Performance Measurement, Analysis, Improvement and Monitoring is 25% of the scored CPPS exam, and it's where many candidates without a quality background lose points. You don't need to be a statistician. You need to recognize a few patterns quickly, often without scratch paper if you test remotely.
The four measure types
| Type | What it tells you | Falls prevention example |
|---|---|---|
| Structure | Whether the resources and setup exist | Number of low beds available per unit |
| Process | Whether the right steps are happening | % of patients with a fall risk assessment within 24 hours of admission |
| Outcome | Whether patients are better off | Falls with injury per 1,000 patient days |
| Balancing | Whether the change caused a problem somewhere else | Restraint use, or loss of mobility in patients kept in bed |
Balancing measures are the ones candidates forget. When a question asks how to confirm that an improvement didn't create new harm, that's the answer.
Two kinds of variation
Every process varies. The key skill is telling which variation matters.
- Common cause variation is the normal, built-in noise of a stable process. To reduce it, you have to change the process itself.
- Special cause variation comes from something outside the usual system, like a new drug shortage or a staffing change. It's a signal worth investigating.
Don't tamper
Reacting to every up-and-down point as if it meant something is called tampering. It usually makes performance worse. If a chart shows only common cause variation, the right move is to redesign the process, not to chase individual data points.
Reading a run chart
A run chart plots data over time with a median line. Four rules, commonly taught by IHI, identify non-random signals:
- Shift: six or more consecutive points all above or all below the median. Points exactly on the median don't count toward or break the run.
- Trend: five or more consecutive points all going up or all going down.
- Too few or too many runs: a run is a series of points on one side of the median. A reference table gives the expected range for your number of points.
- Astronomical data point: a value obviously different from the rest.
Run charts are simple and need fewer data points, which makes them ideal early in an improvement project.
Measurement questions trip up many candidates. Test yourself free.
Try 50 free questionsReading a control chart
A control chart (also called a Shewhart chart) plots data around a center line, usually the mean, with upper and lower control limits typically set about three standard deviations away. Control limits come from the data. They aren't targets or specification limits.
Common rules for special cause include:
- A single point outside a control limit
- Eight or more consecutive points on one side of the center line
- Six consecutive points steadily increasing or decreasing
- Two out of three consecutive points near a control limit (in the outer third)
- Fifteen consecutive points hugging the center line (in the inner third)
Rule sets vary slightly by source, so focus on the logic: a pattern too unlikely to be random noise is a signal.
Which control chart?
| Data | Chart | Example |
|---|---|---|
| Proportion (yes/no per case) | p chart | % of patients receiving VTE prophylaxis |
| Rate with varying opportunity | u chart | Falls per 1,000 patient days |
| Count with equal opportunity | c chart | Medication events per month on a unit with steady volume |
| Individual continuous values | I chart (XmR) | Door-to-balloon time for each patient |
| Rare events | g chart or t chart | Cases or days between retained foreign objects |
Practice questions
Practice question 1
A unit launches a hand-hygiene program. The run chart shows seven consecutive weeks above the baseline median after launch. What does this indicate?
- Random variation; keep monitoring
- A shift, suggesting a non-random improvement
- A trend, suggesting the process is unstable
- An astronomical data point
Show answer
Answer: B. Six or more consecutive points on one side of the median is a shift, a signal that something changed.
Practice question 2
A team reducing catheter use wants to make sure the change isn't causing harm elsewhere. Which is the best balancing measure?
- Catheter days per 1,000 patient days
- CAUTI rate
- Rate of urinary retention requiring reinsertion
- % of catheters with a documented indication
Show answer
Answer: C. Removing catheters earlier could cause retention and reinsertion. That's the unintended consequence to watch. The others are outcome or process measures for the project itself.
Practice question 3
A control chart of monthly falls per 1,000 patient days shows all points within the control limits and no special cause patterns, but the rate is higher than leadership wants. What is the best next step?
- Investigate the highest month to find its cause
- Redesign the fall prevention process
- Change the control limits to match the target
- Wait for a point outside the limits before acting
Show answer
Answer: B. The process is stable but performing poorly. That's common cause variation, so improvement requires changing the system. Investigating one month would be tampering.
Frequently asked questions
A run chart uses a median line and simple rules to detect non-random patterns. A control chart adds statistically calculated control limits, allowing more precise detection of special cause variation.
A measure that checks whether an improvement caused unintended problems in another part of the system.
Six or more consecutive points all above or all below the median.
Common cause is the normal noise of a stable process and requires process redesign to reduce. Special cause comes from something outside the usual system and should be investigated.
Sources
- CBPPS, CPPS Examination Content Outline (published September 2024), linked from IHI’s CPPS examination page.
- CBPPS / PSI Services, CPPS Candidate Handbook (rev. November 2024).
- Perla RJ, Provost LP, Murray SK. The run chart: a simple analytical tool for learning from variation in healthcare processes. BMJ Quality & Safety (2011).
- Provost LP, Murray SK. The Health Care Data Guide (2011).
Exam details change between handbook editions. Always confirm fees, eligibility and policies with CBPPS/IHI before you apply. CPPS Qbank is an independent study resource and is not affiliated with or endorsed by IHI or CBPPS.