You were right, and you lost the room.

From author to expert witness.

On paper, you were an author. You controlled the document, chose the order, led the reader to your conclusion. The last lesson was about wielding that control well. In the room, that role is gone, and mistaking one for the other is the first error. In the room you are not an author presenting, you are an expert witness being examined.

That shift is total. A witness doesn't run the narrative; a witness answers. Questions come at you: from the committee, who genuinely want to understand so they can decide well, and, directly or through the committee, from the manufacturer, who wants to undermine your reconstruction and restore their favourable number. You don't get to deliver your case in your own order and sit down. You get asked things, often pointed things, sometimes hostile things, and every answer is a test, not of whether you're right, but of who you are in that room.

And the room is an adversarial setting: several parties with divergent, partly conflicting goals sit around the same table. The manufacturer wants reimbursement (and stands to profit). Clinicians or patient representatives may want access. The payer wants to control spending. And in the middle sits the committee, wanting one thing above all, a good, defensible decision. Your job is not to be another party pulling toward an outcome. It's to be the one voice in the room they can trust because it isn't pulling anywhere. Understanding what that trust is made of, and how easily it's lost, is everything.

Impartiality is your only currency.

Here's the pivotal insight, and it runs against every competitive instinct you have. In that room, your power does not come from being right. It comes from being impartial, and impartiality, not correctness, is the currency you trade in.

Think about why. The manufacturer is also "right," in their own terms, and their economists are excellent: they'll defend every choice competently. If it were a contest of who argues better, you might win or lose, but you'd be just another skilled arguer. What actually distinguishes your voice, and makes the committee weight it more heavily than the manufacturer's, is one thing the manufacturer can never claim: you have no stake in the outcome. They profit if the drug is funded; you don't. That's why the committee leans toward your reading, not because you're cleverer, but because you're disinterested. Impartiality is the entire source of your special authority in the room.

And that means impartiality is fragile, because it lives entirely in perception. It exists only as long as you look disinterested. The moment you behave like someone with a stake (pushing to win, defending your ego, attacking the manufacturer personally, clinging to your number past what the evidence supports), the committee stops seeing a neutral expert and starts seeing a second combatant. Your one advantage evaporates, not because your analysis got worse, but because you stopped looking like the person whose analysis could be trusted for its neutrality. Everything that follows in this lesson is really one instruction: protect the currency. Do nothing in that room that spends your impartiality, however satisfying it would feel.

Winning the argument is losing the room.

Now the paradox at the heart of it, sharp and counterintuitive: the harder you try to win the argument with the manufacturer, the more certainly you lose the room.

Follow the mechanism. The manufacturer is a party: the committee knows it, and discounts their words accordingly, because parties argue for their interests. You are valuable precisely because you're not a party. But the instant you engage the manufacturer as an opponent to be beaten (scoring points, rebutting to win, treating their case as the enemy of yours), you accept the frame that you and they are two sides of a fight. And once the committee sees two sides fighting, it does exactly what it does with any two advocates: it discounts both. You've dragged your disinterested voice down to the level of the interested one. You didn't lose the argument (you may have won every exchange); you lost your role. You converted yourself from the one person above the fray into one more person in it.

This is why "winning" is a trap. Your instinct, honed by a correct analysis, is to prove you're right by defeating them. But the committee didn't convene you to defeat the manufacturer: they can hear the manufacturer's weaknesses themselves. They convened you to give them an impartial map they can navigate by. Your task in the room is not to beat the manufacturer; it's to make yourself unnecessary as a combatant and indispensable as a neutral judgement: to hand the committee a fair picture and let them decide, rather than fighting a duel in which they're merely spectators. The expert who grasps this stops trying to win, and in not trying to win, keeps the only thing that made their presence worth anything.

Take the stand.

You're in the committee room. Questions come from the manufacturer and the committee: some pointed, some hostile. Choose how to answer, and watch what each answer does to the one thing that matters: the committee's trust in your impartiality. (There's no "clever" answer that wins the argument. There's only the answer that keeps, or spends, your credibility.)

Notice which answers cost you and which paid. Every time you fought (defended your number, matched the manufacturer's hostility, seized the committee's decision), the meter fell, because you looked like a party with a stake. Every time you conceded a fair point, separated fact from judgement, and handed the choice back to the committee, it rose, because that's what only a disinterested expert can do. The winning-sounding answers were exactly the losing ones. You keep the room not by defeating the manufacturer, but by visibly refusing to try, by being, unmistakably, the only person at the table who isn't fighting for a result.

Now you.

The three moves that build credibility.

The hero and the sort weren't random: they're three specific, repeatable moves, each one a way of demonstrating the impartiality you can't just assert. Learn them as moves.

• Move 1: Concede what's fair, and admit what you don't know. When the manufacturer makes a genuine point, say so: "You're right about that." When the data can't settle something, say that too: "That's a real uncertainty; the evidence can't resolve it." This feels like giving ground. It isn't; it's the one thing a party structurally cannot do, so doing it proves you're not one. And it buys belief for everything else: concede where you should, and the committee trusts your firmness where you don't. (This is Lesson 2's calibration, now performed live and under fire.)
• Move 2: Separate fact from judgement, and hand the judgement back. When asked a value question ("is this too expensive?"), don't answer it as if it were factual. Split it: "The ICER is £34,000: that's the fact. Whether that's worth it, given the severity, is your judgement, not mine, and here's what changes under different thresholds." Answering the value question yourself usurps the committee's role and, worse, reveals you as someone with a preferred outcome. Handing it back demonstrates you don't have one. (This is Lesson 2's fact-judgement seam, defended live.)
• Move 3: Answer the question actually asked, directly. Under pressure, the temptation is to retreat into your prepared, strong material regardless of what was asked. Resist it. A short, direct, honest answer to the actual question (even an uncomfortable one) reads as confidence and candour. Dodging into rehearsed narrative reads as evasion, and evasion is what a party does when the truth is inconvenient. Directness itself is a signal of having nothing to hide.

