When Science Meets Bias: The Controversy Over Black Infant Mortality Research

A newly published reanalysis of a highly influential 2020 study on black infant mortality has ignited fierce debate in academic, medical, and legal circles. The original research claimed that Black newborns were more likely to survive when cared for by Black physicians – but serious methodological flaws may have tainted these conclusions and their far-reaching policy implications. This controversy highlights broader questions about how statistics can be manipulated, how research influences policy decisions, and the complex interplay between facts and emotions in public discourse.

The Study That Shook Healthcare Policy

In August 2020, researchers published a study in the prestigious Proceedings of the National Academy of Sciences (PNAS) that examined approximately 2 million births in Florida. The research team, led by Brad N. Greenwood and including University of Minnesota researcher Dr. Rachel Hardman, made a startling claim: Black newborns were significantly more likely to survive when cared for by Black physicians rather than white physicians.

The timing of this research proved critical to its impact. Released during the heightened racial justice discussions following the death of George Floyd in May 2020, the study offered what appeared to be concrete evidence of structural racism within the healthcare system. Major news outlets amplified its findings, and policymakers quickly incorporated the research into arguments for diversifying the medical workforce.

"When Black physicians care for Black newborns, their mortality rate is cut in half," became a commonly cited statistic in discussions about healthcare disparities. The study provided seemingly irrefutable scientific validation for concerns about racial bias affecting medical outcomes, and it aligned perfectly with broader societal conversations about systemic racism.

The study's influence reached the highest levels of government when Supreme Court Justice Ketanji Brown Jackson cited it in her dissenting opinion in the 2023 Supreme Court case Students for Fair Admissions v. Harvard, which addressed affirmative action in university admissions. The research appeared to demonstrate a clear and compelling interest in racial diversity among healthcare professionals – a point that resonated with many advocates for equity in medicine.

Medical schools and healthcare systems began implementing new diversity initiatives, with administrators referencing the study as evidence that increasing minority representation among physicians would directly improve patient outcomes. The research didn't just inform academic discussions; it drove real policy changes affecting medical education, hiring practices, and resource allocation across the healthcare industry.

Yet beneath the surface of this influential study lay methodological problems that would eventually undermine its dramatic conclusions.

The Reanalysis: Uncovering Methodological Flaws

In September 2024, PNAS published a reanalysis of the original data by researcher Robert VerBruggen that called the 2020 study's conclusions into serious question. When VerBruggen applied proper statistical controls – particularly accounting for birthweight, a critical factor in infant mortality – the purported mortality gap between Black newborns treated by white versus Black physicians largely disappeared.

The reanalysis exposed several crucial problems with the original research:

First, the 2020 study failed to adequately control for birthweight, which is perhaps the single most important predictor of newborn survival. Low birthweight babies face significantly higher risks regardless of physician race, and this factor alone could explain much of what the original researchers attributed to physician race.

Second, documents obtained through Freedom of Information Act (FOIA) requests revealed troubling communications between the researchers. These emails suggested that the team deliberately chose statistical approaches that would produce their desired outcome – evidence of racial disparities – rather than letting the data speak for itself.

As physician Vinay Prasad noted in his analysis of the controversy, "The reanalysis shows that when accounting properly for birthweight, there's minimal to no difference in mortality based on physician race." This contradicted the dramatic claims of the original study that had suggested Black physicians cut mortality rates for Black newborns in half.

Furthermore, the design of the original study failed to account for hospital-level factors that might confound the results. Black physicians might be more likely to work at certain hospitals with different resources, patient populations, or practice patterns – all variables that could influence infant mortality rates independent of physician race.

"When researchers cherry-pick methods to arrive at predetermined conclusions, it undermines the entire scientific enterprise," said one academic statistician who reviewed both papers. "The reanalysis suggests this wasn't just an innocent methodological disagreement – it appears the original team was searching for a specific result."

