Objective
We conducted a systematic review to evaluate the influence of race and ethnicity on clinical pregnancy and live birth outcomes after in vitro fertilization (IVF).
Study
We searched PubMed, EMBASE, Web of Science, CINAHL, POPLINE, and Cochrane Central, and hand-searched relevant articles published through July 22, 2015.
Study Appraisal and Synthesis Methods
Two reviewers independently evaluated abstracts to identify studies that compared clinical pregnancy rates and live birth rates for ≥2 racial and/or ethnic groups after nondonor IVF cycles.
Results
Twenty-four studies were included. All 5 US registry-based studies showed that black, Hispanic, and Asian women had lower clinical pregnancy rates and/or live birth rates after IVF, compared with white women. Similarly, most clinic-specific studies reported significant disparities in these primary outcomes, potentially attributable to differences in infertility diagnosis, spontaneous abortion, and obesity. Studies varied with respect to definitions of race/ethnicity, inclusion of first cycles vs multiple cycles for individual women, and collected covariates. Most studies were limited by sample size, inadequate adjustment for confounding, selection bias, and extensive missing data.
Conclusions
Although current evidence points to race and ethnicity, especially black race, as strong predictors of poorer outcomes after IVF, the utility of results is constrained by the limitations described.
Infertility is a global, multiethnic problem that affects 50-70 million couples worldwide. Infertile women of many races and ethnicities benefit from assisted reproductive technologies (ART), particularly in vitro fertilization (IVF), now offered at clinics in >162 countries. However, women who have a live birth using IVF are disproportionately white or Caucasian and are highly educated with middle to high income. Several factors may influence this trend. First, non-white women are less likely to utilize IVF services, despite higher infertility rates among black (7.2%) and Hispanic (6.1%) women, compared with white women (5.5%). IVF is a costly, complex treatment and may be prone to disparities caused by gaps in coverage or differences in cultural beliefs and educational level. Also, race and ethnicity may serve as a surrogate for known risk factors of IVF failure, such as age, obesity, smoking, or uterine abnormalities. Finally, women of racial and ethnic minority groups may be genetically or environmentally predisposed to worse outcomes, independent of other clinical risk factors.
There is relatively low minority representation in IVF clinics, and as a result, literature on ART predominantly reflects the experiences of white women. This is problematic because racial and ethnic groups differ in fertility characteristics and incidence of adverse perinatal outcomes. Disparities found in the outcomes of spontaneously conceived pregnancies are known to persist after adjustment for socioeconomic status and demographic characteristics, suggesting that other biological, environmental, or behavioral factors are at play. These same factors may affect IVF outcomes as well. In fact, black and Hispanic women are more likely to have preterm birth after IVF, compared with white women, and infants born to black, Hispanic, and Asian women after IVF are more likely to have moderate and severe growth restriction. A recent study of ART in the United States reported that, of all the patient characteristics thought to influence ART success, non-Hispanic black race had the strongest negative association with a good perinatal outcome after ART.
Knowledge of disparities in IVF could inform clinical decision-making and other treatment modifications to improve the potential for success in certain populations. Previous reviews examining race/ethnicity and IVF outcomes have lacked systematic methodology or included only data collected at the national level. We therefore conducted a systematic review to report the clinical pregnancy rates and live birth rates (LBRs) among different racial and ethnic groups, using national and clinic-specific data for nondonor IVF. By tabulating success rates for each race/ethnicity group and framing them in a broader context, this review may help clinicians counsel and treat individual patients seeking infertility treatment.
Objective
The aim of this study is to perform a systematic review of the literature to identify studies that compare pregnancy and live birth outcomes for women of different racial and/or ethnic groups.
Materials and Methods
We adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines and performed a search of PubMed, EMBASE, CINAHL, Web of Science, Cochrane Central, and POPLINE on July 22, 2015, using terms for race and ethnicity, IVF treatment, and pregnancy and live birth outcomes ( Supplement A1 ). Our ability to capture race/ethnicity as a variable is inevitably imperfect since no consensus exists on the definitions of race and ethnicity or on how such data should be collected. There are 15 race categories defined by the US Census Bureau (white; black or African American; American Indian or Alaska Native; Asian Indian; Chinese; Filipino; Japanese; Korean; Vietnamese; Native Hawaiian; Guamanian or Chamorro; Samoan; other Pacific Islander; other Asian; and some other race), as well as 2 ethnicity categories (Hispanic or Latino, and not Hispanic or Latino). These, according to the bureau, reflect the social definition of race and ethnicity recognized in the United States. We augmented this list with PubMed Medical Subject Headings and other terms, such as “native,” “aboriginal,” and “of color.” We used the broad categories of white, black, Asian, and Hispanic to simplify presentation of data. If studies used more specific terms, including Caucasian, African American, South Asian, and Indian, we referred to these terms when reporting their findings.
