Digging into TOC Logic

Understanding Transitions of Care (ToC)

What is a Transition of Care?

A Transition of Care represents a complete episode of hospital care - from the moment a patient arrives at the hospital until they are discharged. Episodes aren't limited to your own health system. They include hospitals Zus hears from through national ADT networks (admit, discharge and transfer alerts) and state and regional health information exchanges (HIEs). This might include:

• Single-location stays: Direct admission to inpatient care
• Multi-location episodes: Emergency department visit followed by inpatient admission
• Complex care paths: Transitions between observation, emergency, and inpatient settings

Clinical Value

ToC data gives you a complete picture of each hospital episode, helping you:

• Track patient outcomes across your entire care episode
• Identify readmissions and care transitions
• Monitor length of stay patterns
• Analyze discharge planning effectiveness
• Support quality improvement initiatives

How Episodes Are Defined

Only emergency, inpatient, observation and pre-admit encounters are grouped into episodes. Ambulatory (outpatient) visits never form an episode.

Zus groups related encounters into episodes by time:

• Time window: An encounter that starts within 36 hours of the previous one joins the same episode
• Return ED visits: A new ED visit after the patient has been discharged starts a new episode
• No visit-number grouping: Encounters aren't pre-grouped by shared visit identifiers

Example Scenarios:

• Patient arrives in ED at 2:00 PM, admitted to inpatient at 8:00 PM → Single Episode
• Patient discharged Tuesday, returns to ED Friday → Two separate episodes
• Patient moves from ED to observation to inpatient over 2 days → Single Episode

Key Clinical Information in ToC Data

Episode Timeline

• Start: When the first encounter in this episode began
• End: When the patient was discharged from their final encounter
• Duration: Complete length of stay across all locations

Diagnoses

• Comprehensive diagnosis list from all encounters in the episode
• Primary vs secondary diagnoses when available
• Discharge diagnoses that represent the final clinical picture

Care Locations

• Admitting location: Where the episode began
• Emergency location: ED involvement in the episode
• Discharge location: Final location before discharge
• Location transitions: Movement between departments/units

Admission Context

• Planned vs unplanned admissions
• Emergency admissions: Came through ED
• Direct admissions: Bypassed ED
• Readmission indicators: Inpatient stays flagged as readmissions under the CMS Hospital-Wide Readmission method (see Readmission Detection below)

Behavioral Health Flags

• Automatic detection of episodes involving mental health, behavioral, or substance use conditions
• CCSR coding integration for standardized identification
• Service type: encounters are also flagged as behavioral health based on their service type

Understanding Individual Encounters

What is an Encounter Lens?

An Encounter Lens represents a single, deduplicated healthcare interaction - like one visit to the clinic, one ED visit, or one inpatient stay. It's the building block that ToC episodes are made from.

Encounter Classifications

• Ambulatory (AMB): Outpatient visits, clinic appointments
• Inpatient (IMP): Hospital admissions requiring overnight stay
• Emergency (EMER): Emergency department visits
• Observation (OBSENC): Short-stay observation care
• Pre-admit (PRENC): Pre-admission processing

If Zus can't classify an encounter, usually because it has no codes or other context, it's labeled Unknown. These are usually not patient visits (medication orders, for example), so they're left out of encounter summaries and episodes.

Readmission Detection

Zus flags inpatient (IMP) stays as readmissions using the CMS Hospital-Wide Readmission (HWR) measure method:

• Only the first unplanned readmission within 30 days of an index admission (the original stay) is flagged
• Planned readmissions and same-day transfers aren't counted
• Oncology, rehabilitation and mental health admissions are excluded

See Encounter Lens for the full rules.

This documentation focuses on clinical applications. For technical implementation details, API specifications, or data model information, refer to the Lens documentation.