Is Putting Infants Back to Sleep Enough? : A National Campaign Fails to Reduce Racial and Ethnic Disparities in SIDS – Meaghan Kennedy

Sudden Infant Death Syndrome (SIDS) gained national attention in the U.S. in the early 1990s, when researchers first recognized stomach (prone) sleeping as a significant risk factor for SIDS. Defined as “the sudden death of an infant under 1 year of age, which remains unexplained after a thorough case investigation, including performance of a complete autopsy, examination of the death scene, and review of the clinical history (1),” SIDS was a mysterious, heartbreaking condition that at the time was the second leading cause of infant mortality in the U.S. (2). Following an American Academy of Pediatrics recommendation that all infants be placed on their backs to sleep, the National Institute of Child Health and Human Development (NICHD) announced its Back to Sleep (BTS) campaign in 1994. With nationwide efforts to spread the “back to sleep” message, BTS succeeded in dramatically reducing rates of prone sleeping and SIDS. Between 1992 and 2002, prone sleeping decreased from 70% to 11% and SIDS rates declined by over 50% (3-4).
Despite BTS’s initial success, prone sleeping and SIDS rates have remained constant in recent years and SIDS remains the leading cause of death in infants one month to one year of age (5-6). This slowed progress may reflect the campaign’s failure to effectively reach certain racial/ethnic groups at the highest risk for SIDS. Though SIDS has declined in all racial/ethnic groups, vast inequalities in SIDS rates persist and have actually grown since the BTS campaign began; rates for American Indians/Alaska Natives and African Americans (1.5 and 1.4 per 1000 live births, respectively) remain over two times that of the national rate (0.6 per 1000 live births) (4,7). Though the NICHD has recently begun to recognize this considerable racial/ethnic variation, it has done little to adapt BTS to meet the needs of these high-risk populations. The campaign’s failure to 1) recognize variation in risk factors between and within racial/ethnic groups; 2) incorporate behavioral theories emphasizing the role of social influence on behavior; and 3) use social epidemiology to understand the context for risk, hampers its potential to impact the populations most vulnerable to SIDS.
The BTS Message
Since its inception, BTS has primarily aimed to reduce rates of prone sleeping, with secondary recommendations that caregivers avoid soft bedding and prevent exposure to cigarette smoke. The campaign targets hospitals, physicians, and day care centers as well as the general public via national media campaigns and print materials. BTS’s outreach materials offer general information about SIDS and emphasize the “Safe Sleep Top 10”, a series of steps that caregivers can use to prevent SIDS. Among other recommendations, the “Safe Sleep Top 10” inform caregivers to place infants on their backs to sleep; use a firm sleep surface; keep soft objects and bedding out of the sleeping area; avoid cigarette smoke exposure; keep infants’ sleep areas separate from others’; and prevent overheating during sleep. Though BTS has a variety of outreach resources designed for childcare and healthcare providers, parents, and grandparents, as well as materials specifically designated for African American outreach, the overarching message is unvarying: “Placing your baby on his or her back to sleep is the number one way to reduce the risk of SIDS” (4, 8).
Understanding Inter- and Intra- population Variation in Risk
In using the same ten recommendations and overemphasizing the “back to sleep” message in all of its campaign materials, BTS ignores research documenting significant variability in SIDS risk factors between racial/ethnic groups. For example, while prone sleeping is more prevalent among African Americans than Whites (9-10), recent research suggests that additional risk factors – which are largely underemphasized in the BTS campaign – may significantly affect SIDS risk in African Americans. One study investigating prevalence of unsafe sleep practices in SIDS infants found no difference in sleep position at time of death between African American and non-African American infants. However, there was a higher prevalence of nonstandard bed use (e.g. sofa or adult bed) and bedsharing among African American SIDS infants at time of death, both of which are considered important SIDS risk factors (11). This suggests that bedsharing and nonstandard bed use may underlie elevated SIDS rates in African Americans, countering the emphasis on prone sleep position in BTS’s African American outreach materials.
Several studies show that prone sleeping also may not be the primary risk factor for SIDS among Native Americans. For example, while prone sleeping rates decreased similarly among Caucasians and Native Americans in North Dakota between 1991 and 1996, a corresponding reduction in SIDS was only observed in the Caucasian group (12). Moreover, a study examining SIDS among Native Americans in the Northern Plains region found no differences in prone sleeping or bedsharing between case and control infants; rather, infant clothing layers, number of prenatal and well-baby visits, visits by public health nurses, and maternal alcohol consumption significantly predicted susceptibility to SIDS (13).