We reviewed all abstracts that compared clinical pregnancy rates and/or LBRs after IVF for at least 2 racial and/or ethnic groups. All study designs and languages were eligible for inclusion. We also searched the reference lists of included manuscripts to identify additional relevant manuscripts. Two investigators (L.A.H. and O.C.) independently reviewed all abstracts and excluded animal studies, reviews/metaanalyses, or case reports, as well as studies that did not evaluate IVF, included oocyte donation, did not report pregnancy or LBR, or did not include race/ethnicity. Studies that compared outcomes for one race to those of all other races (eg, white vs non-white) were also excluded. Disagreement about abstract classification was adjudicated. Outcome data were extracted independently by 2 investigators from full-text articles. Details of each included study also were extracted, and we identified whether there was adjustment for potential confounding, and, if so, which variables were adjusted for. Where “not significant” is reported in the tables, it is because the actual P values were unavailable in the full-text articles. No identifiable data were utilized in the course of this project; thus, institutional review board approval was not required.
Materials and Methods
We adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines and performed a search of PubMed, EMBASE, CINAHL, Web of Science, Cochrane Central, and POPLINE on July 22, 2015, using terms for race and ethnicity, IVF treatment, and pregnancy and live birth outcomes ( Supplement A1 ). Our ability to capture race/ethnicity as a variable is inevitably imperfect since no consensus exists on the definitions of race and ethnicity or on how such data should be collected. There are 15 race categories defined by the US Census Bureau (white; black or African American; American Indian or Alaska Native; Asian Indian; Chinese; Filipino; Japanese; Korean; Vietnamese; Native Hawaiian; Guamanian or Chamorro; Samoan; other Pacific Islander; other Asian; and some other race), as well as 2 ethnicity categories (Hispanic or Latino, and not Hispanic or Latino). These, according to the bureau, reflect the social definition of race and ethnicity recognized in the United States. We augmented this list with PubMed Medical Subject Headings and other terms, such as “native,” “aboriginal,” and “of color.” We used the broad categories of white, black, Asian, and Hispanic to simplify presentation of data. If studies used more specific terms, including Caucasian, African American, South Asian, and Indian, we referred to these terms when reporting their findings.
We reviewed all abstracts that compared clinical pregnancy rates and/or LBRs after IVF for at least 2 racial and/or ethnic groups. All study designs and languages were eligible for inclusion. We also searched the reference lists of included manuscripts to identify additional relevant manuscripts. Two investigators (L.A.H. and O.C.) independently reviewed all abstracts and excluded animal studies, reviews/metaanalyses, or case reports, as well as studies that did not evaluate IVF, included oocyte donation, did not report pregnancy or LBR, or did not include race/ethnicity. Studies that compared outcomes for one race to those of all other races (eg, white vs non-white) were also excluded. Disagreement about abstract classification was adjudicated. Outcome data were extracted independently by 2 investigators from full-text articles. Details of each included study also were extracted, and we identified whether there was adjustment for potential confounding, and, if so, which variables were adjusted for. Where “not significant” is reported in the tables, it is because the actual P values were unavailable in the full-text articles. No identifiable data were utilized in the course of this project; thus, institutional review board approval was not required.
Results
Of the 47 abstracts that initially met inclusion criteria, we identified 24 full-text publications that compared clinical pregnancy rates and/or LBRs after IVF for ≥2 racial and/or ethnic groups ( Figure ). Five national studies reported data from the Society for Assisted Reproductive Technologies (SART) Clinic Outcome Reporting System, and 19 site-specific studies reported data from individual IVF clinics ( Tables 1 and 2 ). All studies used white (or Caucasian) women as the referent group, and all included more than twice as many white women as those of other racial or ethnic minority groups. We identified 15 additional studies that met inclusion criteria, but were published as meeting abstracts only with no corresponding full-text publication ( Supplement A2 ). Tabulated results of these studies are in Supplement A3 .
