Welcome to the 51st Annual Congress, and to this session, Anti-Racism in Critical Care Medicine. I will be discussing where is the bias in critical care medicine, the impact of race in critical care medicine. My current role as clinical pharmacy manager is also to be the co-director for a critical care council for a large academic health systems, pharmacy and therapeutics committee, which allows me to promote equity and outcomes. I've been doing a lot of work with the diversity, equity and inclusion committee of SCCM, as with the clinical pharmacy and pharmacology section. I've also been doing work locally with the New York City Society of Health System Pharmacists. I acknowledge that I am by no means an expert in racial issues in the critical care environment, but I'm contributing to the work that must be done. I have no relevant disclosures, but I do speak about veclary on Gilead Speakers Bureau. My objectives will be to review historical events that have led to bias, to introduce areas for potential bias, and to highlight the impact of racial bias in critical care medicine that affects patient outcomes. I must say, as a pharmacist in critical care, and as a race conscious woman, I have not specifically seen intended differences in care overall. But it's important to look and see what is really happening and listen to the concerns of our patients and their families in a compassionate way, considering the backdrop of U.S. history. They say, a quote, history is written by the victors, attributed to Winston Churchill in the 1930s, perfectly describes the colonial spectacular accounts of history in the New World. But history is also told by the literate people capable of taking daily accounts of events, and led to many inaccuracies in the history books within the United States of America. Starting in 1864, post-Senate passing of the 13th Amendment, a whole year passed before the House abolished slavery in January 1865. Black codes were almost simultaneously implemented as early as 1865 in Mississippi, and these were laws establishing codes of conduct for newly freed slaves. Singing, dancing, beating a drum was a punishable crime, and even unemployment was considered vagrancy and subject to jail time. Interracial relationships were obviously punishable by death, and slavery was an inheritable condition where descendants carried the same risk of imprisonment. Education and 1867 efforts ran concurrently, where Northern troops were sent to the South to protect newly freed slaves and preserve the Union of the USA. Many ex-slaves moved West, following the Trail of Tears, to where Native Americans went into quote-unquote Indian territory. The creation of historically Black colleges and universities also began in 1867 to educate Blacks, and communities like in the Greenburg area of Tulsa, Oklahoma coined the Black Wall Street by Booker T. Washington grew. But Jim Crow laws in the South during the 1880s limited such freedoms, and the Ku Klux Klan and white supremacy led fear and intimidation tactics such as lynching of over 4,000 known people. Plessy v. Ferguson, a Supreme Court ruling, justified separate book equal in 1896, while burning and killing of Blacks in the Tulsa, Oklahoma massacre of 1921 continued. Redlining real estate practices of the 1930s showed drawn lines on a map to reflect allowable areas for Blacks to rent and obtain mortgages, creating segregating housing, which aligned with employment and health care. Speaking of which, the Tuskegee Study of Syphilis was also conducted at that time in 1934. Then, the Civil Rights Movement that included some momentous historical events like Brown v. Board of Education in 1940, I'm sorry, 1954, desegregated schools. The Civil Rights Act of 1964 and the Voting Rights Act of 1965 helped to create equality in many sectors, and many civil rights leaders on religious, political, entertainment arenas carried the torch. We see that it's really a complex history that has brought us to this point, where there seem to be many viable concerns leading to the current climate. Fear, frustration, and marginalization leads to mistrust, lack of exposure, clustering of people, economic centralization within the cities, and racialized concentration creates structural racism. Limited access to health care, reluctance to seek medical attention, miseducation, poor health literacy further leads to stereotypes and the formation of bias. And we're here today to discuss health disparities. From living experiences, the natural human occurrence is to form bias. Bias is defined as a strong opinion, inclination toward or against a person, group, place, or thing in an unequal manner, affecting understanding, actions, and decisions. It can either be conscious and explicit, or unknown, unconscious, and implicit. And it's based on our experiences, our education, for example, where we're educated and the strength of such education, environmental factors like system structure or resource allocation, social factors like our culture, our family, or our religion, and of course now media and news outlets influence our unconscious bias. Structural racism further contributes to bias, and it is the totality of ways in which societies foster discrimination via mutually reinforcing inequitable systems. For example, in the housing market, a majority of white people live in neighborhoods composed of at least 75% white, while a majority of black people live in neighborhoods composed of less than 35% white. Never underestimate a zip code, which also zones schools accordingly and maintains areas industry. 