How hard a health system is to run.

Healthcare has always measured size. Beds, revenue, headcount. The Healthcare Complexity Index measures difficulty.

0 to 100 · nine dimensions · always against a stated peer set

0 100 0 Complexity score Against a stated peer set ILLUSTRATIVE
The premise

Same size. Different job.

System A System B 820 beds · $4.1B · 19,400 staff 820 beds · $4.1B · 19,400 staff 1 entity1 state12 payer contracts 31 entities9 states47 payer contracts ILLUSTRATIVE

Two systems can match on every number a board quotes and differ several-fold in how hard they are to hold together.

No benchmark in use can see the difference. So the argument is won on seniority, or lost annually.

“We are above market on pay and I cannot defend it.”
CHRO
“Is it genuinely hard, or are we bad at this?”
CEO & board
“Admin cost is above median. Cut what?”
CFO
“Every metric calls our coverage waste.”
Rural & multi-state
“Our peer group no longer resembles us.”
Post-acquisition
Why it matters

What changes when a market gets a number.

Markets do not move on better opinions. They move when a contested judgment becomes a number anyone can check.

Before · assertion “It is harder here.” After · position “Against these peers, we sit here.”

Consumer credit

Lending once rested on relationship and reputation. One shared score meant any lender could assess the same borrower the same way.

Film and television

Quality was argued title by title. A single public rating gave the whole market one comparable signal to anchor on.

Healthcare’s own precedent

Case Mix Index did this for acuity. “Our patients are sicker” stopped being an assertion and became a filed number.

The parallel is about the function of a shared, independent number, not what it measures. A credit score assesses risk. The Healthcare Complexity Index assesses no quality, creditworthiness or performance of any kind. It describes difficulty, relative to a stated peer set.

How it is built

Records that already exist, to one comparable score.

No surveys. No opinion inputs. A second analyst can re-derive every score independently.

Step 01 Count existing records Step 02 Normalise percentile rank Step 03 Nine dimensions equal weight Step 04 0 / 100 peer set stated ILLUSTRATIVE
Countable
Every indicator computes from records already held. No bespoke surveys.
Peer-relative
No score means anything alone. The peer set travels with it, always.
Equal-weighted
No contestable assumption about which dimension matters most.
Traceable
Every score resolves to the records and formulas that produced it.
Reading a score

One number. Nine drivers underneath it.

010203 040506 070809
Nine dimensions, equal weight. Names withheld pending disclosure decision. Illustrative
Less complex More complex Score 0

Against every US health system in the edition. Answers: how complex are we in national terms?

Both readings are valid and answer different questions. A score published without its peer set is invalid.

See it applied

Follow one executive question.

The Index does not answer it for them. It gives them a defensible position to answer it from.

◎ The Index · measures difficulty
◫ The leader · makes the decision
01 · THE QUESTION

02 · WHY IT STALLS

03 · WHAT THE INDEX ADDS

04 · THE NEW SENTENCE

05 · WHERE IT STOPS

The Index measures difficulty · the leader makes the call
Figures in examples are illustrative. The peer set is stated in every case.
One permitted reading

It finds the line. You decide what crosses it.

Cost above peer median One undifferentiated block Explained by structure Not explained by structure ILLUSTRATIVE

Neither side is coloured as good or bad, because the Index does not declare which side any particular expenditure falls on. That judgment stays with the leader. It simply becomes a judgment rather than a guess.

Read it correctly

What the Index is not.

Every score is comparative and peer-relative. It is never an absolute judgment of a system’s quality or worth.

A verdict

A high score does not indicate poor management. A low score does not indicate efficiency.

A quality score

It does not measure outcomes, predict them, or rate clinical performance.

A ranking of who is better run

Not a performance rating of individuals or of management.

Complete

It sizes countable structure and leaves the emergent and adaptive to judgment.

An operational tool

It does not tell a system what to do next, and never justifies a pay figure.

Portable between peer sets

Two scores from different peer sets cannot be compared directly.

The Index claims rigour and reproducibility. Both weaken in proportion to what is withheld, so any decision to withhold method is made knowingly, by Cornell, with that cost accepted.

Who owns what

A university owns the method. That is the asset.

Cornell ILR Authors the method Owns interpretationPublication authorityThe voice on meaningNamed first, always Spire.AI OpenKnowra Computes the score Data pipelineThree of nine dimensionsThe voice on methodNo unilateral publication

Without a regulator compelling participation, the Index has no authority except reproducibility and independence.

Published free. Never gated behind a commercial action. The commercial relationship is stated openly rather than disguised.

Conceptualised by Cornell’s Institute for Compensation Studies. Powered by Spire.AI OpenKnowra.

Who reads it, and for what

One number, six conversations.

Chief HR officersWhether reward benchmarks set on scale misstate this system’s burden.Not a pay figure
Chief financial officersWhich part of a cost gap is explained by structure, and which is not.Identifies the split only
Boards and CEOsDifficulty separated from competence, so the two can be argued apart.Not a verdict
Workforce plannersWhich dimensions drive a position, and how it moves across editions.Direction, not just level
PolicymakersWhere structural burden concentrates across hardest-to-serve populations.Evidence, not advocacy
Researchers and analystsA documented construct that can be examined, contested and re-derived.Method published

Complexity, not size.

How hard a healthcare system is to run, not how big it is.

Countable structure + Percentile rank + Stated peer set = One dimension score

Nine dimension scores + Equal weight = Healthcare Complexity Index

Conceptualised by Cornell’s Institute for Compensation Studies. Powered by Spire.AI OpenKnowra.