C 2013 Wiley Periodicals, Inc. V
Birth Defects Research (Part A) 97:628–640 (2013)
Racial/Ethnic Disparities in Timing of Death during Childhood among Children with Congenital Heart Defects Wendy N. Nembhard,1* Ping Xu,1 Mary K. Ethen,2 David E. Fixler,3 Jason L. Salemi,1 and Mark A. Canfield2 1
Department of Epidemiology and Biostatistics, College of Public Health, University of South Florida, Tampa, Florida 2 Birth Defects Epidemiology and Surveillance Branch, Texas Department of State Health Services, Austin, Texas 3 Division of Cardiology, Department of Pediatrics, University of Texas Southwestern Medical Center, Dallas, Texas Received 24 May 2013; Revised 27 June 2013; Accepted 2 July 2013
BACKGROUND: Infants with congenital heart defects (CHD) have increased risk of morbidity and mortality. Little is known about racial/ethnic differences in timing of death during childhood. Our intent was to investigate racial/ethnic differences in mortality for CHDs during specific time periods in childhood. METHODS: Texas Birth Defect Registry data were used for a retrospective cohort study with 30,015 singleton infants with a CHD, born January 1, 1999, to December 31, 2007, to non-Hispanic (NH) white, NH-black, or Hispanic women. Texas Birth Defect Registry data were linked to Texas death records to ascertain death. Kaplan–Meier survival probabilities and multivariable Cox-proportional hazard ratios (HRs) and 95% confidence intervals (CIs) were calculated. RESULTS: NH-blacks and Hispanics with specific CHDs had increased mortality during the postneonatal period and early childhood. NH-blacks had increased postneonatal mortality compared with NHwhites for transposition of the great arteries (HR 5 2.4; 95% CI, 1.5–4.0), pulmonary valve atresia without ventricular septal defect (HR 5 4.1; 95% CI, 1.7–9.7), Ebstein’s anomaly (HR 5 8.6; 95 CI, 1.2–61.1), hypoplastic left heart syndrome (HR 5 2.1; 95% CI, 1.2–3.7), coarctation of the aorta (HR 5 2.1; 95% CI, 1.2–3.5), ventricular septal defect (HR 5 2.1; 95% CI, 1.622.8), and atrial septal defect (HR 5 1.4; 95% CI, 1.121.8). Hispanics had increased postneonatal mortality risk for tetralogy of Fallot (HR 5 2.0; 95% CI, 1.1–3.5). Racial/ethnic increases in mortality risk were also observed during infancy and childhood. CONCLUSION: Racial/ethnic differences in mortality were most notably observed during the postneonatal period and early childhood. Future studies should assess factors associated with this disparity in mortality risk for infants C 2013 Wiley Periodicals, Inc. with CHDs. Birth Defects Research (Part A) 97:628–640, 2013. V
Key words: congenital heart defects; survival; mortality; childhood; race; ethnicity; infant
INTRODUCTION Annually, congenital heart defects (CHD) are diagnosed in approximately 1% of all births (Botto et al., 2001) and are the leading cause of mortality among live born infants, the second leading cause among 1- to 4-year-olds and the third leading cause among 5- to 9-year-olds in the United States (Lee et al., 2001; Cleves et al., 2003; Hoffman, 2004; Yang et al., 2006; Lloyd-Jones et al., 2010; Heron, 2011). Infants with CHDs have increased risk of morbidity and mortality (Lee et al., 2001; Cleves et al., 2003; Hoffman, 2004; Yang et al., 2006; Lloyd-Jones et al., 2010) throughout life but especially during infancy and childhood (Nembhard et al., 2008; Gilboa et al., 2010; Khairy et al., 2010). Risk factors associated with increased risk of mortality include pre-
Abbreviations: CHD, congenital heart defects; CI, confidence interval; HLHS, hypoplastic left heart syndrome; HR, hazard ratio; LMP, last menstrual period; NH, non-Hispanic; NDI, National Death Index; TBDR, Texas Birth Defects Registry; TGA, transposition of the great arteries; US, United States; VSU, vital statistics unit; VSD, ventricular septal defect. This study was supported in part by the CDC-funded Texas Center for Birth Defects Research and Prevention (#U50/DD613232), through a cooperative agreement with the Texas Department of State Health Services (DSHS), as well as the Office of Title V & Family Health, Texas DSHS, using Title V Maternal and Child Health Block Grant Funds. *Correspondence to: Wendy N. Nembhard, PhD, Department of Epidemiology and Biostatistics, College of Public Health, University of South Florida, 13201 Bruce B. Downs Blvd., MDC 56, Tampa, FL 33612-3805. E-mail: [email protected] Published online in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/bdra.23169