The reanalysis also highlighted the challenges of properly interpreting observational data. Unlike randomized controlled trials, where researchers can directly test cause and effect, observational studies like the 2020 paper rely on statistical methods to account for confounding variables. When these methods are improperly applied – or selectively chosen to support a hypothesis – misleading conclusions can result.

This methodological debate extends beyond academic quibbling. If the reanalysis is correct, policies implemented based on the original study's conclusions may be addressing the wrong factors in trying to improve Black infant mortality rates – potentially diverting resources from more effective interventions.

From Research to Policy: The Ripple Effects

The controversy surrounding this research exemplifies how flawed studies can rapidly influence policy decisions, court rulings, and public opinion. Once the original study was published and widely publicized, its findings became embedded in numerous policy discussions and decisions despite the lack of scientific replication or thorough vetting of its methods.

Justice Ketanji Brown Jackson's citation of the study in her Supreme Court dissent represents perhaps the most high-profile example of how academic research can shape legal reasoning at the highest levels. "Black babies die at twice the rate of white babies in the United States," she wrote, citing the study as evidence that racial diversity in professions like medicine yields tangible benefits.

The Supreme Court's reliance on this now-challenged research raises difficult questions about how scientific evidence is evaluated in legal contexts. Courts generally lack the specialized knowledge to independently assess the validity of complex statistical studies, instead relying on the peer review process and academic consensus. When that process fails, as it appears to have done in this case, judicial decisions may rest on shaky empirical foundations.

Medical schools and healthcare systems similarly incorporated the study's findings into diversity initiatives and hiring practices. Many institutions cited the research when implementing programs specifically designed to increase the number of Black physicians, arguing that such demographic changes would directly improve patient outcomes.

"This case illustrates why policymakers need to be careful about embracing research findings before they've been thoroughly vetted," noted one health policy expert. "There's often pressure to act quickly on studies that align with prevailing narratives, but sometimes patience and scientific skepticism are warranted."

The timeline between publication and policy implementation was remarkably compressed in this case. Within months of the original study's release, medical schools were citing it in diversity program materials, hospitals were developing new hiring strategies based on its findings, and policymakers were incorporating its conclusions into healthcare equity initiatives.

This rapid translation from research to policy highlights a broader challenge: the difficulty of ensuring that evidence-based policymaking is truly based on sound evidence. When research appears to confirm widely held beliefs or aligns with existing political priorities, it may receive less rigorous scrutiny than studies challenging conventional wisdom.

The infant mortality study controversy also reveals how difficult it can be to correct the record once flawed research has shaped policy. Even after the reanalysis, many of the initiatives launched in response to the original study remain in place, their foundational justification now undermined but their institutional momentum undiminished.

Beyond Statistics: The Human Impact of Research Controversies

Behind the statistical debates lies a sobering reality: Black infant mortality rates in the United States remain consistently higher than those for other racial groups. Regardless of the flawed study, this disparity represents a genuine public health crisis that deserves serious attention from policymakers and medical professionals.

The controversy surrounding this particular research shouldn't obscure the broader reality of health disparities in America. Black mothers are more likely to die from pregnancy-related complications, Black patients often receive less pain medication for identical conditions, and access to quality healthcare remains unequal across racial lines.

"The problem with the flawed study isn't that it highlighted racial disparities in healthcare – those are real and documented," explained one public health researcher. "The problem is that it may have misattributed the cause of those disparities, potentially leading to ineffective solutions."

Real-world consequences flow from such research controversies. If healthcare systems focus primarily on matching Black patients with Black physicians based on the original study's claims, they might neglect more effective interventions addressing social determinants of health, medical resource allocation, or specific clinical practices that contribute to disparities.

For Black parents navigating pregnancy and childbirth, these academic debates translate to practical questions about their healthcare choices. Should they seek out physicians of their own race? Should they worry more about the specific hospital they choose? What factors actually matter most for their baby's health?

"When research like this gets publicized, it affects how patients perceive their care options," noted one maternal health advocate. "If the study was wrong, that means families might have been making decisions based on faulty information – and that has real consequences."

