Objective
The objective of the study was to examine racial and ethnic differences in preterm births in infants conceived by in vitro fertilization (IVF).
Study Design
A retrospective cohort study was conducted of 97,288 singleton and 40,961 twin pregnancies resulting from fresh, nondonor IVF cycles using 2006-2010 data from the Society for Assisted Reproductive Technology Clinic Online Reporting System.
Results
Rates of very early preterm (<28 weeks), early preterm (<32 weeks), and preterm birth (<37 completed weeks) varied across racial and ethnic groups in both singleton and twin pregnancies. In singletons, with white women as the referent, after adjustment of confounding variables, the adjusted odds ratios (ORs) and 95% confidence intervals (CIs) of very early preterm birth, early preterm birth, and preterm birth in black women were 4.8 (95% CI, 4.1–5.7), 3.9 (95% CI, 3.4–4.4), and 2.1 (95% CI, 1.9–2.3). Hispanic women had a significantly lower rate of preterm births as compared with black women and similar or slightly higher rates as compared with white women. Native American women were not at an increased risk of any types of preterm births; Asian women were at a reduced risk of preterm twin births (adjusted OR, 0.8; 95% CI, 0.7–0.9).
Conclusion
There exist notable racial and ethnic disparities in preterm births in infants conceived by IVF, suggesting that mechanisms other than socioeconomic disparities contribute to this difference.
Preterm birth is one of the leading causes of infant morbidity and mortality worldwide. The rate of preterm births has been increasing in most developed countries over the last several decades. Because the etiology for preterm birth is complex and multifactorial, preventive strategies have not been successful in all patients, and much is yet to be known about optimal treatment.
In the United States, there exist great disparities in preterm birth and other pregnancy outcomes among women of different races and ethnicities. In 2010, the preterm rate (birth before 37 completed weeks) was 10.8% among white women, 17.1% among black women, and 11.8% among Hispanic women, respectively. The rates of preterm birth, low birthweight, and infant mortality in black women are approximately double that of white women, and the known risk factors explain only a fraction of this difference.
The increased risks for black women have been attributed in part to poor demographics and socioeconomic status (SES) as measured by income, education, or occupation. However, these explanations have been challenged by the findings that Hispanic women (especially in women of Mexican origin) have been consistently shown to have lower rates of preterm birth, low birthweight, and infant mortality as compared with black women and similar rates as compared with white women, a phenomenon that is referred to as the Hispanic (epidemiological) paradox. Hispanic women in general have a lower SES (eg, lower incomes, lower levels of education, less health insurance, and less prenatal care), similar to black women. However, the rates of preterm birth in Hispanic women, as well as low birthweight and infant mortality, are similar to those of white women (and much lower than their black counterparts).
In the United States, the average cost of in vitro fertilization (IVF) is $12,400 per treatment cycle. Many women require more than 1 treatment cycle, which can increase the cost to as much as $85,000 because of the cumulative cost of failed cycles and medications. This cost is primarily paid from personal sources because the majority of medical insurance carriers do not cover IVF. Thus, women who undergo IVF and other assisted reproductive technologies (ARTs) usually have a relatively higher SES. If SES is an important risk factor for preterm birth, one may expect that there should be fewer racial and ethnic disparities in preterm births in infants conceived by IVF. However, in preterm births resulting from IVF, few studies have specifically examined whether there are racial and ethnic disparities or whether there exists a Hispanic epidemiological paradox. To clarify these questions, we conducted a retrospective cohort study to examine racial and ethnic differences in preterm births of infants resulting from IVF, using data for 2006-2010 births from the Society for Assisted Reproductive Technology Clinic Online Reporting System (SART CORS).
