Answers to Questions.pdf

ANSWERS TO QUESTIONS

REQUEST FOR INFORMATION (RFI)

Medical Investigation Catalog System

1. Will medical review be a multi-phase process (e.g., draft, review, finalize), or is a single review entry sufficient?

ANSWER: Medical consults involve several steps. For primary purposes of tracking status in MEDICS, a medical consult request may be in several states: none (no consult requested), requested but not acknowledged, acknowledged but not assigned, assigned but not complete, and complete. Overall case completion status also is tracked. It is possible for one case to have no, one, or multiple medical consult requests. Several report output types are possible from medical consults;

information on report type is input each time a medical consult is completed.

2. Should the system support structured medical or injury coding using standards such as ICD, AIS, or SNOMED, or will reviews remain primarily narrative?

ANSWER: Customized coding selections with available free text fields are used to track relevant potentially impairing medical conditions and substances. The system currently also supports some manual entry (without integrated searchable database) of other coding information, including AIS, ICD, and narrative injury information. Coding functionality should facilitate data retrieval based on machine-readable fields/categories thereof.

3. Will access to medical records and case materials be limited to assigned users per case, or is visibility determined only by user role?

ANSWER: Permissions are based on application- and case-level roles. Medical officer and administrator application users have broad access to cases and records.

General users of the application have access to cases to which they are assigned;

within these cases, permissions surrounding particular records (such as for viewing and exporting) are based on specific assigned case personnel roles.

4. Should supervisors and reviewers see only the cases within their assigned region or mode, or will they have broader visibility?

ANSWER: Direct supervisors of case personnel may view cases of those personnel, with imputed permissions. Also, case personnel in supervisory roles may be added to individual cases, conveying specific associated permissions. Medical officer and administrator application users have broad visibility of all cases; general users (including supervisors) do not.

5. Should external users authenticate via one-time links, agency login, or user accounts, or will their uploads be handled by internal staff?

ANSWER: External users do not have direct application access. To upload to MEDICS, whitelisted external users authenticate in Box and submit files. An API pulls these files into particular MEDICS cases after validation steps, and associated automatic notifications occur. Non-authenticated external users submitting records do so outside of MEDICS; such records typically are manually associated with cases by medical officer application users, or possibly by clerical personnel who have limited necessary permissions.

6. Will external uploads automatically initiate new cases, or must internal users associate them with existing or new cases?

ANSWER: External uploads pulled into MEDICS for an existing investigation with no existing MEDICS case will open a new MEDICS case using data from SAFTI and will generate certain associated notifications.

7. Will users need to search the full content of PDFs (e.g., keyword search within documents), or just by associated metadata fields?

ANSWER: Ideally, we would like users to search within the document content and metadata fields.

8. Will the system need to support metrics dashboards (e.g., medical coding activity or consult status?

ANSWER: Yes.

9. Will all historical records from MEDICS be migrated, or only those from a specific date range or case type?

ANSWER: Yes. All the historical records will need to be migrated.

10. Will the legacy data from MEDICS be delivered in a structured format (e.g., CSV or database export), or as a mix of files and unstructured metadata?

ANSWER: Files and database exports would be provided. However, ETL work will be needed to ensure the data structure fits proposed solution schema design.

11. Are there any additional data sources outside of MEDICS that need to be included?

ANSWER: No.

12. Will the system need to send data automatically to external systems (e.g., NTSB databases or external agencies), or is export intended as user-driven downloads only?

ANSWER: Yes. We would like the new system to perform automatic uploads to external system using REST APIs.

13. What systems, if any, should the platform integrate with on day one, and what’s the preferred method of data exchange (e.g., API, file transfer)?

ANSWER: The system should upload files to our Docket Management System.

Ideally, the system should also exchange certain data with our subpoena management system.

14. Will internal NTSB administrators manage user roles, case assignments, and workflows directly through the system UI, or should this be centrally managed by IT staff?

ANSWER: Roles management should be handled via the UI and shall be federated EntraID or Okta.

15. Would the client be open to a discovery sprint to review the consistency of the existing metadata and talk through changes NTSB would like to make consistent in the new system?

ANSWER: Yes.