Birth Defects Research (Part A): Clinical and Molecular Teratology 97:628–640 (2013)
TIMING OF DEATH IN CHILDREN WITH CHDS term birth, low birth weight (Curzon et al., 2008), number and type of co-occurring birth defects (Fixler et al., 2010), CHD phenotype, age at operation (Gerling et al., 2009), infant sex (Seifert et al., 2007), type of surgical procedure and more recently, race/ethnicity (Cleves et al., 2003; Garne, 2004; Hoffman, 2004; Chang et al., 2006; Nembhard et al., 2008; Nembhard et al., 2010). Disparities in survival among persons with CHD by maternal race/ethnicity are present overall and for specific CHD phenotypes (Boneva et al., 2001; Lee et al., 2001; Nembhard et al., 2008; Fixler et al., 2010; Flores, 2010; Gilboa et al., 2010; Nembhard et al., 2011). Technological and medical advances in the United States during the past half century have dramatically improved long-term survival for infants born with CHDs; more people with CHDs are living beyond childhood into adolescence and adulthood (Boneva et al., 2001; Nembhard et al., 2008; Gilboa et al., 2010; Khairy et al., 2010). However, the extent of improvement in survival is not universal for all racial/ethnic groups; the improvement in survival is much less among blacks (Boneva et al., 2001; Nembhard et al., 2008; Gilboa et al., 2010). Boneva and colleagues (2001) reported less decline in mortality rates for non-Hispanic (NH) -blacks than NH-whites from 1979 to 1997. Moreover, racial/ethnic disparities in the age at which death occurs are also present. NH-blacks with CHDs die at earlier ages than NH-whites (Boneva et al., 2001; Chan et al., 2012) and are almost 50% more likely to die during infancy and two times more likely to die during early childhood (Nembhard et al., 2008). Hispanic infants with CHDs are 25 to 40% more likely to die during childhood than NH-white with CHDs (Nembhard et al., 2008; Fixler et al., 2010). Gilboa et al. (2010) reported that NH-blacks with CHDs have higher mortality risk during infancy, childhood, and adolescence; however, Hispanics with CHD had no increased mortality risk during the same time periods. Unfortunately, the etiology of racial/ethnic disparities in the occurrence and timing of death among children with CHDs is poorly understood. While it is clear that racial/ ethnic disparities in age at death are present, it is unclear during which time periods in early childhood minority children with CHDs are most at risk for mortality and which CHD phenotypes have racial/ethnic differences in mortality. Elucidating the time periods and phenotypes for which minority children are most at risk of death may provide clearer insight as to the causes of race/ethnic disparities in mortality. Therefore, the purpose of this study was to determine at what age ranges racial/ethnic differences in survival occur and for which specific CHD phenotypes. We hypothesized that NH-blacks and Hispanics would have lower survival rates than NH-whites during each time period and for most phenotypes.
MATERIALS AND METHODS Study Design We conducted a retrospective cohort study using data from the Texas Birth Defects Registry (TBDR), which is maintained by the Birth Defects Epidemiology and Surveillance Branch of the Texas Department of State Health Services. This statewide, population-based, surveillance system actively ascertains information on all structural and chromosomal birth defects diagnosed within 1 year of delivery. TBDR staff review and abstract medical records from deliv-
629
ery units, pediatric hospitals, and birthing centers where affected children are delivered. Using methods described by Forrester and Canfield (2000), TBDR cases are routinely linked to Texas birth certificates filed with the Vital Statistics Unit. More than 99% of live born registry cases from 1999 to 2007 were linked to their Texas birth certificate.
