Bias and Disparities in Neonatal End-of-Life Care

Dr Major-Kincade (MD, MPH, FAAP) is a board-certified neonatologist and pediatrician with more than 20 years of experience in clinical neonatology and health equity advocacy. She recently launched a full-time career in pediatric palliative care and hospice medicine and serves as the Medical Director of Pediatric Palliative Care at Children’s Memorial Hermann Hospital. As a Pampers Womb to World advisory board member, Dr Major-Kincade has also been part of the P&G brand’s philanthropic efforts to improve maternal health outcomes, with particular attention to maternal health equity. Her research interests include perinatal health disparities and inequities, provider-patient communication, implicit and explicit bias in health care, work-life balance, provider well-being, and increasing diversity in the health care workforce.The case is an incoming transport call from a referring Level III neonatal intensive care unit (NICU) to a regional referral Level IV NICU. Some details of the clinical case are omitted or changed to support anonymity. The conversation has been edited for brevity. Verbatim language is presented in quotes.Referring provider (RP): Hi, we have a baby to refer to you. The baby was born at 29 weeks and 3 days and is now 9 days old. After birth, she was noted to have gasping and required intubation in the delivery room. She is currently on endotracheal tube continuous positive airway pressure with pressure support with spontaneous breathing and fraction of inspired oxygen of 0.21. She previously failed extubation after 4 hours owing to her inability to handle secretions. She has no suck or gag reflexes and has tongue movements that may be fasciculations. “The baby still moves, somehow, a little bit, more than at the beginning…I think parents and we; we would like the baby to be evaluated by you guys, especially neuro-wise.”NICU fellow (NF): “We can definitely transport the baby for evaluation by our NICU Neurology team. This does not sound good at all. Have there been conversations about redirecting care at all? Just wondering if the severity or prognosis has been discussed with the family. Of course, we’re happy to transport the baby to see if we can provide the family with information. But I wasn’t sure if that was discussed as a possibility.”RP: “If you pull the tube, the baby is breathing, what do you do with that? I don’t think anyone has really talked to them about withdrawal of support because we were just seeing what happens with the baby, just day by day. I don’t think parents are really the not-do-anything type…The family is well-educated and have a good understanding, but I don’t think it has ever crossed their minds to think about withdrawal. In my experience, the Black population, they don’t go for it. They will just go to the end, you know?”NF: “Well, I think all parents try to make the most loving decisions for their children. Alright, we will be in touch.”How does the description of goals of care and potential redirection strike you?How does the connection between the family’s race and their assumed treatment preferences strike you?Author’s perspective (Dr Younge is a third-year NF, who identifies as a Black and Asian American cis woman): Despite what seemed like severe neurological injury without improvement over a week, it was unclear how forthcoming conversations about prognosis had been with the family. Although I have become more experienced with difficult conversations during my training, I am by no means an expert and am still uncomfortable with these discussions. Perhaps this is something that I never want to be “comfortable” doing. My particular interest in palliative and end-of-life care in the NICU has meant I have actively tried to step into these roles, but I also have the privilege of working in a large academic referral NICU with multiple subspecialists, including pediatric palliative care to support my growth and learning. I recognize that these supports may not be as robust in other units; nonetheless, I think all families deserve direct and honest conversations about prognosis and end-of-life decisions, even if it is in the context of referring to another hospital for more information. I was concerned that the family had not yet had frank conversations about their child’s prognosis and anticipated course of care at transfer, and this news would come as a shock with perhaps avoidable harm to the family.It seemed like assumptions were being made about the parents’ wishes, although it was unclear to me whether this was informed by conversations or other interactions or assumed based on stereotypes and bias. Although I do not think one can ascribe treatment preferences based on one aspect of a person’s identity, I do believe one’s identity, lived experiences, and community inform personal values and treatment preferences. Race is part of identity and can shape these preferences. Perhaps the provider was recognizing something in that but overgeneralizing.Dr Major-Kincade (T.M.K.): My initial response as a Black physician in both the neonatal and palliative care space is triple-fold: sadness, understanding, and anger. This perception that Black families do not typically go for compassionate redirection of care becomes a self-fulfilling prophecy. The literature on the adult population for Black communities especially bears out a higher likelihood of dying with technology, dying in the hospital, and lower use of palliative care and hospice services. When a