The Association between Maternal Parity and Birth Defects Melanie L. McNeese1, Beatrice Josephine Selwyn1, Hao Duong2, Mark Canfield3, and Dorothy Kim Waller *1
BACKGROUND: Previous studies observed that first birth is associated with an increased risk of some categories of birth defects. However, multiple statistical tests were conducted and it was unclear which of these associations would be replicated in a larger study. We used a large database to assess the association between maternal parity and 65 birth defects including birth defects that have not been previously studied. METHODS: Using data from the Texas Birth Defects Registry for years 1999–2009, the risk of a birth defect occurring in a first, third, or fourth or higher birth was compared to the risk of a birth defect occurring in a second birth. RESULTS: Women having their first birth had significantly increased odds of having an infant with 24 of 65 categories of birth defects when compared to women having their second birth. We also observed associations between first birth and an increased risk of five birth defects not previously reported (small penis, preaxial polydactyly, anomalies of the thoracic vertebrae, anomalies of
Introduction First birth is a well-established risk factor for late fetal death and for infants who are small for gestational age (Shah, 2010). Some studies have observed an increased risk of specific categories of birth defects among women having their first birth (Hay and Barbano, 1972; Akre et al., 1999; Kallen, 2002; Carmichael et al., 2003; Pradat et al., 2003; MacMahon, 2006; Carmichael, 2007; Wang et al., 2008; Agopian, 2009; Duong et al., 2012). Although most of these studies assessed a single birth defect, there are two previous studies that assessed the association between maternal parity and a broad spectrum of birth defects (Hay and Barbano, 1972; Duong et al., 2012). In contrast to studies that observed an increased risk of some categories of birth defects among first births, a smaller number of studies have shown a protective effect of first birth, including Hashmi et al., who found multiparous women had a small increase in risk of having children with non-syndromic cleft lip and palate (Sheiner et al., 1999; Pradat et al., 2003; Hashmi et al., 2005). This study used a cross-sectional design to examine the association between four levels of parity (first birth, second birth, third birth, and fourth or higher birth) and
1 Department of Epidemiology, The University of Texas School of Public Health, Houston, Texas 2 Strategic Information Branch, Centers for Disease Control and Prevention, Hanoi, Viet Nam 3 Birth Defects Epidemiology and Surveillance Branch, Texas Department of State Health Services, Austin, Texas
*Correspondence to: D. Kim Waller, P.O. Box 20186, Houston, TX 77225. E-mail: [email protected] Published online 21 February 2015 in Wiley Online Library (wileyonlinelibrary. com). Doi: 10.1002/bdra.23360
C 2015 Wiley Periodicals, Inc. V
the lumbar vertebrae, and sacroccygeal anomalies). Women having their third or fourth or higher birth had significantly increased odds of giving birth to infants with five of 65 birth defects when compared to second births. CONCLUSIONS: Our observations regarding the categories of birth defects that were associated with first births were highly consistent with observations from two previous studies. Research into biological, behavioral, and environmental factors that may increase the risk of specific birth defects among first births is needed to further explore these associations. Birth Defects Research (Part A) 103:144–156, 2015. C 2015 Wiley Periodicals, Inc. V
Key words: nulliparity; multiparity; parity; congenital abnormalities; birth defects
65 categories of birth defects ascertained by the Texas Birth Defects Registry. We sought to confirm the results of previous studies, especially those of Duong et al. (2012) and Hay and Barbano (1972). Unlike the study by Duong et al. which was based on maternal interviews and had a participation rate of less than 70%, this study uses registry data and therefore, does not have the potential for the selection bias that may have been present in the study by Duong et al. The studies by Duong et al. and Hay and Barbano each included a large number of birth defects and included multiple statistical tests, so replication is necessary to determine whether the same types of birth defects are associated with first births across studies or whether the associations observed are due to chance. To this end, this study uses a different and larger database. Furthermore, we sought to assess the association between maternal parity and a greater number of categories of birth defects than were previously assessed. We also sought to assess the association between higher levels of parity and birth defects.