Study Population All live-born, singleton infants diagnosed with a CHD within the first year of life and born January 1, 1999 to December 31, 2007 to NH-white, NH-black, or Hispanic resident women were selected as the study population. CHDs were classified solely based on the British Pediatric Association extension of the International Classification of Diseases, Ninth Edition Clinical Modification (ICD-9-CM) coding system, as modified by the Division of Birth Defects and Developmental Disabilities of the Centers for Disease Control and Prevention and by the TBDR. CHD phenotypes included in the study were: common truncus (745.000–745.010), transposition of the great arteries (TGA; 745.100–745.190), tetralogy of Fallot (745.200 and 746.840), pulmonary valve atresia without ventricular septal defect (VSD) (746.000 without 745.400–745.490), pulmonary atresia with VSD (746.000 with 745.400–745.490), pulmonary valve stenosis (746.010), tricuspid valve atresia/stenosis (746.100), Ebstein’s anomaly (746.200), hypoplastic left heart syndrome (746.700), aortic valve stenosis (746.300), coarctation of the aorta (747.100–747.190), VSD (745.400– 745.490), atrial septal defect (745.510–745.590), and atrioventricular septal defect (745.600–745.690). Of the 3.4 million live-born infants born in Texas during the study period, 33,352 cases were ascertained by the TBDR as having at least one of the 12 selected CHD phenotypes. After excluding 3337 cases (2331 cases were not singleton live births, 923 cases had maternal race/ ethnicity not included in this study, 73 cases had missing date of death, and 468 cases had trisomy 13 or 18; 175 cases had multiple exclusions), our final study population consisted of 30,015 singleton live-born infants.
Study Variables Maternal and infant medical and sociodemographic information were obtained from birth certificates and medical records. Infant variables included birth era (1999–2003 and 2004–2007), sex, gestational age (20–36 or 371 completed weeks), birth weight (
Birth Defects Research (Part A) 97:628–640 (2013)
Racial/Ethnic Disparities in Timing of Death during Childhood among Children with Congenital Heart Defects Wendy N. Nembhard,1* Ping Xu,1 Mary K. Ethen,2 David E. Fixler,3 Jason L. Salemi,1 and Mark A. Canfield2 1
Department of Epidemiology and Biostatistics, College of Public Health, University of South Florida, Tampa, Florida 2 Birth Defects Epidemiology and Surveillance Branch, Texas Department of State Health Services, Austin, Texas 3 Division of Cardiology, Department of Pediatrics, University of Texas Southwestern Medical Center, Dallas, Texas Received 24 May 2013; Revised 27 June 2013; Accepted 2 July 2013
BACKGROUND: Infants with congenital heart defects (CHD) have increased risk of morbidity and mortality. Little is known about racial/ethnic differences in timing of death during childhood. Our intent was to investigate racial/ethnic differences in mortality for CHDs during specific time periods in childhood. METHODS: Texas Birth Defect Registry data were used for a retrospective cohort study with 30,015 singleton infants with a CHD, born January 1, 1999, to December 31, 2007, to non-Hispanic (NH) white, NH-black, or Hispanic women. Texas Birth Defect Registry data were linked to Texas death records to ascertain death. Kaplan–Meier survival probabilities and multivariable Cox-proportional hazard ratios (HRs) and 95% confidence intervals (CIs) were calculated. RESULTS: NH-blacks and Hispanics with specific CHDs had increased mortality during the postneonatal period and early childhood. NH-blacks had increased postneonatal mortality compared with NHwhites for transposition of the great arteries (HR 5 2.4; 95% CI, 1.5–4.0), pulmonary valve atresia without ventricular septal defect (HR 5 4.1; 95% CI, 1.7–9.7), Ebstein’s anomaly (HR 5 8.6; 95 CI, 1.2–61.1), hypoplastic left heart syndrome (HR 5 2.1; 95% CI, 1.2–3.7), coarctation of the aorta (HR 5 2.1; 95% CI, 1.2–3.5), ventricular septal defect (HR 5 2.1; 95% CI, 1.622.8), and atrial septal defect (HR 5 1.4; 95% CI, 1.121.8). Hispanics had increased postneonatal mortality risk for tetralogy of Fallot (HR 5 2.0; 95% CI, 1.1–3.5). Racial/ethnic increases in mortality risk were also observed during infancy and childhood. CONCLUSION: Racial/ethnic differences in mortality were most notably observed during the postneonatal period and early childhood. Future studies should assess factors associated with this disparity in mortality risk for infants C 2013 Wiley Periodicals, Inc. with CHDs. Birth Defects Research (Part A) 97:628–640, 2013. V
