Introduction

Sudden unexpected infant death (SUID), including sudden infant death syndrome (SIDS) and other sleep-related infant deaths, is the leading cause of postneonatal infant mortality in the United States [1, 2]. In 2014, the breakdown of SUID by cause was 44 % SIDS, 31 % unknown cause and 25 % accidental suffocation and strangulation in bed (ASSB). SIDS rates have declined considerably since 1990 whereas ASSB mortality rates have continued to increase since 1998 [2]. Several modifiable and non-modifiable risk factors have been identified for SUID and SIDS. These risk factors include sleeping in the prone and side position, sleeping on a soft surface, exposure to smoking during pregnancy or after birth, overheating, late or no prenatal care, young maternal age, prematurity and/or low birth weight, male sex, and bedsharing [1].

Preventing SUID/SIDS remains an important public healthy priority [1, 2]. In 1994, the “Back to Sleep” campaign was initiated in the United States, a joint effort of the US Public Health Service, the American Academy of Pediatrics, and other organizations. The SIDS rate decreased by >40 % after the institution of the “Back to Sleep” campaign. This was attributed to a decrease in the frequency of prone sleeping from >70 to ~20 % among US infants [1]. While early efforts focused on a safe sleeping position, subsequent improved death scene investigation following the release of the Sudden Unexplained Infant Death Investigation Reporting Form in 1996 [1, 2], provided evidence of other sleep-related risk factors [3–6].

Since the adoption of the “Back to Sleep” campaign, there has been an increasing racial disparity for deaths diagnosed as SIDS, asphyxiation (AS) and undetermined (UD) etiology [1, 2, 7, 8]. In 2009 SUID rates among infants born to American Indian/Alaska Native and non-Hispanic Black mothers were more than twice the rate among infants born to non-Hispanic White mothers [9]. From 1993 to 2007, the National Infant Sleep Position (NISP) Study reported a significant increase in the use of the supine sleep position among all races that plateaued after 2001; however, racial and ethnic disparities persisted, with fewer African American infants being placed in the supine position to sleep [10]. The NISP study also reported that infant bedsharing as usual practice more than doubled from 1993 to 2010. Maternal race or ethnicity was a major factor: compared to White mothers, Black mothers were more than 3 times as likely to bedshare with their infants, Hispanic mothers about 1.5 times as likely, and mothers of other race or ethnicity nearly 2.5 times as likely [11].

Based on increasing evidence, the AAP expanded their safe sleep recommendations to address bedsharing and other modifiable risk factors as well as prone positioning in 2011. Specifically, these recommendations now include supine positioning, use of a firm sleep surface, breastfeeding, room-sharing without bedsharing, routine immunization, consideration of a pacifier, and avoidance of soft bedding, overheating, and exposure to tobacco smoke, alcohol, and illicit drugs [1].

In 2013, the City of Philadelphia, the site of this study, reported 89 sleep-related infant deaths during the period 2009–2010, with overrepresentation of minority infants [12]. The Philadelphia Child Death Review Team found nearly half of the infants who suffered sleep-related deaths were not placed on their back to sleep and, in at least ten cases, death was directly caused by bedsharing [12]. A review of 80 sleep-related infant deaths in the State of Pennsylvania in 2012 had similar findings: 71.2 % of the infants were not in a crib or bassinet, 52.5 % were sleeping with one or more other people and 41.3 % were not sleeping on their back. Approximately two-thirds of these infants were Black [13].

Why do parents choose to place their newborns either in the prone position or in bed with them? Mothers who use the prone position often state that “the baby likes it better and/or sleeps better that way.” [14] Major reasons reported for bedsharing are to calm a fussy infant and to help the infant and/or mother sleep better as well as to facilitate breastfeeding [15].

In summary, sleeping prone and bedsharing are major risk factors for SUID in infants and these risk factors are more prevalent among African American populations. Swaddling is a technique that enhances infant comfort, decreases the startle reflex, reduces spontaneous waking and promotes quiet sleep [16–18]. Although careful instruction is necessary to assure the proper technique and duration of swaddling, the use of swaddling has been suggested as a strategy to encourage infants to sleep in the supine position [18, 19]. In an urban minority population, parents who routinely used swaddling were more likely to place their infant supine when swaddled [20]. Swaddling may also decrease the likelihood of bedsharing, but this has not been explored. The objective of this study was to explore the effect of a postpartum swaddling education intervention on infant sleep practices.

Methods

Study Design

Historical Control Group

From November 2009 through January 2010, we surveyed a convenience sample of mothers of infants 1–3 months of age at a pediatric ambulatory center at an academic medical center that serves a minority, low income population. Inclusion criteria were gestational age equal to or greater than 35 weeks, born at the medical center, and discharged from the nursery. All postpartum women at the medical center received routine education on infant safe sleeping practices, including one-to-one education by a registered nurse, review of safe sleeping practices by a pediatric provider, and a back-to-sleep brochure.

Swaddling Education Intervention Group

From February 2010 through May 2010 we conducted a swaddling education intervention of a cohort of postpartum women, using the same inclusion criteria for historical controls. In addition to the routine education on safe sleeping practices, mothers received a swaddling educational intervention.

