Birth outcomes of Asian-Indian-Americans

International Journal of Gynecology and Obstetrics (2007) 97, 215–220

a v a i l a b l e a t w w w. s c i e n c e d i r e c t . c o m

w w w. e l s e v i e r. c o m / l o c a t e / i j g o

SPECIAL ARTICLE

Birth outcomes of Asian-Indian-Americans G.R. Alexander a , M.S. Wingate b,⁎, J. Mor c , S. Boulet b a b c

University of South Florida, USA University of Alabama at Birmingham, USA University of Hawaii, USA

Received 13 April 2006; received in revised form 15 February 2007; accepted 15 February 2007

KEYWORDS Birth weight; Gestational age; Low birth weight; Preterm; Small-for-gestational age; Infant/neonatal mortality; Fetal growth; Race; Ethnicity; African-American

Abstract Objective: This study examines the maternal characteristics and birth outcomes of infants of U.S. resident Asian-Indian-American (AIA) mothers and compares those to infants of U.S. resident Whites and African-American (AA) mothers. Methods: Single live births to U.S. resident mothers with race/ethnicity coded on birth certificate as AIA, non-Hispanic White, or non-Hispanic AA were drawn from NCHS 1995 to 2000 U.S. Linked Live Birth/Infant Death files. Results: Compared to AAs or Whites, AIAs have the lowest percentage of births to teen or unmarried mothers and mothers with high parity for age or with low educational attainment. After taking these factors into account, AIA had the highest risk of LBW, small-for-gestational age and term SGA births but a risk of infant death only slightly higher than Whites and far less than AAs. Conclusions: The birth outcomes of AIAs do not follow the paradigm that more impoverished minority populations should have greater proportions of low birth weight and preterm births and accordingly greater infant mortality rates. © 2007 International Federation of Gynecology and Obstetrics. Published by Elsevier Ireland Ltd. All rights reserved.

1. Introduction Asian-Americans are a growing population in the U.S. and encompass several distinct groups including Chinese, Filipino, Japanese, Korean, Vietnamese, and Asian-Indian. According to vital statistic data, Asian and Pacific Islanders accounted for 5.2% of all live U.S. births in 2002, an increase ⁎ Corresponding author. Department of Health Care Organization and Policy, School of Public Health, University of Alabama at Birmingham, RPHB 330, 1530 3rd Avenue South, Birmingham, Alabama 35294-0022, USA. Tel.: +1 205 934 6783; fax: +1 205 934 3347. E-mail address: [email protected] (M.S. Wingate).

from 3.4% in 1990 [1–3]. However, even though the percentage of Asian-Indian-American births has nearly doubled since the late 1980s, Asian-Indian-Americans (AIAs) have received relatively less research attention in the United States compared to other groups of “Asian-Americans.” Previous research from Asia and Europe has revealed some distinctive birth outcomes for immigrant Asian-Indian mothers [4–7]. Reports from India indicate that the infants of Asian-Indian mothers tend to have smaller average birth weights and a higher incidence of low birth weight infants than those of Europeans or Americans. Furthermore, infants born in Europe to immigrant Asian-Indian mothers are also lighter on average than their European counterparts [5,6]. Notwithstanding, perinatal mortality rates

0020-7292/$ - see front matter © 2007 International Federation of Gynecology and Obstetrics. Published by Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ijgo.2007.02.017

216

G.R. Alexander et al.