Key words: congenital heart defects; survival; mortality; childhood; race; ethnicity; infant
INTRODUCTION Annually, congenital heart defects (CHD) are diagnosed in approximately 1% of all births (Botto et al., 2001) and are the leading cause of mortality among live born infants, the second leading cause among 1- to 4-year-olds and the third leading cause among 5- to 9-year-olds in the United States (Lee et al., 2001; Cleves et al., 2003; Hoffman, 2004; Yang et al., 2006; Lloyd-Jones et al., 2010; Heron, 2011). Infants with CHDs have increased risk of morbidity and mortality (Lee et al., 2001; Cleves et al., 2003; Hoffman, 2004; Yang et al., 2006; Lloyd-Jones et al., 2010) throughout life but especially during infancy and childhood (Nembhard et al., 2008; Gilboa et al., 2010; Khairy et al., 2010). Risk factors associated with increased risk of mortality include pre-
Abbreviations: CHD, congenital heart defects; CI, confidence interval; HLHS, hypoplastic left heart syndrome; HR, hazard ratio; LMP, last menstrual period; NH, non-Hispanic; NDI, National Death Index; TBDR, Texas Birth Defects Registry; TGA, transposition of the great arteries; US, United States; VSU, vital statistics unit; VSD, ventricular septal defect. This study was supported in part by the CDC-funded Texas Center for Birth Defects Research and Prevention (#U50/DD613232), through a cooperative agreement with the Texas Department of State Health Services (DSHS), as well as the Office of Title V & Family Health, Texas DSHS, using Title V Maternal and Child Health Block Grant Funds. *Correspondence to: Wendy N. Nembhard, PhD, Department of Epidemiology and Biostatistics, College of Public Health, University of South Florida, 13201 Bruce B. Downs Blvd., MDC 56, Tampa, FL 33612-3805. E-mail: [email protected] Published online in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/bdra.23169
Birth Defects Research (Part A): Clinical and Molecular Teratology 97:628–640 (2013)
TIMING OF DEATH IN CHILDREN WITH CHDS term birth, low birth weight (Curzon et al., 2008), number and type of co-occurring birth defects (Fixler et al., 2010), CHD phenotype, age at operation (Gerling et al., 2009), infant sex (Seifert et al., 2007), type of surgical procedure and more recently, race/ethnicity (Cleves et al., 2003; Garne, 2004; Hoffman, 2004; Chang et al., 2006; Nembhard et al., 2008; Nembhard et al., 2010). Disparities in survival among persons with CHD by maternal race/ethnicity are present overall and for specific CHD phenotypes (Boneva et al., 2001; Lee et al., 2001; Nembhard et al., 2008; Fixler et al., 2010; Flores, 2010; Gilboa et al., 2010; Nembhard et al., 2011). Technological and medical advances in the United States during the past half century have dramatically improved long-term survival for infants born with CHDs; more people with CHDs are living beyond childhood into adolescence and adulthood (Boneva et al., 2001; Nembhard et al., 2008; Gilboa et al., 2010; Khairy et al., 2010). However, the extent of improvement in survival is not universal for all racial/ethnic groups; the improvement in survival is much less among blacks (Boneva et al., 2001; Nembhard et al., 2008; Gilboa et al., 2010). Boneva and colleagues (2001) reported less decline in mortality rates for non-Hispanic (NH) -blacks than NH-whites from 1979 to 1997. Moreover, racial/ethnic disparities in the age at which death occurs are also present. NH-blacks with CHDs die at earlier ages than NH-whites (Boneva et al., 2001; Chan et al., 2012) and are almost 50% more likely to die during infancy and two times more likely to die during early childhood (Nembhard et al., 