| Reference | Year | Setting | Race/ethnicity data | Cycle info | Adjustment for confounders |
|---|---|---|---|---|---|
| Baker et al | 2010 | SART CORS (January 2004 through December 2006) | 225,889 Cycles, unknown distribution of white, black, Asian, Hispanic, other/mixed, unknown | Repeated cycles; fresh | a Maternal age, embryos transferred, assisted hatching, ICSI, previous term birth, prior preterm birth, prior spontaneous abortion, infertility diagnosis (male factor, ovulation disorders, diminished ovarian reserve, tubal factors, uterine factors, other factors, unexplained factors) |
| Fujimoto et al | 2010 | SART CORS (2004 through 2006) | 139,027 Cycles White (n = 107,484) Asian (n = 13,671) Black (n = 8903) Hispanic (n = 8969) | Repeated cycles; fresh | a Maternal age, embryos transferred, infertility diagnosis (male factor, endometriosis, PCOS, diminished ovarian reserve, tubal factors, other factors) |
| Seifer et al | 2010 | SART CORS (2004 through 2006) | 158,693 Cycles Fresh: White (n = 120,994) Black (n = 10,354) Cryopreserved: White (n = 25,412) Black (n = 1933) | First cycles and repeated cycles (subgroups); Fresh and cryopreserved; clinics providing ≥50 cycles/y and reporting race for >95% of cycles | a Maternal age, parity, prior spontaneous abortion, embryos transferred, ICSI, infertility diagnosis (tubal factor, male factor, uterine factors, diminished ovarian reserve, other ovarian disorder, endometriosis) |
| Seifer et al | 2008 | SART CORS (1999 through 2000) | 72,272 Cycles Fresh: White (n = 58,459) Black (n = 3116) Cryopreserved: White (n = 10,147) Black (n = 550) | First cycles and repeated cycles (subgroups); fresh and cryopreserved; clinics providing ≥50 cycles/y and reporting race for >95% of cycles | a Maternal age, parity, prior spontaneous abortion, day-3 FSH ratio, embryos transferred, infertility diagnosis (tubal factor, male factor, uterine factor, endometriosis, diminished ovarian reserve, other ovulation disorders, other factors, idiopathic), clinic volume, clinic success rate |
| Purcell et al | 2007 | (a) SART CORS (1999 through 2000) (b) University of California, San Francisco, CA (January 2001 through December 2003) | (a) 27,272 Cycles White (n = 25,843) Asian (n = 1429) (b) 567 Cycles White (n = 370) Asian (n = 197) | (a) First cycles only; from clinics providing ≥50 cycles/y and reported race for >95% of cycles (187 clinics) (b) Repeated cycles | a (a) Maternal age, gravidity, parity, embryos transferred, history of spontaneous or therapeutic abortion, day-3 FSH and E2 levels, ICSI, infertility diagnosis (diminished ovarian reserve, other ovulation disorders, male factor, unexplained) (b) Maternal age, primary or secondary infertility, number of previous IVF attempts, stimulation protocol, prior spontaneous or therapeutic abortion, infertility diagnosis (diminished ovarian reserve, ovulatory dysfunction, or unexplained infertility), days of stimulation, E2 level, endometrial thickness, embryo fragmentation score, physician, difficulty of embryo transfer |
| Reference | Year | Setting | Race/ethnicity data | Cycle info | Adjustment for confounders |
|---|---|---|---|---|---|
| McQueen et al | 2015 | Fertility Centers of Illinois, Chicago, IL (January 2010 through December 2012) | 4045 Women White (n = 3003) Black (n = 213) Asian (n = 541) Hispanic (n = 288) | First cycles only; fresh | a Age, BMI, day-3 FSH, smoking status, infertility diagnosis |
| Kan et al | 2015 | IVF Australia, Sydney, Australia (January 2001 through December 2010) | 2594 Women Asian (n = 522) Caucasian (n = 2072) | First cycles only; fresh | a Age, duration of infertility |
| Rudick et al | 2012 | University of Southern California Fertility Clinic, Los Angeles, CA (January 2006 through August 2009) | 188 Women White (n = 100) Asian (n = 49) Hispanic (n = 26) Other (n = 13) | First cycles only; fresh | a Age, embryos transferred, embryo quality, diagnosis of diminished ovarian reserve |
| Sharara et al | 2012 | Virginia Center for Reproductive Medicine, Reston, VA (January 2004 through December 2009) | 292 Cycles White (n = 238) South Asian (n = 54) | First cycles only; fresh blastocyst transfers (day 5 only) in women age <40 y | No adjustment for observed differences in maternal age, FSH, PCOS |