Striking variability in SIDS exists not only between, but also within racial/ethnic groups. Though SIDS rates are highest among Native Americans, actual rates vary by tribe and region, ranging from a high of 3.5/1000 in the Aberdeen Area of the Northern Plains to a low of 0.78/1000 in the Navajo region of the Southwest (4). Moreover, while prone sleeping does not appear to be a significant risk factor for Native Americans in the Northern Plains, research cites prone sleeping, bedsharing, and nonstandard bed use as the primary contributors to SIDS deaths in Alaska (13-14). Similar variation exists among Hispanics; while Hispanic babies are almost two times less likely to die of SIDS than White babies in the U.S. overall, the rate for Puerto Ricans is twice that of Mexicans and Central and South Americans (15).
BTS’s failure to account for this inter- and intra-group variation in risk may explain its inability to effectively reach certain high-risk populations. Schwartz and Carpenter note the importance of considering variability in risk both between and within populations (16). Citing homelessness, obesity, and racial differences in infant mortality as examples, the authors note that the causes of inter-individual variation within a population, of variation between groups, and of the existence of a condition or disease often differ, necessitating an understanding of variability in risk at multiple levels (16). Given this research, BTS’s dissemination of the same 10 recommendations to all groups – thereby focusing on inter-individual variation within the general population – is predictably ineffective in reaching certain groups. It is crucial that the NICHD look beyond individual-level susceptibility and consider epidemiological research conducted on both the inter-individual and inter-population levels. Such an approach would allow the NICHD to tailor BTS efforts to the needs of high-risk groups and effectively target the risk factors most salient to individual populations.
Looking Beyond Individually-Based Behavioral Theory
BTS’s goal to change individual-level behavioral risk factors through education is guided by the Health Belief Model (HBM), which views an individual’s behavior as the outcome of his of her rational weighing of perceived susceptibility to and severity of a health problem, perceived benefits of taking action, and perceived barriers to taking action (17). Guided by this model, BTS assumes that individuals exposed to campaign materials will consider the costs of SIDS outweighed by the benefits of taking risk-reducing measures and thus will decide to follow the “Safe Sleep Top 10.” Though the HBM is widely used in modern public health, its use in SIDS prevention is misguided; by focusing exclusively on individually-based constructs, the HBM cannot account for powerful socio-cultural influences that may affect SIDS-related risk behavior.
In contrast to the HBM, Social Cognitive Theory (SCT) and Social Network Theory (SNT) look beyond individual-level characteristics, placing additional emphasis on interactions between individuals and their environments. Specifically, SCT describes how individuals’ behavior is influenced by “vicarious learning” or “modeling,” which occurs when people observe others’ behaviors and the consequences of those behaviors (18-19). SNT similarly notes the influence of social surroundings on individuals’ behavior, arguing that characteristics of social networks such as individual centrality/marginality, reciprocity and complexity of relationships, communication patterns, and diversity may serve as important risk or protective factors (18). These theories may be particularly relevant to African American and Native American communities, where social factors heavily influence health behavior. For example, Ashing-Giwa documents the failure of individual-level health behavior models like the HBM to adequately explain breast cancer screening behavior in African American women; specifically, she finds that they overlook social factors such as how interconnectedness with family and friends contributes to African American women’s identity formation (20). Interventions that take advantage of this cultural construct by using African American women who obtain regular screenings as role models for other women have been successful in improving African American women’s breast health (20). Similarly, Native American social networks tend to be interdependent with frequent contact among extended family members, potentially heightening the importance of modeling in these communities (21-22). For instance, studies document the strong influence of grandparents, cousins, and other extended family members on smoking and alcohol use in Native American adolescents (23-24).
