In partnership with

When it all clicks.

Why does business news feel like it's written for people who already get it?

Morning Brew changes that.

It's a free newsletter that breaks down what's going on in business, finance, and tech — clearly, quickly, and with enough personality to keep things interesting. The result? You don't just skim headlines. You actually understand what's going on.

Try it yourself and join over 4 million professionals reading daily.

Check it out

The Need You Were Never Asked About

There's a crucial distinction between what a patient needs and what a patient wants — and it's the one almost no one is allowed to make.

A woman with stage 4 lung cancer that has spread to her brain sits with her family while the oncologist lays out the plan. Four rounds of chemotherapy, starting now. The tone is calm, kind, and certain.

Notice what nobody says. Nobody says the word "need." Nobody has to. The recommendation arrives as the plan. This is what we do next.

The oncologist isn't pressuring anyone, and that's exactly the point. The family feels guided, not pushed. That's the reason it works, and it's also why it keeps happening.

The most consequential decisions in medicine often don't feel like decisions at all. They feel like instructions. And they feel certain.

The options you were never told you had

But the choice in that room was shaped long before it reached the patient. It was shaped by everything that went unsaid, including, often, the option of doing less, on purpose. Had those options and their likely outcomes been laid out plainly, the patient and family could have made a truly informed choice: one rooted in knowledge rather than fear, and backed by the sense that whatever they chose, their physician would support them.

There are questions a patient or a loved one would need to ask, if only they knew the questions existed, or that another path was even realistic:

"When you say this is what we should do, do you mean it will cure the cancer? Or extend my life? By weeks, or by months? What will I suffer to buy that time: the pain, the nausea, the exhaustion? What are the other options, and what happens if I choose one of them instead?"

Or simply: "What would you do if this were your own mother?"

You can't weigh an option you don't know exists or that you're too afraid to ask about. And the silence right after a recommendation is itself a kind of persuasion.

This isn't an uncommon scenario. Consider patients who get stents for stable coronary disease. In one well-known study, the large majority believed the stent would lower their risk of a future heart attack, even though it wouldn't, and even though the cardiologists who referred them generally didn't believe it would either. The trials had already shown as much. The patients simply weren't told.

What "need" actually is

Need is a professional judgment, and it depends on realistic, achievable goals grounded in an evidence-based medical (and sometimes rehabilitative) prognosis.

Is cure a realistic goal? Is more time? Does this person want to die at home, surrounded by the people they love? Each of those goals points to a completely different need. And the goals belong to the patient.

Yes, a patient can want something unrealistic. That's exactly why the conversation is supposed to happen. That’s the back-and-forth where clinician and family trade what each of them knows until both understand enough to make the next best decision. But that conversation requires that the patient actually be given the knowledge. When it doesn't happen, the goals just get filled in by the person writing the treatment plan and making the referrals. That’s the one with the knowledge, and, too often, a financial stake in the answer.

Need should be close to scientific. Instead, it is almost always authored by someone other than the person who has to live, or die, with the consequences.

How necessity gets built without an argument

So how does a "need" get installed without anyone ever arguing for it? It happens through the way a recommendation is delivered.

There's a name for this: a nudge. A nudge is a feature of how a choice is presented that shifts behavior in a predictable direction without removing any options. Medicine runs on them.

The default is the strongest nudge. "We'll get you scheduled for your first round" makes the intervention the path and inaction the deviation. In other words, it’s an opt-out, not an opt-in. Framing is subtler but just as real: "70 of 100 patients do well" and "30 of 100 don't" are the same fact, yet they pull people toward different choices, and survival framing reliably nudges patients toward the more aggressive option.

Authority matters too. Said with enough certainty, a recommendation stops sounding like a recommendation at all. The white coat turns "an option" into "what's done."

But perhaps the most powerful nudge of all is the one I already described: the alternatives left off the table.

This is the layer that can feel uncomfortable to acknowledge. The people with the power to shape all of this, including what gets framed, what gets defaulted, and what gets left unsaid, are very often the same people who benefit financially when the patient says yes to the intervention. This usually isn't malice. Most clinicians barely notice they're doing it, because it's simply how they were trained to practice. Their job, as they understand it, is to treat, to act, to do. But the incentive sits there all the same, and it rarely points toward "less."

