The Disappearing Specialty, Part II

The Vocabulary Heist

Walk into the website of any cancer center built or rebranded in the last five years and you will find palliative care's vocabulary doing work it didn't sign up for. Symptom management. Whole-person care. Goals of care. Supportive care. The words are everywhere. The specialty is harder to find.

This isn't new, and it isn't an accident. We did this to ourselves, deliberately, starting in 2007, in a name change that produced both a methodology for being seen and a slow-motion erasure that has accelerated as the marketing got more sophisticated. Part I of this series argued that virtual-first cancer programs are being marketed as equivalent to comprehensive cancer center care while routing around the accreditation regime that defines what "comprehensive" means. This part is about what happens after the marketing wins.

The First Heist

On November 1, 2007, a palliative care service at a major cancer center changed its name to Supportive Care. Hui and colleagues published the analysis four years later. The findings were a vindication and a warning at the same time. Outpatient referrals shifted earlier in the disease trajectory. Inpatient referrals went up. Patients arrived more willing to engage. The name change worked.

It worked because the word "palliative" was carrying baggage that the field hadn't created and struggles to outrun. Oncologists associated palliative care with end of life, with giving up, with breaking the news that there was nothing more to do. Patients associated it with hospice. The literature documented this stigma in survey after survey. Fadul, Bruera, and colleagues showed in 2009 that medical oncologists at were significantly more willing to refer to a service called "supportive care" than to one called "palliative care," even when the services were identical. The stigma was real, the workaround was effective, and the workaround spread.

By 2018, Hui's national survey found that 35% of NCI-designated cancer centers and 30% of non-NCI cancer centers had adopted "supportive care" as the name of their palliative care program. Outpatient palliative care availability at NCI cancer centers rose from 59% in 2009 to 95% in 2018. The intervention worked. Patients got seen earlier. Time from referral to death lengthened — a marker, in this literature, that referrals were happening at a less catastrophic point in the disease.

Reading the data, you would conclude that the rebranding was an unalloyed success. And in some ways it was. But here is the part that didn't get measured: in trading the word "palliative" for the word "supportive," the field accepted a description of itself that fit comfortably alongside the rest of oncology (Kyle side rant: and we did it by repurposing an existing concept in Oncology that oncologists already do – thereby giving them more reason to think that anyone can just do palliative care), instead of insisting on a description that named what made us different. Supportive care is a generic. It can mean a clinic with three HPM-boarded physicians, an interdisciplinary social worker, a chaplain, and a music therapist. It can mean a nurse navigator who remembers to ask about pain. It can mean ongoing cancer surveillance by an Oncologist. It can mean a wellness app the employer paid four dollars per member per month for. The word doesn't distinguish.

The Second Heist

Twenty years on, the new entrants in employer cancer benefits have figured out something the field probably should have anticipated. If the word "supportive" is generic, then anyone who provides anything that could plausibly be called supportive can claim the label. And once they have the label, they can claim the work. And then they don't have to do the work, because the work is hard to verify from the outside.

Color Health's Virtual Cancer Clinic page does not use the words "palliative care" or "specialty palliative care" anywhere. It does talk about "symptom management." It does talk about "whole-person care." It does have a service called Cancer Connect, which is described as a peer-led mental health support program with published evidence of reducing depression and anxiety scores. None of that is bad. None of that is specialty palliative care. The vocabulary is doing the work the specialty isn't there to do.

Iris by OncoHealth, which actually operates in the same employer space as Color, calls itself "supportive cancer care." Hinge Health for orthopedics, Maven for fertility, Carrum for surgical bundles, Lyra for behavioral health — every employer benefits vendor is positioning itself as the whole-person solution for its vertical. "Whole person" was once a phrase that distinguished palliative care from the disease-focused defaults of subspecialty medicine. Now it is the standard marketing claim for every digital health benefit on the open market.

The trade the field made in 2007 looks different from where we are now. We borrowed the language of supportive care to lower a stigma the field hadn't earned. The new entrants are using the same language, with the same resonance, to claim work they aren't equipped to do. The first move was strategic. The second move is parasitic. And the field has not yet figured out how to push back against the second move without sounding like we are sore losers.

What the Vocabulary Was Doing for Us

The vocabulary wasn't decoration. The words "palliative care" carried a specific clinical meaning: an interdisciplinary specialty trained to manage the physical, psychosocial, and existential dimensions of serious illness across the full trajectory, with formal expertise in pain and symptom management, prognostic communication, family systems care, and end-of-life decision-making. People, who happen to be experts, who would get down with patients and families in the muck and the mire and shield them from just a little bit of the bullshit and buoy their wellbeing in the darkest moments. That definition was hard-won. It distinguished us from oncology, from hospice, from chaplaincy, from social work — not because those fields were less valuable, but because what we did was specifically not what they did.

When the language collapsed into "supportive care" — and then into "whole-person care," "goals of care," "symptom management," and now "wraparound care" — the distinction collapsed with it. A buyer assessing options for a cancer benefit cannot distinguish from the marketing copy whether a vendor has an HPM-boarded clinician on staff, whether they have a referral pathway to specialty palliative care, whether their team includes a chaplain or social worker who has been trained in serious illness care. The buyer reads the words and assumes the work is being done. The work, in many cases, is not.

