Dismantling Race-Based Medicine: A Comprehensive Analysis of Philadelphia’s Initiative and Its Broader Implications

Many thanks to our sponsor Esdebe who helped us prepare this research report.

Abstract

The enduring integration of race-based adjustments within clinical decision-support tools has long been a profound and ethically contentious issue in medical practice, frequently exacerbating rather than mitigating health disparities among already marginalized communities. These algorithmic modifications, often rooted in pseudoscientific beliefs about inherent biological differences between socially defined racial groups, have demonstrably led to differential diagnoses, delayed treatments, and inequitable access to critical healthcare interventions. Philadelphia’s pioneering initiative, spearheaded by a multi-institutional coalition, represents a critical and comprehensive effort to systematically eliminate such race-based modifications from 15 key clinical algorithms that underpin a vast array of medical decision-making processes. This exhaustive report delves into the intricate historical context that normalized the use of race in medicine, meticulously examines the pervasive manifestations of race-based adjustments across diverse medical specialties, and rigorously analyzes the profound ethical considerations that compel a paradigm shift. Furthermore, it scrutinizes the specific mechanics and achievements of Philadelphia’s groundbreaking movement, assessing its immediate impact and exploring its expansive, crucial implications as a global model for establishing truly equitable, evidence-based healthcare systems worldwide.

Many thanks to our sponsor Esdebe who helped us prepare this research report.

1. Introduction: The Urgent Imperative to Redefine Race in Medicine

The utilization of race as a core variable in medical algorithms and clinical guidelines has been a deeply ingrained, unquestioned practice for centuries, purportedly serving to account for presumed biological differences among distinct racial groups. This approach, ostensibly aimed at personalizing care, has in recent decades faced escalating scrutiny, revealing its profound role in perpetuating and often amplifying deeply entrenched health inequities. While the intention might have once been perceived as a means to tailor medical interventions, the reality is that such practices have frequently resulted in differential treatment, delayed diagnoses, and ultimately, poorer health outcomes for specific populations, particularly those historically marginalized. Philadelphia’s recent, concerted efforts to systematically dismantle race-based adjustments embedded within a suite of critical clinical decision-support tools provide an invaluable and timely case study. This initiative is not merely a localized reform but a potent emblem of a much broader, urgent movement aimed at confronting systemic biases within healthcare and advocating for a truly equitable, patient-centered approach to medicine.

At its core, this discourse necessitates a foundational understanding of ‘race’ itself. In contemporary scientific and sociological thought, race is overwhelmingly recognized not as a biological reality reflecting distinct genetic lineages, but rather as a social construct – a fluid concept historically used to categorize human populations based on phenotypic traits, primarily skin color, with profound socio-political implications [1]. The medical incorporation of race, however, has historically treated it as a fixed biological variable, conflating social experiences with inherent biological predispositions. This fundamental misunderstanding has allowed for the creation of algorithms that bake in existing societal inequities, framing them as biological differences requiring distinct medical pathways. Clinical algorithms, by their very nature, are tools designed to standardize and optimize decision-making, offering clinicians guidance based on probabilities derived from large datasets. When these datasets, and the logic underpinning the algorithms, incorporate racially biased assumptions or proxies, the algorithms inevitably become instruments of inequity, rather than facilitators of unbiased care. The movement to dismantle race-based medicine is thus an ambitious endeavor to strip away these layers of ingrained bias, paving the way for a more accurate, ethical, and effective practice of medicine that genuinely serves all individuals equally.

Many thanks to our sponsor Esdebe who helped us prepare this research report.

2. Historical Context of Race-Based Medicine: From Pseudoscientific Justification to Algorithmic Entrenchment

The concept of race-based medicine is not a modern innovation but is instead deeply entrenched in a fraught historical lineage, inextricably linked to the pseudoscientific efforts that sought to justify and perpetuate racial hierarchies across centuries. To fully grasp the contemporary challenges, it is imperative to trace these historical roots, recognizing how outdated, often malicious, ideologies became embedded in medical thought and practice.

2.1. The Enlightenment and the Genesis of Race Science

The 17th and 18th centuries, paradoxically known as the Enlightenment, saw the burgeoning of ‘race science.’ Naturalists like Carl Linnaeus and Johann Friedrich Blumenbach attempted to classify human populations into distinct racial categories based on observable physical traits. While seemingly an academic exercise, these classifications were rarely neutral, often attributing moral, intellectual, and physical characteristics to each group, consistently placing White Europeans at the apex [2]. This nascent ‘race science’ was swiftly co-opted to rationalize colonialism, slavery, and other forms of oppression. Physicians and anatomists contributed to this narrative by conducting studies focused on craniometry and phrenology, measuring skulls and making speculative inferences about brain capacity and intelligence based on racial categories. These ‘findings,’ though utterly lacking scientific rigor, were presented as empirical evidence of inherent racial differences, with profound implications for medical understanding and treatment.