among Asian-Indian infants weighing between 1500 g and 2400 g are significantly lower than their White European counterparts [5]. The paradoxical finding of Asian-Indians having relatively higher low birth weight rates but comparable infant mortality rates suggests an exception to prevailing theories regarding the contribution of low birth weight and its relationship to infant death rates. A somewhat similar conundrum has been identified for Japanese-Americans in the United States whereby their infant mortality rates approach those of U.S. Whites in spite of having markedly higher percentages of low birth weight [8]. Madan et al reported lower mean birth weights and higher rates of small-for-gestational age among infants born to mothers of Asian-Indian origin as compared to infants of White mothers [9]. A later study also demonstrated elevated rates of low birth weight among infants of U.S. resident, foreign-born Asian-Indian mothers as compared to White infants, irrespective of the fact that certain protective factors occurred with a greater frequency among Asian-Indian women [10]. The present study examines the maternal characteristics and birth outcomes of infants born in the U.S. to resident AIA mothers and compares them to infants born to U.S. resident non-Hispanic Whites and non-Hispanic African-American mothers in the United States. Further investigated is whether the birth outcomes of AIAs in the U.S. also exhibited a paradoxical relationship between their measures of low birth weight, preterm and small fetal growth and their infant mortality rates.

2. Methods The data were drawn from the NCHS 1995 to 2000 U.S. Linked Live Birth/Infant Death files [11]. Single live births to U.S. resident mothers, whose maternal race/ethnicity was coded on the birth certificate as either Asian-Indian, non-Hispanic White, or non-Hispanic African-American (Black), were selected for analysis. The selected births were from the states of California, Hawaii, Illinois, New Jersey, New York, Texas, and Washington, as these were the states using the expanded coding of AsianAmericans during this time period. The selected study sample included 3,826,996 births to non-Hispanic White mothers, 1,045,714 births to non-Hispanic African-American mothers, and 102,739 births to AIA mothers. The analysis entailed an initial examination of race/ethnic group variations in maternal risk factors, prenatal care utilization and adverse birth outcomes. Chi square and ANOVA were used to test for significant race/ethnic differences in the rates of proportions or distribution means of maternal socio-demographic characteristics, initiation and adequacy of prenatal care utilization, and birth outcomes. Multiple logistic regression was then employed to calculate odds ratios and 95% confidence intervals for the independent effects of maternal factors on birth outcomes and to estimate the risk of an adverse birth outcomes for AIAs and non-Hispanic African-American, compared to non-Hispanic Whites (the reference group), before and after adjusting for maternal and prenatal care risk factors. Birth weight distributions and fetal growth curves were also visually inspected for the race/ethnic groups.

Table 1 Maternal and prenatal care utilization characteristics by race/ethnicity of mothers, 1995–2000 single live births to U.S. resident mothers Characteristics

Asian-Indian

African-American

White

p-value a

% unmarried % age b 18 % age ≥ 35 % low education b % high education c % primipara d % high parity e % non-U.S.-born % hypertension % diabetes

8.4 0.3 11.4 9.5 68.7 48.7 0.5 95.5 2.3 7.2

67.4 8.2 10.9 20.4 36.2 38.9 6.1 15.2 4.9 2.7

19.9 2.5 17.5 9.4 58.8 42.4 1.7 8.5 4.1 2.5

b .01 b .01 b .01 b .01 b .01 b .01 b .01 b .01 b .01 b .01

Prenatal care utilization f % intensive % adequate % intermediate % inadequate % no care % missing care % 1st trimester # of births

4.4 37.2 40.0 8.6 1.5 8.3 82.0 102,739

5.4 29.9 37.5 13.2 2.8 11.2 72.4 1,045,714

6.2 41.9 38.9 5.7 0.7 6.7 87.7 3,826,996

b .01 b .01 b .01 b .01 b .01 b .01 b .01

a

Chi square used to determine p-values. Less than 12 years of education for adults; for adolescents (b18 years), 2+ years below expected grade level for age. c 13 or more years of education for adults; for adolescents, 2+ years above expected grade level. d Determined by number of previous live births on birth certificate. e One or more previous births for adolescents (N18), 3 or more previous births for 18–21 years; 4 or more previous births for 22–24 years; 5 or more previous births for 25 and older. f Defined by R-GINDEX [12]. Incorporates trimester prenatal care began, number of visits, and gestational age of infant at birth. b