The obvious answer often isn't the evidence

It would be comforting to assume the "obvious" intervention is at least the evidence-based one. Often, it isn't.

For ordinary low back pain, imaging doesn't improve outcomes. This has been studied extensively, yet its overuse tracks with a two- to threefold rise in surgery. A scan often turns up some incidental "abnormality," usually nothing more than the normal wear of being alive, and that finding alone can leave a patient more afraid, more fragile, and more likely to spiral into procedures they never needed. The test manufactures the need.

Arthroscopy for a worn knee performs no better than a placebo procedure, and frequently leads to more surgery. A stent for stable disease carries real risk even when it goes perfectly. About 1 in 100 patients has a heart attack or stroke during or soon after the procedure, on top of the bleeding risk from the procedure itself and the blood thinners that follow.

The takeaway is that a great deal of what we do reflects habit and training far more than the highest-quality, most current evidence. And, lest we forget, doing too much is never free to those of us paying for it. That’s all of us. Overtreatment carries its own harms like cascading tests, anxiety from false alarms, and real physical damage from procedures that were never going to help.

"Medically necessary" and what that was supposed to mean

Which brings me to a phrase we have hollowed out: medically necessary.

There is such a thing as medically necessary care, and it matters enormously. Properly understood, it means care that is genuinely necessary for a person's health and judged against their actual prognosis, both medical and functional, and matched to the achievable, realistic goals shared by the patient, their care partner, and the clinician. That is a real and demanding standard, and one we should all care about: as clinicians, leaders, patients, and policymakers.

In practice, though, "medical necessity" has been bastardized into a billing term. We treat it as "what must be documented and justified to get a claim paid." That is not the same thing as proof that the right clinician applied the best available evidence to this particular person's situation.

Watch it happen in post-acute care. constantly. After a hospital stay, a patient might recover in an inpatient rehab facility, a skilled nursing facility, or at home with home health. In theory, where they go is a medical-necessity decision. But as I've seen in practice, because the overlap in where a patient could appropriately recover is so large, that distinction often barely gets made. The choice turns on who has an open bed, on a case manager's habit, on family logistics. A patient gets sent to a facility not because they medically require that intensity of care, but because their daughter works during the day and can't be home while her mother regains her strength and mobility. Oh, and they can’t afford or refuse to hire short-term help. That is a custodial need, a convenience, a real human problem, for sure. But it is not medical necessity, and we shouldn't pretend it is.

And to be clear, the answer is not "always less." A stroke patient, for example, often genuinely needs the intensity of inpatient rehab to recover the most function possible. That is exactly where they belong! The point is that care should be matched to what this person can realistically achieve, to what genuinely meets the standard of medical necessity, and to the right frequency and intensity of therapy the patient will tolerate and benefit from. Too often, referral habits, preferences, assumptions, misunderstood rules, and bed availability decide it instead.

We are all sharing finite pots of money

This is why the distinction between a need and a want is one that should be getting air time. 

Payers are not bottomless. Whether it's Medicare, Medicaid, a commercial insurer, or an employer's health plan, the money comes from somewhere real, like taxpayers, employers, states, and the premiums the rest of us pay. T

And the pot is under more strain than it has been in a long time. U.S. prescription drug spending hit roughly $806 billion in 2024 and is on track to pass $1 trillion in 2026. A single class of drugs (GLP-1s) ran about $132 billion in 2025 and drove nearly a third of all drug-spending growth, while insurers watched their drug costs jump more than 20% in a single year. Add the surge in acute care, high-cost oncology, neurological, and rare-disease therapies, and you have a system absorbing enormous, legitimate new expenses. And the pressure isn't only pharmaceutical: the single largest driver of projected Medicare spending growth over the next decade is, quite simply, doing more. It’s a  rising volume and intensity of services. In that environment, money wasted on care that wasn’t truly needed is care someone else didn't get.

We seem to get this backwards. As a country, we decided long ago that outside of Medicaid's long-term services and supports, and some VA benefits, we don't cover custodial care. Payers, including CMS, also say they don't cover unnecessary care, the procedure a patient would have declined if they'd understood their options and their likely outcomes, and the non-medically-necessary wants of clinicians or patients. But the fact is, we ALL do pay for those things every day.

