The Top 10 You Cannot Name.

In every hospital strategy document I have been handed in the past five years, the patient appears as an abstraction. “The elderly patient.” “The complex surgical case.” “The multimorbid admission.” Sometimes the strategy reaches the specificity of “the elderly patient with three or more chronic conditions” — which is a population description that still contains, in practice, somewhere between fifteen and forty genuinely distinct operational cohorts. The strategy is written at the level of abstraction that feels appropriate for strategy. The operational reality exists at a level of specificity the strategy never reaches.

The ORBIT layer of hospital operations exists to address this gap. Not as patient-flow methodology, not as case-mix analysis, not as population health. As the discipline of naming — specifically, by diagnosis cluster, by pathway, by referral source — the ten patient cohorts that a department actually serves. A hospital department treats perhaps two thousand distinct presentations annually across the ICD-10 taxonomy. Roughly ten of those presentations, grouped by operational similarity, account for about 80% of the case volume. Naming those ten is not an analytical exercise. It is the foundational act of operational strategy. Without it, every downstream decision — staffing, capital, pathways, coding, referral relationships — is calibrated against an imagined patient rather than a real one.

This is the observation at the foundation of every ORBIT-layer conversation: if the leadership team of a department cannot name its top ten cohorts, specifically, by volume and by margin, the department is not running a strategy. It is running an average.

What naming the ten cohorts actually means.

The naming discipline has a specific output. Not a slide deck. Not a dashboard. A single-page document, maintained quarterly, that lists the ten operational cohorts the department serves, with four specific attributes each: annual case volume, DRG or procedure family the cohort most frequently generates, average margin per case after allocated costs, and the named clinical lead who is accountable for that cohort’s pathway.

For a geriatric department, the ten typically cluster as hip fracture with co-managed geriatric complex treatment, pneumonia in multimorbid elderly, acute-on-chronic heart failure, acute stroke with geriatric rehabilitation, COPD exacerbation, urinary tract infection or urosepsis, cardiac arrhythmia management, acute delirium or confusional state, recurrent falls workup, and post-operative general medical co-management. For an internal-medicine service, the ten cluster differently — probably heart failure, COPD, pneumonia, UTI, GI bleed, DKA, pulmonary embolism, acute coronary syndrome, liver decompensation, and acute kidney injury. For a surgical service, the ten are driven by procedure categories rather than diagnostic categories. Every department has its own ten.

What makes the discipline rare is not the analysis — the data to compile the list exists in every hospital’s coding records. What makes it rare is the commitment to stop updating the list once it is written. The single-page document becomes the operational reference point. Capital decisions are evaluated against it. Staffing is calibrated to it. Pathway redesign starts from it. Controlling reports reference it. The ten cohorts become the grammar of how the department talks about its own work.

Why most hospitals have never done this.

Three structural reasons keep this from happening.

The data lives in the wrong place. Case-mix data sits in the coding department. Clinical pathway ownership sits in the medical directorate. Margin data sits in controlling. No single organisational function has all three, and no one has the mandate to assemble them into a single-page cohort inventory. The data is not missing. The data is fragmented across functions whose routines do not intersect.

The exercise reveals uncomfortable prioritisation. A cohort inventory, honestly compiled, ranks patient groups. Some cohorts will have higher margin than others. Some will have higher clinical complexity. Some will have higher readmission rates. The leadership team will need to decide which cohorts the department optimises for and which it accepts as adequate — a decision most medical directors have been implicitly avoiding their entire career. Naming the ten forces the conversation about priorities that the medical directorate has been able to postpone as long as the strategy was expressed in averages.

It breaks the illusion of comprehensive excellence. Hospital leadership teams, quite reasonably, do not want to be seen publicly admitting that some patient cohorts receive better operational attention than others. The naming exercise, done properly, produces exactly that admission — privately, in the operations meeting, as the basis for real decisions. The discomfort of this admission is why most hospitals do not do the exercise. The productivity of this admission is why the hospitals that do it outperform.

The power-law problem.

When a department does compile the cohort inventory, the result is almost always the same shape: a steep power-law distribution. The top cohort accounts for perhaps 20% of case volume. The top three account for 40%. The top ten account for roughly 80%. The remaining cases are distributed across a long tail of perhaps two hundred smaller groupings, each too small to justify a dedicated pathway but collectively non-trivial in aggregate volume.

The power-law structure has an operational consequence that most leadership teams have not internalised: effort invested in improving the top ten cohorts has roughly four times the operational impact of identical effort invested in the long tail. A pathway redesign for the cohort ranked first, if it improves length-of-stay by 10%, affects perhaps 400 admissions annually. The same pathway redesign for a tail cohort affects perhaps 10. Both are worth doing; they are not worth doing with the same level of leadership attention.

This is the argument for naming: without the inventory, leadership attention is distributed roughly uniformly across the case mix, which is a quarter as efficient as distributing attention proportional to cohort volume. The inventory does not change how patients are cared for — every patient still receives the full clinical standard. The inventory changes how operational improvement effort is prioritised.

What the Schlüchtern ORBIT work showed.

At Main-Kinzig-Kliniken Schlüchtern, the geriatric department built and has maintained the ten-cohort inventory since early 2020 as one of the foundational moves of the operational restructuring. The inventory is reviewed quarterly at the department operations meeting. Pathway improvements over the subsequent five years have been explicitly allocated across the top cohorts rather than distributed uniformly. Under the research programme with Prof. Dr. Rainer Sibbel at Frankfurt School [1,2], the operational outcomes are summarised below; what is most relevant for the ORBIT argument is that the improvements happened because leadership knew which cohorts to concentrate on.

The pattern replicates across specialties. An internal-medicine service that names its top ten makes different capital decisions than one that does not. A cardiology service that knows its top ten staffs call rotations differently. A surgical service that names its top ten negotiates with the ambulance service differently — which is the subject of a later post in this series.

What to do on Monday.

Commission the inventory. This is a two-week exercise for a competent Medizincontroller with access to coding records and cost-allocation data. The output is a single-page document listing the ten largest operational cohorts by annual case volume, with DRG family, average margin, and current clinical pathway owner for each.

Review the document in the next medical-directorate meeting. Do not accept vague aggregates (“the elderly patient group”); require specific clinical clustering by discharge ICD-10 or procedure family. The conversation the document provokes is uncomfortable by design: which cohorts are we structurally good at, which are we adequate at, and which are we losing on? The honest answers to these three questions are the starting material for actual operational strategy.

Assign a named clinical lead to each of the ten cohorts. This is not a new role. It is a formalisation of responsibility that in most hospitals is currently implicit. The lead owns the pathway for that cohort, including length-of-stay, discharge readiness, coding capture, and readmission pattern. Their name goes on the single-page document.

Review the inventory quarterly. Cohort composition shifts as clinical practice evolves and as referral patterns change; a document reviewed once and then filed is worth less than one kept alive in the operational rhythm. The rhythm is the discipline; the document is the artefact.

A department that has done this exercise operates at a different level of operational specificity than one that has not. When the Chefarzt sits down with the Geschäftsführung to request capital or staffing, she can answer the specific question of which cohort the request serves and what the expected volume and margin impact will be. The request becomes evaluable. Decisions become faster. Resources flow toward cohorts where the impact is concrete rather than toward cohorts where the case was simply better presented.

Most hospitals optimise for imaginary patients because the alternative — naming the ten real ones — requires admissions that leadership has been deferring. The top ten are not missing from your hospital. They are already there, already treated, already generating 80% of the case volume. They are just unnamed. The whole ORBIT layer begins with naming them.