The Ladder of Inference as a tool to reduce implicit bias in pediatric clinical practice.

Implicit bias in healthcare professionals is a widespread phenomenon that leads to worse healthcare outcomes for marginalized patient populations. One tool that can help providers identify when biases are impacting the clinical care they are providing and enable them to take corrective action in real time is the "Ladder of Inference" (LOI). The LOI is an instrument that elucidates the process by which we take in information about another person, filter that data through our own interests, needs, perspectives and biases, and then use it to draw conclusions about the individual. These conclusions are often profoundly inaccurate, yet we then act upon them. Thus, we propose the LOI as an "implicit bias detection tool" for neonatal intensive care unit (NICU) clinical practice. We demonstrate on two common NICU cases how utilizing the LOI can uncover mechanisms by which positive and negative feedback loops secondary to unregulated implicit bias lead to a stepwise increase or decrease in the quality of care. As the cases demonstrate, the subtle differences in individual steps up the ladder can lead to care differences of a large magnitude in either direction, hugely positive or detrimentally negative. This shift in the quality of care, then, may contribute to the significant neonatal outcome disparities in infants from minoritized groups. Using the LOI as a practical tool, we demonstrate how it becomes possible to detect one's own implicit biases and thus to consciously monitor the inferences we are making about patients and their families in order to counteract them.

Provider Perceptions on Bereavement Following Newborn Death: A Qualitative Study from Ethiopia and Ghana.

OBJECTIVE: The objective of this study was to explore how clinicians in low- and middle-income countries engage and support parents following newborn death. STUDY DESIGN: Qualitative interviews of 40 neonatal clinicians with diverse training were conducted in Addis Ababa, Ethiopia, and Kumasi, Ghana. Transcribed interviews were analyzed and coded through the constant comparative method. RESULTS: Three discrete themes around bereavement communication emerged. (1) Concern for the degree of grief experienced by mothers and apprehension to further contribute to it. This led to modified communication to shield her from emotional trauma. (2) Acknowledgment of cultural factors impacting neonatal loss. Clinicians reported that loss of a newborn is viewed differently than loss of an older child and is associated with a diminished degree of public grief; however, despite cultural expectations dictating private grief, interview subjects noted that mothers do suffer emotional pain when a newborn dies. (3) Barriers impeding communication and psychosocial support for families, often relating to language differences and resource limitations. CONCLUSIONS: Neonatal mortality remains the leading global cause of mortality under age 5, with the majority of these deaths occurring in low- and middle-income countries, yet scant literature exists on approaches to communication around end-of-life and bereavement care for neonates in these settings. We found that medical providers in Ghana and Ethiopia described structural and cultural challenges that they navigate following the death of a newborn when communicating and supporting bereaved parents.

Perspectives on Resuscitation Decisions at the Margin of Viability among Specialist Newborn Care Providers in Ghana and Ethiopia: A Qualitative Analysis.

BACKGROUND: In high income countries, guidelines exist recommending gestational age thresholds for offering and obligating neonatal resuscitation for extremely preterm infants. In low- and middle- income countries, this approach may be impractical due to limited/inconsistent resource availability and challenges in gestational dating. Scant literature exists on how clinicians in these settings conceptualize viability or make resuscitation decisions for premature infants. METHODS: Qualitative interviews of interprofessional neonatal clinicians were conducted in Kumasi, Ghana, at Komfo Anokye Teaching Hospital and Suntreso Government Hospital, and in Addis Ababa, Ethiopia, at St. Paul's Hospital Millennium Medical College. Transcribed interviews were coded through the constant comparative method. RESULTS: Three discrete major themes were identified. The principal theme was a respect for all life, regardless of the likelihood for survival. This sense of duty arose from a duty to God, a duty to the patient, and a duty intrinsic to one's role as a medical provider. The duty to resuscitate was balanced by the second major theme, an acceptance of futility for many premature infants. Lack of resources, inappropriate staffing, and historically high local neonatal mortality rates were often described. The third theme was a desire to meet global standards of newborn care, including having resources to adopt the 22-25-week thresholds used in high income countries and being able to consistently provide life-saving measures to premature infants. CONCLUSIONS: Neonatal clinicians in Ghana and Ethiopia described respect for all life and desire to meet global standards of newborn care, balanced with an awareness of futility based on local resource limitations. In both countries, clinicians highlighted how wide variations in regional survival outcomes limited their ability to rely on structured resuscitation guidelines based on gestational age and/or birthweight.