Materials and Methods
Study population
The records from the SART CORS dataset used included all singleton and twin births resulting from fresh nondonor oocyte transfers for the years 2006-2010. Data from more than 90% of clinics performing ART in the United States have been collected, verified by the Society for Assisted Reproductive Technology and reported to the Centers for Disease Control and Prevention since 1992 in compliance with the Fertility Clinic Success Rate and Certification Act of 1992 (Public Law 102-493, Oct. 24, 1992). In 2006, maternal height and weight were added as optional data entry fields. Approval for this study was obtained from the Society for Assisted Reproductive Technology and the Louisiana State University School of Medicine and Tulane University institutional review boards.
Records of 141,030 births resulting from fresh nondonor IVF procedures in 2006-2010 were available from SART CORS. The population we studied was limited to cycles in which the length of gestation, calculated by adding 14 days to the number of completed weeks since fertilization, was between 20 completed weeks (140 days) and 44 completed weeks (308 days).
After exclusion of triplets (n = 2302), quadruplets (n = 69), and women with mixed or multiple-race or ethnicity (n = 410), 97,288 singleton and 40,961 twin records were included for analysis.
Definition of race and ethnicity, preterm births, and covariates
In the SART CORS dataset, race and ethnicity were self-reported by the patient as white, black, Hispanic, Asian, and Native (including American Indian or Alaska Native). Women without race/ethnicity information and women who answered with more than 1 race (eg, Hispanic/white) were categorized as unknown for analysis, and their characteristics and rate of preterm births were compared with the total of study population for assessing potential impacts.
Preterm births were classified according to the number of gestational weeks completed as very early preterm (>20 but <28 completed weeks; 140-196 days), early preterm (>20 but <32 completed weeks; 140-224 days), and preterm (>20 but <37 completed weeks; 140-259 days).
Potentially confounding variables included maternal age (ie, <30, 30-39, and ≥30 years old); gravidity; smoking; day of embryo transfer; number of embryos transferred; number of gestational sacs with heart rate; prepregnancy body mass index (BMI) (ie, normal weight [BMI <25.0 kg/m 2 ], overweight [BMI 25.0-29.9 kg/m 2 ], obese [≥BMI 30.0 kg/m 2 ]); prior spontaneous abortions; and infertility diagnosis (ie, male, uterine, and other unexplained factors). The database did not include information about maternal medical complications (hypertension or diabetes), or mode of delivery (vaginal or caesarean section) and whether delivery was spontaneous or iatrogenic.
Statistical analysis
Analysis of variances (ANOVA) was used to compare difference in mean gestational days at delivery between different racial and ethnic groups. A χ 2 test was used to compare the percent of very early preterm, early preterm, and preterm deliveries by race or ethnicity. Using white women as the referent group, the relative risk or odds ratio (OR) and 95% confidence interval (CI) of preterm births were calculated for each of the other racial and ethnic groups. Multivariable logistic regression was used to adjust for the confounders indicated above. The adjusted ORs and their 95% CIs were derived from the coefficients of the logistic models and their SEs. All P values were 2 tailed, and the significance level selected was P = .05. After adjustment, the adjusted ORs were largely identical to crude ORs. Statistical analyses were performed with SPSS 16.0 for Windows (SPSS Inc, Chicago, IL).
Results
In the study population, the overall rate of very early preterm, early preterm, and preterm birth was 1.5%, 2.7%, and 13.3%, respectively, for 97,288 singletons and 4.7%, 10.9%, and 36.9%, respectively, for 40,961 twins. Table 1 shows the characteristics of the study population by race and ethnicity. Compared with other racial/ethnic groups, black women had a higher proportion of women aged 40 years or older, gravidity 3 or greater, and weight in the overweight or obese category but a lower rate of prior preterm births. Asian women had low frequency of smoking during pregnancy and being overweight or obese. Hispanic, Native American, and black women had a higher rate of prior IVF or ART than white and Asian women. There were no significant differences in the number of fresh embryos transferred and the number of sacs with a heart rate among different racial and ethnic groups. All comparisons in Table 1 were statistically significant likely because of the large sample examined.