Postpartum Educational Intervention

A trained research assistant approached eligible postpartum mothers Monday through Friday and invited them to participate in a prospective study of infant sleeping and swaddling. Mothers were given a new swaddling blanket (Miracle Blanket), purchased with study funds. This blanket is lightweight cotton with an ample pocket for the legs, and is commercially available. The research assistant showed mothers how to swaddle the baby and mothers were asked to demonstrate the swaddling back to the research assistant. Mothers were instructed to always place their swaddled infants on their back, in their own crib or bassinet.

Survey

The survey, adapted from the Infant Feeding Practices II (IFPS II) and Advocate Health Care [21, 22], addressed demographics, satisfaction and experience with swaddling, sleeping position (back, side, belly) and sleeping location (crib/bassinet, bedsharing, other). The research assistant surveyed the swaddling and historical groups with the same instrument at their 2 month well child visit.

Data Analysis

Study outcomes were use of swaddling for sleeping, concerns about swaddling, infant sleep position at night, infant sleep location at night, breastfeeding, and pacifier use. Chi square and t-tests were used to compare the groups.

The study was approved by the Institutional Review Board.

Results

One hundred and twenty-one mothers participated in the historical group; 16 refused. Seventy mothers enrolled in the swaddling group; 10 refused. Of those who enrolled in the swaddling group, 40 (57 %) completed the survey at 1–3 months. The study population was predominantly African American with public insurance (Table 1). The groups were similar in demographic profile, except the infants in the control group were younger.

Table 1 Characteristics of study population
Full size table

Compared to the historical group, mothers in the swaddling group were more likely to always swaddle their babies to put them to sleep (52.5 vs. 23.1 %, p < .001) and had fewer concerns about swaddling (5.1 vs. 21.6 %, p = .026) (Table 2). More mothers in the swaddling group reported always placing their baby on the back to sleep at night, but this difference did not demonstrate statistical significance (81.6 vs. 69.4 %, p = .212). Mothers in the swaddling group were significantly more likely to always place their baby in a crib or bassinet to sleep at night (84.6 vs. 66.9 %, p = .042). There were no differences in the rates of breastfeeding or pacifier use between the swaddling and historical groups.

Table 2 Infant swaddling, sleep position and location, breastfeeding and pacifier use
Full size table

Discussion

A swaddling educational intervention increased the use of swaddling in an urban, minority community. However, the impact on safe sleeping practices was limited. Mothers in the swaddling intervention group were significantly more likely to place their baby to sleep in a crib or bassinet, but there was not a demonstrated difference in their use of “back to sleep.” Our study consisted of an educational intervention on safe sleeping practices that utilized a swaddling blanket as a tool to encourage supine positioning of the infant and discourage bedsharing. Our results may be due to the swaddling intervention itself, to the additional education about safe sleeping practices that accompanied the swaddling intervention, or to both.

Swaddling is a simple and popular infant soothing strategy which has been demonstrated to promote quiet sleep in physiologic studies [16, 17]. Special swaddling blankets are increasingly being marketed to the public and introduced in newborn nurseries and neonatal intensive care units. Proper swaddling technique is essential to prevent potential risks including overheating and interference with respiratory effort as well as that of developmental hip dysplasia [1, 16]. A recent meta-analysis based on 4 observational studies demonstrated an increased risk of SIDS for infants swaddled for sleep when compared to unswaddled controls; however, there was significant heterogeneity among the studies, swaddling technique was not well-defined and this analysis did not adjust for the presence or absence of bedsharing. Overall swaddled infants were one and a half times as likely to die from SIDS and this risk increased dramatically depending on sleep position: swaddled infants sleeping prone were more than 12 times as likely to die from SIDS as unswaddled controls, and swaddled infants side sleeping showed a moderate increase over those in the supine position. The risk of SIDS with swaddling increased with the age of the infant and was twice as great for infants over 6 months of age [23]. Recent recommendations are to stress the need to avoid the prone and side sleep position, especially for swaddled infants, and to set an age limit for the use of swaddling and/or to stop swaddling infants as soon as an infant attempts to roll [23, 24].

Our study was conducted in an urban minority population and findings may not be generalizable. The study has a relatively small sample size and limited follow up. Achieving follow up was challenging due to frequent changes in telephone numbers, addresses and health care providers. Additionally, parental responses may have been affected by recall and social desirability bias.

Conclusions

Minority communities experience higher rates of SUID, primarily due to unsafe infant sleeping practices. A postpartum swaddling educational intervention in a minority, urban community increased the use of swaddling but had a limited impact on safe sleeping practices. We found significantly less bedsharing but no significant difference in the use of “back to sleep.” A concerning finding was that even after an postpartum swaddling educational intervention, nearly one in five babies did not sleep on their back, and one in six babies was bedsharing. This high prevalence of unsafe practices shows the need for continued efforts to encourage safe infant sleeping practices in minority communities. Ongoing studies are needed to monitor the safety and effectiveness of swaddling as a tool to promote safe sleeping in infants.