John Fox, MD: Several years ago, we sponsored next-generation sequencing testing for patients who had advanced stage cancer, and the requirement for participation in this was that patients go through an advanced care planning process with a trained psychologist or social worker beforehand. Of the first 24 patients, 5 of them said, “You know what? Not only do I not want this testing, I don’t want any more care.” I think that was an eye-opener for the physicians who were participating in this process, because they thought all of these patients wanted additional care. So, I think we make the erroneous assumption sometimes that patients want to continue with treatment. Until we sit down and have a conversation about what’s important to them, we assume that everybody wants everything, but I think understanding what patients’ goals are is critical.

In fact, in the Temel study, we showed that adding palliative care early improved quality of life and longevity of life without necessarily adding more drugs. That’s where I think we decided that even the patient-reported outcomes we measure are defined by the healthcare delivery system; they’re not defined by patients. It might be a bridge too far for every patient to define their own outcomes, but if we don’t have a system that incorporates or captures patient preferences and goals of care, the likelihood is that we’re going to provide more care than the population of patients wants. Again, that’s my definition of waste: if we provide more care than patients would have wanted had they known their options.

Bhuvana Sagar, MD: I agree. I think education is key—and a good understanding of what their expectation is. Is it really balanced by the evidence? What is the outcome? Because at this point for several cancers, even though we’ve made great advances, we’re still not curing the cancer. The majority of patients continue to die when they have advanced stage cancer. Do they really understand that that’s what they’re up against? What do they get, and what is their survival? Is it a month and a half? Is it 2 months? A lot of times, that is what we’re talking about in trials. Patients may not want to go through a lot of those treatments if they really understand what they’re up against.

John Fox, MD: Interestingly, in the Temel trial out of Massachusetts General Hospital, they looked at concordance rate between the patients’ expectations and the curability of disease. These were all patients who had metastatic disease in palliative care—their understanding was more concordant with the reality that this is not a curable condition. I think the misperception about if their disease is curable may lead them to do things that they might not otherwise do if they knew the real prognosis.

Joseph Alvarnas, MD: I think all of you have articulated why patient-centricity is important to determining what constitutes good care. Lately, we’ve seen a rush to metricize everything. How do you measure whether or not that care was actually delivered in a way that achieves what I think are the very human-centered goals that you’ve articulated? Do we have a way of developing meaningful quality measures or metrics to let us know who does this well and maybe reward them, or who does this poorly and not reward them? How do we do this?

John Fox, MD: In the oncology care model from CMMI, one of the measures in the 13-point care plan is the documentation of what the treatment goal is. Is it curative or palliative? And I just think that if there’s discordance between what the patient’s understanding of their treatment is and what the physician’s expectation is, that’s a measure of poor quality, and it’s a very simple one. I don’t know how you collect information from the patient about what their perception is, but if there’s discordance there, then something is wrong. And the likelihood is that we’re paying for healthcare services that might not be necessary.

Joseph Alvarnas, MD: You raise that question, and it seems that when we look at the nature of the kinds of metrics that people have rushed toward, we see the PQRS measures, which I think are not good when it comes to measuring the quality of cancer care—other than maybe some of the end-of-life measures—and need a little bit more context to be more robust than they are. But if we look at things like the NQF measures for cancer, do we think that someone who got chemotherapy for breast cancer in 119 days got infinitely better care than someone who got chemotherapy at day 121? Are these the kinds of measures that we want to foster? As someone in the payer space or in the health plan space, are these things compelling to you?

Bhuvana Sagar, MD: Ideally, the measures that you really want to focus on are, are we improving their survival? Are we improving their quality of life? Those are probably the best measures that I can think of—and patient-reported outcomes are probably important, as well. I think in the initial iterations of it— where they tried to measure if pain is being quantified and if it is being addressed—there are good process measures to start with, but, ideally, I would like to focus on true outcomes and understand the value of a lot of the things that we are doing. It could be drug therapy, it could be surgical intervention, and it could be a lot of different things that we’re doing to the patient. Admitting them to the hospital: Is that adding any value? That’s not really quality care. They started with some reasonable first steps, but I think we have a long way to go.

John Fox, MD: What’s that phrase? “Not everything that can be measured matters, and not everything that matters can be measured.” I think that’s applicable here. When I think about what outcomes we should measure, we should measure those things that are important to patients, which—not being an oncologist—I would say are probably “Am I living better?” and “Am I living longer?” But rarely in any published clinical trial do we get that information to help inform the patient or the provider. And, in the real world outside of a clinical trial, we’re not measuring those for the most part. The handful of practices that are involved in OCM are measuring some health-related quality of life questions, but, for the most part, that’s an extravagance that nobody invests in.