The seven rooms you cannot see
A map of the closed rooms where healthcare's decisions actually get made, and how each one reaches the bedside.
A conference room in an office park, mid-afternoon, two teams on opposite sides of a long table. One represents a health system, the other a large insurer. They are negotiating unit prices, the rates the insurer will pay for a list of procedures, and the work is slow and unglamorous, conducted in spreadsheets and redlines. By the end of the day they will agree on a number. The word “nurse” is never said aloud in the room. And yet the number they settle on will decide, eighteen months from now, how many nurses are scheduled on a given floor on a Tuesday night, and how much room the budget leaves for anything that is not strictly required.
The people who will live with that number are not in the room. The nurse on that future Tuesday night is not here. Neither is the patient whose call light she will be slower to answer because the floor is one body short. The decision that shapes their evening is being made, reasonably and in good faith, by people who will never meet them.
This is the first thing to understand about healthcare’s closed rooms. They are not sinister. The decisions made in them are rarely malicious and almost never hidden. They are simply closed to the people whose work and lives are shaped most directly by what happens inside them. The teams at the table, often on both sides, have rarely been in a hospital room except as a patient or a visitor.
There is another kind of room entirely. The operating room at three in the morning. The intensive care unit. The recovery floor. These are rooms about the patient, where the most human parts of medicine happen and where the decisions that matter most to a single person get made in real time, by the people standing over the bed. Most of us can enter one kind of room or the other. Almost no one sees both.
I have spent more than two decades moving between them, the clinical rooms and the administrative ones, and I have come to think of the administrative side as seven rooms. Some readers know these rooms perfectly well and have chosen the bedside on purpose; what follows is not an argument that the rooms are where the better work happens, only that what gets decided inside them reaches the bedside whether or not the bedside ever sees the room. Here is the map.
The contracting room
We have just been in the first room. It is worth staying a moment longer, because it shows the pattern the other six repeat. Two things make it consequential. The first is who is absent: the negotiation has a chair for the system and a chair for the payer, and not one for the nurse, the patient, or the clinician whose Tuesday night the number will shape. The second is time. The negotiation itself is often a series of meetings stretched over weeks or months, where a change made at the end can quietly undo something agreed at the beginning, and the gap between the final decision and its effect runs another twelve to eighteen months. By the time the floor feels the squeeze, almost no one traces it back to a number agreed a year and a half earlier in an office park. The decision and its consequence are severed cleanly enough that the room is rarely held to account for either.
The credentialing committee
The second room decides who is allowed to do the work. The physicians already on the medical staff review the files of the physicians, nurse practitioners, and physician assistants who want to join them, and they hold the authority to say yes or no. It is a necessary function and, done well, a careful one; the rigor is the reason the work carries the trust it does. The friction is that the process is structurally slow, gated by primary-source verification and payer enrollment as much as by the committee’s own calendar, and unforgiving of timing. A fully qualified clinician with a clean file can miss a cycle over an administrative detail that has nothing to do with competence, a license that transferred a few days late, a signature that landed after a cutoff. Each name that does not clear in time is a name that is not at the bedside the following month, and in a market already short of clinicians, the cost of that delay falls on the patients who wait and the colleagues who cover the gap. Most patients never picture this room, and it helps determine who is there to care for them.
The executive office
The third room is where workforce planning meets arithmetic. At some point every system runs into the same wall: there are not enough trained clinicians to fill the roles, and the gap is widening as the population ages.
The story most people hear is that fewer young people want clinical careers. The actual problem is closer to the opposite. The binding constraint is training capacity. The 1997 Balanced Budget Act capped the number of Medicare-funded residency positions at roughly their 1996 level, where they stayed essentially frozen for more than two decades until Congress added the first new slots in 2021 and 2023 (AAMC). The pressure shows in the Match: in 2025, more than 47,000 active applicants competed for roughly 40,000 first-year positions, and about 9,500 of them did not match into one (NRMP). On the nursing side, the American Association of Colleges of Nursing reported that schools turned away more than 65,000 qualified applications in 2022 for lack of faculty, clinical training sites, and budget (AACN). That is a different problem than the one most leaders are solving for, and it has different solutions.
The peer review committee
The fourth room is where the clinical and the administrative meet. When a hard case goes to peer review, a physician’s work is examined by colleagues behind a closed door, and the lessons of that case do not stay with the clinician; they become the standard the whole department is held to afterward. A single difficult airway, reviewed honestly, can change the preoperative checklist every colleague follows the next morning. This is clinical governance at the case level, and it is closed for good reasons, including fairness to the clinician and candor in the discussion. It is consequential precisely because the rest of the department inherits whatever the room concludes. There are other rooms like it, the morbidity and mortality conference, the privileging committee, the quality review group that reviews nursing care and medication errors, and I will write about each in time.