11.3% of blacks are unemployed compared to 5.3% of whites, 5% Asians, and 7.4% Hispanics, resulting in 26% of black people being below poverty level, similar to Hispanics but twice that of whites and Asians. Interestingly, urban legend has it that social security number, the fifth digit, of black people is an even number to discourage lenders disproportionately. The criminal system also has six times more incarcerated black males than white males, largely led by the stereotypic association with drugs from the 80s war on drugs, despite equal rates of drug use in both white and black people. Stop and frisk of the 90s targeting black males and incurring crack penalties 100 times harsher than powder cocaine penalties started, prompting mitigation by the Fair Sentencing Act of 2010, which reduced penalties of crack crimes to only 18 times greater than that of powder cocaine crimes. Prescription opioids, in retrospect, is less regulated and punishable and focuses highly on treatment. And finally, resource allocation of quality specialized care may also be limited in certain areas. This presentation by no means discounts the many types of bias that exist. So we do recognize that bias can exist on age, gender, sexual orientation or identity, physical abilities, and race. And it can also be on multiple dimensions or involve multiple minority identities. But we will focus the discussion on race, particularly differences between black and white patients, where most of the data is available, appreciating the fact that many medical references lump black or dark-skinned people and Hispanic or Latinx patients together. Other types of bias that influence medicine involve cognitive bias, which is systematic errors in thinking due to human processing limitations, or the fast-thinking approach, which may account for 36 to 77% of diagnostic errors. Visitation bias involves the selection of articles for publication based on one's own cognitive bias. And patient care biases exist in upwards of 64% of healthcare professionals in a low to moderate level, translating to poor patient-provider interactions, the use of dominant or condescending tone or allowing visitation discordance, delaying clinical decisions, and health inequities that may increase risk factors for severe disease. Here we see that Hispanic patients are twice as likely to be uninsured than black, white, or Asian people. Infant mortality for blacks is twice that of whites, Hispanics, or Asians. Black and Native American people have 5% higher self-assessed poor health. Native Americans have more distress. Black people have twice as much diabetes than white people, and heart disease is highest among black patients according to the National Health Statistics. So these disparities are evident in medicine overall, heavily influenced by structural racism. We see here that the prevalence of hypertension is 20% higher for black patients, diabetes is 20% higher for black patients, and smoking is 5% higher for black patients. So is there an impact in critical care medicine? When looking at stroke mortality, the well-known stroke belt of 2 to 4 times higher mortality in southeastern United States existed since the 1940s. Although stroke mortality has decreased from 1999 to 2016, disparities still exist. The CDC attests to structural racism and racism leading contributions significantly. Since 26% of black residents live in the stroke belt compared to 10% across the U.S., who happen to have an adjusted stroke mortality 40% higher than whites. Similarly, sepsis mortality is higher in those same regions of the United States, but the exact cause cannot be inferred. The U.S. Census delineates color lines according to areas where a large concentration of black or African Americans reside. In urban areas like California, the tri-state area, and Illinois, to name a few. And southern states like Louisiana, Mississippi, Alabama, Georgia, and South Carolina. And this is not as clear for other populations. We see the Asians, the Hispanic X, the Native American and Pacific Islanders, and white patient distributions based on the U.S. Census statistics. It forces us to ask the question resulting in many retrospective analysis of where we have been, especially concentrating after the year 2000 when many efforts to decrease mortality were implemented. Two studies evaluating acute lung injury and acute respiratory distress syndrome in the ICU assessed respiratory management of acute lung injury, assessment of low tidal volume and elevated end-expiratory lung volumes to obviate lung injury, and the fluid catheter treatment trial, otherwise known as the ARMA, Alveoli, and FACT databases from the ARDS network spanning from 1996 to 2008. The first trial by Dr. Aronson looked at 60-day mortality and 28-day vent-free period in all ALI patients greater than 18 years old and concluded that mortality of blacks and Hispanics was 33% and statistically greater than white patients. Adjusted odds ratio of black patients was 1.42, but adjusting for severity crossed the median line. For Hispanic patients, the adjusted odds ratio was 1.94, and adjusting for severity increased the odds to two