| Characteristic | White, % (n = 67,238) | Black, % (n = 4910) | Hispanic, % (n = 6773) | Asian, % (n = 8548) | Native, % (n = 527) | Unknown, % (n = 50,662) | All groups, % (n = 138,659) |
|---|---|---|---|---|---|---|---|
| Maternal age, y | |||||||
| <30 | 18.5 | 16.5 | 19.3 | 14.0 | 18.4 | 15.4 | 17.1 |
| 30-39 | 73.8 | 72.9 | 72.9 | 77.9 | 73.4 | 74.7 | 74.3 |
| ≥40 | 7.6 | 10.5 | 7.8 | 8.1 | 8.2 | 10.0 | 8.6 |
| Gravida | |||||||
| 0 | 48.1 | 36.2 | 45.5 | 52.1 | 49.0 | 46.5 | 47.2 |
| 1 | 26.6 | 25.7 | 23.2 | 25.8 | 24.3 | 26.7 | 26.4 |
| 2 | 13.6 | 16.2 | 14.4 | 12.3 | 10.8 | 14.1 | 13.8 |
| ≥3 | 11.7 | 21.9 | 16.9 | 9.8 | 15.9 | 12.6 | 12.6 |
| Prior preterm birth | |||||||
| 0 | 68.7 | 73.7 | 67.5 | 67.0 | 65.8 | 70.4 | 69.3 |
| 1 | 2.6 | 4.0 | 2.4 | 2.0 | 2.1 | 2.0 | 2.4 |
| ≥2 | 0.3 | 0.7 | 0.4 | 0.1 | 0.2 | 0.2 | 0.3 |
| Unknown | 28.4 | 21.6 | 29.7 | 30.8 | 31.9 | 27.3 | 28.0 |
| Maternal smoking | |||||||
| No | 72.9 | 75.0 | 73.1 | 78.7 | 79.3 | 56.6 | 67.4 |
| Yes | 8.4 | 5.6 | 6.9 | 4.2 | 6.1 | 5.6 | 7.0 |
| Unknown | 18.7 | 19.4 | 20.0 | 17.2 | 14.6 | 37.8 | 25.7 |
| Body mass index | |||||||
| Normal weight | 39.0 | 22.1 | 30.9 | 30.9 | 48.0 | 32.8 | 35.3 |
| Overweight | 14.2 | 21.2 | 15.6 | 15.6 | 11.8 | 17.1 | 12.2 |
| Obese | 10.1 | 20.0 | 10.3 | 10.3 | 4.3 | 13.1 | 8.7 |
| Unknown | 36.8 | 36.7 | 43.2 | 43.2 | 35.9 | 37.0 | 43.9 |
| Fresh embryos transferred | |||||||
| 1 | 8.4 | 7.9 | 6.3 | 10.3 | 7.0 | 8.5 | 8.4 |
| 2 | 61.4 | 59.6 | 60.4 | 57.3 | 57.9 | 58.5 | 60.0 |
| 3 | 22.0 | 23.6 | 22.7 | 22.2 | 23.3 | 21.6 | 22.0 |
| 4 | 5.8 | 6.5 | 7.7 | 6.9 | 8.0 | 7.5 | 6.6 |
| 5 | 2.1 | 2.3 | 2.6 | 3.2 | 3.6 | 3.4 | 2.7 |
| Unknown | 0.3 | 0.2 | 0.4 | 0.2 | 0.2 | 0.4 | 0.3 |
Gestational age at birth (gestational days) and rates of very early preterm, early preterm, and preterm birth varied significantly across racial and ethnic groups ( Table 2 ). In singleton pregnancies, compared with white women, black women delivered approximately 7 days earlier ( P < .001). There were no significant differences in gestational days at birth between white, Hispanic, Asian, and Native American women. The rates of preterm births were also markedly higher in black women than in white and other races/ethnicities. The rates of very early preterm, early preterm, and preterm births were all higher in black women: 5.5%, 8.6%, and 24.0%, respectively, compared with white women, 1.2%, 2.3%, and 12.9%, respectively. With white women as referent, after adjustment for confounding variables, the adjusted ORs and 95% CIs of very early preterm birth, early preterm birth, and preterm birth in black women were 4.8 (95% CI, 4.1–5.7), 3.9 (95% CI, 3.4–4.4), and 2.1 (95% CI, 1.9–2.3). The rate of early preterm and preterm births was slightly higher in Hispanic women, 3.5% and 15.3%, respectively, compared with white women, 2.3% and 12.9%, respectively, with adjusted ORs and 95% CIs of 1.5 (95% CI, 1.3–1.8) and 1.2 (95% CI, 1.1–1.3) ( P < .05). There were no significant differences in rates of preterm births between white, Asian, and Native American women.
