# Why clinics fill with the wrong patients > Pre-admission clinics fill with low-risk patients because slots are allocated in the order bookings arrive rather than by screened risk. The result is a clinic of largely well patients while the frail patient with poor functional capacity gets a short phone call. The constraint is allocation, not capability, so adding clinic capacity does not fix it. *Source: https://getpathways.ai/insights/why-clinics-fill-with-the-wrong-patients* Ask a perioperative lead what their pre-admission clinic needs and the answer is usually more of it — more slots, more nursing hours, another consultant session. It is a reasonable answer to the wrong question, and it is why services that successfully expand their clinic often find the day-of-surgery cancellation rate barely moves. ## The clinic is not the bottleneck Consider what a pre-admission clinic actually is: a fixed quantity of skilled clinical attention, allocated among a much larger booked population. The clinical work done inside it is generally excellent. The problem is upstream of the appointment. Without a mechanism for deciding who should attend, slots are allocated by the only signal available — the order in which bookings arrive. That signal is uncorrelated with risk. The result is predictable and near-universal: a clinic substantially populated by ASA 1 and 2 patients for whom the visit changes nothing, while a frail 82-year-old with poor functional capacity is handled by a fifteen-minute phone call because the diary was full by the time their booking landed. Adding capacity to a system allocating by arrival order produces more appointments distributed the same way. It is genuinely useful — more patients are seen — but it does not preferentially reach the patients whose assessment changes an outcome. ## Why the obvious filters do not work Services that recognise the allocation problem usually reach first for a triage rule based on something already in the booking data. Two candidates present themselves, and both disappoint. **ASA class** is the most common. It is appealing because it is already recorded and it is a clinical judgement rather than an administrative one. But [ASA class](/glossary/asa-physical-status) describes systemic health independent of the procedure, and it is coarse. A large share of the booked population is ASA 2, which includes both a well-controlled hypertensive undergoing a cataract and a patient with a BMI of 38, untreated snoring and a hip replacement ahead of them. Triaging on it fills the clinic with ASA 2 patients, which is roughly where it started. **Procedure type** is the other. It correctly identifies that a hip replacement carries more physiological demand than a cataract, and it is completely blind to the patient. It routes healthy patients having major surgery into clinic and sends unwell patients having minor surgery home. Neither fails because it is a bad measure. They fail because they are single measures, and perioperative risk is a profile. ## What the routing decision actually needs The information that would let a service allocate correctly is not exotic. It is a structured history: current medications verified rather than recalled, functional capacity asked as a concrete question rather than a general one, a sleep apnoea screen, frailty indicators, glycaemic control, and the specific red flags that change an anaesthetic plan. That information exists. The difficulty is that collecting it has historically required the very clinic appointment being allocated — which makes the assessment its own prerequisite. You cannot triage into the clinic on information only obtainable inside it. This is the circularity that keeps the allocation problem in place, and it is the thing worth attacking. Not the size of the clinic; the order of operations. ## Breaking the circularity If a structured history can be taken before the clinic — remotely, at scale, without consuming clinical time — then the routing decision has something to work with. Every booked patient carries a screened profile, and the clinic's finite capacity can be pointed at the patients whose profile warrants it. The clinic does not shrink and its work does not change. What changes is the population inside it. Anaesthetist review goes to red-tier profiles, clinic slots to the moderate tier, telephone review to patients who need a conversation rather than an examination, and everyone else proceeds. ## The second-order effect There is a consequence that tends to surprise services, and it is arguably larger than the throughput gain. Once every booked patient is screened at the point of booking, a set of interventions that were previously impossible become schedulable: [anaemia correction](/glossary/preoperative-anaemia), glycaemic optimisation, [prehabilitation](/glossary/prehabilitation) for patients with low functional capacity. These all require weeks of lead time, and none can be initiated from a clinic appointment three days before surgery. A clinic allocating by arrival order cannot deliver them, not because it lacks the skill but because it meets the relevant patients too late. Screening early turns the wait for surgery from dead time into a window — which is the premise [enhanced recovery](/glossary/enhanced-recovery-after-surgery) programmes are built on, and the reason they are hard to run without it. ## The question worth asking "Do we have enough pre-admission capacity?" is not the diagnostic question. A better one is: of the patients who attended clinic last month, for how many did the visit change the plan? Where that proportion is low, the constraint is allocation. More capacity will be absorbed and the cancellation rate will hold roughly steady, because the patients driving it were never in the room.