provider responds that “they don’t typically go for it,” I wonder if they are considering how often the service is offered or how often the service is denied by omission. How many families are represented in the “in my experience” of those who refused, and, of those, what is their refusal based on? Was it based on prior experience with pediatric palliative care or prior experience with hospice and palliative care with an adult loved one who may have had a traumatic experience? Is the refusal embedded in mistrust, lack of transparency, or delayed care? My feelings of understanding are linked to the other words echoed by the provider, “I don’t think parents are really the not-do-anything type.” The belief that to regoal (ie, redirect) to comfort care means “not doing anything” is a bias unique to palliative care. When, in fact, if provided in line with the family’s goals, wishes, and values, allowing natural death represents the most complete care. I wonder how many times this foundational bias impedes shared decision-making for all families and is possibly further exacerbated by other biases related to racial disparities. Some quick tips and suggested language are provided in Table 1.Most importantly, I am reminded that no culture is a monolith. Each family is unique and has their own lived experiences. I wonder why some families are allowed to explore their responses to different treatment options through the context of their values and others are denied this opportunity under the guise of helping them or avoiding unnecessary discomfort on the part of the provider as well as the family. I wonder how much ambiguity might be removed by naming the bias and addressing that bias through communication skills that, in this case, would facilitate the beginning of a conversation that is sure to come at the receiving hospital. When these initial conversations do not happen, both the family and the receiving team face additional challenges that exacerbate barriers to equitable care. Inequity based on lack of opportunity for shared decision-making is one of the most insidious forms of bias but also one that has the potential for the greatest harm as families unpack and navigate what it means to not have complete information for decision-making.Describe existing literature on racial disparities in neonatal end-of-life care.Discuss possible mechanisms and etiologies for these disparities.Identify the utility and limitations of this data for informing patient care.Implement pathways to provide high-quality, equitable, and antiracist palliative and end-of-life care for NICU families.Neonatal palliative care: Holistic and interdisciplinary care of the infant, parents/guardians, and family members that aims to alleviate and attend to physical, emotional, and spiritual suffering experienced by the infant, their family, and their community. It is a holistic approach to care that can be combined with cure-oriented, disease-modifying care and increase at any time but especially when curative therapies are no longer deemed to be beneficial or appropriate.1Disparities in end-of-life care: Differences among population groups in the way that they access, experience, and receive health care at the end of life that are not informed by preferences (ie, choices based on full and accurate understanding of treatment options and availability of those options).2,3Social drivers (determinants) of comfort: Surrounding individual, social, and environmental conditions that influence whether, to what degree, and in what forms comfort measures are offered to and undertaken by patients and their families.4Goal-concordant care: Care at the end of life that is consistent with the preferences of the patient and their family, specifically in the location of death and intensity of care at the end of life.5Shared decision-making: A spectrum of decisions from those who are heavily directed by the medical team, through decisions made jointly between the patient and family and the medical team, to decisions made independently by the patient or family, with influence from each party in the decision varying based on the clinical situation and the patient or family’s preference.6End-of-life care is some of the most emotional and impactful care that we provide in the NICU. As such, there is a greater risk of bias, prejudice, and stereotypes influencing clinical decisions and ultimately contributing to disparities in end-of-life care.7 Differences in intensity, place of death, and duration of life-sustaining therapies at the end of life may reflect family preferences. Some families may elect to take more intensive paths with an emphasis on life prolongation even when nearing the death of their infant. Individual treatment preferences oriented toward this emphasis may be similar among families with shared racial, ethnic, or cultural identities. Disparities in end-of-life care, by contrast, are differences in access and opportunities to have quality end-of-life care and are attributable to racist policies leading to discrimination and differential access to services. Both differences and disparities in end-of-life care reflect social drivers of comfort, which builds on the concept of social determinants of health in recognition of the individual and socioecological contexts that influence providers and families caring for dying children.4There is no uniform approach to end-of-life care; rather, there is considerable variation in preferences across families. If guided by open and supportive conversations