Salm Ward Trina C, Robb Sara Wagner, Kanu Florence A
Department of Health Promotion and Behavior, College of Public Health and School of Social Work, University of Georgia, 279 Williams St, Athens, GA, 30605, USA.
Department of Epidemiology and Biostatistics, University of Georgia College of Public Health, 101 Buck Road, Athens, GA, 30602, USA.
Matern Child Health J. 2016 Feb;20(2):347-62. doi: 10.1007/s10995-015-1834-7.
To examine: (1) the prevalence and characteristics of bed-sharing among non-Hispanic Black and White infants in Georgia, and (2) differences in bed-sharing and sleep position behaviors prior to and after the American Academy of Pediatrics' 2005 recommendations against bed-sharing.
Georgia Pregnancy Risk Assessment Monitoring System (PRAMS) data were obtained from the Georgia Department of Public Health. Analysis was guided by the socioecological model levels of: Infant, Maternal, Family, and Community/Society within the context of race. Data from 2004 to 2011 were analyzed to address the first objective and from 2000 to 2004 and 2006 to 2011 to address the second objective. Rao-Scott Chi square tests and backward selection unconditional logistic regression models for weighted data were built separately by race; odds ratios (OR) and 95 % Confidence Intervals (CIs) were calculated.
A total of 6595 (3528 Black and 3067 White) cases were analyzed between 2004 and 2011. Significantly more Black mothers (81.9 %) reported "ever" bed-sharing compared to White mothers (56 %), p < 0.001. Logistic regression results indicated that the most parsimonious model included variables from all socioecological levels. For Blacks, the final model included infant age, pregnancy intention, number of dependents, and use of Women, Infant and Children (WIC) Services. For Whites, the final model included infant age, maternal age, financial stress, partner-related stress, and WIC. When comparing the period 2000-2004 to 2006-2011, a total of 10,015 (5373 Black and 4642 White cases) were analyzed. A significant decrease in bedsharing was found for both Blacks and Whites; rates of non-supine sleep position decreased significantly for Blacks but not Whites.
Continued high rates of bed-sharing and non-supine sleep position for both Blacks and Whites demonstrate an ongoing need for safe infant sleep messaging. Risk profiles for Black and White mothers differed, suggesting the importance of tailored messaging. Specific research and practice implications are identified and described.
研究：（1）佐治亚州非西班牙裔黑人和白人婴儿同床睡眠的患病率及特征，以及（2）美国儿科学会2005年发布反对同床睡眠建议前后，同床睡眠和睡眠姿势行为的差异。
从佐治亚州公共卫生部获取佐治亚州妊娠风险评估监测系统（PRAMS）的数据。分析以种族背景下的社会生态模型层次为指导，包括婴儿、母亲、家庭以及社区/社会层次。分析2004年至2011年的数据以实现第一个目标，分析2000年至2004年以及2006年至2011年的数据以实现第二个目标。按种族分别建立加权数据的Rao-Scott卡方检验和向后选择无条件逻辑回归模型；计算比值比（OR）和95%置信区间（CI）。
2004年至2011年期间共分析了6595例（3528例黑人婴儿和3067例白人婴儿）。与白人母亲（56%）相比，报告“曾有过”同床睡眠的黑人母亲（81.9%）显著更多，p<0.001。逻辑回归结果表明，最简约的模型包含了所有社会生态层次的变量。对于黑人，最终模型包括婴儿年龄、怀孕意愿、受抚养人数以及妇女、婴儿和儿童（WIC）服务的使用情况。对于白人，最终模型包括婴儿年龄、母亲年龄、经济压力、伴侣相关压力以及WIC。比较2000 - 2004年和2006 - 2011年期间，共分析了10015例（5373例黑人婴儿和4642例白人婴儿）。发现黑人和白人的同床睡眠率均显著下降；黑人非仰卧睡眠姿势的比例显著下降，而白人则没有。
黑人和白人的同床睡眠率及非仰卧睡眠姿势持续居高，表明持续需要进行安全婴儿睡眠的宣传。黑人和白人母亲的风险特征不同，这表明针对性宣传的重要性。确定并描述了具体的研究和实践意义。
