Evaluation of reproductive histories constructed by linking vital records

Abstract

We used 1.4 million fetal death and birth certificates filed in Georgia between 1980 and 1992 to construct 369,686 chains of two or more reproductive events occurring to the same woman. We evaluated these chains using both information on the certificates and information independently collected in interviews with 1311 women. Overall, 86.6% of the chains had the expected number of events, based on the certificate's information about previous pregnancies. Seventy-nine per cent of the chains had the expected number of events based on the maternal interviews. Consistency between the observed number of events in the chain and the number expected, based either on data from the certificates or from the maternal interviews, was greatest for chains with two or three events. Mothers born in Georgia were more likely to have complete chains than mothers born elsewhere. Among the 551,391 non-linked certificates, 48.7% were the mother's first birth, 40.2% were second or higher-order births to women whose previous pregnancy occurred before 1980, and 11.1% were second or higher-order births to women whose previous pregnancy occurred after 1980. Fetal death and livebirth certificates can be linked to construct pregnancy histories with reasonably low levels of underlinkage and overlinkage.

U.S. Maternally Linked Birth Records May Be Biased for Hispanics and Other Population Groups

“…Adams et al (16) reported the proportion of maternal sets for which the number of records in the set equaled the number of events indicated on the records in the set. This proportion (86% overall) was greater for white race, more education, being married, smaller sets, having the same father listed on all records in the set, and having no fetal deaths or infant deaths in the set.…”

“…Within this framework, linkage error is properly conceptualized as departure from representativeness. The relation of error as departure from representativeness to error as departure from the true linkages, which is the approach customarily taken in the record linkage literature (16, 17), requires further study. The high congruence of the maternal sets produced in the present study with those produced by a traditional Fellegi and Sunter linkage of the same data (17) suggests that the present method produces the true maternal sets as well as a method designed for that purpose.…”

“…Maternal record linkage has generally been conducted with the “competing goals of linking as many records as possible and avoiding incorrect linkages” (16, p. 91). The linkage method used in this paper resolves this conflict into the unified goal of creating a study population of maternal sets that represents the birth and fetal death experience of the source population.…”

“…At least 35 studies using maternally linked datasets from eight U.S. states have been published since development of these datasets began in the mid-1990s. These studies have advanced understanding in several areas of perinatal health, including adverse birth outcomes (6, 10-12), maternal behaviors (13), the quality of vital records data (14, 15), and the quality of maternally linked data (16, 17). …”

Police-Reported Intimate Partner Violence During Pregnancy: Who Is at Risk?

“…An important limitation of this study was the use of birth certificate data to assess demographic factors and maternal reproductive histories, as birth certificate data have limitations in terms of data availability and accuracy (Adams et al, 1997;Buescher, Taylor, Davis, & Bowling, 1993;Clark, Fu, & Burnett, 1997;Dobie et al, 1998;Gould, 1999;McDermott, Drews, Green, & Berg, 1997;Piper et al, 1993). The proportion of missing data in our study was substantial for several demographic as well as a few obstetric history variables; this is particularly important given that at-risk subgroups are overrepresented in records with missing data (Gould).…”