Clinical Preference Card

Surgeon preference cards do not formally exist. The surgical team knows from experience what each surgeon wants for each procedure, but nothing is documented. A new scrub tech assigned to an unfamiliar surgeon must ask colleagues or guess what to pull. When a surgeon is not satisfied with the supplies opened for a case, the team scrambles to find alternatives mid-procedure.

None — AI cannot optimize surgical supply preparation, predict case-level demand, or identify waste because no formal preference card records exist.

Preference cards exist as informal lists — printed sheets taped to cabinet doors or saved in word processing documents. Cards vary in format and completeness. Some list every item with quantities; others list categories without specifics. Cards for the same surgeon may differ between OR suites because each team maintains its own version. There is no single authoritative source for any surgeon's preferences.

AI can display the preference lists that exist, but cannot reliably calculate case-level supply needs because card formats are inconsistent, quantities may be missing, and multiple conflicting versions exist for the same surgeon and procedure.

Preference cards follow a standardized format in a centralized system. Each card specifies the surgeon, procedure type, and a complete item list with quantities, required/optional designation, and acceptable substitutes. The OR team can pull a definitive preference card for any surgeon-procedure combination. But cards are static documents — they are not linked to actual supply consumption records or case outcomes to validate whether the listed items are actually used.

AI can calculate precise supply needs for any scheduled case using standardized preference cards. Can identify items that appear on multiple surgeons' cards for the same procedure. Cannot optimize cards because there is no linkage to what was actually consumed versus what was opened.

Preference cards connect to actual utilization records. Each card links to consumption data from completed cases showing what was opened versus what was used, which items were substituted, and what additional items were requested. The supply chain team can query 'show me Dr. Smith's total knee preference card usage rate — what percentage of listed items are actually used across the last 50 cases?'

AI can perform preference card optimization — identifying items with low utilization rates, recommending quantity adjustments based on actual consumption, and suggesting standardization opportunities where multiple surgeons use equivalent items for the same procedure.

Preference cards are schema-driven entities with full relational modeling. Each card links to the surgeon's profile, CPT procedure codes, item master records with current pricing, historical consumption analytics, case-level cost comparisons, and clinical outcome measurements. An AI agent can navigate from any preference card to the complete clinical, financial, and operational context.

AI can autonomously manage preference card programs — recommending evidence-based card updates, calculating cost impact of proposed changes, identifying value-analysis opportunities, and generating surgeon-specific utilization reports that correlate supply choices with outcomes.

Preference cards are real-time adaptive intelligence. Every surgical case updates the card's utilization profile — items used, items wasted, substitutions made, additional items needed. Cards evolve continuously from real-time case data, automatically surfacing optimization opportunities and adapting to changing practice patterns as surgery progresses.

Fully autonomous preference card intelligence — continuously optimizing card content from real-time case utilization, recommending evidence-based changes, and maintaining cards as living documents that evolve with surgical practice.

Ceiling of the CMC framework for this dimension.