> For the complete documentation index, see [llms.txt](https://docs.concurrence.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.concurrence.com/data/data-architecture.md).

# Data Architecture

Follow the technical path from source access and clinical meaning to shared context, authorized actions, and evidence of completion.

A data integration needs to establish what a source says, which entity it concerns, how a workflow can use it, and what confirms the resulting action. This reading path connects those decisions across Concurrence's data and agent interfaces.

Start with one workflow and its required inputs. Identify the authoritative source for each input and outcome, then follow the boundaries below. [Shared Context and Applications](https://docs.concurrence.com/platform-overview/shared-context-and-applications.md) provides the broader ownership map.

## Read the Data Architecture in Order

| Step                         | Read                                                                                                                                                                                                                                                                  | Decision to make                                                                                    |
| ---------------------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | --------------------------------------------------------------------------------------------------- |
| 1. Scope access              | [Workspaces](https://docs.concurrence.com/data/workspaces.md)                                                                                                                                                                                                         | Which workspace, credentials, and permitted purpose apply?                                          |
| 2. Connect the source        | [Connectors and EHR](https://docs.concurrence.com/data/connectors-and-ehr.md), [Healthcare Interoperability](https://docs.concurrence.com/data/healthcare-interoperability.md), and [Customer Data Intake](https://docs.concurrence.com/data/customer-data-intake.md) | Does this workflow need imported records, dataset files, a live integration call, or a combination? |
| 3. Understand shared state   | [World Model](https://docs.concurrence.com/data/world-model.md)                                                                                                                                                                                                       | How are source identifiers, observations, corrections, and the serving view related?                |
| 4. Define clinical meaning   | [Reusable Clinical Data](https://docs.concurrence.com/data/clinical-data-contracts.md)                                                                                                                                                                                | Which units, codes, measurement times, source evidence, and validation rules make an input usable?  |
| 5. Control use and action    | [Clinical Tools](https://docs.concurrence.com/agent/clinical-tools.md) and [Runtime Safety](https://docs.concurrence.com/operations-and-safety/runtime-safety.md)                                                                                                     | Which identity checks, tool permissions, and approval policies apply to this runtime and action?    |
| 6. Verify the whole workflow | [Patient Context Walkthrough](https://docs.concurrence.com/data/patient-context-walkthrough.md), then [Connect Another Workflow](https://docs.concurrence.com/data/connect-another-workflow.md)                                                                       | What proves the result, who handles an uncertain outcome, and which tested parts can be reused?     |

## Choose the Data Path

<figure><img src="https://3635224444-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FvcLyiHRcwv7g83p6vxAd%2Fuploads%2Fgit-blob-342ee747a58cb6d884274a0ad6c74abe29267415%2Fdata-access-paths.svg?alt=media" alt="Three data paths have different outputs. FHIR import emits world events for asynchronous entity projection. Dataset intake processes registered files and publishes dataset outputs. An integration endpoint returns a tool result. Verify the output of the selected path before using it."><figcaption><p>These paths can serve the same workflow. Each has its own contract and completion evidence.</p></figcaption></figure>

**FHIR import** accepts supported resources and emits source-attributed events. Entity projections make supported data available to readers asynchronously. Inspect import results and the intended entity read separately. Follow the [FHIR implementation guide](https://docs.concurrence.com/developer-guide/platform-api/data-world-model/fhir) for supported resources and import semantics.

**Dataset intake** associates files with a registered contract and version history. Processing and publication have their own status. Choose the dataset's representation and consumer explicitly, and verify that consumer's output. See [Customer Data Intake](https://docs.concurrence.com/data/customer-data-intake.md) and the [intake implementation guide](https://docs.concurrence.com/developer-guide/platform-api/data-world-model/intake).

**Integration calls** let an enabled tool read or change an external system during a workflow. A read can supply a current scheduling result, while a write requests an external change. Interpret the endpoint's result and retain the evidence required by the workflow. Publication into shared context requires its own configured path. See [External Integrations](https://docs.concurrence.com/developer-guide/platform-api/integrations/external-integrations).

## Data and Identity Map

The world-model path connects source identity to retained evidence and derived views. A source record identifies something in its originating system; an entity identifies the modeled resource within a workspace. Events carry observations about that entity and can refer to a related entity. Available provenance and time fields help a reader interpret the evidence.

<figure><img src="https://3635224444-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FvcLyiHRcwv7g83p6vxAd%2Fuploads%2Fgit-blob-6b0bae8c5f784ded4d242ee8b4c7135c43450ffe%2Fdata-identity-model.svg?alt=media" alt="A source record has a source-scoped identifier. Supported resolution associates it with a workspace entity. Source-attributed observations become retained events with entity links, origin and available times; an event can also reference a related entity. Asynchronous projection derives entity views and relationships from retained inputs. A session selects permitted task context from those views and combines it with conversation history and tool results. Source identity, caller verification and action authorization remain separate; an accepted event is not proof that a view is readable."><figcaption><p>Entity views derive from retained inputs. Session context selects permitted facts and combines them with interaction-specific inputs; identity resolution does not itself verify a caller.</p></figcaption></figure>

Supported projection paths determine which facts and relationships appear in a read view. Projection is asynchronous, so event acceptance and reader visibility need separate checks. This map describes the world-model path; dataset outputs and live integration results retain the separate contracts described above. See [World Model](https://docs.concurrence.com/data/world-model.md) for event and projection semantics and [Core Concepts](https://docs.concurrence.com/platform-overview/core-concepts.md#context-current-facts-and-memory) for working context.

## Keep Meaning, Identity, and Permission Explicit

Connection establishes access to data. The [clinical data contract](https://docs.concurrence.com/data/clinical-data-contracts.md) establishes how the implementation interprets it: original values, accepted units and codes, time, missingness, and eligible inputs. Keep that definition with the mapping and calculation tests so another workflow can reuse it.

Source identity associates records with workspace entities. Caller verification governs access to patient context on supported caller-facing paths. Action authorization determines which operation can run. Evaluate each separately: a supplied patient identifier or recognized phone number does not establish all three.

Choose freshness requirements for the decision being made. A historical measurement can be appropriate for a longitudinal calculation, while booking needs a current availability check. [Clinical age, source freshness, and reader visibility](https://docs.concurrence.com/data/clinical-data-contracts.md#freshness-for-the-decision) answer different questions.

## Verify at Each Boundary

For an appointment change, collect evidence that the intended patient context is readable, that the selected action is permitted, and that the scheduling system confirms the requested result. If the result is ambiguous, reconcile the destination before deciding whether a retry is appropriate. A later shared view can carry the confirmed outcome to another workflow.

Use [Connect Data to a Verified Action](https://docs.concurrence.com/developer-guide/guides/data-to-verified-action) for the implementation sequence and [Verify an Integration Action](https://docs.concurrence.com/developer-guide/guides/verify-an-integration-action) for destination confirmation. Record the accepted result and recovery owner in the [Technical Evaluation Packet](https://docs.concurrence.com/platform-overview/evaluation-packet.md).


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.concurrence.com/data/data-architecture.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