The human stakes extend to medical professionals as well. Black physicians, already underrepresented in medicine and often subject to additional scrutiny, found themselves in a complicated position following the original study. While the research highlighted their importance in addressing health disparities, it also potentially reduced their complex clinical skills to their racial identity.

As one Black neonatologist explained, "I don't want patients to think I provide better care because of my race. I provide good care because of my training, experience, and dedication. This study, whether right or wrong, risked essentializing physicians in a way that's problematic."

The controversy also affects public trust in science more broadly. When high-profile studies are later debunked or significantly challenged, it can fuel skepticism about scientific institutions and expertise – a particularly dangerous outcome during an era when evidence-based policy faces numerous challenges.

The Psychology of Statistical Interpretation

This research controversy illuminates broader patterns in how humans process information, particularly statistics that align with or challenge existing beliefs. Psychological research consistently shows that people engage in "motivated reasoning" – evaluating evidence more critically when it contradicts their existing views and accepting confirming evidence with less scrutiny.

The rapid acceptance of the original infant mortality study demonstrates this pattern. For many who already believed that racial disparities in healthcare stemmed from physician bias, the study offered welcome confirmation. The headline finding – that Black babies died at lower rates when cared for by Black doctors – aligned perfectly with narratives about the importance of racial concordance in healthcare.

"Statistics that confirm what we already believe often get remembered and cited without much critical examination," explained one cognitive psychologist. "It feels intuitively right, so the methodological details seem less important."

This pattern isn't unique to any political perspective or demographic group. During the COVID-19 pandemic, for instance, similar dynamics played out across the political spectrum. Different groups embraced or rejected statistics about mask efficacy, vaccine safety, or lockdown impacts based largely on how those statistics aligned with their existing worldviews.

The infant mortality study controversy highlights another psychological phenomenon: the persistence of initial impressions. Even after the reanalysis undermined the original study's conclusions, many continue to cite its findings as if they remain unchallenged. This "continued influence effect" shows how difficult it is to correct misinformation once it has been widely disseminated.

"Our brains aren't designed to constantly update our understanding based on new evidence," noted one neuroscientist. "Once we incorporate a statistic or finding into our mental model of the world, it takes significant effort to dislodge it – especially if that finding feels emotionally or morally significant."

Emotions play a crucial role in how statistics are interpreted. Research on infant mortality inherently triggers strong emotional responses – few subjects matter more than the survival of newborns. These emotions can override critical thinking about statistical methods or research design, leading even sophisticated readers to accept flawed conclusions if they resonate emotionally.

The way findings are framed also significantly impacts how they're received and remembered. The original study's claim that Black physicians cut mortality rates in half for Black newborns created a powerful, memorable statistic that was easily repeatable and emotionally resonant. The reanalysis, with its more nuanced discussion of statistical controls and methodological considerations, proved far less quotable and emotionally impactful.

These psychological patterns help explain why the original study gained such traction despite its flaws, and why correcting the record proves so challenging. They also highlight the need for greater awareness of how cognitive biases shape our interpretation of research findings, particularly on emotionally charged topics involving race, health, and social justice.

Broader Implications for Scientific Integrity

The controversy surrounding the infant mortality study raises profound questions about scientific integrity in an era when research on politically sensitive topics faces intense scrutiny. When studies make dramatic claims about racial disparities, they enter a polarized landscape where findings may be weaponized for political purposes regardless of their scientific merit.

Documents obtained through Freedom of Information Act requests revealed disturbing evidence that the original researchers may have manipulated their analytical approach to produce desired results. These emails suggest they deliberately chose statistical methods that would yield evidence of racial disparities, even when more appropriate methods didn't show significant effects.

This case highlights a growing concern about "p-hacking" and other questionable research practices – where researchers analyze data in multiple ways until they find statistically significant results that support their hypothesis. Such practices undermine scientific credibility and can lead to false conclusions that misinform public policy.

"The replication crisis in science isn't just about honest errors – sometimes it involves researchers putting their thumb on the scale," explained one research methodologist. "When that happens with studies addressing sensitive social issues, the damage extends beyond academia to public trust and policy decisions."