| Csokmay et al | 2011 | Walter Reed Army Medical Center, Washington, DC (January 2003 through December 2008) | 169 Women Fresh: White (n = 68) AA (n = 30) Cryopreserved: White (n = 119) AA (n = 50) | Unspecified cycle number; fresh and cryopreserved blastocyst transfers (subgroup analysis) | No adjustment for observed differences in leiomyoma |
| Gleicher et al | 2011 | Center for Human Reproduction in New York, New York, NY (Dates not available) | 339 Women Caucasian (n = 232) AA (n = 59) Asian (n = 48) | Unspecified cycle number in consecutive IVF patients | a Age, BMI, FMR1 genotype, AMH levels |
| Shuler et al | 2011 | University of Texas Health Science Center, South Texas Fertility Center, San Antonio, TX (January 1998 through January 2008) | 435 Women White (n = 301) Hispanic (n = 134) | First cycles only; fresh | No adjustment for observed differences in tubal infertility |
| Langen et al | 2010 | Stanford University Medical Center, Stanford, CA (January 2005 through December 2006) | 180 Cycles White (n = 112) Asian (n = 68) | First cycles only; Fresh | a Gravidity, BMI, history of IVF, endometrial thickness |
| McCarthy-Keith et al | 2010 | Walter Reed Army Medical Center, Washington, DC Wilford Hall Medical Center, San Antonio, TX and Tripler Army Medical Center, Honolulu, HI (January 2000 through December 2005) | 2050 Cycles White (n = 1280) AA (n = 353) | First cycles only; fresh; in women age <42 y with FSH levels on day 3 or 10 ≤12 mIU/mL | No adjustment for observed differences in infertility diagnosis |
| Dayal et al | 2009 | George Washington University Medical Faculty Associates Fertility and IVF Center, Washington, DC (January 2004 through December 2005) | 251 Cycles Caucasian (n = 180) AA (n = 71) | First cycles only; day-3 embryo transfers | a BMI, uterine fibroids, tubal factor infertility, dose of gonadotropins |
| Shahine et al | 2009 | Stanford University, Stanford, CA (January 2005 through June 2007) | 225 Women Caucasian (n = 145) Indian (n = 80) | First cycles only; day-5 or- 6 blastocyst transfers | a Maternal age, stimulation protocol, PCOS |
| Palep-Singh et al | 2007 | University of Leeds, Leeds, UK (2000 through 2004) | 608 Cycles Caucasian (n = 420) Asian (n = 188) | Unspecified cycle number; fresh | a Maternal age, LH, and gonadotropin dose |
| Feinberg et al | 2006 | Walter Reed Army Medical Center, Washington, DC (1999 through 2003) | 1387 Women Caucasian (n = 974) AA (n = 253) Hispanic (n = 56) | First cycles only; in women age <42 y with FSH levels on day 3 or 10 ≤12 mIU/mL | Presence of uterine fibroids (stratified for AA vs Caucasian) |
| Feinberg et al | 2007 | Walter Reed Army Medical Center, Washington, DC (1999 through 2003) | 1387 Women Caucasian (n = 974) AA (n = 253) Hispanic (n = 56) | First cycles only; in women age <42 y with FSH levels on day 3 or 10 ≤12 mIU/mL | NA |
| Bendikson et al | 2005 | Boston IVF Collaborative Clinics, Boston, MA (1994 through 1998) | 1135 Cycles White (n = 1039) Black (n = 43) Asian (n = 35) Hispanic (n = 18) | First cycles only | No adjustment for observed differences in BMI, gravidity, duration of infertility, infertility diagnosis |
| Nichols et al | 2001 | Greenville Hospital System, Greenville, SC (November 1996 through June 2000) | 358 Cycles White (n = 333) Black (n = 25) | Repeated cycles | a Parity, BMI, tubal factor infertility |
| Sharara and McClamrock | 2000 | University of Maryland, Baltimore, MD (April 1997 through July 1999) | 168 Cycles White (n = 121) Black (n = 47) | First cycles and repeated cycles (subgroups); in women age ≤40 y, (excludes day-3 FSH>10, hydrosalpinges, and intracavitary uterine abnormalities) | No adjustment for observed differences in BMI, infertility diagnosis, duration of infertility, protocol variation (matched for maternal age, early FSH, infertility diagnosis, gonadotropin dose, treatment year) |
| Lashen et al | 1999 | Birmingham Women’s Hospital, Birmingham, UK (1994 through 1997) | 324 Women Caucasian (n = 216) Asian (n = 108) | First cycles only | No adjustment for observed differences in duration of infertility, infertility diagnosis (matched for age, day-3 FSH, gonadotropin dose, indication) |
