The principal finding of our study is that the majority of the 1,400 surveyed families in our birth cohort generally followed pivotal recommendations for SIDS prevention during the infant’s first year of life. This was especially true for the recommendation that infants should sleep in their parents’ sleeping room and that a sleeping bag instead of a blanket should be used. However, we also observed a remarkable proportion of infants who were still placed regularly or even preferably in the prone position for sleeping. The most prevalent deviation from the relevant SIDS guideline in our current analysis was bed-sharing with an adult, which was practiced in a considerable part of the families. For every sixth infant, bed-sharing in the parent’s bed actually was the default sleeping location.
A large majority of parents is well informed regarding measures to prevent SIDS and consequently they mostly plan to abide by these recommendations8th. In the current study, we could show that parents not only know but appear to largely follow the recommendations through the first year after birth. Notably, most infants initially slept in their parents’ bedroom and slowly moved to their own room during the first year. Also, it is encouraging to note that most infants were supplied with a sleeping bag instead of a loose blanket. This finding is in line with Shapiro-Mendoza, who reported a marked trend against the use of blankets13. Regarding other loose items in the infant’s bed, the most frequently used was a baby’s nest and a nursing pillow (U-shaped pillow), which was put into the bed of approximately 20% of the infants. Although the exact attributable risk of such pillows is still unclear, several cases of sudden unexplained infant deaths have been reported in this context14 and the guidelines recommend against the use1. Generally, there should be no loose items at all in the bed that bear the risk of obstructing the infant’s breathing.
To the best of our knowledge there are hitherto no studies which determined the association between bedside sleepers and SIDS risk. Accordingly, there is no evidence for a clear recommendation and hence the SIDS task force of the American Academy of Pediatrics recommends neither for nor against the use of bedside sleepers1. The national German guideline on SIDS prevention does not even mention bedside sleepers4.5. During the period for our birth cohort, an update of the corresponding German SIDS recruitment guideline has been issued4. However, regarding the parameters assessed in our present paper, there have been no relevant changes compared to the previous version5. At least, the US Consumer Product Safety Commission has issued safety standards for bedside sleepers15. As a bedside sleeper seems to become the most common bedroom furniture for infants (at least in our study region), further epidemiological studies on the impact on SIDS are urgently needed.
Unlike for bedside sleepers, data regarding bed-sharing and SIDS are available and several studies showed an increased SIDS risk for infants bed-sharing in an adult’s bed16:17especially in the context of alcohol or drug use or in special situations, like sleeping on a sofa18. Consequently, in most (but not all) current guidelines in Western countries, bed-sharing with an adult is discouraged, ia in the German guideline4, which is relevant for our study collective. However, this topic is subject of the ongoing scientific discussion and other authors emphasize further important aspects of bed-sharing19, eg facilitation of breastfeeding, bonding aspects, or improvement of sleep quality for the infant-parent tryad. Some authors even no longer regard bed-sharing as a modifiable risk factor in its proper sense which can be influenced by simple recommendations20 and suggest a more balanced approach to parental counseling21. However, the aim of our present study was not to question the content of the guideline which is currently applicable to the study population, but to determine the actual implementation by the families.
In a previous study, we found that most parents are well aware that the current SIDS guideline discourages from bed-sharing with an adult and intend to follow the corresponding recommendation not to share the bed with the infant8th. Immediately prior to initial discharge from maternity ward, only 2% of interviewed mothers indicated that their child would be regularly allowed to sleep in the parent’s bed. However, our current analysis of the actual implementation illustrates a considerable gap between intentions and real life behavior with only about one-third of the surveyed families never practicing bed-sharing with the infant at all. These observations are in line with Stromberg et al. who reported that bed-sharing was frequently practiced and that up to 25% of infants under three months almost exclusively slept in the parental bed22. In our current study, overall two thirds of infants co-slept in their parents’ bed at least some of the time. Considering a potential social desirability bias, this rate may well be higher. The subanalysis of the migration balance sheds light on an interesting aspect in this context: while bed-sharing was the only sleeping furniture with a positive net balance against all other items, in summary we observed a particularly pronounced net migration from “own infant’s bed” to “bed-sharing” between birth and four weeks of age. Although a survey of parents’ motivations for this behavior was not part of our present study, other authors analyzed the reasons for bedsharing19 and showed that a major point was facilitation of nighttime care and breastfeeding. Whether the usage of a bedside sleeper will bring a change in this respect and might lead to a more stable avoidance of bed-sharing can only be speculated about and should be addressed in a future study.
