Industry field guide
Continuing care retirement community transition data for AI
A retirement community can provide several levels of service.
- 2022 NAICS
- 623311
- Practice pack
- Example and answer key (JSON)
- Updated
What this guide tests
What transition date is authorized by the supplied document, and does the day-12 packet establish that L2 service has started?
See the specimen and fields · Work through the fictional records
Task and source context
A retirement community can provide several levels of service. A proposal, reservation and effective transfer record may appear in the same file. Medicare provides long-term care information and the Census classification identifies continuing care retirement communities separately. Medicare long-term care overview · U.S. Census 2022 NAICS manual
Invent two levels using abstract labels and a mock unit change. Include a tentative reservation for one date and a signed transfer effective later. Ask which date the documents support and whether a planned service started. Score effective date, authorized sign-off and incomplete evidence independently.
Use original forms and generic locations. Exclude actual resident information, contracts, care needs, payment records and facility access details. Keep care and eligibility decisions with qualified staff.
Industry research
The U.S. Census Bureau’s industry definitions establish which activities these codes cover. The workflow sources in this guide describe the task or record type. Fiund’s pilot below uses fictional material to test a specific decision within that scope.
623311 · Continuing Care Retirement Communities
This U.S. industry comprises establishments primarily engaged in providing a range of residential and personal care services with on-site nursing care facilities for (1) the elderly and other persons who are unable to fully care for themselves and/or (2) the elderly and other persons who do not desire to live independently.
Build a pilot sample
Set up this pilot
Create fictional service-level cards with a reservation, an approved effective date and a missing start event. Grade authorization and actual start separately at the requested cutoff.
Prepare the test packets
- Start with the worked case. Use the fictional record types below as a guide to the source files you need. Give each packet a stable ID and save its source versions with it.
- Change one fact. Make one packet with enough evidence to answer the question and one where the evidence points to a different answer. Keep the other fields fixed so a reviewer can tell what changed.
- Remove the decisive evidence. Give the reviewer a third packet with a plausible near-match in its place. Have them write what can still be established and what evidence would resolve the rest.
Fields the reviewer needs
- Record alias and abstract levels
- Defines the transition.
- Reservation versus effective date
- Separates planned stages.
- Authorization and start-event evidence
- Supports the limited status summary.
Use What health care and social services teams can contribute to AI to turn the specimen into a brief with a defined task, source and review owner.
Fictional exercise
A tentative reservation date is different from an effective service transfer
Two abstract service levels and generic location aliases form an invented administrative timeline. No resident, care need, contract, payment or access detail is represented.
| Record ID | Source | What it records |
|---|---|---|
| RT1 | Proposal and reservation | Fictional record R proposes level L2; tentative reservation at location B starts day 10. Proposal unsigned. |
| RT2 | Signed mock transfer | R transfer from L1/location A to L2/location B effective day 14. Explicitly supersedes proposed start date. |
| RT3 | Service event log | No separate L2 service-start event or completion record supplied. Requested as-of date day 12. |
Question for the model. What transition date is authorized by the supplied document, and does the day-12 packet establish that L2 service has started?
Show the reference answer
RT2 documents an effective transfer date of day 14, replacing RT1's tentative day-10 reservation date. For day 12, that transfer is still future-dated, and RT3 supplies no separate L2 start event. Preserve the signed future transition and lack of current start evidence rather than treat a reserved location or proposal as completed service. The fixture does not evaluate care needs, eligibility or an actual resident's placement.
How to check the result
Ask for the answer, the record IDs that support it and any fact the evidence leaves unresolved. Have a reviewer compare the response against the reference before deciding whether the sample is ready to expand.
Errors to flag in this case
- Using a tentative reservation as the actual service-transfer date.
- Inferring service commencement from a future signed transfer without a matching event.
Change the inputs and run it again
Move the review date to day 15 but retain no start event. Require the transfer's effective date to be recognized while actual service occurrence remains unverified.
Score the decision and the evidence separately
Save the reviewer’s expected answer, supporting record IDs and unresolved fields in a separate answer key. Score the model on the decision, its citations and its handling of the deliberate gap. A correct answer citing a different case counts as an evidence error.
Before expanding collection, have a second reviewer run the same checks on the held-out packet. Keep related versions, views and exports of the same case together so they stay on one side of the training and evaluation split.
Check permission and scope
Use original forms and generic places. Exclude residents, care records, contracts and facility access; qualified staff retain care and eligibility decisions.
Use staff-controlled props and synthetic records. Exclude resident identities, footage, diagnoses, medications and actual care plans.
Describe an authorized archive through Fiund’s existing-data assessment, or describe a new capture capability through the collection brief. The contributor agreement covers the terms for participation.
Sources and further reading
Resources for this workflow
Includes worksheet
What health care and social services teams can contribute to AI
Separate a staff training simulation from a governed clinical or care-data project.
Includes worksheet
Build an AI evaluation sample for service handoffs
Use fictional records to test whether a handoff reached a documented state.
Company and program reviews
Use these reviews to compare sourcing, capture or evaluation approaches. Confirm the applicable program, accepted sources and terms with the provider.
Compare the approach
Protege review for AI teams and data owners
Compare the company’s healthcare data-provider pathway separately from a fictional staff workflow.
Compare the approach
Sapien review: dataset quality review and expert checks
Compare an independent quality review process for an administrative or nonclinical evaluation.
Related industry examples
Industry guide · NAICS 623312
Assisted living nonmedical service preference data for AI
A resident may request a later housekeeping window, while a service schedule and a completion card preserve different information.
Industry guide · NAICS 623220
Recovery residence orientation and acknowledgment data for AI
An orientation packet can be delivered while acknowledgment of a revised version remains pending.
Industry guide · NAICS 623990
Residential care activity kit and inventory data for AI
A residential program can demonstrate an inventory workflow without recording anyone receiving services.