Birth outcomes of Asian-Indian-Americans The maternal risk factors examined in this study included: marital status, adolescent age of mother (b 18 years of age), older aged mothers (N 35 years of age), parity, low and high maternal education attainment, nativity status of the mother (U.S. versus foreign-born), adequacy of prenatal care utilization, and a report of maternal diabetes and/or hypertension. Definitions for the coding on these variables are provided in Table 1. Adequacy of prenatal care utilization was defined by RGINDEX [12]. In addition to neonatal (b 28 days) and infant (b 1 year) mortality rates, the adverse birth outcomes examined included low birth weight (LBW: b 2500 g), very low birth weight (VLBW: b 1500 g), preterm (b 37 weeks gestation), small-for-gestational age (SGA: b 10th percentile of birth weight for gestational age) and term small-for-gestational (SGA birth at 37–41 weeks gestation). Percentiles of the birth weight distribution were calculated for each completed week of gestational age, based on the interval between the recorded dates of last normal menses and the birth. Infants were classified as SGA using the 10th percentile of birth weight values derived from a previously reported 1991 U.S. birth weight for gestational age reference curve [13].

3. Results

217 Table 2 Birth outcomes of Asian-Indian, African-American, and White mothers, 1995–2000 single live births to U.S. resident mothers Birth outcome

AsianIndian

AfricanAmerican

White

Mean birth weight (g) % VLBW a % LBW b Mean gestational age (weeks) % very preterm c % preterm d % SGA e % term SGA f Neonatal mortality rate (NMR) g Infant mortality rate (IMR) h

3170 0.9 8.0 38.86

3158 2.5 11.0 38.49

3438 0.7 4.3 39.11

1.3 8.7 16.4 14.6 2.9

3.8 15.0 14.8 12.3 7.1

1.2 7.6 7.0 6.1 2.7

4.1

11.4

4.4

388.6 37.0 6.5 1.8 2.8

377.9 41.2 9.8 4.1 5.0

410.3 45.0 8.5 2.2 2.4

Gestational agespecific IMR: b 28 28–32⁎ 33–36 37–41 42+

Table 1 presents maternal and prenatal care characteristics by race/ethnic group. Significant (p b .01) differences were evident in the proportion of every maternal characteristic. Compared to African-Americans or Whites, AIAs had the lowest percentage of births to unmarried mothers (8.4%), mothers under age 18 (0.3%) and mothers with high parity for age (0.5%). Over two-thirds of AIA mothers reported high educational attainment, compared to 36.2% of AfricanAmericans and 58.8% of Whites. The vast majority (95.5%) of AIA mothers were non-U.S.-born. The three race/ethnic groups were somewhat similar in their proportions of primiparity, although Asian-Indian mothers were slightly more likely than African-American and Whites to be primiparous. AIA mothers were the least likely to report intensive utilization of prenatal care (4.4%). The birth weight distributions of African-Americans and AIAs were similarly positioned on the birth weight axis but are shifted toward the lower end of the birth weight scale, relative to the White distribution (Fig. 1). Below 1750 g, the

AIA and White distributions were similar, whereas the African-American distribution exhibited an excess proportion of b 1500 g births (as indicated in the inset of Fig. 1). Conversely, between 1750 and 2500 g, the birth weight

Figure 1 Birth weight distribution of White, African-American (AA), and Asian-Indian-American (AIA) mothers, 1995–2000 single live births to U.S. resident mothers.

Figure 2 10th percentile of birth weight for gestational age by race/ethnicity of mother, 1995–2000 single live births to U.S. resident mothers.

Note: All comparisons are significant based on Chi Square (p b.01, except comparison ⁎: pb .05) or ANOVA (for the continuous variables: Fb 01). a Very low birth weight: b1500 g. b Low birth weight: b2500 g. c Very preterm: b 33 weeks gestation. d Preterm: b37 weeks gestation. e Small-for-gestational age: b10th percentile of birth weight for gestational age [13]. f Term SGA: SGA birth born at 37–41 weeks gestation. g Deaths b 28 days/1000 live births. h Deaths b 1 year/1000 live births.

218

G.R. Alexander et al.

