Stanford Health Care

Mixed evidenceQueue eliminationIntake redesignedStep before the decision

FastFax urgent referral fax queue routing

Referral staff do not have to open every incoming fax before provider-declared urgent referrals can enter an expedited work queue. On supported standardized cover sheets, a classifier can detect the referring provider's checked urgent box and select the queue without making a clinical urgency judgment.

Clinical operations · United States

Collections: Human still decides · Queue eliminated · Regulated autonomy

An editorial scene for Stanford Health Care contrasts faxed an external referral to stanford health care and, when applicable, marked the referral urgent on the cover. with faxes an external referral and determines urgency by checking the urgent box on the stanford standardized cover. in the fastfax urgent referral fax queue routing workflow.

Executive brief

The operating-model shift, in one view.

Stanford inserted narrow checkbox detection into fax intake, moving recognized provider-marked urgent referrals into an expedited human queue without asking the model to infer clinical urgency.

AI value · Recall among marked-urgent referrals using the supported standard cover sheet

Verified

FastFax routed 1,059 of the 1,352 to the urgent queue, reported as 78% recall.

The 78% recall applies only to marked-urgent referrals on the supported standard form. It is not recall across all 2,332 marked-urgent referrals and is not a measure of clinical triage.

Before

Faxed an external referral to Stanford Health Care and, when applicable, marked the referral urgent on the cover. → Moved incoming faxes directly into one shared OnBase intake queue containing both urgent and routine referrals. → Opened and read each fax cover sheet to find referrals that the referring provider had…

After

Faxes an external referral and determines urgency by checking the urgent box on the Stanford standardized cover. → Monitors the RightFax queue and sends each new fax to the Stanford FastFax API running on Google Cloud. → Work the urgent and routine queues, verify and process referrals, and handle false positives,…

Human boundary

The referring provider determines and marks urgency. FastFax detects the mark and selects an OnBase queue. Enterprise Contact Center referral staff own referral review and processing, while clinical schedulers and clinical teams retain downstream scheduling, escalation, and clinical decisions.

Why it matters

A narrow model can materially compress a queue without making the consequential decision.

How the work changed

Before

How the work ran before the change.

Step 1 of 4
Referring provider
Faxed an external referral to Stanford Health Care and, when applicable, marked the referral urgent on the cover.
ControlThe referring provider, not Stanford's classifier, determined whether the referral was urgent.
Step 2 of 4
RightFax and OnBase
Moved incoming faxes directly into one shared OnBase intake queue containing both urgent and routine referrals.
ControlNo automated urgent-versus-routine queue split occurred.
Step 3 of 4
Enterprise Contact Center referral management staff
Opened and read each fax cover sheet to find referrals that the referring provider had marked urgent.
ControlHuman review was the first mechanism for identifying the provider's urgency mark.
Step 4 of 4
Referral staff and clinical schedulers
Processed the referral, handled missing or nonstandard information, and made downstream scheduling and clinical.
ControlOperational and clinical authority remained with Stanford staff.

What changed

Staff should not have to open every fax before a provider-declared urgent referral can enter an expedited queue.

Decision rightAI detects priority; humans retain the consequential decision

After

How the same work runs now.

Step 1 of 4
Referring provider
Faxes an external referral and determines urgency by checking the urgent box on the Stanford standardized cover.
ControlUrgency remains a provider declaration; FastFax does not assess the patient's clinical condition.
Step 2 of 4
MuleSoft connector
Monitors the RightFax queue and sends each new fax to the Stanford FastFax API running on Google Cloud.
ControlThe connector inserts the classification service between RightFax and OnBase.
Step 3 of 4
Enterprise Contact Center referral management staff
Work the urgent and routine queues, verify and process referrals, and handle false positives, unsupported forms.
ControlHumans retain referral-processing responsibility; every fax remains in a human-worked queue.
Step 4 of 4
Clinical schedulers and clinical teams
Make scheduling, escalation, and any clinical decisions after referral processing.
ControlFastFax has no clinical decision rights.

Process model built from the published workflow evidence for Stanford Health Care. Every step, actor, and control appears in full below.

Every step, actor, and control

Exception path

Unsupported or old forms, handwritten urgent notations, and differently positioned checkboxes may not be recognized and remain in the routine human-worked queue. This preserves eventual human review but does not prevent delay for a missed provider-marked urgent referral. Referral staff handle false positives and other intake exceptions; public sources do not document a separate automated failover or downtime protocol.