Three moves, one purpose: each is a small, visible proof that you have no side.

The hostile question is a gift.

Now the move that separates a good expert from a great one, because it inverts your instinct completely. When the manufacturer asks a hostile question ("isn't your cautious extrapolation just as arbitrary as our optimistic one?"), every nerve says attack incoming, defend. That instinct is the trap. Defending is what a party does; the moment you defend, you've accepted that you have a position to protect, and you look like the other side. The hostile question isn't an attack to repel. It's a gift, a public opportunity to demonstrate, in front of the committee, exactly how impartial you are.

Here's the ju-jitsu. Don't resist the force of the question; redirect it. Concede the grain of truth in it ("you're right, both extrapolations are assumptions beyond the data"), then hand the decision to where it belongs ("which is why I'm not asking the committee to trust my curve over yours, I showed them both, and the range between, so they can decide which risk they'd rather carry"). The manufacturer tried to drag you into a duel of two extrapolations, yours versus theirs. Your answer refuses the duel and gives the choice to the committee, and it's the committee, not the manufacturer, who comes away satisfied, because you've just treated them as the decision-makers they are. You've used the attack to show you have no case to win, only a map to present.

That reframe transforms the whole experience of being in the room. Hostile questions stop being threats to survive and become the best moments you get, because each one is a chance to show, vividly and against resistance, that you are the only person at the table not trying to win. The manufacturer, by attacking, keeps handing you opportunities to look impartial. Let them. Every hostile question answered this way spends their credibility and builds yours, which is the exact opposite of what they intended, and precisely what serves the committee.

What response best serves you and the committee?

Why this matters for HTA

The committee room is where an assessment is won or lost as a decision, and performing well there is a discipline in its own right, one that inverts ordinary competitive instinct:

• Guard your impartiality above your argument. In the room, being right is necessary but nowhere near sufficient; what gives your voice its weight is that, unlike the manufacturer, you have no stake, and that advantage exists only as long as you visibly behave like someone who has none. Never do the satisfying thing (score the point, defend the ego, seize the value call) if it costs you the appearance of neutrality. The argument is worth less than the impartiality; spend the former to keep the latter, never the reverse.
• Refuse the duel; hand the decision back. Your role is not to defeat the manufacturer (the committee can weigh the manufacturer's case themselves) but to give them a fair map and let them navigate. Separate fact from judgement out loud, return every value question to the committee, and present ranges rather than fighting for a single figure. The expert who tries to win becomes a second advocate; the one who tries to inform stays the one neutral voice, which is the only voice worth having.
• Treat hostility as an opening, not a threat. A hostile question is the best chance you'll get to demonstrate impartiality under fire: concede the grain of truth, credit the other side where it's genuinely strong, and redirect the real decision to the committee. Answering a charge of bias by showing neutrality (not asserting it, and never counter-attacking) turns the attacker's energy into your credibility. The room believes what it sees you do under pressure, not what you claim about yourself.

In the committee room your correctness is assumed and your impartiality is on trial. Every instinct to win the argument is an instinct to spend the one thing that makes your presence worth anything. Concede what's fair, hand back what's theirs to decide, and meet hostility by refusing the fight, and you leave as what they needed: not the winner of an argument, but the one judgement in the room they could trust.

In the room, in one breath.

• The committee room is not a presentation but an adversarial examination: you shift from author (controlling the document) to expert witness (being questioned). Parties with divergent interests sit around the table; you are (or must remain) the one without a stake.
• Impartiality is your only currency. You're weighted above the manufacturer not because you're cleverer but because you're disinterested, and that advantage is pure perception, lasting only as long as you look neutral. Behave like someone with a stake and you become a second advocate, discounted like the first.
• So winning the argument loses the room: the harder you fight the manufacturer, the more you look like the other side, and a committee watching two combatants discounts both. Your job isn't to defeat them but to inform the committee: to hand them a fair map and let them decide.
• Three moves demonstrate the impartiality you can't assert: concede what's fair and admit what you don't know; separate fact from judgement and hand the judgement back; answer the question actually asked, directly. And the hostile question is a gift, a public chance to show neutrality by conceding the grain of truth and redirecting the decision to the committee, never by defending or counter-attacking.

Being right gets you into the room; staying impartial is how you're heard once you're there. The moment you try to win, you become just another voice fighting for an outcome, so don't win. Concede, clarify, and hand the decision back, and be the one person at the table with nothing to gain but a good decision.

You can now take a manufacturer's case apart, rebuild it into an honest report, and defend that report live without forfeiting your neutrality. One thing remains, the fastest, most practical skill of all: the pattern-recognition that lets you glance at a dossier and feel where the problems are before you've done a line of analysis. The red flags that recur in submission after submission, and what each one signals. That's the final lesson, and the end of the course.