These cookies are required for the website to function properly. They ensure security, enable basic features like page navigation, and store user session data. You cannot disable these cookies.
- Healthcare Analytics
- Data model and import order
The tables behind Healthcare Analytics
Healthcare Analytics ships with 17 tables, grouped here by the feature that uses them. Every record your team works with lives in one of them, and you can add fields and tables in Proto.
- Tables
- 17
- Features
- 7, plus shared settings
- Change it
- Add fields and tables in Proto
How the model is organised
Outcome cohorts:
Outcome cohorts, outcome definitions, Organizations, People. Build each cohort revision from its outcome definition, owner, organization and consent basis.
Deidentified outcome facts:
Deidentified outcome facts, evidence documents, metric observations. Load observation-level facts under a recorded consent basis and accept or reject each one.
Metric definitions:
Metric definitions, outcome definitions, source mappings. Approve the numerator, denominator, grain and aggregation method before any result is calculated.
Metric results:
Metric results, metric observations, metric definitions. Keep one result per approved definition and period window, with suppression and invalidity reasons.
Review and evidence:
Review decisions, evidence documents, state history. Record independent reviewer decisions against an exact cohort revision and verify its evidence.
Snapshots and runs:
Source snapshots, source mappings, operation runs, operation steps. Track source freshness, field mappings and every workflow run with its numbered steps.
Follow-up actions:
Follow up actions, People. Assign corrective work raised from published outcomes and close it with a recorded resolution.
Access:
Users holds sign-in accounts with a platform role of View Only, Manager or Admin.
Every table, by feature
- Outcome cohorts5
- Deidentified outcome facts4
- Metric definitions3
- Metric results4
- Review and evidence4
- Snapshots and runs5
- Follow-up actions3
- Shared and settings1
- Outcome cohorts
Measured web properties with lifecycle status and observed outcome fields.
Linked to Organizations, People, Outcome definitions - Outcome definitions
Configured measurement and experiment definitions with effective dating.
Linked to Organizations - Organizations
Customer organizations whose web properties are measured.
- People
Analysts, owners and reviewers belonging to an organization.
Linked to Organizations - State history
Immutable lifecycle transition history for each source.
Linked to Outcome cohorts, People
- Deidentified outcome facts
Consented observation-level session and conversion event facts.
Linked to Outcome cohorts - Evidence documents
Supporting source evidence documents and verification state.
Linked to Outcome cohorts, People - Metric observations
Observation-level metric readings tied to a snapshot and source.
Linked to Outcome cohorts, Source snapshots - Source snapshots
Sourced snapshot captures with coverage window and freshness status.
- Metric definitions
Approved metric definitions with grain, expressions and aggregation method.
Linked to Outcome definitions, People - Outcome definitions
Configured measurement and experiment definitions with effective dating.
Linked to Organizations - Source mappings
Reviewed field mappings between source and target definitions.
Linked to Outcome definitions
- Metric results
Calculated metric results persisted once per approved definition and window.
Linked to Outcome cohorts, Metric definitions, Operation runs - Metric observations
Observation-level metric readings tied to a snapshot and source.
Linked to Outcome cohorts, Source snapshots - Metric definitions
Approved metric definitions with grain, expressions and aggregation method.
Linked to Outcome definitions, People - Operation runs
Workflow execution runs with frozen revision, input hash and result identity.
Linked to Outcome cohorts
- Review decisions
Independent reviewer decisions recorded against an exact source revision.
Linked to Outcome cohorts, People - Evidence documents
Supporting source evidence documents and verification state.
Linked to Outcome cohorts, People - State history
Immutable lifecycle transition history for each source.
Linked to Outcome cohorts, People - Outcome cohorts
Measured web properties with lifecycle status and observed outcome fields.
Linked to Organizations, People, Outcome definitions
- Source snapshots
Sourced snapshot captures with coverage window and freshness status.
- Source mappings
Reviewed field mappings between source and target definitions.
Linked to Outcome definitions - Operation runs
Workflow execution runs with frozen revision, input hash and result identity.
Linked to Outcome cohorts - Operation steps
Numbered steps of an operation run with input/output and receipt state.
Linked to Operation runs - Metric observations
Observation-level metric readings tied to a snapshot and source.
Linked to Outcome cohorts, Source snapshots
- Follow up actions
Assigned corrective and follow-up actions with due dates.
Linked to Outcome cohorts, People - Outcome cohorts
Measured web properties with lifecycle status and observed outcome fields.
Linked to Organizations, People, Outcome definitions - People
Analysts, owners and reviewers belonging to an organization.
Linked to Organizations
Keys and references
Outcome cohorts is the hub. Facts, evidence, observations, results, runs, review decisions, state history and follow-up actions all point at a cohort, and review decisions, runs and state history also record the source revision they acted on. Every cohort needs an organization, an owner in People and an outcome definition. Metric results point at the approved metric definition and the operation run that produced them; metric observations point at the source snapshot.
Organizations, People and outcome definitions carry a unique Code; cohorts carry a Number and a source record id from the source system. Import references by those identifiers, never by a display name that can change. Keep the source reference on each deidentified fact so a reloaded extract matches the existing rows. Load only deidentified data.
Migration order
- Load Organizations, then People and Users, matching People to users by email.
- Load outcome definitions, approved metric definitions with their versions, and source mappings.
- Load source snapshots with coverage dates, record counts and source hashes.
- Load outcome cohorts with their revision, then deidentified outcome facts and evidence documents.
- Load historical metric observations and metric results with their period windows; keep prior review decisions and state history as read-only history.
- Open follow up actions that are still in progress, then reconcile counts and rates with the source before publishing the next cohort.
Set up Healthcare Analytics your way.
Start free, then use Proto to change fields, steps and approvals before your team goes live.