When the system spends on those things anyway, we reflexively call it the payer's failure. But payers want the money spent on what's needed. The failure happens in the room where a want was quietly dressed up as a need, and where no one was asked what they actually wanted in the first place.

That's also what makes this so hard to fix: there's no villain to point at. Everyone who holds the knowledge also carries a default consequence and an associated risk. And almost no one inside the system experiences themselves as doing this. They believe they're following the standard of care, being thorough, and being kind. They think this is what patients want.

What value was always for

This is part of what value-based care was actually for, and what the field mistranslated from the very beginning. It got sold as a way to save money, and I think that's why it hasn't landed the way it should (though I believe it still can). Value-based care, especially in the "downstream" part of healthcare, is partly an attempt to hold the system accountable for whether care matched the patient's real, achievable goals, their understood options, and their likely outcomes, rather than the provider's default plan and framing.

“Value” does not mean “cheaper.” Its definition includes the care a person would actually choose if they were fully informed. And that choice requires understanding their prognosis, every option genuinely available to them, and the outcomes and trade-offs of each, all grounded in the best available evidence. The industry hears the word "value" and reflexively translates it to "discount." That single mistranslation is why we keep failing the woman in that first room.

What matters to you

Back to the room. Listen to the silence. Now you can hear it. "We'll start with four rounds." But toward what goal? Measured against what alternative? And decided by whom? 

The medicine we should want for ourselves, and should insist on for everyone, sounds different. It sounds like: Here's what you're facing. Here's what each path is likely to give you, and here's what it will cost you. Here are all of your options. The need is yours to define, so tell me what matters to you, and let's choose the next step together. The one that fits you.

When a clinician stops implying that you need this, they free themselves to ask a question palliative care has understood the power of for years: What matters to you?

It's the first step any clinician can take tomorrow. And for everyone else in healthcare like the case managers, the rehab teams, the administrators, the people who design and pay for and carry out plans of care, it is the missing perspective. Start there, and "need" stops being something done to a patient. It becomes something decided with them.

Sources & further reading

• The study behind the opening scene. Most people receiving chemotherapy for incurable cancer don't understand it can't cure them. Weeks JC, et al. "Patients' Expectations about Effects of Chemotherapy for Advanced Cancer." New England Journal of Medicine, 2012;367(17):1616–1625. nejm.org/doi/full/10.1056/NEJMoa1204410

• Why patients think a stent does more than it does. The study that found patients believe stents prevent heart attacks while the cardiologists referring them generally don't — a textbook informed-consent gap. Rothberg MB, Sivalingam SK, Ashraf J, et al. "Patients' and Cardiologists' Perceptions of the Benefits of Percutaneous Coronary Intervention for Stable Coronary Disease." Annals of Internal Medicine, 2010;153(5):307–313. acpjournals.org/doi/10.7326/0003-4819-153-5-201009070-00005

• The evidence that stents don't help in stable disease. Three landmark trials, ending with the first sham-controlled one. COURAGE: Boden WE, et al. NEJM, 2007;356(15):1503–1516. ISCHEMIA: Maron DJ, et al. NEJM, 2020;382(15):1395–1407. ORBITA: Al-Lamee R, et al. The Lancet, 2018;391(10115):31–40. nejm.org/doi/full/10.1056/NEJMoa070829 · nejm.org/doi/full/10.1056/NEJMoa1915922 · imperial.ac.uk/news/182879/study-heart-stents-stable-angina-highlights

• The risk a stent carries even when it works. A plain-language summary from Harvard, with cardiologist William Boden: roughly a 1% risk of heart attack or stroke during or soon after the procedure, plus bleeding from the procedure and the blood thinners that follow. "Does a coronary stent make sense for stable angina?" Harvard Health, 2024. health.harvard.edu/heart-health/does-a-coronary-stent-make-sense-for-stable-angina

• What a "nudge" actually is. The concept comes from Thaler RH and Sunstein CR, Nudge: Improving Decisions About Health, Wealth, and Happiness (2008). For how it operates in medicine — including the ladder from light "information framing" to powerful "defaults" — see the AMA Journal of Ethics: "Designing Nudges for Success in Health Care," 2020. journalofethics.ama-assn.org/article/designing-nudges-success-health-care/2020-09

