795-805.
[2] Shapiro-Mendoza C K ,Kimball M,Tomashek KM Anderson RN, Blanding S. US infant mortality trends attributable to accidental suffocation and strangulation in bed from 1984trough 2004. Pediatrics 2009;123 :533-539.
[3] Pasquale-Styles MA,Tackitt P L,Schmidt CJ. Infant death scene investigation and the assessment of potential risk factors for asphyxia. J Forens Sci 2007;4 :924-929.
[4] Weese-Mayer DE, Ackerman MJ, Marazita ML, Berry-Kravis EM. Sudden infant death syndrome: review of implicated genetic factors. Am J Med GenetA 2007;143:771-788.

Conflict of Interest:

None declared

Thank you for your comments and the opportunity to clarify the purpose and findings of our study. The primary purpose of our study was to "compare differences in the sleep environments for younger (birth through 3 months) and older (ages 3 months to 364 days) infants who experienced sleep-related deaths."[1]

Both letters list variables that were not included in our analysis, including attributes of the adult bedsharer (e.g., use of alcohol, tobacco, and/or sedating substances), whether the infant was unattended, and feeding method. All of these are important considerations. However, the purpose of the study was to determine differences in sleep environments at different ages. There has been no prior evidence to suggest that the attributes of the adult who bedshares with an infant varies from early infancy to late infancy. Unfortunately, data on postnatal smoke exposure was, in our opinion, not reliable; the proportion of missing data was sufficiently high that we questioned the validity of multiple imputation to account for that. Our data source did not have information on whether the infant was breast fed or formula fed. However, we re-analyzed the data to help determine if the inclusion of prenatal exposure to maternal tobacco, illicit drugs, heavy alcohol use, or misuse of over-the-counter drugs into the regression models changed our findings for adult bed sleep place and bed-sharing. It did not. After adding those factors to the regression models, deaths in the younger age group were still associated with bedsharing (OR 1.9, 95% CI 1.7-2.1, P<0.001). Deaths in the younger age group also continued to be associated with sleeping in an adult bed (compared to sleeping in a crib, bassinet, or playpen) (OR 1.6, 95% CI 1.4-1.8, P<0.001) [in that model, prenatal maternal misuse of over-the-counter medications was removed due to Hessian singularities.] In those regression models, prenatal maternal tobacco use was associated with bedsharing (OR 1.3, 95% CI 1.1-1.5, P<0.01) and adult bed sleep place (OR 1.2, 95% CI 1.0-1.4, P<0.01); the other additional factors were not significant.

The authors of both letters discuss the dangers of sofas and recliners. Sofas, recliners, and similar furniture were not included in the "adult bed" category because the risk of an infant sleeping on a sofa is different from that of an infant sleeping on a bed; sofas, recliners, and similar furniture were included in the "other" category. We absolutely agree that sofa sleeping is an extremely hazardous practice for infants. Indeed, we have conducted a separate analysis of infant deaths that occur on sofas and hope to publish that soon. However, we disagree with Bartick et al's assertion that "the literature clearly shows that most accidental smothering/entrapment deaths involve sofas, recliners, or chairs." In the study referenced by Bartick et al, while 16% of deaths occurred on sofas, 38% occurred on beds.[2]

The authors of both letters state concerns that there was not a control group in this study. For studies involving SIDS and other sleep- related infant deaths, case-control studies have been the gold standard, as it is impossible to conduct randomized controlled trials. We acknowledge that this is not a case-control study. However, all risk factors considered in this study were examined for both study groups, younger and older infants. In addition, since the purpose of this study was to compare differences in risk factors between younger and older infants (and not to compare the degree of risk for any specific factor), our study used the older infant group as the comparison group.

Smith et al assert that SIDS and "smothering" are conflated in our study, that this study "focuses almost exclusively on the 'other sleep- related causes,'" and that we aggregated "safe and un-safe situations" together. We disagree. SIDS comprised 35% of the study population, while 26.8% and 38.2% were attributed to accidental suffocation and strangulation in bed (ASSB) and undetermined, respectively. Ultimately, our goal is to reduce infant mortality, not just from SIDS, but from all sleep-related infant deaths. As the rates of non-SIDS sleep-related infant deaths (including ASSB and Undetermined) continue to rise, there is some degree of diagnostic shifting with regards to these deaths, and most of these deaths occur in hazardous sleep situations, it is important to identify the factors that may contribute to these preventable deaths, regardless of the "official" cause of death. Furthermore, the primary statistical test conducted in this study examined the odds of having a risk factor present (i.e., "safe situation") compared to not having the risk factor present (i.e., "unsafe situation").

