Development of a Test of Residents' Ethics Knowledge for Pediatrics (TREK-P)

Abstract

Background Professionalism is one of the Accreditation Council for Graduate Medical Education's core competencies. Residency programs must teach residents about ethical principles, which is an essential component of professionalism. Objectives We aimed to formally develop a valid and reliable test of ethics knowledge that effectively discriminated among learners in pediatric residency training and to improve methods for measuring outcomes of resident education in medical ethics. Methods We created an instrument with 36 true/false questions that tested knowledge in several domains of pediatric ethics: professionalism, adolescent medicine, genetic testing and diagnosis, neonatology, end-of-life decisions, and decision making for minors. All questions and their correct answers were derived from published statements from the American Academy of Pediatrics Committee on Bioethics. We invited a range of participants from novices to experts to complete the test. We evaluated the instrument's reliability and explored item discrimination, omitting 13 items with the least discriminatory power. Score differences between the 3 categories of examinees were evaluated. Results The 23-item test, completed by 54 participants, demonstrated good internal reliability (Kuder-Richardson 20 statistic = 0.73). The test was moderately difficult and had a mean overall score of 17.3 (±3.3 standard deviation). Performance appropriately improved with degree of expertise: median scores for medical students, postgraduate year-3 residents, and ethicists were 15 (65%, range, 11–19), 19 (83%, range, 14–23), and 22 (96%, range, 20–23), respectively. Ethicists' scores were significantly higher than those of medical students (P < .001) and residents (P = .007). Moreover, residents performed significantly better than medical students (P = .001). Conclusions We developed a standardized instrument, entitled Test of Residents' Ethics Knowledge for Pediatrics (TREK-P), to evaluate residents' knowledge of pediatric ethics. The TREK-P is easy to administer, reliably discriminates among learners, and highlights content areas in which knowledge may be deficient.

Knowledge of Pediatric Ethics: Results of a Survey of Pediatric Ethics Consultants

“…Forty respondents (40%) scored 20 or higher, thereby achieving the level of high performance reported in previous work (Kesselheim et al 2012). Post hoc analysis of this high-performing subgroup revealed that scoring 20 or above on the TREK-P was more common (55% vs. 33%, p D 0.04) among ECs who received salary support than among ECs without salary support and among ECs with, as compared to those without, formal training (47% vs. 22%, p D 0.02).…”

“…Because of a skew in the distribution of TREK-P scores in this sample, the nonparametric Wilcoxon rank sum test (Kruskal-Wallis test with >2 comparison groups) was used to compare the distribution of TREK-P scores between the following groups: (1) CHA hospitals versus COTH hospitals; (2) formal training versus informal training; (3) high volume (>10 per year) versus low volume (10 or fewer per year) consult services; (4) budget; (5) salary support; and (6) geographic region. Post hoc analyses compared these same groups for the difference in proportion having TREK-P scores 20, a marker of particularly high performance identified in earlier work (Kesselheim et al 2012). …”

“…Prior experience with the TREK-P also revealed low scores on items related to genetic testing and diagnosis (Kesselheim et al 2012). Despite published guidance from, among other professional groups, the American Academy of Pediatrics (2013) and the American College of Medical Genetics and Genomics, this domain of medical ethics remains challenging even for pediatric ECs.…”

“…The Test of Residents' Ethics Knowledge for Pediatrics (TREK-P) is a valid and reliable test to measure knowledge about principles of pediatric ethics (Kesselheim, McMahon, and Joffe 2012). This scale is comprised of questions addressing common ethical dilemmas in adolescent medicine, genetic testing and diagnosis, neonatology, end-oflife decisions, and decision making for minors.…”

“…Thirteen items with low discriminatory power were removed through item reduction, yielding a 23-item test with good internal reliability (Kuder Richardson Formula 20 [KR-20] D 0.73). Performance significantly improved with degree of expertise, and a score below 20 of 23 items correct (87%) was proposed as a minimal score threshold to identify residents who could benefit from additional educational efforts (Kesselheim, McMahon, and Joffe 2012).…”