A black woman in a mask sitting in a waiting room filling out a form on a clipboard.

Healthcare providers and scientists are reexamining race as an explanatory variable for genetic and biological differences. Since race is a sociopolitical construct, using it as a biological system of classification can introduce systematic differences in care and exacerbate racial and ethnic health disparities (CLN, March 2023).

The increased awareness in medicine about the limitations of race-based medical algorithms led to the medical community removing race coefficients from several widely used clinical calculations, including the estimated glomerular filtration rate equations and the vaginal birth after cesarean success calculator (1, 2). Both race-based algorithms had the potential to compromise care for Black patients and lead to worse outcomes.

Now the clinical laboratory community is examining the potential for patient harm due to another race-based variable: maternal serum screening (MSS) biomarkers.


MSS provides risk estimates for fetal aneuploidies including Trisomy 21 (Down syndrome), Trisomy 18 (Edwards’s syndrome), and Trisomy 13 (Patau syndrome), as well as or open neural tube defects (ONTD) (3). Patients with increased MSS risk estimates undergo invasive amniocentesis or chorionic villus sampling (CVS) testing to provide a definitive diagnosis (3).

As a screening test, MSS paradigms seek to optimize sensitivity in order to maximize detection. Minimizing false positive results is also important, due to the risk of miscarriage associated with amniocentesis and CVS and the negative emotional toll that false positive results can have on the patient.

Healthcare providers perform first-trimester screening between approximately 11 and 14 weeks of gestation. They combine ultra-sonographic detection of translucent fluid collection behind the fetal neck, the nuchal translucency, with maternal serum human chorionic gonadotropin (hCG) and pregnancy-associated plasma protein A (PAPP-A) measurements. In second-trimester screening, also known as the quad screen, providers perform maternal serum hCG, alpha-fetoprotein (AFP), unconjugated estriol (uE3), and dimeric inhibin A (DIA) measurements between 15 and 21 weeks of gestation (3).

Clinical laboratories calculate multiples of median (MoMs) for each biomarker by dividing the patient concentration by the median concentration of the biomarker in the local population. The pattern of increase or decrease in the biomarker MoMs correlates with the risk for trisomies and ONTD. Age-specific risks are modified with likelihood ratios for the serum biomarker MoMs using Bayes’ theorem, with further adjustment for demographic and clinical variables that have been shown to affect MSS biomarker concentrations, including maternal race, weight, diabetes, in vitro fertilization, multiple gestations, and smoking status.


The College of American Pathologists (CAP) requires accredited clinical laboratories to calculate separate MSS biomarker medians for Black and White pregnant patients. Alternatively, if sufficient data is not available, labs must apply a correction factor for patients from the less common races. CAP also requires clinical laboratories to consider using median values specific to other races and ethnic groups if significant differences in biomarker concentrations exist between these groups in the screening population. Labs must document the reasoning for the inclusion or exclusion of the race adjustments (for each biomarker) and the definition the lab used to identify which patients receive the race adjustment.

Notably, the American College of Medical Genetics, in their 2019 technical bulletin on maternal serum AFP screening, also recommends race-based adjustments for AFP MoMS for the detection of ONTD.

CAP requirements for race-specific MSS biomarker medians are based on studies that have demonstrated increased concentrations of PAPP-A, AFP, and hCG and decreased DIA concentrations in Black compared to White patients. The intent of calculating MSS risk scores with race- and ethnic-specific adjustments is to improve accuracy. Indeed, researchers have reported population-level differences between racial and ethnic groups across multiple but not all studies (1, 4–7).

However, the application of these adjustments is based on an inaccurate premise: that race and ethnicity describe genetically and socioeconomically homogenous groups. Further, implementing these adjustments requires patient self-identification or assignment of race by healthcare providers, although there is no gold standard for categorizing individuals by race.

Cell-free DNA (cfDNA) screening is a more sensitive and specific method for aneuploidy detection compared to MSS and does not incorporate patient race. Circulating cfDNA in the maternal serum consists of a fraction of placental DNA known as the fetal fraction (cffDNA). cfDNA in the maternal circulation is analyzed for aneuploidy, microdeletions, and large copy number changes.

