In the hospitals within which I have worked, tensions often exist between advisory and referral models of consultation. The referral approach, with consultants assuming care for some aspects of care, appears to be more frequent in tertiary academic hospitals and where discrete, technical skills (such as the ability to perform endoscopy) are required. Hospitalists tend to prefer consults in the advisory role, when possible, which keeps them in the loop in terms of decision making. Many subspecialists, at least in academic medical centers, seem to prefer directly assuming some aspects of care. For example, most oncologists I know, when consulting in the hospital on a patient with a cancer diagnosis, minimally involve the ward team in decisions regarding treatment options. They may write orders directly in the chart. Even where they do not write orders themselves, they typically instruct the ward team as to what to do. While this makes some sense, given the specialized knowledge and skillset of oncologists, such a directive role may give rise to resentment on the part of hospitalists. Hospitalists will readily admit that they are not experts in oncology (and they have no desire to write chemo orders). However, especially in advanced illness the relevant question is often less what tumor-related treatment to pursue than whether it should be pursued and what alternatives might exist. On this point they tend to believe they have something to contribute. At the extreme, specialty consultants may treat hospitalists “like interns,” whose job is to take care of the “scut work,” reserving the “real work” for themselves. Similarly, many cardiology consultants, when asked their opinion about whether cardiac catheterization is advisable, will go ahead and schedule catheterization, if thought appropriate, without discussing this first with the ward team.

My experience is that neither model is ideal for palliative care consults; a good consult requires a delicate blend of approaches. For example, ward teams may wish for us to take over certain tasks, such as talking with patients and families about care options, but would be offended if we discussed certain medical treatment options with which they might not feel comfortable. For example, I have recommended dexamethasone for a patient with pain due to a compression neuropathy. However, the primary care team, worried about immunosuppressive side effects, did not wish to take this recommendation. Had I prematurely mentioned the medication to the patient, this would have put the primary care team in an awkward position.

Modeling certain skills to care team members, such as assessing goals of care, can be an efficient and effective way of teaching skills. Of course, this requires coordination in the timing of meetings with patients and families, and it is time consuming for the observing team members. When possible, however, this is a superb way to help other clinicians understand what we do, and skill development is promoted. Where not possible, try to summarize what you did in your consult, not just the result. This may provide some indirect modeling. Explaining the “why” of your recommendations also promotes better understanding. Less tangibly, if other clinicians can witness some success from your interventions, this may promote an interest on their part in learning how you “did your magic.”

The greater the overlap between palliative care and other specialties, the stronger yet less obvious this resistance can become. For example, geriatricians, who overlap greatly with palliative care in their work with chronically ill and dying elders, may overtly view palliative care practitioners as allies in a cause serving such patients (as we believe we are). However, they may be slow to recognize the need for palliative care services for geriatric patients distinct from the care they provide, thinking that they ought to be able to provide such care for elders based on their credentialing, even if their training was deficient in specific palliative care skills. Similar resistance can be found among pain management specialists, who may view palliative care consults as competition. Early in our process of establishing a consult team, we found, somewhat to our surprise, that some hospitalists became quite irate at finding us talking with their patients. One time, an attending physician asked us why we were discussing care options with his patient. We informed him that his resident team had requested this on the consult. “Well, I was not informed of this!” he said angrily. Although one would think this was a problem of team miscommunication, I doubt the same attending physician would have been offended to find an infectious disease consultant at the bedside of a patient with an exotic infection, as he would readily acknowledge the specialist’s expertise. At least in our hospital, residents often request consults without first informing the attending physician. So what was going on? I think these very good physicians felt that in discussing certain things such as goals of care and preferences we were infringing on their areas of expertise (working with the whole patient) and thus felt an affront. We have since built in the step of ensuring that the attending physician is aware of the consult before we perform it.

Reflect for a moment on the prior palliative care note: “The unit of care is the patient, the family, and the care team.” Of course, within this group, the patient is the star of the show, but sometimes he or she is doing just fine—at least in terms of suffering. It may be family members or perhaps members of the care team who are suffering the most. Sometimes, we become the “patient,” if we get especially stressed out or a particular case really gets to us. Most often, there is more than one patient, one person, who is in need of attention and healing.