Adapting user-centered design principles to improve communication of peer parent narratives on pediatric tracheostomy.

BACKGROUND: Parents who have to make tracheostomy decisions for their critically ill child may face forecasting errors and wish to learn from peer parents. We sought to develop an intervention with peer parent narratives to help parents anticipate and prepare for future challenges before making a decision. METHODS: To ensure that the intervention reflects parents' needs (rather than experts' opinions), we adapted a user-centered design (UCD) process to identify decision-critical information and refine the presentation format by interviewing parents who had tracheostomy decision making experience. Phase 1 (n = 10) presented 15 possible forecasting errors and asked participants to prioritize and justify the problematic ones. It also asked participants to comment on the draft narratives and preferred delivery mode and time of the intervention. Phase 2 (n = 9 additional parents and 1 previous parent) iteratively collected feedback over four waves of user interviews to guide revisions to the informational booklet. RESULTS: Phase 1 revealed that parents wanted information to address all forecasting errors as soon as tracheostomy becomes an option. They also highlighted diverse family situations and the importance of offering management strategies. The resulting prototype booklet contained five sections: introduction, child's quality of life, home care, practical challenges, and resources. Feedback from Phase 2 focused on emphasizing individualized situations, personal choice, seriousness of the decision, and caregiver health as well as presenting concrete illustrations of future challenges with acknowledgement of positive outcomes and advice. We also learned that parents preferred to use the booklet with support from the care team rather than read it alone. CONCLUSIONS: A UCD process enabled inclusion of parental perspectives that were initially overlooked and tailoring of the intervention to meet parental expectations. Similar UCD-based approaches may be valuable in the design of other types of patient communications (e.g., decision aids).

Provider perspectives on Asram in Ghana.

Neonatal mortality is one of the leading causes of under-five mortality globally, with the majority of these deaths occurring in low- and middle-income countries. In Ghana, there is a belief in an array of newborn conditions, called Asram, that are thought to have a spiritual, rather than physical, cause. These conditions are predominantly managed by traditional healers as they are considered unable to be treated by allopathic medical providers. Through a series of semi-structured qualitative interviews of medical providers in Kumasi, Ghana, conducted in July-August 2018, this study sought to elucidate perspectives of allopathic medical providers about Asram, including the perceived implications of traditional newborn care patterns on newborn health and higher-level neonatal care. Twenty health care providers participated and represented a tertiary care hospital and a district hospital. Medical providers were universally aware of Asram but varied on the latitude they gave this belief system within the arena of newborn care. Some providers rationalized the existence of Asram in the backdrop of high neonatal mortality rates and long-standing belief systems. Others highlighted their frustration with Asram, citing delays in care and complications due to traditional medical treatments. Providers utilized varying approaches to bridge culture gaps with families in their care and emphasized the importance of open communication with the shared goal of improved newborn health and survival. This study describes the importance of providers being aware of socio-cultural constructs within which pregnant women operate and suggests a focus on the shared goal of timely and effective newborn care in Ghana.

Evaluating the Use of a Decision Aid for Parents Facing Extremely Premature Delivery: A Randomized Trial.

OBJECTIVE: To assess decisional conflict and knowledge about prematurity among mothers facing extreme premature delivery when the counseling clinicians were randomized to counsel using a validated decision aid compared with usual counseling. STUDY DESIGN: In this randomized trial, clinicians at 5 level III neonatal intensive care units in the US were randomized to supplement counseling using the decision aid or to counsel mothers in their usual manner. We enrolled mothers with threatened premature delivery at 220/7 to 256/7 weeks of gestation within 7 days of their counseling. The primary outcome was the Decisional Conflict Scale (DCS) score. One hundred mothers per group were enrolled to detect a clinically relevant effect size of 0.4 in the Decisional Conflict Scale. Secondary outcomes included knowledge about prematurity; scores on the Preparedness for Decision Making scale; and acceptability. RESULTS: Ninety-two clinicians were randomized and 316 mothers were counseled. Of these, 201 (64%) mothers were enrolled. The median gestational age was 24.1 weeks (IQR 23.7-24.9). In both groups, DCS scores were low (16.3 ± 18.2 vs 16.8 ± 17, P = .97) and Preparedness for Decision Making scores were high (73.4 ± 28.3 vs 70.5 ± 31.1, P = .33). There was a significantly greater knowledge score in the decision aid group (66.2 ± 18.5 vs 57.2 ± 18.8, P = .005). Most clinicians and parents found the decision aid useful. CONCLUSIONS: For parents facing extremely premature delivery, use of a decision aid did not impact maternal decisional conflict, but it significantly improved knowledge of complex information. A structured decision aid may improve comprehension of complex information. TRIAL REGISTRATION: Clinicaltrials.gov: NCT01713894.