Unfortunately, many patients cannot afford cfDNA prenatal screening, and insurance plans do not universally cover the testing. The Coalition for Access to Prenatal Screening only reports that several leading national commercial insurance plans cover cfDNA prenatal screening for those considered "high risk,” and seven states and the District of Columbia Medicaid programs do not cover cfDNA prenatal screening at all, even for high-risk pregnancies (8). Furthermore, lower cffDNA has been observed in African American and South Asian individuals compared to White individuals (9). This suggests that MSS may remain an important prenatal screening tool for under-insured and uninsured patients.


Race-based adjustments do not account for genetic or socioeconomic heterogeneity within racial and ethnic groups. To demonstrate how genetic and socioeconomic heterogeneity within racial and ethnic groups can confound MSS screening results, we can examine the results of one of the studies referenced by CAP that explored gestational age-dependent effects of maternal and pregnancy characteristics, including maternal race, on free β-hCG and PAPP-A (4).

The study was performed in a cohort of 27,908 singleton pregnancies from the United Kingdom (UK) and 125,461 singleton pregnancies from Denmark with normal fetal karyotypes or that resulted in the birth of a phenotypically normal neonate (4). The study did not describe the method the authors used to assign race to the participants. Gestational age-dependent increases in PAPP-A concentrations were statistically different in Black participants compared with White participants (~5-6% higher) in both the UK and Denmark cohorts. Gestational age-dependent increases in β-hCG concentrations were statistically higher in Black compared to White participants in the Denmark cohort (~3-4% higher), but no differences in the weekly change of β-hCG concentrations were reported  in Black participants of the UK cohort compared to White participants.

In exploring possible reasons for the discrepancy in rate of change of β-hCG concentrations between the Black participants in the two different cohorts, the authors state that "maternal serum levels of PAPP-A and free β-hCG may vary according to the broad categories of racial origin and within such categories, including the exact country or, indeed, the tribe of origin." The "Afro-Caribbean racial origin" subset of the UK cohort originated from the Caribbean, in contrast to the Denmark cohort, where the majority originated from Africa.

A recent retrospective cohort study examined AFP concentrations in 27,710 pregnant patients that underwent MSS at the University of Washington in Seattle. The study found no statistically significant difference in raw maternal serum AFP concentrations, or weight- and gestational age-adjusted AFP concentrations, between Black and non-Black patients (1). Using a regression model to remove race adjustment, the study found no statistically significant difference between Black and non-Black patients in median raw maternal serum AFP values, nor in median maternal serum AFP MoM. The results of this study conflict with early studies that demonstrated higher concentrations of AFP in Black individuals compared with White individuals in study cohorts in East London, the UK, and Farmington, Connecticut (5, 7).

Why do racial differences in MSS biomarkers appear to be study- and population-dependent? Ancestral African populations exhibit a large degree of genetic heterogeneity. Forced and voluntary migration of people of African ancestry and subsequent admixture with individuals of other geographic ancestries has resulted in further genetic heterogeneity between people racially categorized as Black. In other words, the Black racial category describes people from multiple modern day geographic locations that are genetically diverse but treated as a single biological category when healthcare providers use race-based adjustments to generate a clinical result.

Socioeconomic differences between individuals categorized in the same racial or ethnic group between different populations may also contribute to conflicting observations between studies.


While multiple physiologic and clinical factors reportedly impact MSS biomarker concentrations (e.g., maternal weight, smoking status, diabetes) and are incorporated into aneuploidy risk estimates, there has been limited research conducted on the impact of social and structural factors.

Systemic racism significantly contributes to the development and the continuation of racial health disparities (10). In the context of scientific research, Lett et al. recommend that race should not be used as a measure for biological differences, but rather as a proxy for exposure to systemic racism (10).