| Mahmud et al | 1995 | Oxford University, Oxford, UK (April 1987 through December 1993) | 132 Women White (n = 88) Indian (n = 44) | First cycles and repeated cycles (subgroups) | No adjustment for observed differences in duration of fertility, primary vs secondary infertility, infertility diagnosis (matched for maternal age, BMI, treatment year) |
Two large studies allowed us to characterize national IVF success rates for the 4 main racial/ethnic groups of white, black, Asian, and Hispanic. Both studies used SART data for 2004 through 2006 yet differed slightly in inclusion criteria and covariates. Baker et al incorporated other/mixed and unknown categories but neglected to specify the overall race/ethnicity distribution. They reported lower clinical pregnancy rates for black, Asian, and Hispanic women (32.2%, 31.2%, and 37.5%, respectively), compared with white women (40.5%), as well as lower LBRs per pregnancy for black, Asian, and Hispanic women (75.1%, 81.5%, and 82.1%), compared with white women (83.7%). As expected, Fujimoto et al reported essentially the same rates, but adjusted for fewer covariates and found that only Asian ethnicity was significantly predictive of a lower clinical pregnancy rate. Both studies attributed the lower likelihood of live birth to the increased risk of pregnancy loss in black, Asian, and Hispanic women.
Black/African American
Similar to the studies by Baker et al and Fujimoto et al, 2 studies by Seifer et al confirmed lower clinical pregnancy rate and LBR among black women, regardless of prior ART treatment ( Table 3 ). In the study with the most extensively adjusted model, the significant effect of race persisted after adjusting for treatment and patient factors, as well as clinic-level variables such as clinic volume and overall clinic pregnancy rate.
| Reference | CPR, % | Significance | LBR, % | Significance | ||
|---|---|---|---|---|---|---|
| Black | White | Black | White | |||
| Fresh cycles | ||||||
| Baker et al | 32.2 | 40.5 | OR, 0.76; CI, 0.72–0.80 | 75.1 a | 83.7 | OR, 0.62; CI, 0.56–0.68 |
| Fujimoto et al | 32.0 | 40.1 | OR, 1.09; CI, 0.99–1.2 | 75.0 b | 83.7 | OR, 0.62; CI, 0.56–0.68 |
| Seifer et al | ||||||
| (No prior ART) | 29.3 | 38.3 | P < .001 | 76.9 b | 84.8 | RR, 1.31 c ; CI, 1.26–1.37 |
| (Prior ART) | 25.0 | 32.4 | P < .001 | 71.0 b | 81.6 | RR, 1.33 c ; CI, 1.24–1.42 |
| Seifer et al | ||||||
| (No prior ART) | 27.7 | 33.6 | P < .001 | 20.7 c | 28.4 | RR, 1.24 d ; CI, 1.12–1.36 |
| (Prior ART) | 22.1 | 28.9 | P < .001 | 15.7 c | 23.7 | RR, 1.38 d ; CI, 1.20–1.57 |
| Sharara and McClamrock | 19.1 | 42.1 | P < .01 | 14.9 c | 38.8 | P < .01 |
| McCarthy-Keith et al | 46.1 | 52.6 | RR, 0.88; CI, 0.78–0.99 | 33.7 c | 45.7 | RR, 0.74; CI, 0.63–0.91 |
| Csokmay et al | 40 | 50 | P = .39 | 16.7 c | 39.7 | OR, 0.30; CI, 0.10–0.89 |
| Feinberg et al | 39.5 | 42.6 | RR, 0.93; CI, 0.78–1.1 | 29.6 c | 35.8 | RR, 0.83; CI, 0.67–1.02 |
| McQueen et al | 24.4 | 36.2 | OR, 0.63; CI, 0.44–0.88 | 16.9 c | 30.7 | OR, 0.50; CI, 0.33–0.72 |
| Gleicher et al | 10.2 | 25.9 | OR, 0.27; CI, 0.10–0.72 | – | – | – |
| Dayal et al | 34 | 28 | P = .48 | 25 c | 24 | P = .99 |
| Nichols et al | 70.8 | 48 | OR, 3.3; CI, 1.3–8.6 | – | – | – |
| Bendikson et al | – | – | – | 20.1 c | 18.3 | Not significant |
| Cryopreserved cycles | ||||||
| Seifer et al | 31.8 | 31.8 | P = .43 | 71.8 b | 80.3 | RR, 1.10 d ; CI, 1.00–1.21 |
| Seifer et al | 22.7 | 20.2 | P = .145 | 16.5 c | 16.0 | RR, 1.03, P = .718 |
| Csokmay et al | 42.0 | 39.5 | P = .86 | 28.0 b | 30.2 | P = .85 |
a Per clinical intrauterine gestation + heterotopic
b Per clinical intrauterine gestation
