Watch enough interviews with health ministers, agency heads and foundation directors and a pattern appears. They are asked what they are proud of. They are asked about their priorities, their vision, their partnerships, what keeps them up at night. They are asked, occasionally, about a scandal. They are almost never asked about the ordinary, structural choices that define their jobs: what they stopped, who lost, what they got wrong and what would change their minds.
This is not because interviewers are lazy. It is because the questions that would reveal how a health system is really being run are the ones that every party to the conversation has a reason to avoid. The leader avoids them because honest answers are costly. The interviewer avoids them because access depends on the leader coming back. The audience rarely demands them because it has never heard them answered and does not know what it is missing.
My argument is that the absence of these questions is not a minor gap in public discourse. It shapes policy. When leaders know they will only be asked about additions, launches and ambitions, they manage for additions, launches and ambitions. The questions a system asks in public become the incentives it runs on in private.
Here are seven that are almost never asked, why each is avoided, and what an honest answer would reveal.
1. What did you stop funding last year?
Every health budget is a set of choices, and most of them are choices not to fund something. Yet leaders are nearly always presented through what they started: a new programme, a new hospital, a new partnership.
The question is avoided because stopping things creates a visible constituency of losers, while starting things creates a constituency of winners. A programme that ends has staff, contractors, patients and political sponsors who will complain. A programme that is quietly underfunded until it fades creates far less noise. So the honest answer is often “we didn’t formally stop anything; we let several things starve.” That answer is uncomfortable, which is exactly why it is useful.
What it would reveal: whether the organisation has any real priority-setting process at all. A leader who can name what was stopped, and why, is running a budget. A leader who can’t is running a list of commitments that grows every year until the money runs out and choices are made by default, usually by whoever shouts loudest or whoever controls procurement.
2. What did your last failed programme teach you?
Health systems run experiments constantly. Many fail. The public record of those failures is thin, because failure is rarely declared. Programmes are “restructured”, “integrated into routine services” or “transitioned to a new phase.”
The question is avoided because the leader answering it is often the person who approved the programme, or whose predecessor did and whose party still claims it. Admitting failure carries career risk and, in donor-funded contexts, the risk that funders will draw broader conclusions. There is also a genuine difficulty: most programmes partly fail, and separating what failed from what worked takes an honesty that a press conference does not reward.
What it would reveal: whether the organisation learns. The content of the answer matters less than whether one exists. A leader who can say “we assumed community health workers would be paid through district budgets, and in practice those payments were three to six months late, so we moved them to a central payroll” is describing a system that updates. A leader who has no answer is describing a system where failures are buried, and therefore repeated.
3. Who loses under your reform?
Almost every serious health reform redistributes something: money, power, patients, status. Moving care from hospitals to primary care means hospitals lose budget. Introducing strategic purchasing means some providers lose contracts. Adding a service to a benefit package without adding money means something else is rationed.
Leaders present reforms as win-win because reforms with named losers attract organised opposition before they are passed. And interviewers rarely press, partly because asking “who loses?” sounds hostile, as if one were opposing the reform.
What it would reveal: whether the reform has been thought through politically, and whether it will survive. A reform whose losers are unnamed has not been negotiated with them, which means the negotiation will happen later, during implementation, often through quiet non-compliance. The leader who can say “specialist hospitals will lose about a fifth of their outpatient income, and here is what we are offering them in return” has a reform. The leader who says “everyone benefits” has a press release.
4. What data would change your mind?
This is the question that separates a position from a belief. If a health leader supports a policy, there should be some pattern of evidence that would lead them to drop it. If there is none, the policy is not evidence-based, whatever the speeches say.
It is avoided because naming a falsifying condition creates a commitment that can be checked later. It gives critics a measuring stick. It also exposes, uncomfortably often, that no monitoring system exists that could even produce the relevant data.
What it would reveal: first, whether the leader is holding the position for reasons they can articulate. Second, whether the system is collecting the information needed to know if the policy works. A minister who says “if facility deliveries in the pilot districts haven’t risen within two years, we will redesign the incentive” has made a promise that can be tracked. That is precisely why it is rarely said.
5. What are you doing that the evidence doesn’t support?
Every health system spends money on things with weak evidence: screening programmes that persist out of habit, drugs that stay on formularies because removing them is politically hard, training models that have never been evaluated, facilities built for reasons that had little to do with need.
Leaders avoid this question because the honest answer implicates their own budget. It also invites the follow-up, “then why are you still doing it?”, to which the true answer is usually some combination of politics, contracts and inertia.
