Addressing Disability Bias in Neuroprognostication

“Has anyone spoken to family? Do they understand the terrible prognosis?”Diego, a three-year-old with trisomy 21, was admitted to the pediatric intensive care unit (ICU) after a near-drowning event requiring prolonged resuscitation. Concerns about cerebral edema were growing.“Will they want extreme measures? I sure hope they don’t. His brain injury is devastating…I mean, even if he survives, he’ll just be suffering.”The medical teams were discussing prognosis. Four days after his injury, Diego had not demonstrated any purposeful movements. Imaging showed widespread changes consistent with hypoxic-ischemic brain injury, and his electroencephalogram showed diffuse slowing.“What are his chances? He’ll never walk and probably never talk. At best, even if he survives, he’ll be trapped in there, dependent on others for everything. He was already significantly disabled—and now this—some fates are worse than death.”In a conference room, the medical teams sat down with Diego’s parents to discuss his prognosis. Yet, after the neurologist spoke, the parents offered a radical reorientation: “Will he know when his family is around? Will he be able to hear us sing to him? Can he still be happy?”This vignette (which is an anonymized composite of cases) captures a common phenomenon where parents and medical teams have different and, at times, conflicting perspectives about prognosis. While these differences highlight challenges in pediatric neuroprognostication—for example, uncertainty of recovery potential in a developing brain, differences in family preferences for receiving prognostic information after a neurologic injury or diagnosis, and difficulty of effectively conveying medical information1—we believe that Diego’s case reveals a fundamental flaw in neuroprognostication: the reliance on ableist heuristics that frame disability as loss and deficiency. To enhance care delivery and equity of care, we propose a challenge to ableist assumptions and offer an alternative, capabilities-based framework for neuroprognostication.Clinical approaches to neuroprognostication often focus on how an injury or disease deviates from a well-child norm. We argue that this approach represents an implicit form of ableism, defined as “a network of beliefs, processes and practices that produce a particular kind of self and body” such that disability “is cast as a diminished state of being human.”2 In Diego’s case, this approach generates multiple deficit-focused statements regarding ambulation (“he’ll never walk”), communication (“never talk”), and other capacities (“dependent on others”). This insistence on normalcy has been criticized by disability scholars:3 by framing prognostication around how a child differs from a well-child norm, a sense of deficiency and loss is implied. This perspective is rooted in the medical model of disability, which views disability as a problem residing within an individual that medicine aims to minimize or eradicate.4 The medical model is commonly accompanied by language that both reflects implicit ableist attitudes5,6 and inserts negatively-valenced value judgements into prognostic communication7 (eg, “devastating injury,” “terrible prognosis,” or the extreme “fate worse than death”). Although these linguistic qualifiers can represent a clinician’s (or at times, parents’) reflexive emotional response to difficult circumstances, centering them in prognostic communication perpetuates negative disability bias. The bias introduced by a clinician’s choice of words is further complicated by clinician bias toward pessimistic prognostication, especially early in the disease course.8,9Compounding the challenges for neuroprognostication is the absence of relevant outcome data. Much of the longitudinal data guiding neuroprognostication uses clinician- or researcher-defined outcomes lacking family perspectives.10,11 The frequent use of composite outcomes, which often conflate neurodevelopmental impairments with death, has been justifiably criticized.12 These limitations make the terminology used in medical prognostication difficult to relate to caregiver questions.13 Even if clinicians want to predict how Diego’s injury will impact his capacity to delight in play or love his family, empirical data are lacking, and relevant examples often only reflect personal anecdotes.Social and cultural factors can heighten the challenge of discussing disability during neuroprognostication. For example, parental perspectives on disability often differ from clinical definitions: in one study, only 12% of parents of formerly preterm children clinically classified to have severe neurodevelopmental impairment agreed with the classification.14 This propensity toward divergent conceptions of the capacities of children with neurologic impairment is captured in a qualitative study of pediatric ICU providers: “Oh. Holy smokes. I’m completely off base here’… [the] kid is capable of x, y, and z, and [I] didn’t think they could get past a, b, and c.”15Additionally, clinicians and parents frequently have different values and perceptions about quality of life in neurological injury,16–18 reflecting different baselines of conceptualizing disability. This discrepancy, which is closely related to the well-described “disability paradox” (where people with disabilities report significant limitations alongside good quality of life, whereas health care workers and/or society at large perceive worse quality of life than what people with disabilities report19), may exacerbate implicit