Training Pediatric Residents in Evidence-Based Ethics: An Exploratory Clinical Trial ♦ 161

Abstract

Background: There is an increasing need to train physicians to make sophisticated ethical decisions based on a critical assessment of patient prognosis. Physicians often have an overly pessimistic view of the outcome of premature infants and a limited ability to evaluate neonatal and follow-up studies. Purpose: We assessed whether a brief educational intervention would increase housestaff knowledge, ethical sophistication, and epidemiologic skills needed to address treatment decisions for marginally viable newborns. Hypotheses: A brief educational intervention for pediatric housestaff prior to their neonatal intensive care unit (NICU) rotation will: a) augment their knowledge base as assessed by the number of housestaff who substantially overestimate rates of death and cerebral palsy among extremely premature infants; b) augment the ethical sophistication of housestaff as assessed by the number of specific ethical issues considered in recommending whether to provide intensive care to marginally viable infants; and c) augment the epidemiologic skills of housestaff as assessed by the methodologic criteria used to evaluate the quality of follow-up studies. Methods: A clinical trial involving 64 pediatric housestaff was performed with alternating control and intervention housestaff rotations in the NICU. A 45-60 min discussion involving neonatal case studies and literature review was provided to intervention housestaff before their Parkland NICU rotation. Intervention and control groups were evaluated at the end of their rotation, using an 8-item questionnaire scored by two faculty members blinded to treatment group. Results: Intervention housestaff were less likely than controls to substantially overestimate mortality (23% vs 55%) and morbidity (46% vs 74%) for extremely premature infants in our center;(p<0.05). Intervention housestaff tended to cite more methodologic criteria to evaluate follow-up studies than did controls; (p=0.052). The intervention did not increase the range of ethical criteria considered in treatment recommendations for marginally viable infants. Cerebral palsy rates (as compared to those reported in the methodologically strongest follow-up studies) were substantially overestimated by a high percentage of both intervention and control housestaff (70% vs 87%); (p=0.123). Conclusions: Our intervention had limited success. A more intensive intervention may be required to have a large impact on housestaff skills in addressing difficult ethical issues for marginally viable infants.