Thematic Analysis of Interprofessional Provider Perceptions of Pediatric Death.

PURPOSE: Though provider and patient perceptions of death are characterized in the adult population literature, there is limited information related to providers' perceptions in pediatric and neonatal patients. The purpose of this study was to better understand how interprofessional care team members perceive and experience neonatal and pediatric end-of-life situations. DESIGN AND METHODS: This survey questionnaire was administered to interprofessional providers following their participation in an institutional workshop, as part of an ongoing institutional effort to improve end-of-life experiences for patients/family and providers. Interprofessional care providers completed an electronic survey consisting of closed-ended and one open-ended question to elicit their perceptions of their participation in end of life care for a recent neonatal/pediatric patient in the period before the child's death. RESULTS: The qualitative analysis of 306 free-text responses commenting on the deaths of 138 patients, contained within 880 completed mixed-method surveys, is described. Thematic analysis of the free text discovered three primary themes from the data: favorable aspects of the death experience, unfavorable aspects of the experience, and combined favorable and unfavorable aspects. Four subthemes contributed to the themes; namely, language, parental presence, trust/rapport in provider relationships and inclusion in decision-making, communication, and culture. CONCLUSIONS: Multiple factors contribute to how interprofessional care providers perceive end-of-life care experiences for neonatal/pediatric patients. The same death may be perceived differently by different providers. PRACTICE IMPLICATIONS: Understanding favorable and unfavorable aspects of providing end-of-life care will support strategies to provide resources, education and support to facilitate coping and resiliency in care providers.

Resident Attitudes on Ethical and Medical Decision-Making for Neonates at the Limit of Viability.

OBJECTIVE: This study aims to identify pediatric resident knowledge and attitudes on current practices and ideal gestational age (GA) thresholds for offering and mandating resuscitation, and the role of influencing factors in decision-making. STUDY DESIGN: Pediatric residents were assessed via electronic survey at a large academic institution. RESULT: A total of 62% of the residents identified 23 weeks as the lower threshold for resuscitation, despite 84 and 89% reporting that practices are inconsistent and unclear, respectively. Only 21% identified 24 weeks as the latest GA that parents may decline. The majority disagreed with our current practices, identifying older GA as appropriate for all thresholds. They reported scientific evidence as undervalued, and attending physicians' personal beliefs as overvalued in decision-making. CONCLUSION: Our residents recognize decision-making for extremely preterm infants consistent with general guidelines for management based on population outcomes, but attribute these decisions to physicians' personal beliefs. Preferences for higher GA thresholds for resuscitation may reflect disproportionate pessimism about these patients or diverse values regarding autonomy.

Equity, inclusion and cultural humility: contemporizing the neonatal intensive care unit family-centered care model.

Existing NICU family centered care models lack the key elements of equity, inclusion and cultural humility. These models were conceived to support families during the stressful life event of an infant's NICU admission. Their development, however, occurred prior to recognition of the medical field's systematic shortcomings in providing equitable care and their impact on outcome disparities for marginalized communities; thus, they do not include cultural or equitable healthcare considerations. Given the significant neonatal care inequities for marginalized groups, incorporating the experience of these patients in a targeted manner into family centered care frameworks is of critical importance to ensure culturally humble and thus more just and equitable treatment. Here, we review past approaches to NICU family centered care and propose a novel, updated framework which integrates culturally humble care into the NICU family centered care framework.

Ethical challenges in first-in-human trials of the artificial placenta and artificial womb: not all technologies are created equally, ethically.

Artificial placenta and artificial womb technologies to support extremely premature neonates are advancing toward clinical testing in humans. Currently, no recommendations exist comparing these approaches to guide study design and optimal enrollment eligibility adhering to principles of research ethics. In this paper, we will explore how scientific differences between the artificial placenta and artificial womb approaches create unique ethical challenges to designing first-in-human trials of safety and provide recommendations to guide ethical study design for initial human translation.