Materials and Methods The Texas Birth Defects Registry (TBDR) includes cases of birth defects from the entire state, monitoring approximately 380,000 births annually and tracking over 1,000 distinct categories of structural birth defects and chromosomal disorders. All women who reside in Texas at the time of delivery and all infants with a structural or chromosomal defect diagnosed prenatally or within one year of birth are included in the Registry. Diagnoses considered normal variants (e.g., hip abduction and anal fissure) are excluded from the Registry. The TBDR includes cases of live births, fetal deaths, and elective terminations irrespective of the gestational age at the end of the pregnancy. Texas is divided into 11 public health regions, all of which are included in the TBDR.
BIRTH DEFECTS RESEARCH (PART A) 103:144–156 (2015)
Trained staff use active case ascertainment to identify cases from 200 birthing hospitals and licensed birthing centers in Texas. When a potential case is identified, the medical record is abstracted and reviewed by individuals trained in identifying and coding birth defects. Cases meeting certain criteria (e.g., those with more than one birth defect or with a genetic etiology) are reviewed and coded by a board certified geneticist working with the TBDR (Texas Birth Defects Epidemiology and Surveillance, 2006). Coding is based on the British Pediatric Association extension to the codes of the International Classification of Diseases (ICD codes). Case information is updated if pediatric visits or examinations indicate a new or modified diagnosis before the first birthday. By linking registry data to birth certificates and fetal death certificates, the TBDR obtains the following information relevant to this study: mother’s age, ethnicity, education, smoking status, and parity. Previous studies have shown that these variables are accurately reported on birth certificates (Piper et al., 1993; Zollinger et al., 2006). There were 4.2 million live births in Texas between January 1, 1999 and December 31, 2009. Data were obtained from the TBDR and Texas vital statistics for all births during this time, including all cases of major birth defects. All twins and other multiple gestations were excluded from this study because the parity of multiple births is inaccurately recorded in vital statistics records (Waller et al., 2003). Births in which the mother’s parity was missing were also excluded (0.4%). Although pregnancy terminations are included in the TBDR, pregnancy terminations that occur early in gestation (
BACKGROUND: Previous studies observed that first birth is associated with an increased risk of some categories of birth defects. However, multiple statistical tests were conducted and it was unclear which of these associations would be replicated in a larger study. We used a large database to assess the association between maternal parity and 65 birth defects including birth defects that have not been previously studied. METHODS: Using data from the Texas Birth Defects Registry for years 1999–2009, the risk of a birth defect occurring in a first, third, or fourth or higher birth was compared to the risk of a birth defect occurring in a second birth. RESULTS: Women having their first birth had significantly increased odds of having an infant with 24 of 65 categories of birth defects when compared to women having their second birth. We also observed associations between first birth and an increased risk of five birth defects not previously reported (small penis, preaxial polydactyly, anomalies of the thoracic vertebrae, anomalies of
Introduction First birth is a well-established risk factor for late fetal death and for infants who are small for gestational age (Shah, 2010). Some studies have observed an increased risk of specific categories of birth defects among women having their first birth (Hay and Barbano, 1972; Akre et al., 1999; Kallen, 2002; Carmichael et al., 2003; Pradat et al., 2003; MacMahon, 2006; Carmichael, 2007; Wang et al., 2008; Agopian, 2009; Duong et al., 2012). Although most of these studies assessed a single birth defect, there are two previous studies that assessed the association between maternal parity and a broad spectrum of birth defects (Hay and Barbano, 1972; Duong et al., 2012). In contrast to studies that observed an increased risk of some categories of birth defects among first births, a smaller number of studies have shown a protective effect of first birth, including Hashmi et al., who found multiparous women had a small increase in risk of having children with non-syndromic cleft lip and palate (Sheiner et al., 1999; Pradat et al., 2003; Hashmi et al., 2005). This study used a cross-sectional design to examine the association between four levels of parity (first birth, second birth, third birth, and fourth or higher birth) and