Given these findings, research citing advice from health care professionals as an important predictor of infant sleep position in the general population (25) may not be as applicable to African American and Native American communities, where friends and family may strongly influence sleep position. Several studies involving primarily African American communities link social influence with choice of sleep position. For example, one study found that advice from an important female friend or relative was more predictive of infant sleep position than advice from a doctor or nurse (26). Another demonstrated that receiving or reading a BTS brochure had no effect on sleep position, but infants whose grandparents were in the home were more likely to be placed prone (27). Thus, advising health care professionals to recommend nonprone sleep positions and disseminating brochures from a national health institute as BTS does may not have any significant effect on SIDS risk behavior among African Americans and Native Americans. Instead, BTS messages may need to come from within communities, using local leaders as role models and taking advantage of existing social networks to spread SIDS risk reduction methods through families.
Contextualizing Risk: Why are Certain Groups Differentially Exposed?
In addition to the HBM, BTS’s emphasis on risk factor elimination also likely stems from the widely-accepted “triple-risk hypothesis” for SIDS, which states that SIDS occurs when an infant experiences three simultaneous conditions: an underlying vulnerability in control of vital functions; a critical developmental period in this control; and an exogenous stressor (e.g. prone sleeping) that aggravates the underlying vulnerability. Based on this hypothesis, epidemiologists conclude that they can reduce SIDS mortality by eliminating exogenous stressors – by following steps like the “Safe Sleep Top 10” – without understanding the source of the underlying vulnerability (4). This reflects modern epidemiology’s driving “black box” paradigm, wherein multiple individual-level causes are linked to health outcomes directly without considering other mediating factors (28). Though the “black box” paradigm is the standard in modern epidemiology, its isolation of individual-level risk factors from mediating contextual influences is flawed and may explain BTS’s failure to eliminate racial/ethnic disparities in SIDS.
Link and Phelan question this use of individual-level risk factors without a deeper understanding of the contextual factors that expose individuals to risk (29). Specifically, they criticize the emphasis on proximal causes of health problems – such as the “Safe Sleep Top 10” – while the more distal social determinants of disease are largely ignored. Moreover, they argue that some social factors, particularly those involving access to resources that limit exposure to risk, are fundamental causes of disease; as a result, targeting individual-level intervening mechanisms will not effectively eliminate a health problem (29). Recent criticism of the “black box” paradigm builds on Link and Phelan’s arguments, citing a need for a multidisciplinary social/ecological epidemiology that accounts for the macro-level social, historical, cultural, economic, and political factors that shape health in addition to the individual and molecular/genetic levels that are currently emphasized (28, 30).
Using an ecological approach that contextualizes risk is essential to understand inter-individual and inter-population variation in vulnerability to SIDS. In addition to the individual-level modifiable behavioral risk factors targeted by BTS, SIDS is associated with low socioeconomic status (SES), low maternal education, reduced prenatal care, single parenthood, young maternal age, and unemployment, which are pervasive among the groups most susceptible to SIDS (4,7,31). Native American reservations are plagued by poverty and unemployment (32), while in Washington D.C. – where over 70% of children are African American and over a third live in poverty – SIDS occurs at almost double the national rate (33, 34). Lack of access to adequate health care may be one mediating factor affecting the relationship between SES and SIDS; however, research indicates that the relationship between socioeconomic variables and poor health outcomes often persists after controlling for health care access and other individual-level factors such as risk behaviors, income, and employment status (35). Thus, other factors extending beyond individual-level characteristics may mediate the relationship between SES and health problems such as SIDS. Among Native Americans and African Americans, historical loss, distrust of the medical system, racial segregation, and discrimination may be important mediating factors (36-38). Still other characteristics of the isolated rural areas and the poor, racially-segregated urban communities in which many Native Americans and African Americans live – such as transportation, crime, and housing conditions – may also contribute to SIDS disparities (35).
A fundamental flaw in the theory driving BTS is the NICHD’s lack of understanding of the role these and other contextual factors may play in shaping exposure to risk. Confronted with slowed declines in SIDS and widening racial/ethnic disparities in the past 10 years, the NICHD has recently made efforts to better understand the underlying causes of disparities in SIDS. However, in a section highlighting racial/ethnic disparities in its 2001 publication Targeting Sudden Infant Death Syndrome (SIDS): A Strategic Plan, the NICHD misguidedly responds to research that shows a persisting excess risk for SIDS among African Americans and American Indians after adjusting for certain socioeconomic factors: “This observation suggests that biological markers – e.g., susceptibility genes – may play important roles in the pathogenesis of SIDS in specific racial and ethnic groups; certain genes predisposing infants to premature birth or sudden death may prove to be more prevalent in some populations than others” (4). Instead of looking to historical, cultural, or other socio-physical forces whose influence on SIDS are not yet understood, the NICHD appeals exclusively to genetics to explain persisting racial/ethnic disparities (35). Until the NICHD moves beyond micro- and individual-level approaches, acknowledges the role of these macro-level forces, and incorporates an understanding of context into BTS, racial and ethnic disparities in SIDS will inevitably persist.
