Every Inquiry Finds Culture
Every Inquiry Finds Culture
Why fifteen years of healthcare reviews keep reaching the same conclusion and why the remedy keeps failing
A pattern emerges in major reviews of healthcare failure across the United Kingdom, and once you notice it, it is hard to stop seeing it.
Something goes badly wrong. Patients are harmed, sometimes over years, often after staff had raised concerns that were not acted on. An independent inquiry is commissioned. It takes evidence, reconstructs the timeline, and reports — carefully, humanely, and usually at considerable length.
And almost always, somewhere near the centre of its findings, is the word culture. A culture that prioritised targets and finances over care. A culture that did not welcome concerns. A culture of defensiveness, of poor teamworking, of failure to listen. The recommendations follow: leadership development, values work, programs to encourage speaking up, a renewed commitment to openness.
Then, some years later, another service in another trust fails in recognisably similar ways, and another inquiry finds the same thing. More than one inquiry chair has said so explicitly: that the lessons of an earlier review had not been learned, and that the failings they were now describing had been described before.
That repetition is the most important finding in the whole body of work, and it is rarely treated as one.
What needs saying first
Before going further, one thing has to be clear, because the argument that follows can easily be heard as something it isn’t.
Patients were harmed. Some of the people responsible behaved in ways that should have consequences and, in some cases, did. Nothing here suggests otherwise. Explaining why a failure recurs is not the same as excusing the people involved, and an argument that blurred those two deserves to be dismissed.
The distinction that matters is simple enough to state: individuals are accountable for what they had the authority to do, and the structure is accountable for the rest. The inquiries have been rigorous about the first. The question is whether their remedies have addressed the second.
What the inquiries are actually describing
Here is the difficulty with culture as the conclusion of an inquiry.
What investigators observe is behaviour: staff across a service acting defensively, avoiding raising concerns, prioritising the measured over the important. That pattern is real, and “culture” is the word being used to name it.
But the pattern is not the culture. Culture, in any useful sense, is not a shared attitude. It is the organisation’s real answer to a practical question: what actually pays off here, and what costs you. And that answer is not written in the values statement. It is expressed through the architecture – who decides what, who carries the consequence, what happens to the person who raises a concern, what the board chooses to measure. People learn it by watching what the architecture does, and then behave accordingly.
So the behaviour an inquiry observes is the visible end of a chain. The architecture expresses what the organisation actually rewards; the people inside it read that correctly; the behaviour follows.
And that explains the repetition. The usual remedies – values work, leadership development, encouragement to speak up – act on the stated culture, on what the organisation says it values. They leave untouched the architecture through which the real answer is expressed. For a while, while attention is high and the inquiry is fresh, the stated values and the behaviour align. When attention moves on, the architecture goes on expressing what it always expressed, people go on reading it correctly, and in time another inquiry finds the same behaviour and calls it the same thing.
What sits underneath the word
The interesting thing is that the inquiries themselves usually document the architecture in detail. They just file it under culture.
Read the findings closely, and the same features recur.
Decisions made far from where their consequences land.
Resource allocation, staffing levels, bed management and service configuration are decided at executive and board level, sometimes in response to pressures set nationally. The consequences are borne on the ward, where the people carrying them have no authority over the decisions that created them.
Accountability without the authority to discharge it.
A ward manager or clinical lead is answerable for safety in a service whose staffing, equipment and pathways are set elsewhere. When something goes wrong, their name is attached to it. The decision that produced it usually has no name attached.
Raising a concern costs the person who raises it.
This is the most consistent finding across the whole body of work, and it is almost always treated as a problem of attitude. But it has a clear mechanism. The person who raises a concern disrupts a schedule, implicates colleagues, and creates work for managers who are already overloaded. They are often thanked at first. Some weeks later, if the concern persists and nothing has moved, the conversation begins to be about them – their persistence, their tone, their tendency to escalate. The unresolved problem stays where it was. The discussion shifts to the one element in the situation that someone present had the authority to act on: the person.
Everyone watching learns from that. They do not stop raising concerns entirely. They raise them in a form that obliges nobody to act: later, more hedged, stripped of the detail that would have required a response. The system fills with correctly recorded observations that change nothing, and from the board’s side it looks like a functioning process.
Oversight that measures what is easy rather than what matters.
Boards receive performance dashboards, and dashboards report what can be counted quickly: waiting times, finances, activity. The things that would reveal a failing service: the quality of escalation, the proportion of concerns that led to a change, what happens to people who raise them, are rarely on the dashboard, because they are hard to measure and uncomfortable to see.
None of these is a matter of attitude. Each is a feature of how authority, consequence and information are arranged. This is the architecture through which an organisation shows what it actually values. And each will produce the behaviour the inquiries describe, reliably, in any organisation built that way, whatever its stated values.
Why the remedy keeps being values work anyway
If the architecture is visible in the evidence, why do the recommendations keep addressing attitudes instead?
Partly because what investigators encounter most directly is how the place feels to the people in it: they interview staff, and staff describes their experience in terms of atmosphere and relationships.
But mostly because of what each kind of remedy asks of whom. A values recommendation asks the staff to behave differently: to speak up, to listen, to lead with compassion. It is adoptable by the organisation as it stands. An architectural recommendation asks the people with authority to give some of it up: to move decisions closer to where their consequences land, to make raising a concern cost the raiser less and the ignorer more, to put on the board’s dashboard the things that would embarrass the board.
Only one of those can be accepted without anyone senior losing anything. It is not surprising that it is the one most often chosen, and it is not a matter of bad faith. It is simply the arithmetic of who is asked to pay.
What would actually change something
Three things, none of which is a program.
Put the authority where the accountability is.
For any role that answers for safety, ask whether the person can alter what produces it: the staffing, the equipment, the pathway. Where they cannot, either move the authority to them or move the accountability to whoever holds it. A named owner without authority is not ownership. It is exposure.
Make raising a concern cheaper than ignoring one.
At present, the costs run the other way. A concern should create an obligation on someone identifiable to respond, within a defined time, with a record of what they decided and why. And the treatment of people who raise concerns should be something the board actually sees, not whether a policy exists, but what happened to the last several people who used it.
Change what the board measures.
Add to the dashboard the proportion of raised concerns that led to a decision, the time taken to act on them, and how often the same issue is raised twice. These are readable; they cannot easily be managed, and they reveal what the current measures conceal.
The finding that matters
The inquiries have been right about what they found. The word they used for it named the behaviour accurately and located its cause in the wrong place, and the remedies followed the word.
Fifteen years of values programs have produced a health service in which almost everyone can describe the right culture, and which keeps producing the wrong behaviour. That is not evidence that culture change is hard. It is evidence that the programs were aimed at what the organisation says it values rather than at the architecture through which it shows what it actually values.
Behaviour changes when that changes. Unlike a statement of values, an architecture can be redesigned by people with the authority to do it, if they are prepared to use it.