1 Department of Epidemiology, The University of Texas School of Public Health, Houston, Texas 2 Strategic Information Branch, Centers for Disease Control and Prevention, Hanoi, Viet Nam 3 Birth Defects Epidemiology and Surveillance Branch, Texas Department of State Health Services, Austin, Texas
*Correspondence to: D. Kim Waller, P.O. Box 20186, Houston, TX 77225. E-mail: [email protected] Published online 21 February 2015 in Wiley Online Library (wileyonlinelibrary. com). Doi: 10.1002/bdra.23360
C 2015 Wiley Periodicals, Inc. V
the lumbar vertebrae, and sacroccygeal anomalies). Women having their third or fourth or higher birth had significantly increased odds of giving birth to infants with five of 65 birth defects when compared to second births. CONCLUSIONS: Our observations regarding the categories of birth defects that were associated with first births were highly consistent with observations from two previous studies. Research into biological, behavioral, and environmental factors that may increase the risk of specific birth defects among first births is needed to further explore these associations. Birth Defects Research (Part A) 103:144–156, 2015. C 2015 Wiley Periodicals, Inc. V
Key words: nulliparity; multiparity; parity; congenital abnormalities; birth defects
65 categories of birth defects ascertained by the Texas Birth Defects Registry. We sought to confirm the results of previous studies, especially those of Duong et al. (2012) and Hay and Barbano (1972). Unlike the study by Duong et al. which was based on maternal interviews and had a participation rate of less than 70%, this study uses registry data and therefore, does not have the potential for the selection bias that may have been present in the study by Duong et al. The studies by Duong et al. and Hay and Barbano each included a large number of birth defects and included multiple statistical tests, so replication is necessary to determine whether the same types of birth defects are associated with first births across studies or whether the associations observed are due to chance. To this end, this study uses a different and larger database. Furthermore, we sought to assess the association between maternal parity and a greater number of categories of birth defects than were previously assessed. We also sought to assess the association between higher levels of parity and birth defects.
Materials and Methods The Texas Birth Defects Registry (TBDR) includes cases of birth defects from the entire state, monitoring approximately 380,000 births annually and tracking over 1,000 distinct categories of structural birth defects and chromosomal disorders. All women who reside in Texas at the time of delivery and all infants with a structural or chromosomal defect diagnosed prenatally or within one year of birth are included in the Registry. Diagnoses considered normal variants (e.g., hip abduction and anal fissure) are excluded from the Registry. The TBDR includes cases of live births, fetal deaths, and elective terminations irrespective of the gestational age at the end of the pregnancy. Texas is divided into 11 public health regions, all of which are included in the TBDR.
BIRTH DEFECTS RESEARCH (PART A) 103:144–156 (2015)
Trained staff use active case ascertainment to identify cases from 200 birthing hospitals and licensed birthing centers in Texas. When a potential case is identified, the medical record is abstracted and reviewed by individuals trained in identifying and coding birth defects. Cases meeting certain criteria (e.g., those with more than one birth defect or with a genetic etiology) are reviewed and coded by a board certified geneticist working with the TBDR (Texas Birth Defects Epidemiology and Surveillance, 2006). Coding is based on the British Pediatric Association extension to the codes of the International Classification of Diseases (ICD codes). Case information is updated if pediatric visits or examinations indicate a new or modified diagnosis before the first birthday. By linking registry data to birth certificates and fetal death certificates, the TBDR obtains the following information relevant to this study: mother’s age, ethnicity, education, smoking status, and parity. Previous studies have shown that these variables are accurately reported on birth certificates (Piper et al., 1993; Zollinger et al., 2006). There were 4.2 million live births in Texas between January 1, 1999 and December 31, 2009. Data were obtained from the TBDR and Texas vital statistics for all births during this time, including all cases of major birth defects. All twins and other multiple gestations were excluded from this study because the parity of multiple births is inaccurately recorded in vital statistics records (Waller et al., 2003). Births in which the mother’s parity was missing were also excluded (0.4%). Although pregnancy terminations are included in the TBDR, pregnancy terminations that occur early in gestation (