We agree that infants who are breastfed are at lower risk for SIDS and other sleep-related infant deaths than those who are not breastfed, and we wholeheartedly support breastfeeding promotion, not only because it reduces the risk of SIDS, but because it is the optimal nutrition source for infants. However, there is evidence that the risk of bedsharing for these infants is higher than the risk of room sharing without bedsharing. A recent meta-analysis of 5 large case-control studies (with 1472 SIDS cases and 4679 controls) examined the association between bedsharing and SIDS, with particular emphasis on lower-risk babies, i.e., those who were breastfed and whose mothers did not smoke. Even among these lowest risk infants, bedsharing infants were at a fivefold higher risk for SIDS up until the age of 3 months, when compared to infants who slept in the same room as their parent(s) but did not bedshare.[3]

Finally, we wholeheartedly agree with Bartick et al's statement that "public health efforts must address the reality that tired parents must feed their infants at night." While bedsharing facilitates breastfeeding, it is not essential for successful breastfeeding. Furthermore, we believe that the AAP recommendation that the infant sleep within arm's length in a bedside bassinet, portable crib, or crib, can provide optimal protection against sleep-related deaths, as it allows for easy access to the infant for breastfeeding and comforting without bedsharing. The goal should be for all infants to breastfeed (unless there are medical contraindications) and for no infants to bedshare.

Jeffrey D. Colvin, MD, JD

Vicki Collie-Akers, PhD

Christy Schunn, MSW

Rachel Y. Moon, MD

1. Colvin JD, Collie-Akers V, Schunn C, Moon RY. Sleep Environment Risks for Younger and Older Infants. Pediatrics 2014;134:e406-e412.

2. Blair PS, Sidebotham P, Evason-Coombe C, Edmonds M, Heckstall- Smith EM, Fleming P. Hazardous cosleeping environments and risk factors amenable to change: case-control study of SIDS in south west England. BMJ 2009;339:b3666.

3. Carpenter R, McGarvey C, Mitchell EA, et al. Bed sharing when parents do not smoke: is there a risk of SIDS? An individual level analysis of five major case-control studies. BMJ open 2013;3(5): pii:e002299.

Conflict of Interest:

None declared

The article by Colvin et al, "Sleep Environment Risks for Younger and Older Infants," incorrectly concludes that bedsharing is the top risk factor for sleep-related deaths.

The authors fail to include any of the other major known risk factors for these deaths in their analysis besides sleep location and position. By far the other leading risk factors for SIDS are maternal smoking1, infants sleeping unattended,1 and formula feeding.2 The risk of infant death from bedsharing, it is strongly potentiated by maternal prenatal and/or postnatal smoking.3 The use of alcohol and sedating substances by parents/caregivers poses the greatest risk of suffocation death3 but was not included in the study. The smoking variable was actually available in the authors' dataset, and their failure to utilize it is inexplicable and concerning, as it would have likely significantly altered their results.

The authors also failed to distinguish adult beds from far riskier types of sleeping surfaces, although the literature clearly shows that most accidental smothering/entrapment deaths involve sofas, recliners, or chairs.3

Because the authors did not use a control group, or population statistics regarding the prevalence of all factors stratified by age and smoking status at a minimum, it is impossible to draw any conclusions about the cause of sleep-related deaths from their study. Furthermore, 25% of the infants in this study actually died in cribs.

Increasing evidence shows that one of the major risk factors for SIDS is infant formula feeding, but infant feeding was not reported in this paper. A 2010 cost analysis links formula feeding to 911 excess infant deaths per year, 447 of which are from SIDS.4 The best, most recently conducted meta-analysis on SIDS and infant feeding method indicates that the number of SIDS deaths related to formula feeding is likely much higher than in the 2010 study2.

