Harville Emily W, Jacobs Marni, Shu Tian, Breckner Dorothy, Wallace Maeve
Department of Epidemiology, Tulane School of Public Health and Tropical Medicine, 1440 Canal St Ste 2000, New Orleans, LA, 70112, USA.
Division of Biostatistics and Study Methodology, Children's National Health System, Washington, DC, 20010, USA.
Matern Child Health J. 2018 Jun;22(6):858-865. doi: 10.1007/s10995-018-2460-y.
Introduction Researchers in perinatal health, as well as other areas, may be interested in linking existing datasets to vital records data when the existence or timing of births is unknown. Methods 5914 women who participated in the Bogalusa Heart Study (1973-2009), a long-running study of cardiovascular health in childhood, adolescence, and adulthood, were linked to vital statistics birth data from Louisiana, Mississippi, and Texas (1982-2010). Deterministic and probabilistic linkages based on social security number, race, maternal date of birth, first name, last name, and Soundex codes for name were conducted. Characteristics of the linked and unlinked women were compared using t-tests, Chi square tests, and multiple regression with adjustment for age and year of examinations. Results The Louisiana linkage linked 4876 births for 2770 women; Mississippi linked 791 births to 487 women; Texas linked 223 births to 153 women; After removal of duplicates and implausible dates, this left a total of 5922 births to 3260 women. This represents a successful linkage of 55% of all women ever seen in the larger study, and an estimated 65% of all women expected to have given birth. Those linked had more study visits, were more likely to be black, and had statistically lower BMIs than unlinked participants. Discussion Linking unrelated study data to vital records data was feasible to a degree. The linked group had a somewhat more favorable health profile and was less mobile than the overall study population.
引言 在围产期健康以及其他领域的研究人员，当出生情况或时间未知时，可能会有兴趣将现有数据集与生命记录数据相链接。方法 5914名参与博加卢萨心脏研究（1973 - 2009年）的女性被链接到来自路易斯安那州、密西西比州和德克萨斯州（1982 - 2010年）的生命统计出生数据。该研究是一项针对儿童、青少年和成年人心血管健康的长期研究。基于社会保险号、种族、母亲出生日期、名字、姓氏以及名字的语音相似编码进行确定性和概率性链接。使用t检验、卡方检验以及对年龄和检查年份进行调整的多元回归对已链接和未链接女性的特征进行比较。结果 路易斯安那州的链接为2770名女性链接到4876例出生记录；密西西比州为487名女性链接到791例出生记录；德克萨斯州为153名女性链接到223例出生记录；在去除重复项和不合理日期后，总共为3260名女性留下5922例出生记录。这代表在更大规模研究中所有曾参与研究女性的55%成功实现链接，预计所有已生育女性的65%实现链接。那些被链接的女性有更多的研究访问次数，更可能是黑人，并且与未链接的参与者相比，她们的体重指数在统计学上更低。讨论 将不相关的研究数据与生命记录数据相链接在一定程度上是可行的。与整个研究人群相比，已链接组的健康状况略好且流动性较低。