Decision authority

Before
#	Actor	Action	Control
01	Referring provider	Faxed an external referral to Stanford Health Care and, when applicable, marked the referral urgent on the cover sheet.	The referring provider, not Stanford's classifier, determined whether the referral was urgent.
02	RightFax and OnBase	Moved incoming faxes directly into one shared OnBase intake queue containing both urgent and routine referrals.	No automated urgent-versus-routine queue split occurred.
03	Enterprise Contact Center referral management staff	Opened and read each fax cover sheet to find referrals that the referring provider had marked urgent.	Human review was the first mechanism for identifying the provider's urgency mark.
04	Referral staff and clinical schedulers	Processed the referral, handled missing or nonstandard information, and made downstream scheduling and clinical workflow decisions.	Operational and clinical authority remained with Stanford staff.

After
#	Actor	Action	Control
01	Referring provider	Faxes an external referral and determines urgency by checking the urgent box on the Stanford standardized cover sheet or by using another form or notation.	Urgency remains a provider declaration; FastFax does not assess the patient's clinical condition.
02	MuleSoft connector	Monitors the RightFax queue and sends each new fax to the Stanford FastFax API running on Google Cloud.	The connector inserts the classification service between RightFax and OnBase.
03	FastFax image preprocessor and machine-learning checkbox classifier	Processes the cover-sheet image and detects whether the urgent checkbox is marked at a known location on a supported standardized form.	The model detects a form mark; it does not infer clinical urgency, diagnose, prioritize from clinical content, or override the referring provider.
04	MuleSoft connector and OnBase	Use the API result to place a recognized provider-marked urgent fax in the urgent OnBase queue and other or unrecognized faxes in the routine OnBase queue.	Queue selection accelerates human attention but does not authorize downstream clinical action.
05	Enterprise Contact Center referral management staff	Work the urgent and routine queues, verify and process referrals, and handle false positives, unsupported forms, handwritten urgency, and other exceptions.	Humans retain referral-processing responsibility; every fax remains in a human-worked queue.
06	Clinical schedulers and clinical teams	Make scheduling, escalation, and any clinical decisions after referral processing.	FastFax has no clinical decision rights.

Work that left the path

Opening every supported standard-form fax before a provider-marked urgent referral can be placed in an expedited queue
Mixing all recognized provider-marked urgent referrals with routine referrals in one undifferentiated work queue
Using referral staff as the first-pass checkbox detector on every supported standardized cover sheet

Human role before

Enterprise Contact Center referral staff manually opened every fax in a single mixed queue to find provider-marked urgent referrals, then referral staff and clinical schedulers performed downstream processing and action.

Human role after

Referral staff work separate urgent and routine queues and still verify, process, and resolve exceptions. Clinical schedulers and clinical teams retain scheduling, escalation, and clinical authority. The redesign changes the order in which some faxes receive human attention; it does not transfer clinical judgment to the model.

AI role

FastFax preprocesses fax images and classifies whether a referring provider checked the urgent box at a known location on a supported standardized cover sheet. Its output selects the urgent or routine OnBase work queue. It does not infer urgency from clinical content or determine what care a patient needs.

Outcomes

Recall among marked-urgent referrals using the supported standard cover sheet

Verified

The early operational snapshot contained 1,352 provider-marked urgent referrals using the standard cover sheet.FastFax routed 1,059 of the 1,352 to the urgent queue, reported as 78% recall.

Early post-go-live operational snapshot reported by Stanford SEAL after the December 2022 launch. · Subset of 1,352 standard-form urgent referrals within 26,921 total faxes; excludes the 42% of marked-urgent referrals using unsupported cover-sheet formats.

The 78% recall applies only to marked-urgent referrals on the supported standard form. It is not recall across all 2,332 marked-urgent referrals and is not a measure of clinical triage.

Precision of the urgent queue in the early operational snapshot

Verified

Before FastFax, urgent and routine referrals entered one shared queue, so there was no classifier-created urgent queue.FastFax routed 1,290 faxes to the urgent queue; 1,059 were provider-marked urgent and 231 were nonurgent, reported as 82% precision.

Early post-go-live operational snapshot reported by Stanford SEAL after the December 2022 launch. · 1,290 faxes routed to the urgent queue within 26,921 processed faxes.

Stanford's page labels the 231 nonurgent faxes as an 18% false positive rate, but 231 divided by 1,290 is the false-discovery share, complementary to the reported 82% precision; this record preserves the reported counts and precision without generalizing to clinical accuracy.

Mean queue processing time in the early operational snapshot

Verified

Contemporaneous routine-queue faxes averaged 152 minutes.Urgent-queue faxes averaged 38 minutes.