Perinatal palliative care in sub-Saharan Africa: recommendations for practice, future research, and guideline development.

Worldwide, sub-Saharan Africa has the highest burden of global neonatal mortality (43%) and neonatal mortality rate (NMR): 27 deaths per 1,000 live births. The WHO recognizes palliative care (PC) as an integral, yet underutilized, component of perinatal care for pregnancies at risk of stillbirth or early neonatal death, and for neonates with severe prematurity, birth trauma or congenital anomalies. Despite bearing a disproportionate burden of neonatal mortality, many strategies to care for dying newborns and support their families employed in high-income countries (HICs) are not available in low-and-middle-income countries (LMICs). Many institutions and professional societies in LMICs lack guidelines or recommendations to standardize care, and existing guidelines may have limited adherence due to lack of space, equipment, supplies, trained professionals, and high patient load. In this narrative review, we compare perinatal/neonatal PC in HICs and LMICs in sub-Saharan Africa to identify key areas for future, research-informed, interventions that might be tailored to the local sociocultural contexts and propose actionable recommendations for these resource-deprived environments that may support clinical care and inform future professional guideline development.

Risk of death at home or on hospital readmission after discharge with pediatric tracheostomy.

OBJECTIVE: To evaluate outcomes of patients discharged home following tracheostomy, including the timing and place of death for non-survivors. STUDY DESIGN: We retrospectively reviewed medical records of infants undergoing tracheostomy between 2006 and 2017, within the first year of life for congenital or acquired neonatal conditions. RESULTS: Of the 224 patients discharged after tracheostomy, 127 (57%) required home mechanical ventilation (MV). Overall, 40 (18%) patients died (65% were on MV); 38% of the deaths occurred at home and 63% at a subsequent hospitalization. Having tube feeding was identified as significantly associated with increased mortality on multivariate analysis. Having a tracheostomy for upper airway obstruction was the only variable significantly associated with increased risk of death at home on multivariate analysis. CONCLUSIONS: Having tube feeding was associated with increased risk of death overall and having the tracheostomy for obstructive airway conditions was associated with death occurring at home.

Resuscitation decisions in fetal myelomeningocele repair should center on parents' values: a counter analysis.

In our response to, "Parental request for non-resuscitation in fetal myelomeningocele repair: an analysis of the novel ethical tensions in fetal intervention" by Wolfe and co-authors, we argue that parental authority should guide resuscitation decision-making for a fetus at risk for preterm delivery as a complication of fetal myelomeningocele (fMMC) repair. Due to the elevated morbidity and mortality risks of combined myelomeningocele, extreme prematurity, and fetal hypoxia, parents' values regarding the acceptability of possible outcomes should be elicited and their preferences honored. Ethical decision-making in these situations must also consider the broader context of the fetal-maternal dyad. Innovations in fetoscopic approaches to fMMC repair may pose additional complexity to these resuscitation decisions.

Validity and Reliability of the Modified Attitudes About Drug Use in Pregnancy Scale.

OBJECTIVE: To contemporize the Attitudes About Drug Abuse in Pregnancy questionnaire, keep the length of the modified scale brief to promote use, and test the psychometric properties of the modified scale among perinatal nurses. DESIGN: Cross-sectional survey. SETTING: Four hospitals in the Midwestern United States. PARTICIPANTS: Registered nurses who worked in perinatal units (N = 440). METHODS: We collected data from participants using survey methods. Seven experts in perinatal substance use research and clinical care informed scale modifications. We used a split-sample design involving maternal-newborn units (labor, postpartum) and newborn-focused units (NICU, pediatrics). We evaluated construct validity using factor analysis and reliability using Cronbach's alpha. We tested for differences between units using analysis of variance and Tukey's post hoc honest significant difference test of pairwise differences. RESULTS: The final modified scale included 13 items that loaded on one factor and showed internal consistency reliability in both samples (α = .88-.91). We found a statistically significant difference in mean score between NICU and pediatric units; however, the absolute difference was small and likely not clinically significant. CONCLUSIONS: The Modified Attitudes About Drug Use in Pregnancy scale has initial evidence for validity and reliability, was updated to reflect current terminology in the field, and is a pragmatic tool for use in research.