Evidence suggests blanket advice against bedsharing may be ineffective and may be counterproductive, directly contributing to infant deaths in at least some cases. A survey of nearly 5,000 US mothers found that sleepy parents are taking nighttime feedings to the sofa to avoid falling asleep with their infants in bed, and large numbers of these parents are unintentionally falling asleep there.5 Experts agree that sofas pose a much higher risk for infant death than beds.1,3 Parents of two SIDS babies exactly described exactly this scenario, unaware that couches are far more dangerous places for infants to sleep than parental beds.3

The conclusions of this flawed study may be used to support the American Academy of Pediatrics' recommendation against all bedsharing. Instead, public health efforts must address the reality that tired parents must feed their infants at night somewhere, and that sofas are very dangerous for all infants, but especially for those of parents who are smokers or under the influence of alcohol or drugs. Public health messages must also acknowledge that breastfeeding infants and formula feeding infants do not sleep the same way when bedsharing, and that formula feeding infants have a higher risk of death.

References

1. Moon RY. SIDS and other sleep-related infant deaths: expansion of recommendations for a safe infant sleeping environment. Pediatrics. Nov 2011;128(5):1030-1039.
2. Hauck FR, Thompson JM, Tanabe KO, Moon RY, Vennemann MM. Breastfeeding and reduced risk of sudden infant death syndrome: a meta-analysis. Pediatrics. Jul 2011;128(1):103-110.
3. Blair PS, Sidebotham P, Evason-Coombe C, Edmonds M, Heckstall-Smith EM, Fleming P. Hazardous cosleeping environments and risk factors amenable to change: case-control study of SIDS in south west England. BMJ. 2009;339:b3666.
4. Bartick M, Reinhold A. The burden of suboptimal breastfeeding in the United States: a pediatric cost analysis. Pediatrics. May 2010;125(5):e1048-1056.
5. Kendall-Tackett K, Cong Z, Hale T. Mother-Infant Sleep Locations and Nighttime Feeding Behavior: U.S. Data from the Survey of Mothers' Sleep and Fatigue. Clinical Lactation. 2010 2010;1(Fall).

Conflict of Interest:

None declared

The article by Colvin et al, "Sleep Environment Risks for Younger and Older Infants," incorrectly concludes that bedsharing is the top risk factor for sleep-related deaths.

The authors fail to include any of the other major known risk factors for these deaths in their analysis besides sleep location and position. By far the other leading risk factors for SIDS are maternal smoking[1], infants sleeping unattended,[1] and formula feeding.[2] The risk of infant death from bedsharing, it is strongly potentiated by maternal prenatal and/or postnatal smoking.[3] The use of alcohol and sedating substances by parents/caregivers poses the greatest risk of suffocation death3 but was not included in the study. The smoking variable was actually available in the authors' dataset, and their failure to utilize it is inexplicable and concerning, as it would have likely significantly altered their results.

The authors also failed to distinguish adult beds from far riskier types of sleeping surfaces, although the literature clearly shows that most accidental smothering/entrapment deaths involve sofas, recliners, or chairs.[3]

Because the authors did not use a control group, or population statistics regarding the prevalence of all factors stratified by age and smoking status at a minimum, it is impossible to draw any conclusions about the cause of sleep-related deaths from their study. Furthermore, 25% of the infants in this study actually died in cribs.

Increasing evidence shows that one of the major risk factors for SIDS is infant formula feeding, but infant feeding was not reported in this paper. A 2010 cost analysis links formula feeding to 911 excess infant deaths per year, 447 of which are from SIDS.[4] The best, most recently conducted meta-analysis on SIDS and infant feeding method indicates that the number of SIDS deaths related to formula feeding is likely much higher than in the 2010 study.[2]

Evidence suggests blanket advice against bedsharing may be ineffective and may be counterproductive, directly contributing to infant deaths in at least some cases. A survey of nearly 5,000 US mothers found that sleepy parents are taking nighttime feedings to the sofa to avoid falling asleep with their infants in bed, and large numbers of these parents are unintentionally falling asleep there.[5] Experts agree that sofas pose a much higher risk for infant death than beds.[1,3] Parents of two SIDS babies exactly described exactly this scenario, unaware that couches are far more dangerous places for infants to sleep than parental beds.[3]

The conclusions of this flawed study may be used to support the American Academy of Pediatrics' recommendation against all bedsharing. Instead, public health efforts must address the reality that tired parents must feed their infants at night somewhere, and that sofas are very dangerous for all infants, but especially for those of parents who are smokers or under the influence of alcohol or drugs. Public health messages must also acknowledge that breastfeeding infants and formula feeding infants do not sleep the same way when bedsharing, and that formula feeding infants have a higher risk of death.