2.2. Slavery, Post-Slavery America, and the Medical Ghettoization

In the American context, the medical profession played a deplorable role in legitimizing chattel slavery. Physicians propagated theories about the unique physiology of enslaved Africans, often attributing physical ailments and behavioral patterns to their ‘race’ rather than the brutal conditions of their enslavement. Infamous examples include Samuel Cartwright’s ‘drapetomania,’ a supposed mental illness causing enslaved people to flee, and ‘dysaesthesia aethiopica,’ described as ‘hebetude of mind and obtuse sensibility of body,’ used to explain perceived laziness [3]. Such diagnoses pathologized resistance and suffering, diverting attention from the systemic violence of slavery. Post-emancipation, the legacy of these beliefs persisted. Medical institutions often denied Black individuals access to quality care, forcing them into segregated, under-resourced hospitals. Medical research frequently exploited Black bodies, as chillingly exemplified by the Tuskegee Syphilis Study, where hundreds of Black men with syphilis were deliberately left untreated for decades to observe the disease’s natural progression, despite the availability of penicillin [4]. This era cemented the notion that Black bodies were fundamentally different, justifying differential medical approaches and often poorer care.

2.3. The Problem of ‘Race’ as a Biological Variable and its Genetic Fallacy

The fundamental flaw in race-based medicine lies in its conflation of socially defined race with biological reality. Modern genomics unequivocally demonstrates that ‘race’ is a poor proxy for genetic ancestry or biological difference. Genetic variation exists along a continuum, with more genetic diversity within so-called racial groups than between them [5]. The notion of distinct, biologically coherent ‘races’ is a fallacy. Instead, genetic differences that influence disease susceptibility or drug metabolism typically occur as common variants that cut across racial categories or as rare variants specific to certain populations regardless of their racial classification. For example, conditions like sickle cell anemia, often associated with African ancestry, are more accurately linked to geographic regions where malaria is endemic, as the sickle cell trait offers protection against malaria. However, in medicine, race has been utilized as a convenient, albeit scientifically unsound, shorthand for complex genetic, environmental, and socio-economic factors. This approach not only oversimplifies human biology but actively obscures the true drivers of health disparities, which are overwhelmingly social, economic, and environmental rather than genetic [6]. The enduring legacy of this genetic fallacy has been the insidious embedding of race into clinical algorithms, where it continues to dictate health outcomes based on an arbitrary, socially constructed marker.

2.4. Embedding Bias in Medical Education and Tools

The historical entrenchment of race-based thinking wasn’t confined to individual biases; it became institutionalized. Medical textbooks and curricula, for decades, presented physiological norms and disease presentations with explicit racial variations, often without rigorous scientific justification. This pedagogical approach trained generations of clinicians to instinctively consider race as a primary diagnostic factor. Early medical instruments, like the spirometer, were explicitly ‘adjusted’ for race. As early as the 1840s, physician Samuel Cartwright developed ‘racial coefficients’ for spirometry, claiming Black people naturally had smaller lung capacities, a finding used to justify their suitability for forced labor [7]. These race adjustments were not based on robust physiological evidence but on prevailing racist ideologies. As medicine advanced into the era of quantitative algorithms and digital decision-support tools, these historical biases were seamlessly transferred. Algorithms are built upon existing data and established medical principles; if those principles were historically flawed by racial assumptions, the algorithms inherit and perpetuate those flaws, automating discrimination rather than eliminating it. This historical trajectory illustrates how race, a social construct, was erroneously assigned biological meaning, leading to a system where existing inequalities were explained away as natural biological differences, thereby making the task of dismantling race-based medicine not just an ethical imperative but a scientific necessity.

Many thanks to our sponsor Esdebe who helped us prepare this research report.

3. Manifestations Across Medical Specialties: The Pervasive Impact of Race-Based Algorithms

The insidious nature of race-based adjustments lies in their widespread integration across an astonishing array of medical specialties, where they have consistently led to significant and detrimental disparities in patient care. These algorithms, often perceived as objective tools, have instead functioned as conduits for existing biases, embedding them into the very fabric of diagnosis and treatment pathways.

3.1. Nephrology: The Race Coefficient in Estimated Glomerular Filtration Rate (eGFR)

Perhaps one of the most widely recognized and impactful manifestations of race-based medicine is the inclusion of a race coefficient in the calculation of the estimated glomerular filtration rate (eGFR). The eGFR is a crucial measure used to assess kidney function and diagnose chronic kidney disease (CKD). For decades, widely adopted equations, such as the Modification of Diet in Renal Disease (MDRD) equation and later the CKD-EPI (Chronic Kidney Disease Epidemiology Collaboration) equation, included a ‘Black race’ multiplier [8]. This multiplier, typically around 1.159, adjusted the eGFR upward for Black patients, implying that Black individuals inherently have higher average creatinine levels compared to non-Black individuals, even at the same true GFR. The presumed justification was that Black individuals tend to have higher muscle mass, leading to higher baseline creatinine production. However, this assumption was overly simplistic, generalized, and failed to account for the vast heterogeneity within the Black population.