Moving Forward: A Multidisciplinary, Eco-Social Back to Sleep
Given the considerable variation in risk within and across racial/ethnic groups and the range of contextual factors that may influence SIDS rates, it is essential that the NICHD employ a multidisciplinary approach as it works to reduce disparities in SIDS. Though BTS had a dramatic impact on infant sleep practice and SIDS rates in its initial years, emphasizing individual-level behavioral risk factors limits its efforts to eliminate SIDS as a health problem. Instead, the NICHD should follow an eco-social epidemiologic approach to understand not only how risk varies between individuals and groups but also why people are differentially exposed to risk. Addressing the context in which individuals experience risk and tailoring BTS to address those contextual factors in different populations is critical to have any meaningful effect on disparities – only then can the fundamental causes of SIDS be understood and the persistent relationship between race/ethnicity and SIDS be eliminated.
REFERENCES
(1) Willinger M, James LS, Catz C. Defining the Sudden Infant Death Syndrome (SIDS): Deliberations of an Expert Panel Convened by the National Institute of Child Health and Human Development. Pediatric Pathology 1991; 11:677-684.
(2) CDC. Infant Mortality -- United States, 1993. MMWR 1996; 45:211-215.
(3) American Academy of Pediatrics, Task Force on Infant Sleep Position and Sudden Infant Death Syndrome. The Changing Concept of Sudden Infant Death Syndrome: Diagnostic Coding Shifts, Controversies Regarding the Sleep Environment, and New Variables to Consider in Reducing Risk. Pediatrics 2005; 116:1245-1255.
(4) National Institute of Child Health and Human Development. Targeting Sudden Infant Death Syndrome (SIDS): A Strategic Plan. Rockville, MD: National Institute of Child Health and Human Development, 2001.
(5) National Institute of Child Health and Human Development. SIDS Rate and Sleep Position, 1988-2003. Rockville, MD: National Institute of Child Health and Human Development. http://www.nichd.nih.gov/sids/upload/SIDS_rate_backsleep_03.pdf
(6) CDC. QuickStats: Leading Causes of Neonatal and Postneonatal Deaths – United States, 2002. MMWR 2005; 54:966.
(7) Pickett KE, Luo Y, Lauderdale DS. Widening Social Inequalities in Risk for Sudden Infant Death Syndrome. American Journal of Public Health 2005; 95:1976-1981.
(8) National Institute of Child Health and Human Development. SIDS: “Back to Sleep” Campaign. Rockville, MD: National Institute of Child Health and Human Development. http://www.nichd.nih.gov/sids/
(9) Ottolini MC et al. Prone Infant Sleeping Despite the “Back to Sleep” Campaign. Archives of Pediatric and Adolescent Medicine 1999; 153:512-517.
(10) Shields LB et al. Risk Factors Associated with Sudden Unexplained Infant Death: A Prospective Study of Infant Care Practices in Kentucky. Pediatrics 2005; 116:e13-e20.
(11) Unger B et al. Racial Disparity and Modifiable Risk Factors Among Infants Dying Suddenly and Unexpectedly. Pediatrics 2003; 111:e127-e131.
(12) McCulloch K et al. Prevalence of SIDS Risk Factors: Before and After the “Back to Sleep” Campaign in North Dakota Caucasian and American Indian Infants. Clinical Pediatrics 2000; 39:403-410.
(13) Iyasu S et al. Risk Factors for Sudden Infant Death Syndrome among Northern Plains Indians. JAMA 2002; 288:2717-2723.
(14) Gessner BD, Ives GC, Perham-Hester KA. Association between sudden infant death syndrome and prone sleep position, bed sharing and sleeping outside an infant crib in Alaska. Pediatrics 2001; 108:923-927.
