Hospital Planning Models

Most building types let you design outward from their rooms. A hospital does not. The number of departments, the intensity of their interdependence, and the fact that a wrong early move is almost impossible to undo mean you must commit to an organising diagram first and resolve the rooms inside it. This diagram is often called the parti - the big idea that says, in the crudest possible terms, where the acute departments sit, where the beds go, how the public arrives, how supplies and waste move, and where the building can later grow.

The model does a great deal of silent work. It decides whether your sickest, most equipment-heavy departments - emergency, operating theatres, intensive care, imaging - can cluster tightly together on one well-serviced level (the so-called hot floor), or whether they end up scattered and linked by long, slow journeys. It decides how much daylight reaches inpatients and staff. It decides whether the structural and servicing systems can run cleanly or have to bend around an awkward shape. And, crucially, it decides whether the hospital can expand in twenty years without demolishing itself - because medicine changes faster than buildings, and a hospital that cannot grow becomes obsolete while it is still new.

> Choose the model badly and you bake in crossed flows, dark wards, starved services and a building that cannot change - flaws no amount of later detailing will cure.

Think of the model as the skeleton and the departments as the organs. You would not design a body by perfecting a kidney and then asking where the ribs go. You establish the frame - its proportions, its circulation spine, its capacity to carry load and services - and then let the organs find their right places within it. The rest of Module 1 is that sequence in full: the model here, the departments and their adjacencies next (Lesson 1.2), the segregated flows that thread through it (1.3), and the grid that makes it buildable and adaptable (1.4).

Hospital models are not a rigid taxonomy - real hospitals blend them - but four recurring families are worth knowing by name, because each is a distinct answer to the nearness-daylight-growth problem.

Block, or podium-and-tower. The heavy, deep, equipment-dense departments (emergency, imaging, theatres, labs, kitchens) occupy a wide, low podium; the repetitive inpatient wards stack in a slender tower above it. Architects sometimes call this 'a matchbox on a muffin' - a thin ward slab on a broad diagnostic-and-treatment base. It is compact, it keeps the hot floor tight, and it suits dense urban sites. Its risks are daylight to the deep podium and the vertical dependence of the whole hospital on a few lift cores.

Pavilion. The oldest model - separate low blocks, historically to give each ward cross-ventilation and sun and to isolate infection before antibiotics. It gives superb daylight and natural air and is easy to build in phases, but it sprawls across a large site and makes journeys between departments long, which is why pure pavilion plans are rare for acute hospitals today and commoner on generous campuses.

Street, or mall. A long internal 'hospital street' - a top-lit public spine - with departments branching off it like shops off an arcade. Wayfinding becomes intuitive (find the street, then the door), and the hospital can grow by extending the street or adding new wings to it. It needs length and width of site, and the street itself is a large piece of circulation to build and condition.

Racetrack (and its cousin, the nucleus). A planning pattern for the ward and for deep departments: a continuous double-corridor loop wraps a serviced central core, with rooms on the daylit perimeter and clinical support, lifts and dirty utilities in the windowless middle. It is efficient and keeps travel short, but inner rooms get no window, so it is used where that is acceptable. Most large hospitals are a hybrid - a podium-and-tower massing, organised along a street, with racetrack wards in the tower.

No model is 'correct'; the right one falls out of the project's constraints. Four questions usually decide it.

How big is the site, and how dense is it? A tight urban plot in a city centre pushes you toward the compact podium-and-tower - you build upward because you cannot build outward. A generous peri-urban or rural campus invites a lower, more spread street or pavilion arrangement, with room on the ground for the hot floor and for future wings.

What is the climate? In much of India, daylight and natural ventilation are assets to capture and harsh afternoon sun and monsoon a thing to manage. A shallow-plan, naturally-ventilated pavilion or street ward can reduce mechanical load in ambient areas; a deep, sealed podium depends heavily on its services and standby power. The model you pick quietly sets the hospital's lifelong energy bill.

How large and acute is the hospital? A small primary or community health centre does not need, and cannot afford, an elaborate model - a simple, clear, single-storey block is often right. A large acute or super-specialty hospital must concentrate its emergency-imaging-theatre-ICU hot floor, which favours models that let those departments share a level and a services strategy.

How will it grow and change? This is the question most often underweighted and most expensive to get wrong. A good model has obvious, pre-planned directions to expand - the street extends, a new tower rises on a pre-built podium, a wing is added - without cutting off daylight, blocking the ambulance route or forcing demolition. Ask of every model: where does the next phase go, and what does building it do to the hospital running beside it? Treat any numbers you are handed - efficient depths, travel distances, bed-per-floor targets - as typical planning guidance to test against this brief, this site and the current code, not as universal rules.

Whatever massing you choose, the move that makes a hospital actually work - and keep working for decades - is to lay a disciplined planning grid under the whole thing. A grid is simply a repeating dimensional rhythm of structural bays and, above the ceilings, a matching zone for services. Get it right and departments of very different kinds - a ward, a lab, an office - can occupy the same bays; partitions can move; a department can expand into its neighbour; and the services can be distributed and maintained along clear, repeating routes.

The grid is what lets the model be more than a frozen sculpture. A podium-and-tower whose tower floors share one grid can be re-partitioned from a twelve-bed bay into single rooms, or from a ward into offices, without touching the structure. A street plan whose wings all step off the same grid can grow a matching wing years later. Without a grid, every change becomes a demolition.

This is the deep link between this first lesson and the last one in the module (1.4): the planning model is the visible shape, the grid is the invisible discipline that keeps the shape alive. The two are chosen together. A good healthcare architect does not fall in love with a picturesque massing and then fight the structure and services into it; they develop the model and the grid as one idea, so that nearness, daylight, flows, buildability and future change all resolve on the same set of lines.

The specific dimensions of that grid - the bay size, the floor-to-floor height, the depth of the service zone - are engineering decisions made with your structural and MEP engineers against the equipment, the spans and the current code. What matters at this stage is the principle: choose a clear model, and put a rigorous, generous grid beneath it. Everything the rest of the course builds - the departments, the flows, the systems - lands more easily on a building that was organised this way from its first diagram.