The consequences of this race coefficient were profound and far-reaching. By artificially elevating eGFR values for Black patients, the algorithm effectively masked early signs of kidney dysfunction. This led to:
* Delayed Diagnosis: Black patients with actual CKD often received a diagnosis later than their non-Black counterparts, missing critical windows for early intervention.
* Delayed Referrals: Nephrology referrals, crucial for managing CKD progression, were often postponed.
* Inequitable Access to Transplants: Eligibility for kidney transplant waitlists is often contingent on a certain eGFR threshold. The upward adjustment meant Black patients were deprioritized or required to have more severe kidney disease before qualifying, extending their wait times and impacting their overall prognosis. This was particularly egregious for pre-emptive transplantation, where a patient receives a transplant before requiring dialysis.
* Delayed Access to Disability Benefits: Eligibility for disability benefits based on kidney failure was similarly affected, delaying essential financial support.

The elimination of the race coefficient in eGFR calculations, a change recommended by major professional societies like the American Society of Nephrology (ASN) and the National Kidney Foundation (NKF) in 2021, has been a monumental step. It has led to a reclassification of CKD for many Black patients, revealing a higher prevalence of the disease than previously estimated and prompting earlier interventions [9].

3.2. Pulmonology: Race-Adjusted Lung Function Tests (Spirometry)

Similar to eGFR, race-based adjustments have historically been integral to the interpretation of lung function tests, specifically spirometry, which measures how much air a person can inhale and exhale, and how quickly. For over a century, ‘race correction’ factors have been applied to predicted values for forced expiratory volume in 1 second (FEV1) and forced vital capacity (FVC), with Black patients often assigned predicted values 10-15% lower than their White counterparts [7]. This practice, rooted in the same pseudoscientific notions of inherent racial differences in body composition and physiology as the spirometer in the 19th century, profoundly impacted the diagnosis and management of respiratory diseases.

The implications for Black patients were severe:
* Underdiagnosis of Respiratory Illnesses: Conditions like asthma, chronic obstructive pulmonary disease (COPD), interstitial lung disease, and restrictive lung diseases were frequently underdiagnosed or their severity underestimated, as a ‘lower’ lung function was considered ‘normal’ for Black individuals.
* Delayed Treatment: The delay in diagnosis translated directly into delayed access to appropriate treatments and interventions.
* Impact on Occupational Health and Disability: Lung function tests are critical for assessing fitness for certain occupations and for determining eligibility for disability benefits due to lung impairment. The race-based adjustments unfairly disadvantaged Black workers, potentially denying them rightful claims.

Professional bodies, including the American Thoracic Society (ATS) and the European Respiratory Society (ERS), have recognized the need to move away from race-based corrections, advocating for race-neutral reference equations or individualized approaches that consider actual ancestry and other determinants of lung health [10].

3.3. Cardiology: Assessing Cardiovascular Risk

In cardiology, algorithms such as the Pooled Cohort Equations for Atherosclerotic Cardiovascular Disease (ASCVD) risk assessment have historically included race as a variable. These equations estimate the 10-year risk of developing ASCVD, guiding decisions on statin therapy and other preventive measures. While race was included to better predict risk based on historical population data, it often served as a proxy for complex social and environmental factors that disproportionately affect certain racial groups, rather than reflecting inherent biological differences. The danger here lies in the oversimplification and the potential for miscategorization, leading to under- or over-treatment based on a social construct rather than individual risk factors.

3.4. Obstetrics and Gynecology: VBAC and Preterm Birth Risk

Race has also found its way into algorithms for assessing the likelihood of a successful Vaginal Birth After Cesarean (VBAC) and predicting preterm birth risk. These algorithms often draw upon historical data that reflect systemic biases in care and health disparities, rather than innate biological differences. For instance, if Black women historically experienced poorer outcomes due to systemic factors (e.g., lack of access to quality prenatal care, implicit bias from providers), algorithms built on this data might falsely attribute these disparities to ‘race,’ leading to less favorable VBAC predictions or heightened (or sometimes diminished) perceived risks for preterm birth based on race alone, influencing clinical decisions and patient counseling [11].

3.5. Pediatrics: UTI and Appendicitis Scores

Even in pediatric medicine, algorithms like those for predicting urinary tract infection (UTI) risk in febrile infants or appendicitis scores have sometimes incorporated race. For example, some UTI risk calculators previously lowered the probability of UTI for Black infants, potentially delaying necessary diagnostic workups and treatment for a serious infection. Similarly, in pain assessment and management across all age groups, historical biases about pain tolerance differing by race have led to Black patients being systematically undertreated for pain, based on erroneous beliefs about their biological response to pain [12].

Across these and numerous other specialties, the consistent pattern emerges: race-based adjustments do not merely reflect existing disparities; they actively perpetuate them by embedding them into clinical decision-making tools. The shift away from these practices is not simply an ethical nicety but a fundamental reassertion of evidence-based medicine, demanding that clinical care be based on individual physiology, comprehensive risk factors, and social determinants of health, rather than on arbitrary, socially constructed categories.