Figure 3 Birth weight-specific infant mortality by race/ethnic groups, 1995–2000 single live births to U.S. resident mothers.

distribution of African-Americans and AIAs was comparable while lower percentages of LBW were observed for Whites. Table 2 displays rates and proportions of selected adverse birth outcomes by race/ethnic group. Whites demonstrated the lowest proportions of LBW and preterm deliveries, and African-Americans had the highest proportions. The very preterm and VLBW percentages of AIAs were more similar to Whites. AIAs exhibited the highest percentages of small-forgestational age (16.4%) and term SGA (14.6%), followed by African-Americans and then by Whites. Visual examination of the 10th percentile birth weight–gestational age distribution suggests that AIAs and African-Americans had similar fetal growth patterns, with the AIA birth weight percentiles at term being slightly lighter than those of African-Americans (Fig. 2). At the 50th and 90th percentiles of birth weight for gestational age, AIAs experienced lower birth weights than African-Americans and Whites for every gestational age (data not shown). Neonatal mortality rates for AIAs (2.9 neonatal deaths per 1000 live births) were only slightly higher than Whites (2.7). The infant mortality rate of AIAs (4.1) was less than half that of African-Americans (11.4) and slightly less than that of Whites (4.4). Within gestational age categories above 28 weeks, AIAs had the lowest gestational age-specific infant mortality rates. African-Americans had a survival advantage for gestational ages b 28 weeks. Birth weightspecific infant mortality rates by race/ethnic group are displayed in Fig. 3. AIAs had the lowest infant mortality risk between 1000 and 3500 g.

Table 3

Table 3 presents unadjusted and adjusted odds ratios for each birth outcome among AIAs and African-Americans, as compared to the White reference group. Compared to Whites, both AIAs and African-Americans had significantly increased risks for most adverse birth outcomes, the exception being the mortality indicators for AIAs. For AIAs, these risks increased after adjustment, yet risks for AfricanAmericans decreased after adjustment for maternal sociodemographic and prenatal care characteristics. The opposite was observed for African-Americans. For example, AIAs had an unadjusted risk of 1.15 [CI 1.12–1.17] for preterm birth, compared to an unadjusted risk of 2.14 [CI 2.13–2.16] for African-Americans. When adjusted, the risk of preterm for AIAs increased to 1.39 [CI 1.36–1.43], while African-American adjusted risk decreased to 1.91 [CI 1.89–1.92]. Both AIAs and African-Americans had a significantly higher risk of LBW than Whites. However, when the maternal and prenatal care factors were added to the logistic regression model, the adjusted odds ratios and confidence intervals revealed that AIAs had the greatest risk of LBW compared to Whites. The odds of a SGA and term SGA birth were highest among AIAs, within both the unadjusted and adjusted models. The unadjusted odds of a neonatal or infant death were not significantly different for Whites and AIAs. However, the adjusted odds of neonatal and infant mortality were only slightly significant. For infant mortality, the unadjusted risk among AIAs was 0.93 (CI: 0.84–1.02), and the adjusted odds were 1.18 (CI: 1.06–1.31). Regardless of adjustment, African-Americans had a greater than twofold risk of neonatal and infant mortality when compared to Whites and AIAs.

4. Discussion The study results indicate that AIA mothers demonstrated lower proportions of several adverse socio-demographic risk characteristics as compared to non-Hispanic White and African-American mothers, although the proportion of mothers with reported diabetes among AIAs was more than twice that of Whites or African-Americans. Theoretically, the relatively lower proportions of mothers with high-risk maternal socio-demographic risk factors among AIAs should be predictive of proportionally lower rates of LBW and SGA, as compared to Whites and African-Americans. However, this was not the case, as these results revealed a higher incidence

Unadjusted and adjusted risks for adverse birth outcomes, 1995–2000 single live births to U.S. resident mothers

