In his 1889 valedictory address to the University of Pennsylvania (later published as “Aequanimatas”), William Osler famously celebrated imperturbability as the essential element of a physician’s character. In defining this core trait, Osler set a high bar, describing imperturbability as “coolness and presence of mind under all circumstances, calmness amid storm, clearness of judgment in moments of grave peril….” But while acknowledging that achieving equanimity could be elusive, and perhaps beyond the reach of many, he also outlined a plan by which the average physician might move asymptotically toward this goal. That plan involved the steady acquisition of biomedical knowledge and experience:

In a true and perfect form, imperturbability is indissolubly associated with wide experience and an intimate knowledge of the varied aspects of disease. With such advantages he [sic] is so equipped that no eventuality can disturb the mental equilibrium of the physician; the possibilities are always manifest, and the course of action clear.

More than 125 years later, most trainees would likely be pleased to be described as “imperturbable” on their clinical evaluations. Yet the path to equanimity in 2016 may have less to do with the acquisition and retention of scientific knowledge (though that remains important) and more to do with the ability to steer patients toward goals consistent with their values even in the face of clinical uncertainty. For the modern physician, tolerance for ambiguity has become not just a core element of character but a fundamental clinical skill.

There is perhaps no area of medicine more fraught with ambiguity than end-of-life care. Part of the challenge stems from difficulty defining fundamental terms. Patients’ wishes are determinative, but when and how should they be elicited? Physician aid-in-dying may be a slippery slope, but when does compassionate care glissade into euthanasia? Futile care should be avoided, but how do we know with sufficient certainty that an intervention is futile? In this issue of JGIM, Dzeng and colleagues take on the issue of medical futility from the perspective of the young physicians persuaded (or coerced) to deliver it.1 In interviews with 22 residents and fellows training in three institutions, the authors found a striking degree of moral distress among trainees who felt they had been forced to participate in futile care. These findings raise two further questions. The first—how do we support trainees in coping with the negative emotions that can arise from participating in futile care?—is well addressed in the article. The second—why do our laws and institutions continue to foster the provision of care that most physicians would be loath to accept for themselves?—is a deeper question that should provide grist for future research and debate.

Ideally, medical journals should disseminate reliable evidence and thus contribute to a net reduction in the amount of ambiguity present in the universe. While JGIM aspires to this ideal, we sometimes find ourselves in the uncomfortable position of publishing evidence that is seemingly contradictory. This is one of those times. In this issue, Lee et al. report results from an Internet-based survey of retail pharmacy users.2 A surprisingly high percentage of respondents (18 %) reported contacting their physician by Facebook during the past 6 months, despite institutional policies that discourage use of social media for clinical communication. In contrast, the study by Jenssen et al. found that only 3 % of their nationally representative sample of adults were willing to discuss health goals by Facebook.3 (It seems reasonable to infer that even fewer were actually using Facebook for this purpose.) These disparate results may result from differences in sampling frames, survey methods, and response rates—and support two conclusions, one clinical and one scientific. The clinical conclusion is that health systems should continue to implement secure web-based portals that allow physicians and patients to communicate beyond the clinic walls. If these systems can be safely linked to popular social media sites such as Facebook, so much the better. A corollary is that insurers must devise ways to pay for physicians’ time as they utilize these new approaches. The scientific conclusion is that what seem like arcane differences in survey methodology can lead to huge discrepancies in results.

Finally, by raising questions about the value of specialty care, a final article injects ambiguity into the age-old reverence for expertise. Ricardo et al. report that among patients with mild-moderate renal insufficiency, prior nephrology specialty care was associated with more frequent treatment of chronic kidney disease complications and use of medications affecting the renin-angiotensin axis but not with long-term outcomes.4 This study should not be taken to invalidate Osler’s adulation for “an intimate knowledge of the varied aspects of disease.” A sturdy body of evidence in fact supports the premise that specialists (including both nephrologists and psychiatrists) are more likely than generalists to deliver guideline-concordant care on a condition-by-condition basis. The article likely signals something more subtle. Nephrologists no doubt contribute much to the care of patients with kidney disease, but their evidence-based armamentarium, even when fully deployed, is likely weaker (in terms of producing observable benefits through established process-outcomes links) than we would like. Furthermore, as Ricardo et al. speculate, well-supported generalists, such as those working within academic centers and multispecialty groups, may frequently deliver quality of care approaching levels achieved by their subspecialty colleagues. Similarly, psychiatrists contribute essential expertise to the care of patients with more severe forms of mental illness. However, artificial divisions between the medical and mental health sectors (including separate medical records, different standards for confidentiality, and mental health “carve outs”) tend to muddle communication, raising the probability of medical error.

About 2300 years before Osler, Hippocrates is said to have described the physician’s task in this way: “Life is short, and art long; the crisis fleeting; experience perilous, and decision difficult.” While it is admittedly difficult to remain imperturbable in the face of these immutable challenges, it is our duty—and privilege—to try.