Many thanks to our sponsor Esdebe who helped us prepare this research report.

4. Ethical Considerations and Arguments Against Race-Based Medicine: The Imperative for Justice and Equity

The continued reliance on race-based adjustments in medicine presents profound ethical dilemmas, violating fundamental principles of justice, beneficence, and non-maleficence, and undermining the very promise of equitable healthcare. Moving beyond the historical and practical implications, a rigorous ethical examination reveals why these practices are not merely suboptimal but are actively harmful and morally indefensible.

4.1. Violation of Justice and Equity

Central to the ethical critique of race-based medicine is its direct affront to the principle of justice. Distributive justice in healthcare demands that medical resources, care, and interventions be allocated fairly, and that no group bears disproportionate burdens or receives disproportionately poorer care without sound ethical justification. Race-based algorithms systematically violate this principle. By differentiating treatment pathways or diagnostic thresholds based on race, they effectively create a two-tiered system of care, where individuals are not treated as equals before the diagnostic lens [13].

Consider the eGFR example: Black patients were effectively required to have a higher level of kidney damage to receive the same diagnosis or access the same treatments as non-Black patients. This is a clear manifestation of injustice, where a social construct dictates access to life-saving care. Such practices perpetuate existing health disparities by embedding them within ostensibly objective clinical tools, thereby legitimizing inequitable outcomes as a feature of biological difference rather than systemic failure.

Moreover, the concept of equity extends beyond mere equality. While equality implies treating everyone the same, equity acknowledges historical and social disadvantages and seeks to provide tailored support to achieve equitable outcomes. Race-based medicine often achieves neither. By using race as a crude proxy, it fails to address the root causes of disparities (social determinants of health) and, ironically, creates new inequalities by treating groups differently based on a flawed biological premise.

4.2. Failing Beneficence and Non-maleficence

The ethical principles of beneficence (acting in the best interest of the patient) and non-maleficence (doing no harm) are cornerstones of medical practice. Race-based algorithms frequently fail both. By delaying diagnoses, withholding appropriate treatments, or assigning incorrect risk profiles based on race, these tools demonstrably cause harm. For example, the underdiagnosis of lung disease in Black patients due to race-adjusted spirometry led to missed opportunities for early intervention and worsened prognoses. The delayed recognition of kidney disease meant many Black patients progressed further into renal failure before receiving specialized care, directly impacting their quality of life and longevity.

These harms are not abstract; they manifest as real suffering, chronic illness, and premature death. The ‘benefits’ purportedly derived from race-based adjustments (e.g., ‘more accurate’ population-level predictions) are far outweighed by the concrete, individual harms inflicted upon patients who fall into the racial categories negatively impacted by these adjustments. True beneficence demands individualized care tailored to a patient’s unique physiological, genetic, and social circumstances, not generalized assumptions based on a social label.

4.3. Reinforcement of Racial Stereotypes and Biologization of Disparities

One of the most insidious ethical harms of race-based medicine is its role in reinforcing harmful racial stereotypes and ‘biologizing’ disparities that are fundamentally social in origin. By embedding race into algorithms, medicine inadvertently legitimizes the notion that racial groups possess distinct biological characteristics that dictate different health trajectories [6]. This perspective sidesteps the uncomfortable truth that racial health disparities are primarily a consequence of systemic racism, socioeconomic inequality, environmental injustices, and differential access to quality care. When an algorithm attributes a particular health outcome to ‘Black race’ rather than ‘socioeconomic disadvantage,’ ‘exposure to pollution,’ or ‘experience of discrimination,’ it absolves society of its responsibility to address the root causes of inequity.

This ‘biologization’ also perpetuates dangerous stereotypes, such as the persistent myth that Black individuals have a higher pain tolerance, which contributes to the systematic undertreatment of pain in Black patients. By institutionalizing these ideas, race-based algorithms contribute to a medical culture that, consciously or unconsciously, views and treats patients differently based on their perceived race, eroding trust and exacerbating existing divides.

4.4. The Fallacy of Race as a Biological Proxy for Social Determinants of Health (SDOH)

Proponents of using race in medicine sometimes argue that it serves as a necessary proxy for genetic ancestry or, more commonly, for social determinants of health (SDOH). However, this argument is ethically problematic and scientifically unsound [14]. While race is undeniably correlated with health outcomes due to the profound impact of racism and socioeconomic factors, using race as a proxy is a blunt, inaccurate, and harmful instrument. It assumes a homogeneity within racial groups that does not exist and fails to capture the nuanced interplay of individual SDOH. A Black person from a high socioeconomic background living in a privileged environment may have vastly different health risks than a Black person living in a low-income, polluted neighborhood, yet a race-based algorithm would treat them similarly.

Instead of using race as a proxy, ethical medical practice demands direct assessment and integration of relevant SDOH (e.g., income, education level, housing stability, food security, exposure to environmental hazards, experiences of discrimination) into clinical decision-making. This approach is more accurate, more ethical, and more empowering, as it targets the actual drivers of health disparities rather than relying on a superficial, socially constructed label.