15) American Lung Association. State of Lung Disease in Diverse Communities: 2007. American Lunch Association, 2007.
http://www.lungusa.org/site/pp.asp?c=dvLUK9O0E&b=308853
(16) Schwartz S, Carpenter KM. The Right Answer for the Wrong Question: Consequences of Type III Error for Public Health Research. American Journal of Public Health 1999; 89:1175-1179.
(17) Edberg M. Individual Health Behavior Theories (pp.35-49). In: Edberg M. Essentials of Health Behavior: Social and Behavioral Theory in Public Health. Boston, MA: Jones and Bartlett Publishers, 2007.
(18) Edberg M. Social, Cultural, and Environmental Theories (Part I). (pp.51-64). In: Edberg M. Essentials of Health Behavior: Social and Behavioral Theory in Public Health. Boston, MA: Jones and Bartlett Publishers, 2007.
(19) Bandura A. Health Promotion by Social Cognitive Means. Health Education and Behavior 2004; 31:143-164.
(20) Ashing-Giwa, K. Health Behavior Change Models and Their Socio-Cultural Relevance for Breast Cancer Screening in African American Women. Women and Health 1999; 28:53-70.
(21) Harrison et al. Family Ecologies of Ethnic Minority Children. Child Development 1990; 61:347-362.
(22) MacPhee D, Fritz J, Miller-Heyl J. Ethnic Variations in Personal Social Networks and Parenting. Child Development 1996; 67:3278-3295.
(23) Swaim RC, Oetfing ER, Thurman PJ et al. American Indian adolescent drug use and socialization characteristics: a cross-cultural comparison. Journal of Cross-Cultural Psychology 1993; 24:50-70.
(24) Kegler MC, Cleaver VL, Yazzie-Valencia M. An Exploration of the Influence of Family on Cigarette Smoking Among American Indian Adolescents. Health Education Research 2000; 15:547-557.
(25) Willinger M et al. Factors Associated With Caregivers’ Choice of Infant Sleep Position, 1994-1998, The National Infant Sleep Position Study. JAMA 2000; 283:2135-2142.
(26) Colson et al. Barriers to Following the Supine Sleep Recommendation among Mothers at Four Centers for the Women, Infants, and Children Program. Pediatrics 2006; 118:e243-e250.
(27) Moon RY, Omron R. Determinants of Infant Sleep Position in an Urban Population. Clinical Pediatrics 2002; 41: 569-573.
(28) Susser M, Susser E. Choosing a Future for Epidemiology: I. Eras and Paradigms. American Journal of Public Health1996; 86:668-673.
(29) Link BG, Phelan J. Social Conditions as Fundamental Causes of Disease. Journal of Health and Social Behavior 1995; Extra Issue: 80-94.
(30) Krieger N. Theories for social epidemiology in the 21st century: and ecological perspective. International Journal of Epidemiology 2001; 30:668-677.
(31) Blair PS et al. Major Epidemiological Changes in Sudden Infant Death Syndrome: A 20-Year Population-Based Study in the UK. The Lancet 2006; 367: 314-319.
(32) U.S. Census Bureau Geography Division. Population, Land Area, and Poverty Data for American Indian and Alaska Native Areas. Washington, D.C.: U.S. Census Bureau. http://www.census.gov/geo/www/ezstate/aianapov.html
(33) Moon RY, Oden RP, Grady KC. Back to Sleep: An Educational Intervention with Women, Infants, and Children Program Clients. Pediatrics 2004; 113:542-547.
(34) DC KIDS COUNT Collaborative for Children and Families. Every Kid Counts in the District of Columbia, 13th Annual Fact Book 2006. Washington, D.C.: DC KIDS COUNT Collaborative for Children and Families.
(35) Haan M, Kaplan G, Camacho T. Poverty and Health: Prospective Evidence from the Alameda County Study. American Journal of Epidemiology 1987; 125:989-998.
(36) Whitbeck LB, Chen X, Hoyt DR, Adams GW. Discrimination, historical loss and enculturation: culturally specific risk and resiliency factors for alcohol abuse among American Indians. Journal of Studies on Alcohol 2004; July:409-18.
(37) Williams DR and Collins C. Racial Residential Segregation: A Fundamental Cause of Racial Disparities in Health. Public Health Reports 2001; 226:404-416.
(38) Williams DR. Racial/Ethnic Variations in Women’s Health: The Social Embeddedness of Health. American Journal of Public Health 2002; 92:588-597.