The compensation committee
The fifth room is where senior executive pay gets set, and most of the decision is made before anyone sits down. A consultant builds the number over weeks against a hand-picked peer group and a target percentile, and a compensation committee of a few board members, usually with backgrounds in business or finance, meets to ratify it, often in under an hour. The deliberation stays inside the room. Even what eventually becomes public can understate the total, since incentive structures, deferred compensation plans, and other legitimate vehicles reduce what is reportable in a single year. The figure that does surface, for a nonprofit system, sits on a tax form almost no one reads, and only for a handful of the highest-paid executives.
I am not arguing the nonprofit construct is broken. I am arguing it has features that let compensation outrun the mission’s reach into the community, and that the conversation we are not having is about the features themselves. The numbers we can see point at the tension. At twenty-two major nonprofit medical centers, the wage gap between hospital chief executives and registered nurses rose from 23 to 1 in 2005 to 44 to 1 in 2015, while inflation-adjusted nurse pay barely moved (Marcus et al., Clinical Orthopaedics and Related Research), and research on nonprofit systems has found that most of the growth in chief executive pay tracks the size of the organization a leader runs more closely than the quality of the care it delivers (PLOS One). One state-level analysis went further still: a North Carolina report found the 2019 pay of eleven nonprofit hospital chief executives equal to the salaries of 572 registered nurses (NC State Health Plan).
The succession room
The sixth room decides who comes next. The question of who replaces a chief executive when she retires gets answered long before the rest of the organization knows there is a question, and the way it gets answered tells you what a board actually values. I have said in rooms like this one that we are successful when those we develop put us out of a job. That line tends to cool a room, because most leaders still measure success by how long they hold the seat rather than by who they have ready to take it.
The deal room
The seventh room is where a community hospital becomes a chain asset. Mergers and acquisitions are how a great deal of American healthcare is now organized, and the logic of a given deal is usually sound on its own terms. Often it is more than sound; the capital and the capability a larger partner brings are sometimes the only thing that keeps a struggling hospital’s doors open, and I have changed my own mind over the years about how much good the right partnership can do. What the model on the table almost never prices is the local discretion that lets a hospital do something for its community that would never clear a system-wide spreadsheet, along with the slower effects on the credentialing pipeline and the staffing ratios this map has been tracing. Decision rights move upward and outward, toward the standard and away from the floor, and that trade is real whether or not anyone in the room names it. Those effects show up eighteen months after the signing, in rooms the deal room never has to sit in.
The force that runs through all of them
There is an eighth space I have deliberately left off the numbered list, because it does not behave like the others. The rulemaking process, where a federal or state agency sets a payment rule through notice and comment, acts on several of the seven at once. A change to how procedures are paid for reshapes the contracting room and the deal room together. A change to how training is funded reaches straight into the executive office. Regulation is less a room you enter than a current that runs underneath the others, and it moves the water in all of them.
What this map is for
A word on what this is not, because the discipline matters and is worth saying aloud so a reader can hold me to it. This is not a map for gossip. It does not name and blame. It will not trade on outrage to keep your attention, and it will not critique former colleagues or employers in public, because that work belongs in private conversation. What it will do is describe, plainly and specifically, what happens in the rooms most readers never enter, and trace the line from each of them to the bedside, where the consequences land. I will credit the people whose work I respect by name. I will sit with complexity rather than pretend it away. And when I change my mind about something I have argued for here, I will say so and explain why.
The reason to draw the map at all is that the people in the clinical rooms and the people in the system rooms are, more often than either side admits, working toward the same thing and unable to see each other do it. The family in a surgical waiting room cannot see the contracting room that set the staffing on the floor where their father will recover. The board in the compensation committee rarely sees the three-in-the-morning version of the hospital it governs. Almost no one in American healthcare can stand in all seven of these rooms and the clinical ones too, and the cost of that is a system whose halves keep deciding each other’s conditions without ever meeting.
The number agreed in that office park takes effect long after anyone remembers the meeting. The nurse who is one body short on a Tuesday night will not know why, but she will be the first to feel it, and her floor is where a year-and-a-half-old number finally becomes legible as a real night with a real patient. The people who set the conditions she works under will never see that floor at the hour it is hardest. Most of these rooms close before the consequence arrives, and open again before anyone connects the two. The slow work of this newsletter is to connect the decision to the floor it lands on, so that the people in the rooms and the people at the bedside are at least describing the same system.
From the room you can’t see, the voice you need to hear.
Each of the seven rooms gets its own piece. Subscribe at pastthedoor.com to read them as they publish.