• Why "70 survive" and "30 die" lead to different choices. The classic study showing that framing identical outcomes as survival versus mortality flips treatment decisions — in patients and physicians alike. McNeil BJ, Pauker SG, Sox HC Jr, Tversky A. "On the Elicitation of Preferences for Alternative Therapies." New England Journal of Medicine, 1982;306(21):1259–1262. nejm.org/doi/full/10.1056/NEJM198205273062103

• Habit, not evidence, driving how much care patients get. Within the same organization, low-value-service rates varied dramatically across primary care physicians — and almost none of that variation could be explained by anything measurable about the doctors themselves. Schwartz AL, Jena AB, Zaslavsky AM, McWilliams JM. "Analysis of Physician Variation in Provision of Low-Value Services." JAMA Internal Medicine, 2019;179(1):16–25. jamanetwork.com/journals/jamainternalmedicine/fullarticle/2717501

• Imaging for ordinary low back pain — and the harm it can do. Routine imaging doesn't improve outcomes, tracks with a two- to threefold rise in surgery, and can leave patients more fearful and worse off after an incidental finding. Flynn TW, Smith B, Chou R. "Appropriate Use of Diagnostic Imaging in Low Back Pain: A Reminder That Unnecessary Imaging May Do as Much Harm as Good." Journal of Orthopaedic & Sports Physical Therapy, 2011;41(11):838–846. jospt.org/doi/10.2519/jospt.2011.3618

• Knee surgery that works no better than a fake one. Two sham-controlled trials found arthroscopy for a degenerative knee performed no better than placebo surgery. Moseley JB, et al. "A Controlled Trial of Arthroscopic Surgery for Osteoarthritis of the Knee." NEJM, 2002;347(2):81–88. Sihvonen R, et al. "Arthroscopic Partial Meniscectomy versus Sham Surgery for a Degenerative Meniscal Tear." NEJM, 2013;369(26):2515–2524. nejm.org/doi/full/10.1056/NEJMoa013259 · nejm.org/doi/full/10.1056/NEJMoa1305189

• How little evidence guides where patients recover. A systematic review of post-stroke rehabilitation found minimal evidence to determine which setting — inpatient rehab versus skilled nursing — is right for an individual patient, despite heavily overlapping populations. Alcusky M, et al. "Post-Acute Care Setting, Facility Characteristics, and Post-Stroke Outcomes: A Systematic Review." Archives of Physical Medicine and Rehabilitation, 2018. pmc.ncbi.nlm.nih.gov/articles/PMC5874162

• Where the "volume and intensity" point comes from. MedPAC's finding that growth in the volume and intensity of services is the single largest driver of projected Medicare spending growth over the next decade. MedPAC, Report to the Congress: Medicare and the Health Care Delivery System, June 2026, Chapter 1. medpac.gov

• The pressure on the shared pot. U.S. prescription drug spending reached $805.9 billion in 2024 (up 10.2%) and is projected to top $1 trillion in 2026, with GLP-1s alone around $132 billion and nearly a third of all drug-spending growth — while insurers' drug costs rose more than 20% in a single year. ASHP national trends report (via AJHP): news.ashp.org/.../us-drug-spending-up-10-in-2024-with-weight-loss-drugs-remaining-top-driver · 2026 projection: pharmexec.com/view/us-prescription-drug-spending-set-exceed-trillion-2026-report · Insurer trend: statnews.com/2025/12/08/health-insurers-spending-more-on-prescriptions-glp-1-drugs-cited

• Where "What matters to you?" comes from. The reframe from "What's the matter?" to "What matters to you?" as the heart of shared decision making. Barry MJ, Edgman-Levitan S. "Shared Decision Making — The Pinnacle of Patient-Centered Care." New England Journal of Medicine, 2012;366(9):780–781. nejm.org/doi/full/10.1056/NEJMp1109283

*Disclaimer: All opinions and ideas expressed in this article are solely mine and none represent a recommendation or should be viewed as advisement of any kind to anyone to do anything.*