Birth outcome

Very preterm Preterm VLBW LBW SGA Term SGA Neonatal mortality Infant mortality

Asian-Indian

African-American

Unadjusted

Adjusted

1.12 1.15 1.31 1.92 2.60 2.65 1.06 0.93

1.38 1.39 1.55 2.48 2.99 2.98 1.23 1.18

(1.06–1.18) (1.12–1.17) (1.23–1.40) (1.87–1.96) (2.56–2.65) (2.61–2.70) (0.94–1.19) (0.84–1.02)

a

(1.30–1.47) (1.36–1.43) (1.45–1.67) (2.42–2.55) (2.93–3.05) (2.92–3.05) (1.08–1.40) (1.06–1.31)

Unadjusted

Adjusteda

3.39 2.14 3.65 2.72 2.31 2.18 2.63 2.60

2.76 1.91 3.18 2.27 1.85 1.76 2.29 2.01

(3.35–3.44) (2.13–2.16) (3.59–3.71) (2.71–2.75) (2.29–2.32) (2.17–2.20) (2.55–2.71) (2.54–2.67)

(2.71–2.80) (1.89–1.92) (3.11–3.23) (2.25–2.29) (1.83–1.86) (1.74–1.77) (2.21–2.37) (1.96–2.07)

a Adjusted for: marital status, maternal age, maternal educational attainment, parity, nativity of mother, prenatal care utilization, diabetes, and hypertension. White is the reference group.

Birth outcomes of Asian-Indian-Americans of LBW and SGA. Moreover, while the average birth weight of infants of AIA mothers is similar to that of African-Americans and approximately 270 g lighter than Whites, their infant and neonatal mortality risks are markedly lower than those of African-American infants and comparable to White infants, even after controlling for socio-demographic characteristics. This suggests that other unmeasured factors, both sociodemographic and otherwise, may potentially influence their birth outcomes. The results of this study are similar to those of European and U.S.-based investigations [6,7,9,10]. Infants of British Asian-Indian mothers have been shown to have different birth weights, growth patterns, and gestational periods as compared to British White mothers. As reported by Dawson, moderately low birth weight Indian infants had better crude perinatal mortality rates than Caucasians, while the overall mortality rates of Asian-Indian infants were slightly, but not significantly, higher than Whites, suggesting that AsianIndian babies considered “low birth weight” may not behave clinically like similar White infants [5]. Correspondingly, Gould reported lower rates of birth weight-specific mortality among Asian-Indian infants as compared to Whites, despite a higher incidence of LBW [10]. The differences between White Europeans and AsianIndians have been explained in European literature through various theories, including genetic variations and maternal malnutrition [4,7]. A connection between birth weight and maternal size may be attributable to certain genetic factors related to the shortness or smaller size of the mother caused by undernourishment occurring during childhood [7]. Alternatively, the differential patterns of growth observed among Asian-Indian infants may be attributable to a different body habitus among this ethnic group and may be due to genetic factors, not suboptimal growth [9,14]. The results of this investigation, focusing mainly on sociodemographic characteristics, provide no evidence against these theories, leaving them open for further investigation. This study's findings are circumscribed by factors related to secondary data analysis, including the limited number of vital record variables available for investigation and the potential miscoding and/or misclassification of the variables utilized. Similar to Hispanic-Americans, the birth outcomes of AIAs have been referred to as a paradox, since these outcomes do not follow the paradigm that more impoverished minority populations should have greater proportions of low birth weight and preterm births and accordingly greater infant mortality rates [15,16]. Instead, AIAs are similar to Japanese-Americans in having higher LBW and SGA rates than their socio-demographic characteristics would suggest, but still maintaining relatively equivalent infant mortality as Whites [8,17]. Samoans are yet another ethnic group that does not fit conventional thinking due to having relatively lower LBW rates but higher infant mortality rates compared to White infants [17]. The increasing number of these paradoxical observations begs the question of whether these are indeed paradoxes or whether the traditional paradigm is flawed. Rather than label these populations as paradoxes because they do not meet existing expectations, it may be time for researchers to seriously reevaluate the limitations and generalizability of their theories regarding the determinants of ethnic diversity in

219 perinatal outcomes. Birth weight serves as a good proxy indicator of fetal maturity, but may not be equally valid as an indicator of mortality risk for different ethnic groups [18]. The results of this study suggest that there should be less emphasis on the use of low birth weight for ethnic comparison. Very low birth weight and very preterm birth rates may be better universal indicators of infant mortality risk regardless of ethnic group. Understanding how ethnic differences in birth weight translate into disparate risks is a critical step toward assuring equal and better quality health care to all groups.