The smaller site-specific studies had some conflicting results. Sharara and McClamrock published the first evidence in 2000 that black women had lower clinical pregnancy rate and lower LBR per cycle, as well as lower implantation rate than white women. The patients were matched based on age, diagnosis, and other factors; however, black women had higher body mass index (BMI), longer duration of infertility, higher tubal factor infertility, and higher peak estradiol levels, potentially contributing to poorer outcomes. Three other studies conducted at military ART centers found disparities even in these “equal-access-to-care” settings. The study by McCarthy-Keith et al examined 2050 cycles from 3 IVF centers and found significantly lower clinical pregnancy rate and LBR for black women, compared with white women. A smaller study by Csokmay et al also found that black race was significantly predictive of lower LBR. The third military study by Feinberg et al reported a “clinically significant” difference with lower LBR for black women, but this did not reach statistical significance. Of note, these studies did not adjust for observed differences in infertility diagnoses known to influence IVF success. Feinberg et al stratified by uterine leiomyoma, but this did not affect the results. In contrast, Gleicher et al controlled for fragile X mental retardation genotypes, as well as age, BMI, and antimüllerian hormone (AMH) level, and found that decreased rates of pregnancy in black women persisted. A recent large study conducted by McQueen et al found significantly lower clinical pregnancy rate and LBR after adjustment for age, BMI, follicle-stimulating hormone (FSH), smoking status, and infertility diagnosis, demonstrating a LBR among black women that was almost half that of white women (16.9% vs 30.7%).
Nichols et al was the only study reporting a higher clinical pregnancy rate in African American women by comparing 24 African Americans to 273 whites. Almost all black women in this study were undergoing their first IVF cycle compared with 72% of whites. Dayal et al found no significant differences in clinical pregnancy rate, LBR, or implantation rate after controlling for tubal or uterine factors, BMI, and gonadotropin dose, but their study was underpowered and could detect only a 27% difference in pregnancy rates. Similarly, a study by Bendikson et al had a small numbers of minority patients and found no differences in unadjusted LBRs or miscarriage rates.
For women who used cryopreserved embryos, SART data showed that black women had lower LBRs after transfer of frozen-thawed embryos than white women, although clinical pregnancy rates were not different. Analysis of a military center data by Csokmay et al revealed no differences in clinical pregnancy rate, LBR, or the spontaneous abortion rate for frozen-thawed cycles only. Of note, this study did find significant differences in the LBR after the completion of a fresh embryo transfer and a subsequent frozen-thawed transfer, with an overall LBR for whites of 62.2% compared with only 37.5% for African American women.
Asian/Indian
The 3 SART studies of Asian race reported worse clinical pregnancy rate and LBR for this subgroup, independent of other patient- and treatment-related factors. The study of Purcell et al, which used a 1999 through 2000 SART data set, found lower clinical pregnancy rate and LBR per cycle for Asian women than white women, after controlling for gravidity, parity, day-3 FSH and E2 levels, as well as the variables used in Baker et al and Fujimoto et al. The same study evaluated cycles from a California IVF clinic with a relatively large Asian patient population and confirmed that Asian race was associated with lower clinical pregnancy rate than that of white women. Access to individual clinic data allowed them to include additional variables in their model, such as primary vs secondary infertility, endometrial thickness, and difficulty of transfer.
Two studies published in 2015, including that of McQueen et al conducted in Chicago, IL, and of Kan et al conducted in Sydney, Australia, evaluated the outcomes of >500 Asian women ( Table 4 ). McQueen et al found lower clinical pregnancy rate and LBR for Asian women compared with white women, after adjustment for key factors such as age, BMI, and infertility diagnosis, although the analysis did not account for observed differences in stimulation time and peak E2 levels. Kan et al identified differences in clinical pregnancy rate and LBR that were no longer significant after controlling for age and duration of infertility, as Asian women tended to be older and waited longer to pursue IVF.
