Residents as Research Subjects: Balancing Resident Education and Contribution to Advancing Educational Innovations.

Background: Research in education advances knowledge and improves learning, but the literature does not define how to protect residents' rights as subjects in studies or how to limit the impact of their participation on their clinical training. Objective: We aimed to develop a consensual framework on how to include residents as participants in education research, with the dual goal of protecting their rights and promoting their contributions to research. Methods: A nominal group technique approach was used to structure 3 iterative meetings held with the pre-existing residency training program committee and 7 invited experts between September 2018 and April 2019. Thematic text analysis was conducted to prepare a final report, including recommendations. Results: Five themes, each with recommendations, were identified: (1) Freedom of participation: participation, non-participation, or withdrawal from a study should not interfere with teacher-learner relationship (recommendation: improve recruitment and consent forms); (2) Avoidance of over-solicitation (recommendation: limit the number of ongoing studies); (3) Management of time dedicated to participation in research (recommendations: schedule and proportion of time for study participation); (4) Emotional safety (recommendation: requirement for debriefing and confidential counseling); and (5) Educational safety: data collected during a study should not influence clinical assessment of the resident (recommendation: principal investigator should not be involved in the evaluation process of learners in clinical rotation). Conclusions: Our nominal group technique approach resulted in raising 5 specific issues about freedom of participation of residents in research in medical education, over-solicitation, time dedicated to research, emotional safety, and educational safety.

Provider Perspectives Regarding Resuscitation Decisions for Neonates and Other Vulnerable Patients.

OBJECTIVES: To use structured surveys to assess the perspectives of pediatric residents and neonatal nurses on resuscitation decisions for vulnerable patients, including neonates. STUDY DESIGN: Pediatric providers were surveyed using scenarios for 6 critically ill patients of different ages with outcomes explicitly described. Providers were asked (1) whether resuscitation was in each patient's best interest; (2) whether they would accept families' wishes for comfort care (no resuscitation); and (3) to rank patients in order of priority for resuscitation. In a structured interview, each participant explained how they evaluated patient interests and when applicable, why their answers differed for neonates. Interviews were audiotaped; transcripts were analyzed using thematic analysis and mixed methods. RESULTS: Eighty pediatric residents and neonatal nurses participated (response rate 74%). When making life and death decisions, participants considered (1) patient characteristics (96%), (2) personal experience/biases (85%), (3) family's wishes and desires (81%), (4) disease characteristics (74%), and (5) societal perspectives (36%). These factors were not in favor of sick neonates: of the participants, 85% reported having negative biases toward neonates and 60% did not read, misinterpreted, and/or distrusted neonatal outcome statistics. Additional factors used to justify comfort care for neonates included limited personhood and lack of relationships/attachment (73%); prioritization of family's best interest, and social acceptability of death (36%). When these preconceptions were discussed, 70% of respondents reported they would change their answers in favor of neonates. CONCLUSIONS: Resuscitation decisions for neonates are based on many factors, such as considerations of personhood and family's interests (that are not traditional indicators of benefit), which may explain why decision making is different for the neonatal population.

Changes in perinatal hospital deaths occurring outside the neonatal intensive care unit over a decade.

AIM: Perinatal deaths occurring outside the neonatal intensive care unit (NICU) are rarely recorded in outcome studies, despite having a direct impact on perinatal statistics. Our aim was to investigate the timing and modes of perinatal deaths that occurred outside the NICU and changes over time. METHOD: We reviewed all perinatal deaths from 22 weeks of gestation onwards, without NICU admissions, during two periods in a Canadian tertiary mother and baby hospital and categorised deaths according to nine specific categories. RESULTS: There were 444 perinatal deaths that satisfied the inclusion criteria. The total number of perinatal deaths increased from 2000 to 2002 (n = 197) and 2007 to 2010 (n = 247). The proportion of foetuses alive at the time of their mother's hospital admission, but then stillborn, decreased. There was a significant increase in terminations for congenital anomalies in the second cohort and a decrease in deaths following induction of labour and comfort care for foetal anomalies. CONCLUSION: Approaches to end-of-life care changed between the two study periods. Paediatricians should be aware of the epidemiology of perinatal mortality in their own practice, as it has a direct impact on the denominator in NICU outcome studies.

Later rather than sooner: the impact of clinical management on timing and modes of death in the last decade.

AIM: Technological and clinical advances have reduced neonatal deaths, and this study explored how the mode and timing of neonatal deaths has changed in a tertiary neonatal intensive care unit (NICU) over 10 years. METHODS: We carried out a retrospective chart review on NICU deaths in 2000-2002 and 2007-2010, categorising deaths and compared the timing, cause and mode of death in the two cohorts. RESULTS: We analysed 204 neonatal deaths and found that the average age at death doubled from 9.71 days in 2000-2002 to 18.8 days (p = 0.014) in 2007-2010 and that the number of deaths in the first 48 h of life fell from 52% to 29% (p < 0.001). Mode of death and ethical decision-making was similar. In both cohorts, 26% of patients who died had a do not resuscitate order and 9% of the 151 patients without an order died while receiving cardiopulmonary resuscitation. Most neonates received medication to keep them comfortable and their use was similar in both cohorts. CONCLUSION: Changes in neonatal management have led to a reduction in early deaths and an increase in age at death. The impact of later deaths on families and healthcare providers deserves further research.

End-of-life decisions for extremely low-gestational-age infants: why simple rules for complicated decisions should be avoided.

Interventions for extremely preterm infants bring up many ethical questions. Guidelines for intervention in the "periviable" period generally divide infants using predefined categories, such as "futile," "beneficial," and "gray zone" based on completed 7-day periods of gestation; however, such definitions often differ among countries. The ethical justification for using gestational age as the determination of the category boundaries is rarely discussed. Rational criteria used to make decisions regarding life-sustaining interventions must incorporate other important prognostic information. Precise guidelines based on imprecise data are not rational. Gestational age-based guidelines include an implicit judgment of what is deemed to be an unacceptably poor chance of "intact" survival but fail to explore the determination of acceptability. Furthermore, unclear definitions of severe disability, the difficulty, or impossibility, of accurately predicting outcome in the prenatal or immediate postnatal period make such simplistic formulae inappropriate. Similarly, if guidelines for intervention for the newborn are based on the "qualitative futility" of survival, it should be explicitly stated and justified according to established ethical guidelines. They should discuss whether newborn infants are morally different to older individuals or explain why thresholds recommended for intervention are different to recommendations for those in older persons. The aim should be to establish individualized goals of care with families while recognizing uncertainty, rather than acting on labels derived from gestational age categories alone.