The peer review process, designed to catch such problems, failed in this case. This raises questions about whether traditional academic gatekeeping functions effectively when research touches on politically charged topics where reviewers themselves may hold strong prior beliefs.

Several journals have faced similar controversies in recent years, publishing studies on contentious topics that later required correction or retraction. These incidents suggest a need for more rigorous review processes, particularly for research that makes dramatic claims about sensitive social issues.

"Peer review works best when reviewers approach manuscripts with healthy skepticism," noted one journal editor. "But when findings align with reviewers' own views on important social issues, that skepticism sometimes gets suspended."

The controversy also highlights tensions between advocacy and objectivity in research. Scientists studying health disparities often genuinely want to improve outcomes for marginalized groups. This laudable goal can sometimes conflict with the scientific imperative to follow evidence wherever it leads, even if the results don't support preferred narratives.

Academic institutions face challenging questions about how to respond when researchers appear to have manipulated data or analysis to support predetermined conclusions. Should there be formal investigations? Retractions? Consequences for researchers who violate scientific norms?

"We need mechanisms to correct the scientific record that don't depend on political pressure," argued one research ethicist. "The infant mortality study controversy became polarized so quickly that it became difficult to have a dispassionate discussion about the actual evidence."

The case also demonstrates the need for data transparency. The reanalysis was only possible because the original data were accessible to other researchers. Without this transparency, flawed findings might have continued influencing policy indefinitely without correction.

Moving Forward: Science, Policy, and Public Discourse

The controversy surrounding this research points toward several constructive paths forward for science, policy, and public discourse. Rather than simply discarding all research on racial disparities or uncritically accepting any study that identifies such disparities, a more nuanced approach is needed.

First, the scientific community must strengthen safeguards against methodological manipulation. More rigorous pre-registration of study designs, greater transparency about analytical decisions, and stronger peer review processes can help ensure that research on sensitive topics maintains the highest standards of integrity.

"Pre-registering analyses before seeing the data makes it much harder to 'find' the results you want," explained one statistician. "It forces researchers to commit to their methods based on scientific principles rather than desired outcomes."

Second, policymakers need more sophisticated frameworks for evaluating research evidence, particularly when making decisions about complex social issues. Single studies, no matter how dramatic their findings, should rarely drive major policy changes without corroboration and careful scrutiny.

"Evidence-based policymaking doesn't mean jumping on the latest headline-grabbing study," noted one policy analyst. "It means carefully weighing multiple lines of evidence, understanding methodological limitations, and acknowledging uncertainty."

Third, media outlets should improve how they report on scientific research, particularly studies addressing politically charged topics. Greater attention to methodological details, inclusion of independent expert perspectives, and follow-up coverage when studies are challenged can help prevent misleading narratives from taking hold.

Fourth, the public deserves better tools for evaluating statistical claims. Improving statistical literacy and critical thinking skills can help citizens navigate an information landscape filled with competing data points and contradictory research findings.

"Most people aren't going to become statisticians," said one education expert. "But they can learn to ask basic questions about how research was conducted and whether findings have been replicated."

Finally, the controversy highlights the need for humility from all sides in debates about complex social phenomena. Racial disparities in healthcare outcomes result from multiple interacting factors, and simplistic explanations – whether attributing differences entirely to physician bias or dismissing structural factors altogether – rarely capture the full reality.

"The truth about healthcare disparities is more complex than either side of this debate might suggest," concluded one health equity researcher. "We need robust research methods, honest engagement with evidence, and a commitment to following data wherever it leads – even when the conclusions challenge our prior beliefs."

The infant mortality study controversy ultimately serves as a case study in how science, policy, and public discourse interact in our polarized era. By learning from this example – recognizing both the reality of health disparities and the importance of methodological rigor in studying them – we can build a stronger foundation for addressing persistent inequities in healthcare outcomes.

This post contains affiliate links. If you purchase through these links, I may earn a commission at no extra cost to you.