Timeliness of End-of-Life Discussions for Blood Cancers

dents and students can derive an ambivalent approach to value. Particularly in academic centers where resources seem limitless, attending physicians often suggest additional tests that expand exhaustive differential diagnoses. This builds on trainees’ gnawing fear of missing something to establish thoroughness as an unrestrained virtue. The pressures of individualized quality metrics along with compressed inpatient evaluations to minimize length of stay may further reinforce this mindset and encourage defensive testing. These practices contradict and displace discussions of value. The research letter by Pierce et al3 in this issue of JAMA Internal Medicine underscores the unrealized potential of rounds to teach high value. Pierce et al observed 168 patient encounters on internal medicine rounds and recorded how often an attending physician invoked the American College of Physicians’ test-ordering appropriateness criteria.3 Attending physicians only discussed appropriate test ordering during 35 (20.8%) of 168 patient encounters. They most often brought up whether a test affected care (23 encounters [13.7%]) and rarely discussed whether a test would cause harm (4 encounters [2.4%]). There is no standard on how often attending physicians should discuss value with trainees, and conversations that residents initiated about value were not captured in the study. However, the findings are in line with surveys of program directors, medical students, and residents who all noted missed opportunities for faculty to model these skills.4 How might educators change the culture of rounds from rewarding meticulousness to celebrating moderation? Just as we challenge trainees to explain their diagnostic reasoning, so too should we probe them to justify their use of tests and resources. Attending physicians should model the confidence needed to enact a stepwise diagnostic workup. In addition to using evidencebased guidelines, we should reemphasize history and physical examination findings to refine and guide diagnostic workup. Finally, we should study such interventions to determine whether they affect practices as well as attitudes.