times that of white patients. Vent-free days for blacks was equal to that of whites and statistically greater than Hispanic patients. The author concluded that black and Hispanic patients have a higher risk of death, possibly because black patients have a higher severity of illness and potential for a delay in diagnosis, and Hispanic patients may have a language discordance. The second trial by Dr. Cangellaris looked at 60-day mortality and vent-free and organ failure-free days at 28 days or discharge between Duffy null and Duffy positive patients of African descent. The Duffy antigen receptor for chemokines or IL-6 or DARK was studied because of an associated pathogenesis in ALI animal models. They looked at all patients over 18 with genetic data on record and found that DARK deficiency polymorphism is present in 67% of patients, and there was no clinical severity of illness differences between Duffy null and Duffy positive patients, but the 60-day mortality of Duffy null was 38% and significantly greater than Duffy positive patients at 21%, even with similar infectious causes. There was 2.6 increased 60-day odds of death in the Duffy null patients, even with 8 fewer vent-free days and 4.5 fewer organ failure-free days. The author concluded that Duffy or DARK deficiency in people of African descent may pay a role in ALI and is associated with worse outcomes in African Americans. Similar outcomes exist with African Americans without genetic sampling to validate this potential genetic polymorphism. There were two studies looking at community-acquired pneumonia outcomes in the ICU, one from a cohort of VA data, another from a prospective observation from the Genetic and Inflammatory Markers of Sepsis database. The VA study looked at length of stay and 30-day mortality in men over 65 predominantly, according to the Infectious Disease Society of America and the American Thoracic Society guidelines for CAP. They found that in ICU, African Americans had a shorter adjusted hospital length of stay with a hazard ratio of 0.84, and that there was a decreased adjusted 30-day mortality with a hazard ratio of 0.82. Dr. Freit concluded that African Americans had better outcomes with equal access, similar quality, and similar severity of illness. Dr. Maher's study looked at CAP treatment differences in the ED and ICU in patients over 18 with CAP. White patients were generally older, but the Charleston Comoridi Index was similar between groups. They found that there was a similar admissions at about 86%, but Black patients were primarily admitted to large academic centers with a decreased likelihood of receiving guideline directed antibiotics within four hours. There was significant hospital effect and similar care otherwise when adjusting for clustering. The author concluded that there was similar care and outcomes between Black and White patients. The racial differences in survival after in-hospital cardiac arrest from the National Registry of Cardiopulmonary Resuscitation Database looked at survival to discharge, return of spontaneous circulation greater than 30 minutes, or post-resuscitation survival and neurological status in patients over 18 with in-hospital V-fib and V-tach arrest. They found that survival to discharge of Blacks was 25% and statistically significantly less than White patients at 37%. ROSC of Black patients was 56% and ROSC in White patients was 67%. There was clustering of Black patients in hospitals that tended to have lower survival overall. So Dr. Chan and colleagues concluded that Black patients have a 27% lower relative risk and a 12% lower adjusted risk of survival, and no clinician bias was found. The recommendation was to improve survival in underperforming hospitals. There's much more data for sepsis-related mortality. Three studies from the Database of Healthcare Costs and Risk of Sepsis in the National Three studies from the Database of Healthcare Costs and Utilization Projects sponsored by the Agency for Healthcare Research, AHRQ, identified patients by ICD-9 codes for sepsis and severe infections. Dr. Dombrowski studied patients in New Jersey in 2002. Dr. Sandoval studied California patients in 2011. Dr. Jones studied 18 states' worth of patients in 2004 and 2011. All studies looked at patients greater than 18 years old hospitalized for sepsis. The occurrence and outcomes of sepsis influence of race found that hospitalizations of Black was statistically greater than Whites, especially in the 35 to 44-year-old range, with a relative risk of 4.4. The adjusted mortality of Blacks was statistically greater than Whites, and the case fatality was the same. The length of stay of Black patients was greater than White patients, and the ICU admissions were greater for Blacks than Whites. The author concluded that adjusted mortality is higher for Blacks than Whites, but the case fatality is similar. Association between race and case fatality rates in hospitalization for sepsis found that the case fatality of Whites is actually statistically greater than Blacks and Hispanics or Asian patients, and even logistic regressions decreased the odds risk of mortality in Blacks, Hispanics, and Asians. The hospital mortality of Whites was two times that of Blacks, Hispanics, and the same for Asians, and the mortality difference of hospitals was statistically significant. The case fatality was lower