Device continuity timeline input guide

Use one row for each distinct loss-of-function episode, not one row per visit. Do not include names, contact details, serial numbers, health notes, or other identifying information. This experimental tool is not a secure service for personal records and is not clinical or repair advice.

All headers shown in example.csv are required, even if optional values are blank. case_id is unique and non-identifying. device_type is a broad class. status is open or restored. loss_at is required. request_at, assessment_at, part_ordered_at, part_received_at, and appointment_at are optional observed milestones. restored_at is the time the person or authorized source confirmed function returned; an appointment is not restoration. open_observed_through_at is the latest time unresolved status was actually confirmed; this is required to calculate any unresolved duration. Enter timestamps with Z or an explicit UTC offset (for example 2026-10-01T08:00:00-07:00). Blank means missing or not applicable, not zero.

An unresolved case observed before the cutoff is last known open, not known open at the cutoff. An open observation after the cutoff does not establish the status at the cutoff and is indeterminate. Durations assume continuous loss of function between the reported start and observation, which an isolated later check cannot prove. Intermittent recovery, loaners, repeat breakdowns, and clinical findings need a richer protocol. The demo is fictional and does not establish real-world benefit.