Quality Improvement Methods and Tools
This domain tests whether you can select and apply an improvement method, run disciplined tests of change, and use the right tool for diagnosis versus display. Expect questions on the Model for Improvement, PDSA, DMAIC, Lean, driver diagrams, and control chart interpretation.
Core rules and concepts
- Model for Improvement: three questions (what are we trying to accomplish; how will we know change is an improvement; what changes can we make that will result in improvement) plus PDSA ramps.
- PDSA: state a prediction in Plan; run a small test; Study the data against the prediction; Act to adopt, adapt, or abandon. Test on one patient, one nurse, one day before scaling.
- DMAIC: define, measure, analyze, improve, control. Best for well-characterized problems with available data and a need for statistical rigor.
- Lean: remove waste (defects, waiting, overproduction, transportation, inventory, motion, overprocessing, unused talent), map value streams, standardize work, use 5S and kaizen.
- Diagnostic tools: flowchart, Pareto chart (focus on the vital few), fishbone, 5 whys, affinity diagram. Design tools: driver diagram (aim, primary drivers, secondary drivers, change ideas) and FMEA.
- Control chart signals: a point beyond a control limit, a shift (for example eight consecutive points on one side of the centerline), a trend of six or more increasing or decreasing points, and two of three points beyond two sigma.
Worked example
Medication reconciliation errors run at 22%. A team builds a driver diagram: accurate medication list, patient engagement, and handoff communication as primary drivers. Their first PDSA tests pharmacist-led reconciliation on one unit for two weeks with ten patients, predicting a 50% reduction in errors. After Study shows errors fell to 12% but time per patient doubled, they adapt the process (technician obtains history, pharmacist reviews) and test again before spreading. Reasoning: the prediction makes the test falsifiable; the small scale limits risk; the adaptation addresses a balancing measure (staff time) before spread.
Traps that cost marks
- Running PDSA without a written prediction or without Study.
- Starting with a full-scale rollout instead of incremental tests.
- Calling a trend from a run chart with too few data points.
- Treating common-cause variation as special cause and tampering with a stable process.
- Using a fishbone diagram as the entire analysis instead of verifying causes with data.
- Skipping the Control phase in DMAIC, so gains decay.