For example, systemic racism leads to racial differences in social determinants of health such as food availability, built neighborhood environment including green space, and proximity to pollutant reservoirs. Despite the availability of methods to assess how social and structural factors might interact with disease (10), researchers scarcely employ them to interrogate racial differences in disease biomarkers, disease incidence and prevalence, or health outcomes (10).

A new expert consensus report of the National Academies of Science, Engineering and Medicine urges researchers to directly evaluate the environmental factors or exposures that may affect genetic and genomic studies, rather than relying on population descriptors such as race as a proxy where possible. When the use of race as a proxy for environmental factors or exposure is unavoidable, the report recommends that researchers clearly identify how race is employed and why the use of race is relevant to the study.

Our understanding of race, genetics and social determinants of health have evolved significantly since the studies that provided the basis for the inclusion of race in MSS were performed. As such, a reappraisal of this practice is long overdue.

Christina C. Pierre, PhD, DABCC, FADLM, is a clinical assistant professor in the department of pathology and laboratory medicine at the Perelman School of Medicine at the University of Pennsylvania in Philadelphia, and a clinical chemist and section director of clinical chemistry and coagulation testing at Penn Medicine Lancaster General Hospital in Lancaster, Pennsylvania. +Email: [email protected]

Octavia M. Peck Palmer, PhD, FADLM, is an associate professor at the University of Pittsburgh School of Medicine and division director of clinical chemistry at the University of Pittsburgh Medical Center in Pittsburgh. + Email: [email protected]


  1. Burns NR, Kolarova T, Katz R, Ma K, Delaney S. Reconsidering race adjustment in prenatal alpha-fetoprotein screening. Obstet Gynecol 2023; doi:10.1097/AOG.0000000000005045
  2. Delgado C, Baweja M, Crews DC, et al. A unifying approach for GFR estimation: Recommendations of the NKF-ASN Task Force on Reassessing the Inclusion of Race in Diagnosing Kidney Disease. Am J Kidney Dis 2021; doi:10.1053/J.AJKD.2021.08.003
  3. Dashe JS. Aneuploidy screening in pregnancy. Obstet Gynecol 2016; doi:10.1097/AOG.0000000000001385
  4. Ball S, Ekelund C, Wright D, et al. Temporal effects of maternal and pregnancy characteristics on serum pregnancy-associated plasma protein-A and free β-human chorionic gonadotropin at 7–14 weeks’ gestation. Ultrasound Obstet Gynecol 2013; doi: 10.1002/UOG.11209
  5. Benn P, Clive J, Collins R. Medians for second-trimester maternal serum alpha-fetoprotein, human chorionic gonadotropin, and unconjugated estriol; differences between races or ethnic groups. Clin Chem 1997; doi: 10.1093/clinchem/43.2.333
  6. Cowans NJ, Spencer K. Effect of gestational age on first trimester maternal serum prenatal screening correction factors for ethnicity and IVF conception. Prenat Diagn 2013; doi: 10.1002/pd.4010
  7. Watt H, Wald N, Smith D, Kennard A, Densem J. Effect of allowing for ethnic group in prenatal screening for Down’s Syndrome. Prenat Diagn 1996; doi: 10.1002/(SICI)1097-0223(199608)16:8<691::AID-PD946>3.0.CO;2-2
  8. Coalition for Access to Prenatal Screening Coverage Scorecards. Accessed November 30, 2022. https://www.ptonline.com/articles/how-to-get-better-mfi-results
  9. Deng C, Liu S. Factors affecting the fetal fraction in noninvasive prenatal acreening: A review. Front Pediatr 2022; doi: 10.3389/FPED.2022.812781
  10. Lett E, Asabor E, Beltrán S, Cannon AM, Arah OA. Conceptualizing, contextualizing, and pperationalizing race in quantitative health sciences research. Ann Fam Med 2022; doi:10.1370/AFM.2792/-/DC1

More in This Series

In the March 2023 issue of CLN, the nuances and complexities surrounding the use of race in medicine were highlighted in the article, “The Past, Present, and Future of Race in Medicine.” Read more at www.myadlm.org/cln/health-equity-diversity-and-inclusion