Early post-go-live operational snapshot reported by Stanford SEAL after the December 2022 launch. · Urgent and routine OnBase queues in the enterprise fax-referral workflow.

This is a contemporaneous comparison between differently prioritized queues, not a randomized causal estimate, and it measures processing at this workflow step rather than referral-to-appointment time or patient outcomes.

Average urgent-referral processing time in the peer-reviewed case

Company-reported

About 33 hours before FastFax.About one hour after FastFax; the accompanying editorial describes the post-intervention average as about 63 minutes.

Peer-reviewed case published 2026-04-15, describing the implementation and the January-June 2023 pilot with continued post-pilot use. · Urgent externally faxed referrals in Stanford Health Care's enterprise referral-management workflow.

The published summary does not establish referral-to-appointment reduction, improved clinical outcomes, or improved safety. The approximately one-hour measure should not be conflated with the earlier 38-minute urgent-queue snapshot or applied to FastFax 2.0.

Effective coverage of all provider-marked urgent referrals in the early snapshot

Inference

The snapshot reported 2,332 total provider-marked urgent referrals, including unsupported forms.The disclosed counts imply that 1,059 divided by 2,332, or approximately 45%, of all provider-marked urgent referrals reached the urgent queue.

Derived from the early post-go-live operational snapshot. · All 2,332 provider-marked urgent referrals among 26,921 faxes in the disclosed snapshot.

This is a catalog inference from disclosed counts, not a metric stated by Stanford. It assumes the reported 1,059 true positives are the urgent referrals reaching the urgent queue and highlights the material coverage loss from unsupported forms.

What leaders can reuse

A narrow model can materially compress a queue without making the consequential decision. FastFax worked by detecting a provider's explicit priority mark and moving matching work forward, while referral staff and clinicians retained action and authority.

Anti-pattern

Calling checkbox detection clinical triage, reporting 78% recall as if it covered all marked-urgent referrals, or using original FastFax metrics to imply referral-to-appointment, safety, patient-outcome, or FastFax 2.0 performance.

Questions

01Is the model detecting an authorized human's explicit decision, or making a new consequential judgment?
02What percentage of the full priority population is covered after unsupported forms and other out-of-scope inputs are included?
03Which form versions, layouts, and handwritten conventions are outside the validated boundary?
04What happens operationally when a priority item is missed, and how quickly can a human recover it?
05Are classifier recall, queue precision, queue-processing time, and end-to-end patient access being measured and communicated as distinct outcomes?
06Who retains authority for downstream scheduling, escalation, and clinical action?

Portability conditions

A high-volume intake process in which an authorized upstream actor already declares priority explicitly
Standardized forms with stable mark locations and sufficient labeled examples for each supported form version
Integration points that can route every item into either an expedited or routine human-worked queue
Named human owners for referral review, exception resolution, downstream scheduling, and clinical decisions
Monitoring that reports performance separately for supported forms, unsupported forms, and the full intake population
Form-version governance and retraining processes that detect old, nonstandard, handwritten, or repositioned urgency indicators
A safe fallback that preserves eventual human review when the model cannot recognize the form or mark

Reputation risk

low

Evidence and authority

What the public record supports.

Current · updated 2026-09-06

1 primary operational source and 2 peer-reviewed publications; workflow metrics are scoped to supported forms and urgent-fax processing.

Bundle 1.0.0 · reviewed 2026-09-06 · stable ID ba3aaddbae2a7271

Related transformations

Sources

FastFax Uses Machine Learning to Triage Urgent Patient ReferralsStanford Emerging Applications Lab, Stanford Medicine · primary
Optimizing Enterprise Referral Processing through Automated Fax TriageNEJM Catalyst Innovations in Care Delivery · 2026-04-15 · peer reviewed
Culture and Leadership Are Essentials for InnovationNEJM Catalyst Innovations in Care Delivery · 2026-04-15 · peer reviewed

Read the evidence, freshness, caveat, and version policy.

The operating-model shift, in one view.

How the work changed

Outcomes

Recall among marked-urgent referrals using the supported standard cover sheet

Precision of the urgent queue in the early operational snapshot

Mean queue processing time in the early operational snapshot

Average urgent-referral processing time in the peer-reviewed case

Effective coverage of all provider-marked urgent referrals in the early snapshot

What leaders can reuse

What the public record supports.

Related transformations

Commercial cost-proxy risk ranking

Automated LVO detection with parallel transfer notification

Advance Alert Monitor

Cytora headless risk digitization feeding Z-Horizon+ underwriting workbench

Sources