Quality Improvement and Patient Safety
This domain tests measurement, improvement methods, and safety-event response. Questions often ask you to choose a measure type, interpret variation, or select the appropriate analysis after a harm event.
Core rules and concepts
- Donabedian framework: structure (resources), process (what is done), and outcome (results). Good measures link process to outcome and include balancing measures for unintended effects.
- Model for Improvement asks three questions: what are we trying to accomplish, how will we know change is an improvement, and what changes can we make. PDSA cycles test changes on a small scale before spread.
- Use run charts or statistical process control to distinguish common-cause from special-cause variation. A run of six or more points on one side of the median suggests a shift.
- Root cause analysis follows serious harm; failure mode and effects analysis prospectively examines high-risk processes. Both should produce owned, measurable actions.
Worked example
A unit targets falls per 1,000 patient days. Baseline is 3.2. After a bundle of hourly rounding, bed alarms, and medication review, the rate falls to 2.6 over six months. A run chart shows six consecutive points below the baseline median. The team verifies data definitions and case mix, checks a balancing measure (call-light response time), and then spreads the bundle. Reasoning: a run of six points supports a non-random shift, but only confirmed measurement and balancing data justify spread.
Traps that cost marks
Blaming individuals for system failures; declaring improvement from a single before-after comparison; chasing common-cause variation with special-cause fixes; omitting balancing measures; writing action plans without owners or dates; confusing RCA with FMEA.