in Blacks and Hispanics even after adjusting for in-hospital differences. After adjusting for clustering of Black patients in fewer hospitals, the in-hospital care does not account for disparities in sepsis outcome, and pre-hospital care should be the focus. Lastly, racial disparities in sepsis-related in-hospital mortality looked at sepsis hospitalizations and in-hospital mortality. They found that sepsis diagnosis increased significantly from 2004 to 2013, but fortunately, sepsis mortality rates decreased in the same year. Mortality for Blacks and Asians was significantly higher than that of Whites and Hispanics in 2004, while mortality for Blacks, Asians, and Hispanics was significantly higher than that for Whites in 2013. Adjusting for patient and hospital-level characteristics attenuated disparities for Black and Hispanic patients, but not Asians. Adjusting for patient and hospital characteristics yielded similar rates for all except Asian and other patients. Further sepsis mortality was assessed by Dr. Chaudhary's analysis of Vizient consortium data of mostly-affiliated hospital of 30 states in 2012 to 2014. 249 hospitals met criteria in 7 states. After adjusting for age, gender, risk of mortality, year, and randomization, the author found that hospitalization of Blacks was actually less than Whites, with a significant odd ratio, but greater in Blacks greater than 30 years old. The adjusted mortality of community-acquired sepsis and healthcare-associated sepsis was lower in Black patients, and there was a decreased odds of mortality in Blacks than White patients. Dr. Chaudhary concluded that there was a lower sepsis mortality in Black patients. Dr. Moore looked at Black-White racial disparities in sepsis, a prospective analysis of the reasons for geographic and racial differences in stroke that regards database. The objective of this longitudinal study was to characterize racial differences in sepsis, first hospitalization for serious infection, and hospitalization for sepsis overall. In stroke patients over 45 years old, they found that Blacks had a lower rate of hospitalization for serious infection, and adjusting for risk factors maintained an odds ratio of 0.65. Secondarily, they found that the same for hospitalization for sepsis among patients with similar rates of severity of illness, ICU admission, hospital mortality, and long-term mortality. Dr. Moore concluded that Black participants have a decreased risk of sepsis due to decreased rates of hospitalization for sepsis, but similar odds once hospitalized. They did admit that this study had poor sensitivity and specificity for sepsis in a longitudinal format. The epidemiology of pulmonary embolism, racial contrasts in incidents and in-hospital fatality was a retrospective cohort from data of the National Center for Health Statistics compressed mortality file of discharges. The objective was to, based on ICD-9 and 10 codes, to find explanations for racial disparities in sepsis and hospitalization for sepsis overall. The objective was to, based on ICD-9 and 10 codes, to find explanations for Black and White disparity in PE mortality. They found that the overall mortality decreased from 45% to 33%. Within each time period, the incidence was statistically significantly higher for Blacks at 47% and women at 62%. But the mortality of Blacks was also higher and the mortality of men was higher than that of women. Dr. Schneider concluded that case fatality rates were similar, but unable to determine the ICU specific disparities. Despite the lack of clinical data and non-coded demographic data being unavailable, and ED patients and in-transit data was also included in the study. Opening this discussion for pregnancy data is not solely ICU, but very important because there is a phenomenal disparity in pregnancy-related mortality, enough that the CDC tracks the statistics in a publicly transparent source. Since the rise in mortality of 1986, reporting became available in 1987, showing that within one year of pregnancy, Black women die three times as much as Hispanic, non-Hispanic Whites, Asian, or Pacific Islanders, and 1.5 times more than Asian Americans or Alaskan Natives. These deaths are mostly attributed to poor access of care, quality of care, chronic disease, but the CDC also considers structural racism and biases fundamentally. The breakdown shows cardiovascular conditions as the biggest contributor to pregnancy-related death, followed by infection or sepsis and other non-cardiovascular medical conditions. So it becomes harder to tease out ICU-specific mortality, as many patients may not need ICU admissions, but one can look or consider looking at sepsis. This may diminish the power or make it difficult to obtain power in a frequentist approach study. Race and insurance status and risk factors for trauma mortality, too, was a retrospective cohort to determine effects of race and insurance on mortality from 2001 to 2005. All adults with blunt trauma were included, and they found that both race and insurance affected mortality, in that Black patients and Hispanic patients had a statistically higher mortality, and uninsured patients had a statistically higher mortality than insured patients. So it makes sense that insured White patients had the lowest mortality compared to others in the group. Overall, ICU mortality was assessed in the effect