Medical Device Safety and Risk Management
This domain asks you to quantify and control harm from medical devices across the lifecycle, using ISO 14971 logic and hospital-specific risk ranking.
ISO 14971 vocabulary
A hazard is a potential source of harm; a hazardous situation exists when people, property or environment are exposed; harm is injury or damage to health. Risk combines severity of that harm with the probability of its occurrence. Risk control follows a fixed hierarchy: inherently safe design, then protective measures in the device or process, then information for safety such as labeling and training. Residual risk is judged against defined acceptability criteria, and production plus post-production surveillance feeds the loop.
Hospital-side ranking
Clinical engineering scores devices for PM frequency and inspection depth using physical risk to patient and staff, maintenance requirements, incident history, regulatory relevance and downtime impact. The score sets the interval; purchase price does not.
Analytic tools
FMEA/FMECA scores severity, occurrence and detection, with RPN = S x O x D. Fault tree analysis works top-down from a defined top event. Both should end in verified actions, not a spreadsheet.
Worked example
A syringe-pump failure mode scores S=9 (overinfusion), O=4, D=6, giving RPN 216. A battery failure scores S=4, O=7, D=3, giving 84. The overinfusion mode gets design or process action first even though it is less frequent, because severity drives priority; a detection score of 6 warns that current controls will often miss the failure.
Traps
- Treating RPN as absolute; severity 9-10 modes need action regardless of the number.
- Calling every event use error when the device failed.
- Ignoring residual risk once controls are in place.
- Confusing hazard with harm in written analyses.