that seek to share medical information, assess a family’s understanding, and ascertain values and goals, the differences seen among racial groups in end-of-life care may reflect goal-concordant care arrived at through shared decision-making. On the other hand, insufficient training on behalf of providers to hold such conversations may contribute to end-of-life care disparities. All families are harmed when such conversations do not occur, but this harm occurs more frequently and deeply in families from marginalized and underserved backgrounds. As seen in this case, in which clinical uncertainty collides with complex and high emotions, avoiding such conversations can lead providers down a path based on assumptions or “choosing your own adventure,” wherein race and ethnicity blindly impact treatment decisions.Growing evidence in pediatric and neonatal literature mirrors findings of racial differences and disparities in palliative and end-of-life care seen in adult literature. This includes disparities in the mode of death and care intensity experienced at the end of life, referrals to palliative care, and opportunities for conversations around redirection of care. On the one hand, differences may be justifiable if they respect the autonomy of families and reflect their goals and values. However, assumptions and limited exploration of goals and values may lead to unjustifiable disparities. In this article, we discuss possible mechanisms that may lead to observable differences based on informed choice vs those that create inequities in end-of-life care. We also highlight some caveats and cautions when interpreting the available literature, such as overstating the role of mistrust and concerns with using outcomes for non-Hispanic white families as reference or baseline.Over the past several decades, the circumstances of death have changed in the NICU, with deaths following withdrawal of life-sustaining therapies becoming more common than deaths following unsuccessful resuscitative efforts.8 Therefore, there is a greater need and opportunity to support and prepare families to consider decisions to redirect care. Evidence suggests that mode of death in the NICU may differ by race and ethnicity, with infants from minoritized racial groups experiencing higher-intensity care at the end of life. In a multicenter registry of deaths of premature infants, de-escalation or non-escalation of care occurred more often for non-Hispanic white infants than for non-Hispanic Black infants or Hispanic infants.9 In a retrospective review of deaths in the Children’s Hospital Neonatal Network Database, compared with white infants, Black infants had decreased odds of non-escalation at the end of life.10 Among infant decedents who received palliative care consultation in NICUs in the Southern United States, the majority of infants died after withdrawal of life-sustaining interventions; however, Black infants were more likely to receive cardiopulmonary resuscitation before death compared with white infants.11Factors underlying the racial and ethnic differences in intensity of care at the end of life remain poorly understood in the neonatal population. Observations that infants from racially marginalized groups experience higher-intensity care toward the end of life may reflect various combinations of familial wishes, presumptions on behalf of providers, availability of pediatric palliative care services, and intersections of socioeconomic variables that encompass social drivers of comfort. In a multicenter study of newborns delivered between 22 and 24 weeks’ gestation, Black compared with white birthing parent race was associated with a greater likelihood of attempted resuscitation even when controlling for center variability.12 In a prospective study of at with anticipated the majority of on and with an emphasis on and may lead some especially those from marginalized groups to toward of intensive care before that racial and ethnic differences in end-of-life and palliative care are by family preferences a of racist policies and health care is evidence that the and quality of to consider redirection of care may by patient In a study from the Neonatal among infants born between 22 and weeks’ gestation, white families were more likely than Black or Hispanic families to have of redirection of In a more study from this of infants born before 29 weeks’ gestation, Black and Hispanic families were likely to have of redirection of care and redirection of care compared with white In contrast, in a retrospective study of infants who died in the NICU, there were of to or life-sustaining medical therapies between Black and white families. parents more often to life-sustaining therapies compared with Black although this not and most of to redirect care for their toward Perhaps with greater of death, there is greater equity in conversations to life-sustaining However, when there is more as in the case of infants born conversations may be equitable in all available paths to in the the of these conversations may lead to a self-fulfilling of higher-intensity end-of-life care for infants from racially minoritized groups that may not with family preferences. When pediatric palliative care may be to family especially in conditions and in for the end of life. about death and dying and prognosis may more frequently and among families to palliative care However, evidence suggests that access to this of support by with receiving lower of palliative care referrals and compared with white