of race and ethnicity on outcomes among patients in the intensive care unit. In patients over 18 in the ICU for greater than four hours, they saw that Asians had the highest mortality and Blacks had the lowest. The odds ratio of Asians was 1.37 and Hispanics was 1.4. Adjusting the odds ratio crossed the median line for Asians and Hispanics and made it equal to that of Whites. The ICU length of stay for Blacks and Hispanics was statistically higher than that of Whites at two days, but adjusting the odds made only the Black patients have a higher 0.68 day length of stay. There was no differences in hospital characteristics. The author concluded that there was similar mortality and ICU length of stay in all patients despite greater severity and physiologic derangement in Blacks and older with more male patients in the White group. Difficulties lie in identifying race. In all these studies, there was a slight difference, but since 2004, self-reporting became the standard method of identification. The major issue lies in the fact that race is a social construct. For example, in 1865, race was Northern or Southern while Blacks were considered property. Since that time, physiologic color Black versus regional distinctions like African American could be used, but doesn't really capture immigrants in this large melting pot. Plus, there's no genetic basis for these designations and the association can cause social, political, or economic consequences. So, this lends to some background to the fact that only 46% of CPP members disclosed their race and ethnicity, while 37% of SECM disclosed their race and ethnicity. Unifying terminology attempts to classify people historically marginalized by health-harming discrimination. The term POC or POC, people of color, was controversial because Black people were called colored during the pre-civil rights era, and it's difficult to merge a diverse group of 573 Indian nations. So, the name streamlined in 2020 to BIPOC, Black, Indigenous, and People of Color, to convey a shared marginalized experience of two groups. URIM, or Underrepresented in Medicine, primarily focuses on pipeline education and training categorization and to foster a more equitable workforce, reflective of the U.S. population, while peer-defined is the newest terminology, forcing administrators and leaders to consider people who have been traditionally excluded because of their ethnicity and race. Many unanswered questions exist, but in summary, high-quality prospective data illustrating disparities are unavailable and probably unethical. Existing data addresses important questions without clinical detail. Most crude mortality differences diminish once adjusted, and a large heterogeneity of data makes it difficult to accurately make a conclusion. There's many unanswered questions, again, that exist. Such questions like, why do Black patients have higher mortality in their sixth decade of life than whites in their seventh or eighth? But it's a good thing, and they are hypothesis generating. There is potential for publication bias here, but there is active and intentional work to be done. Restorative efforts to improve diminished inequities involve us. Diversifying the workforce provides different cultural perspectives that public health data shows may lead to less bias and better patient-provider interactions. In the structurally advancing pipeline programs, elementary and middle schools' outreach mentorship and advocacy help bridge some gaps and allow individuals to invest in the community. Enhancing self-awareness and education, plus approaching cultural competence with humility, helps encourage lifelong commitments to self-reflection and mutual exchange of cultural imbalances. Considering expanding our social networks involves personal commitment, but may truly provide value. And finally, consider incorporating social epidemiologists, medical anthropologists, social scientists, and historians in the conversation. The importance of diversifying the workforce builds on equity, which sometimes may not considered an issue, but it really depends on the individual perspective, which I can't emphasize enough. Black Man in a White Coat is a New York Times bestseller and Time magazine top 10 book, where Dr. Tweedie describes all the issues he personally faced in chapters like People Like Us, Charity Care, Inner City Blues, Confronting Hate, Maintaining Status Quo, and Do Not Appear Insecure When Doctors Discriminate, The Color of HIV-AIDS. So, efforts must be intentional and involve the community, med schools, editorial boards, and research hierarchy. For the future, commit to resolve health inequities, continually reflect on injustice, improve precision on geographic ancestry in relation to race, open dialogue, educate actively, and consider visionary medicine as a tool for teaching medical students. And it kind of has a role play aspect where people can use their imagination and produces futures that preserve health and racial justice for all. Dr. Holloway, an American historian and 21st president of Rutgers University, quotes, if we are a country that realizes its ambitions, its aspirations, we are a country that must be honest with itself to tell the truth about its own history.

racial bias

critical care medicine

impact

patient outcomes

structural racism

racial disparities

healthcare workforce