This is the failure mode the field set up in 2007. The fix that worked then doesn't work now, because the people running the play in 2026 aren't paying tribute to specialty palliative care while they borrow our language. They're substituting for us.

Who Is Filtered Out, and Why It Matters

The advanced-cancer population in the United States is not demographically the same as the early-stage cancer population. SEER and CDC data show that Black Americans are diagnosed with metastatic disease at higher rates across multiple cancer types — for prostate, the metastatic-at-diagnosis rate is roughly twice that of white Americans. Hispanic Americans face later-stage diagnosis disparities for breast, cervical, and colorectal cancers. Older adults make up a disproportionate share of patients with advanced disease. Patients with multiple chronic conditions, lower health literacy, and lower-resource living situations are more likely to present with advanced disease, more likely to face barriers to coordinated care, and more likely to need specialty palliative care.

The employer-sponsored benefits market does not see most of these patients. Working-age employer-insured populations skew younger, healthier, and whiter than the general population. By the time a patient progresses to advanced disease and needs the work specialty palliative care does most, they have often left employment, transitioned to disability, aged into Medicare, or been picked up by a different insurance arrangement entirely. The new cancer benefits programs are not failing those patients accidentally. They are not designed to see them.

This is what makes the vocabulary heist a particular kind of erasure. The work specialty palliative care is best at — managing complex symptoms in advanced disease, navigating prognostic uncertainty, supporting families through bad news and worse decisions, accompanying patients to the end — happens in populations the new market doesn't reach. The new market then borrows the vocabulary that describes that work and uses it to sell to a different population that doesn't need as much of it. The patients who do need it are left to a system that increasingly looks underfunded, fragmented, and slow next to the venture-capital-backed virtual clinics that have absorbed the language without doing the work.

If the field's vocabulary has been redirected to serve people who don't need our specialty, in marketing aimed at buyers who are not paying for the patients who do, that is a kind of structural invisibility the 2007 rebranding did not anticipate. It is also the kind of invisibility that is hard to argue against, because the rebranding is the field's oft-official position. Some of us told the world to call us supportive care. We can't be surprised that the world is now using that word for things that aren't us.

Where I Might Be Wrong

Three places this argument could be weaker than I think.

The 2007 rebranding really did work for a long time. Patients got seen earlier. Referrals went up. The argument that we should have insisted on "palliative care" as the name and absorbed the stigma cost is easier to make in retrospect than it would have been to make in the room. Stigma is a real harm. The field's leaders made a defensible call with the data they had.

The new entrants may not be substituting for specialty palliative care so much as failing to address it at all. The vocabulary heist framing implies a more deliberate replacement than may be happening. It is possible that companies like Color simply don't think about advanced-disease care because it isn't the population they serve, and the vocabulary is being deployed not to substitute for our work but to describe their own (real, if narrower) work in language that resonates with employer buyers. The end result for our visibility is similar. The intent is different, and the difference matters when we're deciding how to push back.

The marketing critique might be smaller than I'm making it. If the field's actual clinical access to advanced-disease patients in safety-net and community settings is what matters, then the language used by employer benefits vendors selling to a different population is mostly irrelevant. The fight that matters is over Medicare and Medicaid coverage of specialty palliative care, not over employer benefits marketing copy. I take this point. I think the language matters because language shapes what gets funded, what gets included in quality measures, and what gets taught in training programs. But the empirical question of how much it matters is open.

What Comes Next

The first thing the field can do is stop being polite about this. Naming a vendor's claim that they provide "whole-person cancer care" or "supportive care" or "goals-of-care discussions" as marketing rather than care isn't impolite. It is honest. We can do it without insulting anyone. We can do it without making it personal. We can do it by asking, in every public conversation about a new cancer benefit, the question: who is the HPM-boarded clinician in the model, what is the referral pathway to specialty palliative care, and what is the published proportion of advanced-disease patients receiving that care within weeks of advanced diagnosis?

The second is to give up on the idea that the next vocabulary fix will solve the problem. The field has been on the lookout for the next term that will lower stigma and drive referrals. The lesson of the last twenty years is that whatever word we choose, the market will adopt it within a decade and the same dynamic will replay. Stigma reduction is a real goal. It is not the same goal as specialty visibility. The two have been conflated, and we should stop conflating them.

The third — and this is where Part III goes — is to recognize that the next vocabulary collapse isn't going to happen in marketing copy. It's going to happen in training data. The AI systems being deployed to drive the next generation of cancer care are learning what cancer care is from the same sources that produced "wraparound care" and "supportive cancer care." The algorithm doesn't know what the words used to mean, and it isn't going to learn from us if we aren't there.

The Disappearing Specialty is a three-part R&R series on how specialty palliative care is being structurally disappeared from the next generation of cancer care.

Part I — When the Floor Disappears
Part II — The Vocabulary Heist (you are here)
Part III — No Signal for the Work (next Tuesday)

I am a palliative care physician, educator, and professional strategery expert known for turning rounds into rants and rants into teaching points.