Acknowledgments This work was supported in part by DHHS, HRSA and MCHB grants MCJ-9040.

References [1] Martin JA, Hamilton BE, Sutton PD, Ventura SJ, Menacker F, Munson ML. Births: final data for 2002. Natl Vital Stat Rep 2003;52(10):1–113. [2] Census 2000 Summary File/[United States] prepared by the U.S. Census Bureau, 2001. [3] Census of Population and Housing, 1990: Summary Tape File 1 (United States), Bureau of the Census—Washington: The Bureau, 1991. [4] Meire HB, Farrant P. Ultrasound demonstration of an unusual fetal growth pattern in Indians. Br J Obstet Gynaecol 1981; 88(3):260–3. [5] Dawson I, Golder RY, Jonas EG. Birthweight by gestational age and its effect on perinatal mortality in white and in Punjabi births: experience at a district general hospital in West London 1967–1975. Br J Obstet Gynaecol 1982;89(11):896–9. [6] Clarson CL, Barker MJ, Marshall T, Wharton BA. Secular change in birthweight of Asian babies born in Birmingham. Arch Dis Child 1982;57(11):867–71. [7] Davies DP, Senior N, Cole G, Blass D, Simpson K. Size at birth of Asian and White Caucasian babies born in Leicester: implications for obstetric and paediatric practices. Early Hum Dev 1982;6(3):257–63. [8] Mor JM, Alexander GR, Kogan MD, Kieffer EC, Ichiho HM. Similarities and disparities in maternal risk and birth outcomes of White and Japanese-American mothers. Pediatr Perinat Epidemiol 1995;9(1):59–73. [9] Madan A, Holland S, Humbert JE, Benitz WE. Racial differences in birth weight of term infants in a northern California population. J Perinatol 2002;22(3):230–5. [10] Gould JB, Madan A, Qin C, Chavez G. Perinatal outcomes in two dissimilar immigrant populations in the United States: a dual epidemiologic paradox. Pediatrics 2003;111(6 Pt 1): e676–82. [11] 1995, 1996, 1997, 1998, 1999, 2000 Birth Cohort Linked Birth/ Infant Death Data Sets, NCHS CD-ROM Series 20, Numbers 12a, 14a, 15a, ASCII Version, US Dept Health and Human Services, Centers for Disease Control and Prevention. [12] Alexander GR, Kotelchuck M. Quantifying the adequacy of prenatal care: a comparison of indices. Public Health Rep 1996;111(5):408–19. [13] Alexander GR, Himes JH, Kaufman R, Mor JM, Kogan MD. A United States national reference for fetal growth. Obstet Gynecol 1996;87(2):163–8. [14] Mathai M, Thomas S, Peedicayil A, Regi A, Jasper P, Joseph R. Growth pattern of the Indian fetus. Int J Gynecol Obstet 1995;48(1):21–4.

220 [15] Markides K, Corei LJ. The health of Hispanics in the southwestern United States: an epidemiologic paradox. Public Health Rep 1986;101(3):253–65. [16] Fuentes-Afflick E, Hessol NA, Perez-Stables EJ. Testing the epidemiologic paradox of low birth weight in Latinos. Arch Pediatr Adolesc Med 1999;153:147–53.

G.R. Alexander et al. [17] Alexander GR, Kogan MD. Ethnic differences in birth outcomes: the search for answers continues (commentary). Birth 1998;23(3):210–3. [18] Alexander GR, Tompkins ME, Allen MC, Hulsey TC. Trends and racial differences in birth weight and related survival. Matern Child Health J 1999;3(1):71–9.