What it would reveal: the size of the gap between the evidence-based system the leader describes and the one they actually run. It would also reveal where the easiest efficiency gains are. Disinvestment from low-value care is one of the least politically popular and most financially significant things a health system can do. A leader willing to name candidates publicly has already done half the work.
6. Where are your own figures weakest?
Health leaders routinely cite coverage rates, facility numbers, enrolment totals and outcome trends. They almost never volunteer which of those numbers they trust least.
The question is avoided because it undermines the leader’s own claims of progress. It also exposes the data system: numbers built from facility reports that are filled in from memory at the end of the month, denominators based on census estimates that are a decade old, enrolment figures that count people who have never used the scheme.
What it would reveal: how much of the public story is resting on sand. A leader who says “our immunisation coverage figures are reliable in urban districts and I would not stake much on them in the north, where the population estimates are out of date” is being more useful than one who cites a single national percentage to one decimal place. The honest answer also tells funders and researchers where to invest in measurement.
7. What would you do differently if you knew you would not be reappointed?
This one is sharper, and it is about time horizons. Health reform pays off over ten or twenty years. Leaders serve for three or four, sometimes far less. The rational thing for a leader to do is prioritise what shows results within their tenure.
The question is avoided because it names that incentive directly and asks the leader to acknowledge that their current choices are shaped by it. Very few people will say that on the record.
What it would reveal: which long-term investments are being deferred. Answers tend to point at unglamorous, slow-return work: health information systems, workforce pipelines, purchasing agencies, supply chains. Knowing what a leader would do if freed from the electoral or appointment clock is a map of what the system is systematically neglecting.
The incentive underneath
All seven questions share a structure. They ask about subtractions rather than additions, about uncertainty rather than confidence and about the long term rather than the tenure. They are the questions that make a leader look less like a visionary and more like a manager of trade-offs, which is what a health leader actually is.
The people with the most power to ask them, journalists with access and moderators at high-profile events, are also the people whose access depends on not asking them too hard. A conference that embarrasses a minister will struggle to get the next one. A journalist who corners an agency head may find the next interview goes to a friendlier outlet. So the system settles into a comfortable equilibrium of soft questions, and that equilibrium teaches leaders that a public career in health can be built without ever explaining a trade-off.
The strongest objection
The strongest objection is that these questions are unfair, not in content but in setting. A minister cannot say, in public, which programmes she is letting starve without triggering a crisis in those programmes. An agency head cannot admit his coverage figures are weak without jeopardising funding that depends on them. Honesty in public can be punished in ways that make honesty in private impossible. Some of the most effective health leaders operate precisely by not saying what they are doing until it is done.
That objection is real, and it should shape how the questions are asked. But it proves less than it seems. Most of these questions can be answered at a level of generality that protects specific programmes while still revealing how decisions are made. “We stopped two vertical training programmes and folded the budget into district supervision” names no one’s job. And the objection cuts the other way too: if leaders cannot talk about trade-offs in public at all, the public never learns that trade-offs exist, and every future reform starts from the fiction that everything can be funded at once.
A worked example (hypothetical)
Imagine a national health insurance agency head at a public forum. The usual question: “What are your priorities for the coming year?” The answer, predictably, is expansion of enrolment, digitisation of claims and stronger partnerships with private providers.
Now imagine the question is: “Your scheme added three new services to the benefit package last year. Which existing service has had its effective funding reduced as a result?” A candid answer might be that the tariff for outpatient visits was not raised in line with inflation, so providers are being paid less in real terms for primary care, and some have started asking patients for informal top-ups.
That single answer tells the audience more about the scheme than an hour of priorities. It explains why patients report paying at clinics despite being covered. It shows where the scheme’s financial pressure is being absorbed. And it gives primary care providers, researchers and legislators something specific to act on.
Who can change the questions
Moderators and journalists can agree in advance with each other, not with the speaker, on one subtraction question per session. Asked routinely, it stops being an ambush and becomes normal.
Conference organisers can build sessions around these questions explicitly, and brief speakers ahead of time that they will be asked. Warning people removes the ambush without removing the question.
Legislators and oversight bodies can put them into formal budget hearings, where the setting already expects scrutiny.
Funders can ask grantees and government partners for them in writing: what did you stop, what failed, what would change your plan. Then they can reward honest answers instead of punishing them.
Health leaders themselves have the most power of all. The first minister who answers “what did you stop funding?” without being asked will set a standard that others will be measured against.
The takeaway
The questions a health system asks its leaders in public become the things those leaders manage for. As long as leaders are asked only about what they are adding, they will keep adding, and the real decisions about what gets cut, who loses and what failed will keep happening where no one can examine them.
Which question would you most like to hear a health leader answer honestly, and in what setting would they be most likely to do it? Share it in the comments.