2008). Hispanic infants with CHDs are 25 to 40% more likely to die during childhood than NH-white with CHDs (Nembhard et al., 2008; Fixler et al., 2010). Gilboa et al. (2010) reported that NH-blacks with CHDs have higher mortality risk during infancy, childhood, and adolescence; however, Hispanics with CHD had no increased mortality risk during the same time periods. Unfortunately, the etiology of racial/ethnic disparities in the occurrence and timing of death among children with CHDs is poorly understood. While it is clear that racial/ ethnic disparities in age at death are present, it is unclear during which time periods in early childhood minority children with CHDs are most at risk for mortality and which CHD phenotypes have racial/ethnic differences in mortality. Elucidating the time periods and phenotypes for which minority children are most at risk of death may provide clearer insight as to the causes of race/ethnic disparities in mortality. Therefore, the purpose of this study was to determine at what age ranges racial/ethnic differences in survival occur and for which specific CHD phenotypes. We hypothesized that NH-blacks and Hispanics would have lower survival rates than NH-whites during each time period and for most phenotypes.
MATERIALS AND METHODS Study Design We conducted a retrospective cohort study using data from the Texas Birth Defects Registry (TBDR), which is maintained by the Birth Defects Epidemiology and Surveillance Branch of the Texas Department of State Health Services. This statewide, population-based, surveillance system actively ascertains information on all structural and chromosomal birth defects diagnosed within 1 year of delivery. TBDR staff review and abstract medical records from deliv-
629
ery units, pediatric hospitals, and birthing centers where affected children are delivered. Using methods described by Forrester and Canfield (2000), TBDR cases are routinely linked to Texas birth certificates filed with the Vital Statistics Unit. More than 99% of live born registry cases from 1999 to 2007 were linked to their Texas birth certificate.
Study Population All live-born, singleton infants diagnosed with a CHD within the first year of life and born January 1, 1999 to December 31, 2007 to NH-white, NH-black, or Hispanic resident women were selected as the study population. CHDs were classified solely based on the British Pediatric Association extension of the International Classification of Diseases, Ninth Edition Clinical Modification (ICD-9-CM) coding system, as modified by the Division of Birth Defects and Developmental Disabilities of the Centers for Disease Control and Prevention and by the TBDR. CHD phenotypes included in the study were: common truncus (745.000–745.010), transposition of the great arteries (TGA; 745.100–745.190), tetralogy of Fallot (745.200 and 746.840), pulmonary valve atresia without ventricular septal defect (VSD) (746.000 without 745.400–745.490), pulmonary atresia with VSD (746.000 with 745.400–745.490), pulmonary valve stenosis (746.010), tricuspid valve atresia/stenosis (746.100), Ebstein’s anomaly (746.200), hypoplastic left heart syndrome (746.700), aortic valve stenosis (746.300), coarctation of the aorta (747.100–747.190), VSD (745.400– 745.490), atrial septal defect (745.510–745.590), and atrioventricular septal defect (745.600–745.690). Of the 3.4 million live-born infants born in Texas during the study period, 33,352 cases were ascertained by the TBDR as having at least one of the 12 selected CHD phenotypes. After excluding 3337 cases (2331 cases were not singleton live births, 923 cases had maternal race/ ethnicity not included in this study, 73 cases had missing date of death, and 468 cases had trisomy 13 or 18; 175 cases had multiple exclusions), our final study population consisted of 30,015 singleton live-born infants.
Study Variables Maternal and infant medical and sociodemographic information were obtained from birth certificates and medical records. Infant variables included birth era (1999–2003 and 2004–2007), sex, gestational age (20–36 or 371 completed weeks), birth weight (
