Two hospitals in Zeeland agree they must cooperate to survive. The boards agree. The insurer agrees. The region agrees.
In September 2025, Adrz and ZorgSaam presented a plan to divide care between their locations. Kidney-cancer surgery to Goes. Breast-cancer surgery and heart catheterisation to Terneuzen. Neurology organised jointly. Care would move in both directions.
By mid-2026, Follow the Money reported that little of that plan had been realised. Medical specialists at ZorgSaam had dug in, concerned that Zeeuws-Vlaanderen would lose a full hospital. The insurer's director publicly called the situation embarrassing for Zeeland's citizens. What still stands is the kidney-cancer step — built around a surgical robot that can only be in one place. Hardware forced that decision. Governance did not.
This is not a story about two boards that failed. Each did what a board is built to do: represent its own organisation and bring its professionals' concerns to the table. Neither had the authority to end the disagreement — because in a cooperation, nobody does. That is the subject of this article.
The trade press offered the usual explanations: emotions, interests, urgency not yet high enough. All true. None of them explains why the outcome is so predictable. The sector describes the same difficulty in general terms: in June 2026, after visiting seven oncology care networks, the IGJ found professionals strongly committed to appropriate care while the organisational and financial basis of those networks remained fragile. In my own work with boards, I meet the pattern repeatedly — different regions, different people, different relationships, the same result.
When a failure repeats regardless of who is involved, the cause is rarely the people. It is the structure.
The mechanism nobody thinks about
Inside a single organisation, a disagreement has somewhere to go. Two department heads who cannot agree both report, eventually, to someone with the authority to decide.
Most conflicts never travel that far. Both parties know they could. The existence of a decision point disciplines every negotiation below it. People compromise and move on — not because they are reasonable by nature, but because prolonging the fight has a predictable end: someone above them will settle it, and neither side controls how.
Hierarchies are not valuable because they create agreement. They are valuable because they give disagreement a legitimate way to end.
This mechanism is so reliable that nobody thinks about it. It appears in no governance code. It works silently, the way a foundation works — noticed only when it is missing.
What cooperation deletes
When two hospitals cooperate, that mechanism does not exist. The people on either side share no common superior at any level. Every effective organisation contains a final point of legitimate authority. A cooperation, by default, contains none.
Cooperation deletes the decision point. Nobody notices. Because nobody realised it was carrying the system.
"But we have a joint steering group." A steering group is a table, not a decision point. It can decide only when both sides already agree — which is exactly when nobody needs it. The question is not whether a shared body exists. The question is whether it can end a conflict the parties genuinely disagree about, and whether both organisations will accept the outcome as binding.
Usually it cannot. The script is recognisable. Nobody says no; meetings simply end without a decision. Issues circle — raised in the steering group, referred to a working group, returned to the steering group. "Alignment" becomes a permanent agenda item.
Meanwhile, each hospital's own hierarchy does exactly what it was built to do: it backs its own people. A board that hears its specialists' concerns carries those concerns to the table. It cannot settle them there. Two well-functioning hierarchies, facing each other, produce a deadlock that neither can end.
Even the party with real leverage stands outside the structure. An insurer can threaten to contract a treatment at one location only. That is pressure on the deadlock, not a route through it.
Hospitals already contain this gap
It would be comfortable to treat this as a network problem — something that happens between organisations. It is closer to home.
In most Dutch general hospitals, a large share of the medical specialists are not employees. They are organised in a medisch specialistisch bedrijf that contracts with the hospital. A board that wants to move care does not instruct its specialists. It negotiates with them.
A hospital, in other words, already contains a governance interface before it ever joins a network. That is why cooperation deadlocks so often wear a clinical face. Specialists who dig in are not showing a character flaw. They are a group with real power, real concerns, and no structured place to decide them. A cooperation between two hospitals with two medical staffs is not one interface. It is at least three.
Writing it down is not distrust
Cooperation agreements tend to be precise about intentions — which care moves where, which volumes, which timelines — and silent about authority. Who may put a change on the table? Who must respond? And when the parties disagree: where does the conflict go?
The silence is no accident. Writing down authority feels like distrust. We cooperate because we trust each other; specifying who decides would signal that we don't.
That instinct is exactly backwards. Unspecified authority does not produce equality. It produces permanent negotiation, in which every question is settled by seniority, persistence and personal relationships — and every settlement lasts until the next scarce resource. That negotiation, not the written agreement, is what burns through the goodwill a partnership runs on.
Trust starts cooperation. Clear authority keeps it alive.
Building what the network lacks
Inside one organisation, the decision point comes free with the hierarchy. In a cooperation, you must build it. There are several ways to do this: a joint board with real authority over defined questions; an escalation ladder both boards commit to in advance, ending somewhere binding; arbitration agreed for named types of conflict — before any conflict exists.
The form matters less than the property all forms share: a conflict that cannot be resolved at the table has a known, pre-agreed place to go. And both sides knew it when they signed.
Such a mechanism is rarely used. Like its counterpart inside the organisation, it works mostly by existing. Negotiations conclude in the shadow of a real decision point. Negotiations without one circle.
Three questions for any board in a cooperation
For every cooperation your organisation depends on, three questions tell you whether the decision point exists.
Who may initiate — which party can put a change on the table in a way the other must take up?
Who must respond — within what time, with what obligation?
Where does deadlock go — and is the outcome binding?
If the honest answer to the third question is "we would work it out together," the cooperation is running on goodwill — and look at who "we" turns out to be. Questions the boards leave unanswered do not go away. They travel downward, and become the daily negotiation of the people at the interface, who carry the stress of a conflict they were never given the authority to end.
Boards know how to govern organisations. The next challenge is learning how to govern the spaces between organisations. That requires something every organisation already possesses internally — and every network must consciously create:
a legitimate place where disagreement can end.
Sources
Follow the Money, Hoe landelijk afgedwongen samenwerking tussen Zeeuwse ziekenhuizen compleet escaleerde.
Inspectie Gezondheidszorg en Jeugd (Dutch Health and Youth Care Inspectorate), review of oncology care networks, June 2026.
Author's note: I write from the perspective of governance and organisational design, not from inside hospital practice. My work concerns how authority and accountability travel through organisations and networks. Dutch hospital care is a sharp example of this, not an exception: wherever execution depends on cooperation across organisational boundaries, the place where disagreement can end disappears.
About the author: Robert Teunissen founded CaC Management in Geneva in 2007. He has more than thirty years' experience advising executive boards and supervisory boards on governance, organisational design and leadership — including at the World Health Organization, the European Commission, Nestlé and Swiss Life. More at cac.management.