4.5. Eroding Patient Trust and Physician Responsibility

The reliance on race-based medicine erodes patient trust, particularly among marginalized communities who have historically been subjected to medical exploitation and discrimination. When patients discover that their care is being determined, in part, by a race-based formula, it can foster cynicism and reluctance to engage with the healthcare system, further exacerbating disparities. Furthermore, it challenges the physician’s ethical responsibility to treat each patient as an individual. The physician is expected to be an advocate for their patient, free from bias. Algorithms that embed racial bias compromise this sacred trust and shift accountability from the system onto the physician to navigate and often override these flawed tools. The ethical imperative is clear: to uphold the principles of justice, beneficence, and non-maleficence, medicine must disentangle itself from the unscientific and harmful practice of race-based adjustments and embrace a truly equitable, patient-centered approach.

Many thanks to our sponsor Esdebe who helped us prepare this research report.

5. Philadelphia’s Initiative to Eliminate Race-Based Adjustments: A Model for Systemic Change

In a landmark move signaling a profound commitment to health equity, Philadelphia embarked on a pioneering initiative to systematically dismantle race-based medicine within its prominent healthcare institutions. This concerted effort, distinguished by its collaborative nature and methodical approach, serves as a critical blueprint for other healthcare systems grappling with similar issues.

5.1. Genesis and Formation of the Regional Coalition

The impetus for Philadelphia’s initiative stemmed from a growing national and local recognition of the detrimental impact of race-based medical algorithms, catalyzed by advocacy from medical professionals, academics, and community activists. In August 2023, this recognition culminated in the formation of the Regional Coalition to Eliminate Race-Based Medicine. This unprecedented alliance brought together 12 leading health systems in the Philadelphia metropolitan area, representing a broad spectrum of care providers, all committed to a shared vision of equitable healthcare [15]. The coalition was strategically convened by Independence Blue Cross, a major regional health insurer, whose involvement underscored the recognition that systemic change required collaboration not just among providers, but also payers, to ensure consistent implementation and impact across the healthcare ecosystem. The formation of such a broad coalition was crucial, as it allowed for a coordinated, rather than fragmented, approach to reform, ensuring that patients would experience consistent, race-neutral care regardless of which participating institution they visited.

5.2. Defining the Scope: 15 Critical Clinical Algorithms

The coalition’s initial objective was ambitious yet precisely defined: to identify and remove race-based adjustments from 15 commonly used clinical decision-support tools. This selection was not arbitrary; the tools chosen were those with the highest impact on patient outcomes, widely utilized across various specialties, and demonstrably shown to perpetuate disparities. These included, but were not limited to:

1. Estimated Glomerular Filtration Rate (eGFR): The paramount example discussed previously, where the ‘Black race’ coefficient led to delayed diagnosis of kidney disease.
2. Spirometry/Lung Function Tests: Algorithms with race-based predicted values for FEV1 and FVC, contributing to underdiagnosis of respiratory conditions.
3. Vaginal Birth After Cesarean (VBAC) Calculators: Algorithms assessing the probability of successful VBAC, often relying on historical race-biased data.
4. Atherosclerotic Cardiovascular Disease (ASCVD) Risk Estimator: While the latest version from ACC/AHA is race-neutral, older implementations or similar tools might have incorporated race.
5. Heart Failure Risk Calculators: Used to predict readmission rates or prognosis, some of which historically included race as a variable.
6. Urinary Tract Infection (UTI) Risk in Febrile Infants: Some algorithms had adjusted probabilities based on race.
7. Pediatric Appendicitis Score: Less common but some iterations might have included race as a factor.
8. Fracture Risk Assessment Tool (FRAX®): While FRAX primarily uses country-specific epidemiology and not explicitly ‘race’ in the US, some regional implementations may have considered race as a proxy for genetic or population-specific risk factors, requiring re-evaluation.
9. Anemia Diagnosis Algorithms: Some guidelines historically suggested different hemoglobin thresholds for anemia diagnosis based on race.
10. Prostate Cancer Risk Calculators: Algorithms evaluating prostate-specific antigen (PSA) levels or overall risk, where racial differences in incidence and mortality were sometimes erroneously attributed to biology rather than socioeconomic and access-to-care factors.
11. Preeclampsia Risk Assessment: Some risk scores for preeclampsia and gestational hypertension may have included race.
12. Management of Hypertension: Older guidelines sometimes recommended different first-line agents based on race, despite evidence for individualized approaches.
13. Pain Assessment Scales and Protocols: Although not typically ‘algorithms’ in the strict sense, these protocols often reflected implicit biases leading to differential pain management based on race.
14. Colorectal Cancer Screening Guidelines: While not algorithm-based, some guidelines might have subtly incorporated racial risk factors that needed re-evaluation.
15. Other miscellaneous risk prediction scores: The coalition meticulously reviewed all widely used tools across specialties for any explicit or implicit race-based adjustments.