| Variable | n | Gestational days, mean (SD) | % preterm birth <28 wks, aOR (95% CI) a | % preterm birth <32 wks, aOR (95% CI) | % preterm birth <37 wks, aOR (95% CI) | |||
|---|---|---|---|---|---|---|---|---|
| Singletons | ||||||||
| White | 46,790 | 270.3 (18.1) | 1.2 | Referent | 2.3 | Referent | 12.9 | Referent |
| Black | 3587 | 261.7 (29.4) | 5.5 | 4.8 (4.1–5.7) b | 8.6 | 3.9 (3.4–4.4) b | 24.0 | 2.1 (1.9–2.3) b |
| Hispanic | 4700 | 268.2 (21.2) | 1.9 | 1.6 (1.3–2.1) c | 3.5 | 1.5 (1.3–1.8) c | 15.3 | 1.2 (1.1–1.3) c |
| Asian | 6279 | 270.3 (18.0) | 1.3 | 1.1 (0.9–1.4) | 2.5 | 1.1 (0.9–1.3) | 12.0 | 0.9 (0.9–1.0) |
| Native | 369 | 269.5 (20.2) | 1.6 | 1.4 (0.6–3.1) | 3.0 | 1.3 (0.7–2.3) | 12.7 | 1.0 (0.7–1.3) |
| Missing | 35,563 | 270.1 (19.6) | 1.5 | 1.3 (1.2–1.5) a | 2.7 | 1.2 (1.0–1.3) | 12.8 | 1.0 (0.9–1.0) |
| All groups | 97,288 | 269.8 (19.4) | 1.5 | 2.7 | 13.3 | |||
| P value | < .001 (ANOVA) | < .001 (χ 2 ) | < .001 (χ 2 ) | < .001 (χ 2 ) | ||||
| Twins | ||||||||
| White | 20,265 | 247.8 (23.2) | 4.2 | Referent | 10.3 | Referent | 62.4 | Referent |
| Black | 1316 | 238.1 (34.4) | 12.7 | 3.3 (2.7–3.9) b | 21.5 | 2.4 (2.1–2.7) b | 69.8 | 1.4 (1.2–1.6) c |
| Hispanic | 2049 | 246.8 (25.5) | 5.9 | 1.4 (1.2–1.8) c | 12.2 | 1.2 (1.0–1.4) | 63.9 | 1.1 (1.0–1.2) |
| Asian | 2258 | 250.1 (23.2) | 3.9 | 0.9 (0.8–1.2) | 9.1 | 0.9 (0.8–1.0) | 55.9 | 0.8 (0.7–0.9) c |
| Native | 157 | 244.2 (24.4) | 4.5 | 1.1 (0.5–2.3) | 13.4 | 1.3 (0.8–2.1) | 63.9 | 1.1 (0.8–1.7) |
| Missing | 14,916 | 247.9 (24.7) | 4.7 | 1.2 (1.0–1.3) | 10.8 | 1.1 (1.0–1.1) | 60.7 | 0.9 (0.9–1.0) |
| All groups | 40,961 | 247.6 (24.4) | 4.7 | 10.9 | 36.9 | |||
| P value | < .001 (ANOVA) | < .001 (χ 2 ) | < .05 (χ 2 ) | < .001 (χ 2 ) | ||||
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