References

1. Moon RY. SIDS and other sleep-related infant deaths: expansion of recommendations for a safe infant sleeping environment. Pediatrics. Nov 2011;128(5):1030-1039.
2. Hauck FR, Thompson JM, Tanabe KO, Moon RY, Vennemann MM. Breastfeeding and reduced risk of sudden infant death syndrome: a meta-analysis. Pediatrics. Jul 2011;128(1):103-110.
3. Blair PS, Sidebotham P, Evason-Coombe C, Edmonds M, Heckstall-Smith EM, Fleming P. Hazardous cosleeping environments and risk factors amenable to change: case-control study of SIDS in south west England. BMJ. 2009;339:b3666.
4. Bartick M, Reinhold A. The burden of suboptimal breastfeeding in the United States: a pediatric cost analysis. Pediatrics. May 2010;125(5):e1048-1056.
5. Kendall-Tackett K, Cong Z, Hale T. Mother-Infant Sleep Locations and Nighttime Feeding Behavior: U.S. Data from the Survey of Mothers' Sleep and Fatigue. Clinical Lactation. 2010.

Conflict of Interest:

None declared

"Sleep Environment Risks for Younger and Older Infants" (Colvin et al, Pediatrics 2014-0401) fails to add any useful information to the SIDS/infant mortality dialogue. Like many other papers on sleep-related risks, this article lacks clear, unambiguous definitions of critical variables, omits significant documented risk factors, and conflates SIDS and other unrelated forms of infant death, drawing conclusions that the data do not support. Specifically: (1) The bed-partner could be human or animal, drunk or sober, smoker or not, adult or child, caring mother or inattentive other. A baby sleeping with an inebriated uncle is different from a baby sleeping next to his sober breastfeeding mother. (2) "Adult bed" included adult bed, waterbed, adult mattress, bunk bed, child's bed, sofa bed, and air mattress. (We can only presume that sofas and recliners are included. They are common shared-sleep choices, especially for mothers who have been told their bed is unsafe.) A baby sleeping on a soft sofa cushion next to that inebriated uncle is different from a baby sleeping with his breastfeeding mother in a firm, uncluttered bed. The categories "Adult bed" and "person" (defined as "sleeping on the chest or in the arms of another person") were combined as one category, though one is inanimate and of varying safety, and the other is animate and of varying safety. (3) Infant feeding method is never mentioned, despite the fact that formula-feeding is a risk factor for SIDS and affects the mother's behavior during bedsharing. (4) The words "smoking" and "smoke" do not appear in the article, yet smoking during pregnancy and in a baby's household is a significant risk for SIDS. The authors acknowledge that "because there is no comparison group, risk cannot be determined." Without a control group, actual risk cannot be calculated. (5) SIDS and smothering are conflated. Most SIDS risks have nothing to do with sleep location, while smothering and suffocation are entirely location-related. The distinction is especially important when bedsharing is discussed. SIDS is linked to physiological problems; suffocation results from environmental problems. (6) Bedsharing is cited as "the predominant risk factor for younger infants" without reference to other significant risk factors or data from comparison groups. (7) The study refers to "sudden infant death syndrome and other sleep-related causes of infant mortality" but focuses almost exclusively on the "other sleep-related causes" The combining of safe and unsafe situations in single categories is distressing but not new. The studies on which this article relies - the Ohio child death review for example*, - tend to do the same, a shortcoming that the authors acknowledge but accept. The conflation of SIDS and suffocation issues is also distressing but not new. The current review, like many before it, fails to provide adequate guidance to health care professionals and is potentially damaging to families making decisions without full information about their options. * Presentation by Tessie Pollack, Ohio Department of Health to Ohio Collaborative to Prevent Infant Mortality, March 2014. http://www.odh.ohio.gov/~/media/ODH/ASSETS/Files/cfhs/Infant%20Mortality/collaborative/presentations/3-11-14/Tessie%20Safe%20Sleep%20presentation%20for%20OCPIM%203%2011%2014.ashx

Conflict of Interest:

None declared