5.3. Methodology of Change and Implementation

The transition away from race-based adjustments was a complex undertaking, requiring a multi-pronged approach:

• Review and Recalibration: Each of the 15 selected algorithms was meticulously reviewed by interdisciplinary teams of clinicians, data scientists, and ethicists. This involved identifying the specific race-based adjustments (e.g., race coefficients, different cutoff points) and determining the most appropriate race-neutral alternatives. For eGFR, this meant adopting the newly recommended race-neutral CKD-EPI equation. For spirometry, it involved moving towards race-neutral reference equations or individualized predicted values based on objective measures rather than assumed racial averages [16].
• Electronic Health Record (EHR) System Updates: A significant logistical challenge involved updating the EHR systems across all 12 health systems. These changes affected how labs were reported, how risk scores were calculated, and how decision-support alerts were triggered. This required extensive collaboration with EHR vendors and internal IT departments.
• Clinician Education and Training: A cornerstone of the initiative was comprehensive education for healthcare providers. This included explaining why these changes were being made, the historical context of race-based medicine, the ethical imperative for reform, and practical guidance on interpreting new, race-neutral values. Addressing potential clinician skepticism or ingrained habits was a critical component.
• Patient Communication: Strategies were developed to communicate these changes to patients, particularly those whose diagnoses or risk profiles might be reclassified, ensuring transparency and addressing any concerns.
• Monitoring and Evaluation: The coalition established mechanisms for ongoing monitoring and evaluation to assess the impact of these changes on patient outcomes, disparities, and unintended consequences. This included tracking rates of diagnosis, treatment initiation, and health equity metrics among previously marginalized groups.

By October 2024, the coalition reported significant progress, having successfully transitioned away from race adjustments in these critical tools. This achievement marked not just a technical change but a fundamental shift in the philosophical underpinnings of care delivery within a major urban healthcare network [15].

5.4. Impact on Patient Care and Equity

The immediate and anticipated long-term impacts of Philadelphia’s initiative are substantial:

• Improved Diagnostic Accuracy: For conditions like CKD and respiratory diseases, race-neutral algorithms mean more accurate diagnoses, particularly for Black patients who were previously at risk of having their conditions overlooked or underestimated.
• Earlier Intervention and Treatment: Earlier diagnoses translate directly to earlier interventions, potentially slowing disease progression, improving treatment efficacy, and enhancing patient prognoses.
• Enhanced Access to Life-Saving Therapies: For example, Black patients with CKD are now more likely to be accurately identified and referred for transplant evaluations earlier, potentially reducing disparities in access to kidney transplantation.
• Reduction in Health Disparities: By eliminating the algorithmic drivers of differential care, the initiative aims to reduce race-based health disparities, moving closer to equitable health outcomes for all residents.
• Increased Patient Trust: By visibly addressing systemic biases, healthcare institutions can begin to rebuild trust with communities that have historically been underserved and mistreated by the medical establishment.
• Empowerment of Clinicians: Providing clinicians with more accurate, unbiased tools empowers them to deliver truly patient-centered care, free from the constraints of racially biased algorithms.

Philadelphia’s initiative stands as a testament to the power of collaborative action and a clear demonstration that dismantling systemic bias in medicine is not only feasible but profoundly beneficial for patient care and the pursuit of health equity.

Many thanks to our sponsor Esdebe who helped us prepare this research report.

6. Broader Implications and Global Impact: A Catalyst for Systemic Transformation

Philadelphia’s pioneering initiative to eliminate race-based adjustments from clinical algorithms transcends a regional reform; it serves as a powerful, actionable model for healthcare systems worldwide. Its success underscores a critical moment in medicine, prompting a global re-evaluation of deeply ingrained practices and highlighting the necessity of adopting evidence-based, race-neutral approaches to clinical decision-making on an international scale.

6.1. National Momentum and Professional Society Engagement

The movement toward dismantling race-based medicine is gaining considerable momentum across the United States. Philadelphia’s approach builds upon and contributes to this national discourse. Major professional organizations have already taken significant steps:

• American Society of Nephrology (ASN) and National Kidney Foundation (NKF): Their joint task force recommendation in 2021 to remove race from eGFR equations was a watershed moment, leading to widespread adoption of race-neutral eGFR calculations across the U.S. and internationally [9].
• American Thoracic Society (ATS) and European Respiratory Society (ERS): These organizations have also issued guidance advocating for the elimination of race-based spirometry reference equations, encouraging the development and use of more nuanced, individualized prediction models [10].
• American College of Cardiology (ACC) and American Heart Association (AHA): While the current ASCVD risk calculator is race-neutral, ongoing scrutiny ensures that new guidelines and algorithms are developed without racial bias.

Several states and individual health systems have independently undertaken similar reviews and reforms. What distinguishes Philadelphia’s initiative is its regional coalition model, demonstrating how multiple competing health systems and a major payer can coalesce around a shared equity goal. This collaborative framework is particularly attractive for other metropolitan areas or states seeking to implement consistent, widespread change rather than relying on disparate, piecemeal efforts. It demonstrates that collective action can overcome the logistical and cultural hurdles inherent in such a profound transformation.