research suggests that palliative care may be for NICU patients and their of pediatric palliative care services, including access to as well as palliative care skills for neonatology providers, would likely disparities seen in end-of-life quality of goals of care and other difficult conversations may be limited by in NICU In a of NICU training in communication Despite this of being or for leading family to discuss goals of In of for an anticipated delivery at gestation, and neonatology discussed clinical and available treatment options most of the However, they goals and values to shared decision-making. is opportunity to improve conversations through communication this the for Medical recently with around as a for neonatology be further with attention to equity in end-of-life care and social drivers of choice may have a impact on our communication with families. In family for infants, the words or often than families and typically when families Medical and can also with NICU parents that such as and or no were poorly especially among parents with lower A understanding of prognosis may lead families to higher-intensity care when the medical team may this path is and may contribute to racial disparities in end-of-life the intersections of race and other socioeconomic including language and choice has the potential to exacerbate or racial disparities in end-of-life in the previously literature is that white were as the reference or in all Although such highlight racial differences among groups and may be for outcomes such as infant or birth, they may not as well for interpreting outcomes related to end-of-life care. When we the experience of to be the we risk that the of this or end-of-life care. for end-of-life care are in all groups but especially in because this aspect of health care is deeply We be concerned that the of the are not based on the community in a to pediatric palliative care quality of the who the as white and were palliative care This suggests that the of families and especially those from marginalized have not been into the of end-of-life care. A of we would be to treatment preferences for families through care and be to those preferences with goal-concordant of health care and medical providers is an for disparities in end-of-life However, the role of in this way the role of racist policies that The perspective of marginalized groups as as providers of the to be and use our to create antiracist policies our of communication, including a lack of transparency, can or exacerbate the family in the that they were being information or limited options in the care of their to the mistrust is the of patients deemed to be to palliative the end of life or in the of a providers may become or about a family’s to redirect care or their decision to further disease-modifying such as and are common in such and may our and care, when we that the family is care options that do not with our personal preferences. We to how a family will the and of any but we that the family is with the for their Perhaps we can take these to and reflect on our own values and about the and of available treatment options and how we discuss these options with I would highlight a different perspective of end-of-life care and death in the hospital through the lived experience, of Black families. In the the provider Black families to go for it As providers often this to mistrust, but I want to to think about death and dying for the Black community in the United death and dying of because there were limited that for of and when they provide care, it was often that the hospital you are may be a hospital in which some family members were denied care. this family even died the or at after being denied to the hospital. some marginalized death in the hospital may a does death at vs death in the death in the hospital one is previously it a prior Perhaps perhaps but we be not to death at in a that may not with the family’s values, and prior experiences, which may not have been with a that to comfort care for some families does not but represents doing they to life but suffering in with their and not this path for some families may not doing they to provide a death and a of many without As noted the way to and of at the end of life is with care that the family and does not the biases of the health care equitable care to NICU families equitable care at the end of life. skills in news and the values and goals of families may disparities in end-of-life care. As we seek to and these we be to that a family will a particular path or that there is an or of care in any and to the many of a lived experience and the socioecological context that may a family’s decision in this care can provide equitable care to families their NICU can take individual and their interdisciplinary and to their skills as palliative care providers and create to provide more equitable and end-of-life care for all families the individual providers can seek further training to support their communication skills in goals of care and their can and family to discuss and that may have been in by the of social drivers of comfort can be in and and for end-of-life care. When palliative care are referrals to these can be to equitable of and Neonatal Palliative for Care in the case The family was and, with interdisciplinary support from NICU providers, NICU the Palliative Care and social made the difficult and loving decision to their baby to natural of the of prior and after the death of a infant, including palliative care.