ableism in prognostication. Moreover, people living with disability can demonstrate a “response shift” in their self-rated quality of life, where personal and social adaptation leads to higher reported quality of life over time.20 Taken together, these social and culture dimensions ensure that the concept of disability is never set in stone (and that talking about disability is rarely straightforward). Instead, the meaning of disability can change over time, and is often be shaped by culturally constructed definitions of health, well-being, beauty, suffering, and quality of life.21–24As a final related point, it is important to note that pediatric neuroprognostication has been criticized for discriminating against intellectual disability. Pediatric bioethics arose in part because of the “Baby Doe” cases, where treatment was withheld from infants with trisomy 21 on claims about poor quality of life.25–27 Recently, the accuracy of other diagnoses historically referred to as “lethal” has been challenged, with calls to liberalize medical and surgical interventions for children with genetic conditions like trisomy 13 and 18.28–30 These shifts in medicine reflect larger cultural trajectories toward disability justice which is a movement and accompanying set of principles that focus on societal transformation beyond disability rights to full societal access and inclusion.31 The framework of disability justice reminds the field of pediatrics that the category “disabled” is just as much a product of [1] inadequate social and cultural supports and [2] negatively biased practices of clinical care, as it is from impairments in a child’s body.32Together, the reliance on a well-child norm, limitations of existing outcome data, and bias around quality of life in neurologic injury combine to form an ableist baseline in prognostication. An ableist baseline assumes that what needs to be predicted and communicated after neurologic injury is the degree of deviation from a neurotypical or “ideal” child. As medicine’s ability to care for critically ill children improves, there is an urgent need for a disability-inclusive approach to neuroprognostication.In contrast to a medical model focused on losses compared with “normal” children, we propose an approach centered on capabilities. A capabilities-based approach (CBA) focuses on describing and cultivating fundamental human capacities.33 A CBA spotlights what people can do, rather than what they cannot. A CBA views children with neurological injuries in light of central capabilities that all humans share: for example, maintaining bodily integrity and homeostasis, relating and connecting to others, delight in movement, and perceiving the world through multiple senses.33A CBA offers a novel way to conceptualize and practice pediatric neuroprognostication. We hypothesize that integrating a capabilities lens into neuroprognostication will result in (1) active debiasing of neuroprognostication, (2) adjustment of language used in prognostic communication, and (3) reworking of assumptions commonly used in the collection of prognostic outcome data.First, clinicians should use prognostication frameworks to actively debias (eg, ouR-HOPE13,34) and reorient prognostication to family values (eg, ALIGN35). Use of functional assessments designed for children with disabilities (eg, Pediatric Evaluation of Disability Inventory [PEDI]) can provide a natural strengths-based foundation for discussing baseline function and skills.36 Clinicians should also develop disability humility37 and strive for disability cultural competence38—an understanding of the behaviors, knowledge, attitudes, and policies that contribute to lived experiences of people with disabilities.Second, a CBA uses language to reframe prognosis. It weaves into the idea of prognosis a conversation about what supports a particular child needs to meaningfully engage with their environment. Table 1 illustrates examples of how clinicians may shift from deficit-oriented to capabilities-based language across multiple domains of neuroprognostication.Third, the assumptions used in data collection around prognostic outcomes must evolve. The mismatch between clinician- and researcher-derived outcomes and the reported experiences of families represents a form of epistemic injustice39—by omitting the testimony of patients and caregivers, disability bias in medicine is reinforced.5,6 For example, a hyperfocus on speech ignores expansive forms of nonverbal communication.40 To enable a CBA in prognostication, research on outcomes of neurologic impairment must both critically examine the types of data studied and collected, as well as incorporate qualitative data from families and patients with disabilities.41,42Integrating a CBA in prognostication offers a holistic avenue for clinicians to share knowledge with patients and families. Similar to calls for use of strengths-based approaches43,44 and appreciative inquiry45 in health care, this broader perspective focuses on children’s capabilities—alongside necessary conversations about medical and functional impairments—to promote a more supportive approach to their care. This integration requires systems-level changes in our training, research agendas, and clinical conversations.We are grateful to have had critical review from Lindsey Topping-Schuetz, mother of Owen, who has years of lived experience as a parent of a child with a complex diagnosis and accompanying disability. We would also like to thank Dr David Magnus and Dr Talia Shear for their critical reviews of the manuscript.