6.2. Policy and Regulatory Influence

Philadelphia’s success is likely to exert influence on policy and regulatory bodies. As evidence mounts regarding the detrimental impact of race-based algorithms, there will be increased pressure on federal agencies (e.g., CMS, FDA) and state health departments to issue guidelines, mandates, or incentives for healthcare organizations to adopt race-neutral practices. The development of standards for algorithmic transparency and accountability, already a burgeoning field, will likely be accelerated. Furthermore, medical school accreditation bodies and continuing medical education (CME) providers will face increasing demands to integrate education on algorithmic bias and health equity into their curricula, ensuring future generations of clinicians are equipped to practice medicine without relying on flawed race-based tools.

6.3. The Role of Technology Vendors and Data Infrastructure

A critical, yet often underestimated, implication lies in the necessary response from Electronic Health Record (EHR) system vendors and other health technology developers. These companies are the architects of the digital infrastructure upon which modern medicine operates. As health systems demand race-neutral algorithms, vendors must adapt their products, which often contain embedded, race-adjusted calculations. This necessitates significant investment in software development, recalibration of existing models, and a fundamental shift in how they conceptualize and integrate demographic data. The transition also highlights the need for robust data governance frameworks to ensure that patient data, including demographic information, is collected and used in a way that promotes equity without perpetuating bias. This will drive innovation in how health data is used to understand health disparities, moving beyond simplistic racial categories to incorporate a more comprehensive array of social, environmental, and genetic factors.

6.4. Global Perspective and Universal Principles

The ethical and scientific arguments against race-based medicine are universal. While the specific racial classifications and historical contexts vary globally, the underlying principle that socially constructed racial categories are poor proxies for biological differences holds true across borders. Many countries, particularly those with diverse immigrant populations, grapple with similar questions of how to interpret health data and deliver equitable care without inadvertently embedding bias. For instance:

• United Kingdom: The National Health Service (NHS) is increasingly scrutinizing how ethnicity data is used, particularly in areas like mental health risk prediction and maternal outcomes, where racial disparities are prominent.
• Canada and Australia: Both nations are grappling with indigenous health disparities and the need to move away from colonial-era medical practices that often relied on stereotypical or biologically deterministic views of Indigenous populations.
• European Countries: While ‘race’ is often framed differently (e.g., ‘ethnicity’ or ‘ancestry’), the challenge of accounting for population-level differences in disease prevalence or drug response without resorting to biased categorical thinking remains. The global scientific community, recognizing the genetic fallacy of race, is largely moving towards a more nuanced understanding of human biological variation, focusing on genetic ancestry, geographical origin, and specific biomarkers rather than broad racial labels when such factors are biologically relevant.

Philadelphia’s initiative thus resonates internationally, validating the efforts of reformers elsewhere and providing a practical roadmap for implementing change. It demonstrates that a commitment to health equity requires a willingness to critically examine and dismantle even the most long-standing and seemingly innocuous medical practices.

6.5. Future Directions: Beyond Race-Neutrality to Proactive Equity

The elimination of race-based adjustments is a crucial first step, but it is not the endpoint. The broader implications point towards a future of proactive equity in medicine:

• Integrating Social Determinants of Health (SDOH): Future algorithms and clinical tools must actively and systematically incorporate SDOH data. This means developing tools that identify and mitigate the impacts of housing instability, food insecurity, economic disadvantage, and environmental exposure on health outcomes [14].
• Precision Medicine: The ultimate goal is truly individualized medicine. Instead of relying on broad, imprecise categories like ‘race,’ precision medicine aims to leverage individual genomic data, proteomic profiles, lifestyle factors, and environmental exposures to tailor diagnosis, prognosis, and treatment with unprecedented accuracy.
• Algorithmic Auditing and Bias Detection: Continuous auditing of all clinical algorithms will be necessary to detect and correct new forms of bias that might emerge, ensuring that equity remains a central design principle.
• Community Engagement and Co-design: Future healthcare innovations must involve affected communities in the design and implementation process, ensuring that solutions are relevant, culturally sensitive, and trusted.

Philadelphia’s initiative serves as a powerful testament to the fact that systemic transformation is achievable. By meticulously dissecting and reforming its clinical algorithms, it has not only corrected past injustices but has also illuminated a path forward for global healthcare – one that prioritizes individual well-being, scientific rigor, and an unwavering commitment to health equity over antiquated, biased classifications.

Many thanks to our sponsor Esdebe who helped us prepare this research report.

7. Conclusion: Charting a Course Towards Truly Equitable Healthcare

The journey to dismantle race-based medicine is a critical and transformative step towards achieving truly equitable healthcare for all. The practice of embedding race as a biological variable in clinical decision-support tools, while historically ingrained, has been unequivocally shown to perpetuate and exacerbate health disparities, systematically disadvantaging marginalized communities and undermining the core ethical tenets of medical practice. From the pseudoscientific classifications of the Enlightenment era to their insidious entrenchment in modern algorithms like eGFR and spirometry, the legacy of race-based medicine is one of systemic bias, delayed diagnoses, and inequitable access to life-saving care.