In summary, we have to acknowledge that the majority of the parents in our sample (and in many others) are unwilling to follow the current guidelines regarding bed-sharing. An assessment of the possible reasons for this deviation from official advice was not part of our study, but at least the significant association between bed-sharing and breastfeeding suggests an important partial explanation. In general, when condering our data on bed-sharing, it is important to realize that our study sample does not represent a high-risk collective for SIDS, such as low-birth weight or premature infants. It will be subject of the scientific debate, whether future recommendations for parents in such a collective should be less categorically against bed-sharing but modified towards a risk-graded recommendation. Considering that two-thirds of families in our survey obviously did not follow the blanket recommendation not to bed-share, it appears that the current practice of counseling is not working. Whether a more targeted approach, focusing on specific high-risk infants, might be overall more effective needs to be clarified in future studies. Regardless of this thought for future advancement of SIDS counseling, all parents should be educated by the health care professionals about the potential benefits and the risks of bedsharing.
In the National Infant Sleep Position StudyColson and colleagues showed that between the years 1993 and 2000 the percentage of parents who exclusively placed their child on the back was continuously rising23. However, since 2001 this trend has reached a plateau. While the majority of mothers in our study had the intention to never actively put their infant in the prone position for sleeping, the number of infants being actually placed in prone position increased over the first year of life. Already at the age of four weeks, approximately one third of the surveyed infants was placed in a prone position sometimes or more often. However, the occasional prone position is of particular concern, as infants who are not accustomed to this position carry a particularly high risk of SIDS24. As most parents were aware of SIDS prevention measures after birth8thbut their good intentions were not consistently implemented during the first year of life, we consider that these recommendations need to be addressed repeatedly to increase the level of actual implementation.
Among the environmental factors for SIDS, exposure to tobacco smoke is an important modifiable risk factor25,26,27. An analysis of a large longitudinal cohort from 2012 found that 43% of German children under 17 years have at least one parent who smokes28. In our cohort, almost a quarter of the children were born into households with a smoker, reflecting a significant number of infants who are still going to be exposed to this avoidable risk factor for SIDS. Moreover, we have to realize that questions regarding parental smoking behavior are especially prone to social desirability bias. Thus, the actual number may even be higher than reported by the parents in this study.
Our study has some limitations. First, there is a potential selection bias as parents had to have sufficient German language skills to be included. Since ethnic differences can impact sleeping habits in families29,30,31, Studies in different cultural collectives may yield different results. Furthermore, parents with higher educational status are overrepresented in the KUNO-Kids birth cohort9. This circumstance led to a difference in some socio-economic characteristics between the drop-outs and the families that remained in the cohort. Specifically, families with only one child, an older age of the mother, a mother with higher education level, a mother with full-time employment before birth of the child, and a mother without a migration background were more likely to continue participating in the study (for details see Brandstetter et al.9). Therefore, when interpreting our data, it is important to keep in mind that, conversely, families with lower socio-economic status and/or with a migration background are significantly underrepresented. In the context of SIDS prevention this aspect has to be considered, as there seems to be an association between SIDS risk behavior and low socio-economic status16:32. Another limitation of our analysis is that we included primiparae as well as multiparae. However, for mothers with already one older child (or several older children as well), the prior exposure to SIDS recommendations and the experience with their implementation is likely to influence the implementation with the current child, particularly if the mother attempted to comply with guidance previously and found it difficult to do so. A further point, which has to be considered when interpreting our data, is that infants very likely might have moved from one place to the other during the night. Nevertheless, in order to keep the questionnaires as simple as possible for the parents and thus to achieve a better response quality, in this survey we focused on asking about the default primary sleeping place where the baby slept most of the time. Therefore, we might have underestimated the role of some secondary sleeping places (especially during daytime), for example the frequency of bed-sharing. Moreover, as the KUNO-Kids birth cohort is an interdisciplinary study covering multiple domains of pediatric health and development, we were forced to limit data collection in our substudy to an appropriate number of items. Hence, we focused on key aspects of SIDS prevention but were unable to assess some detailed aspects that might also have been of interest, such as the differentiation between nighttime and daytime sleep or the combination of different risks, such as bed-sharing plus smoking or alcohol consumption. Finally, there was a notable dropout of families from the baseline interview after birth to the one-year questionnaire. This may have further selected the most conscientious families who may be more prone to follow recommendations. In summary, the actual situation with respect to adherence to SIDS prevention recommendations in the first year of life might be even worse than reported in this study.