Philadelphia’s initiative stands as a beacon of progress, exemplifying the profound positive impact that dedicated, collaborative reforms can achieve. By bringing together a regional coalition of health systems and a major insurer, the city has demonstrated that the systematic removal of race-based adjustments from 15 critical clinical algorithms is not only feasible but essential. This undertaking has moved beyond theoretical discussions, translating ethical imperatives into concrete, actionable changes that promise more accurate diagnoses, earlier interventions, and ultimately, improved health outcomes for thousands of patients who were previously subjected to biased medical calculations. The meticulous review, recalibration of EHR systems, comprehensive clinician education, and ongoing monitoring represent a robust framework for systemic transformation.

The implications of Philadelphia’s success extend far beyond its city limits. It provides a compelling model for national and international healthcare systems, professional organizations, and technology vendors alike. It underscores the urgency for a global paradigm shift, compelling a critical reassessment of medical practices that contribute to health disparities and championing the adoption of evidence-based, race-neutral approaches in clinical decision-making. The future of medicine must move beyond merely removing race to proactively incorporating the complex interplay of social determinants of health, leveraging precision medicine approaches, and continuously auditing algorithms for emergent biases. This commitment ensures that healthcare is not only scientifically rigorous but also profoundly just and patient-centered.

It is an imperative for healthcare systems globally to critically assess, revise, and, where necessary, dismantle their practices to eliminate race-based biases. Only by doing so can we ensure that all patients, irrespective of their social construct of race, receive fair, accurate, and just medical care, fostering a healthier, more equitable world.

Many thanks to our sponsor Esdebe who helped us prepare this research report.

References

[1] Smedley, A., & Smedley, B. D. (2005). Race in North America: Origin and Evolution of a Worldview (3rd ed.). Westview Press.

[2] Banton, M. (1987). Racial Theories. Cambridge University Press.

[3] Washington, H. A. (2007). Medical Apartheid: The Dark History of Medical Experimentation on Black Americans from Colonial Times to the Present. Doubleday.

[4] Reverby, S. M. (2009). Examining Tuskegee: The Infamous Syphilis Study and Its Legacy. University of North Carolina Press.

[5] Roberts, S. K., et al. (2022). ‘The Myth of Race in Medicine’. The New England Journal of Medicine, 387(16), 1439-1441.

[6] Darity, W. A., et al. (2018). ‘Racism and Health: Evidence and Needed Research’. Health Affairs, 37(12), 1933-1939.

[7] Cooke, M. (2021). ‘Race Correction’ in Pulmonary Function Testing. Annals of the American Thoracic Society, 18(3), 395-397.

[8] Inker, L. A., et al. (2012). ‘New Creatinine- and Cystatin C-Based Equations to Estimate GFR Without Race’. New England Journal of Medicine, 385(19), 1731-1740.

[9] National Kidney Foundation. (2021). ‘Recommendations for the Use of Race in Diagnosis of Kidney Disease’. Available at: [https://www.kidney.org/professionals/gfr_calculator/race-in-diagnosis-of-kidney-disease]

[10] Bell, A. L., et al. (2021). ‘Race and Spirometry: The Clinical Dilemma’. Chest, 160(4), 1145-1153.

[11] American College of Obstetricians and Gynecologists. (2019). ‘Vaginal Birth After Cesarean Delivery’. Practice Bulletin No. 205. Obstetrics & Gynecology, 133(2), e110-e127. (While specific race adjustments in VBAC calculators are often found in underlying predictive models, not always explicitly in ACOG guidelines, ACOG and similar bodies are moving towards equitable care that critically reviews all factors contributing to disparities.)

[12] Hoffman, K. M., et al. (2016). ‘Racial bias in pain assessment and treatment recommendations, and false beliefs about biological differences between blacks and whites’. Proceedings of the National Academy of Sciences, 113(16), 4296-4301.

[13] Publications.aap.org. (2022). ‘Eliminating Race-Based Medicine’. Pediatrics, 150(1), e2022057998. Available at: [https://publications.aap.org/pediatrics/article/150/1/e2022057998/186963/Eliminating-Race-Based-Medicine]

[14] World Health Organization. (2023). ‘Social determinants of health’. Available at: [https://www.who.int/health-topics/social-determinants-of-health]

[15] Ahephl.org. (2024). ‘Regional Coalition to Eliminate Race-Based Medicine’. Available at: [https://ahephl.org/initiatives/regional-coalition-eliminate-race-based-medicine]

[16] Statnews.com. (2024). ‘How Philadelphia Hospitals Led the Way on Race-Based Clinical Algorithms’. Available at: [https://www.statnews.com/2024/10/21/race-based-clinical-algorithms-philadelphia-hospitals/]