Certified Case Manager (CCM) guideStudy & preparation
CCM Quality Measures and PDSA: Measure the Result, Then Test the Change
Short answer
For CCM quality scenarios, distinguish resources, care actions and health results: AHRQ classifies these as structure, process and outcome measures. Add a balancing measure to look for unwanted consequences of a change. Use Plan-Do-Study-Act to prepare a small test, carry it out, interpret the findings and decide what to change next; a better before-and-after percentage alone does not prove causation.
Are you measuring resources, actions or results?
For CCM quality scenarios, classify the measure by what it records: AHRQ distinguishes structure, process and outcome measures. Structure describes capacity or resources, process describes care actions, and outcome describes health effects; AHRQ's PDSA guidance adds balancing measures to check for unwanted consequences of a change.
The Commission's July 2026 Certification Guide, checked November 11, 2026, includes quality indicators, data and performance improvement in its exam content. See the CCM domain and format guide for where this material fits. The AHRQ sources used here were checked November 14, 2026.
Consider a fictional case management program that helps clients attend follow-up visits after discharge. Every program detail, time interval, count, percentage and target below is invented teaching material, not an official CCM question, clinical standard or reported performance statistic. The proposed measures are examples for classification, not validated instruments.
| Measure type | What the fictional program could record | What it establishes |
|---|---|---|
| Structure | Whether a shared follow-up tracking system is available | A resource exists; it does not establish that staff use it |
| Process | Percentage of eligible clients completing a follow-up visit within seven days | The defined care action occurred |
| Outcome | Percentage reporting improved ability to carry out usual daily activities at 30 days, among those with both initial and follow-up ratings | A reported change in functioning, with the stated measurement limits |
| Balancing | Average staff time spent scheduling per eligible client | Whether the change adds scheduling workload |
A completed visit is a process measure even though it is a desirable result of scheduling work. It does not, by itself, establish improved health or functioning. Ask what the measure actually counts before choosing a label. For practice in this area, 55 Quality and Outcomes Evaluation and Measurements questions are planned for CCM prep.
What makes a percentage interpretable?
Start with the numerator, denominator and time window. In the fictional program, define the process measure as:
Clients completing the visit within seven days ÷ all eligible clients in that week's group × 100.
Suppose 12 clients are eligible, staff reach 10, and eight complete the visit within seven days. Two calculations are possible:
- 8 ÷ 12 × 100 = 66.7%: completion among all eligible clients, rounded to one decimal place.
- 8 ÷ 10 × 100 = 80%: completion among clients staff reached.
Both calculations are arithmetically correct. Only the first answers the program's defined question. Excluding the two people staff could not reach makes the percentage higher and changes the population being described.
Keep the definition consistent when comparing periods. If an earlier report included all eligible clients and a later report included only contacted clients, do not call the difference an improvement without resolving that mismatch.
Apply the same check to the fictional outcome measure. Clients without a 30-day rating are missing observations, not automatically clients whose functioning failed to improve. Record how many ratings are missing alongside the percentage. A result among respondents does not describe every enrolled client.
Before interpreting a dashboard, write a sentence completing: “This percentage is the share of ___ who ___ within ___.” If the report cannot fill those blanks, you need more information.
Why add a balancing measure?
AHRQ's Model for Improvement job aid uses balancing measures to check for negative consequences. A project can improve its main measure while creating a problem elsewhere.
For this fictional program, suppose scheduling time rises from an average of eight to 12 minutes per eligible client after a new checklist is introduced. That increase does not automatically mean the checklist should be abandoned. It gives the team a tradeoff to examine alongside visit completion.
The important distinction is the measure's purpose. Here, staff time is selected to detect an unwanted workload effect of improving follow-up. A second completion percentage would give another view of the main process; it would not answer the workload question.
For a study scenario, name the consequence you want to detect before selecting its measure. “More measurement” is less useful than “check whether the change increases scheduling work.” At launch, CCM prep is planned to explain every answer, which helps you test that kind of reasoning.
What belongs in each PDSA step?
PDSA means Plan, Do, Study, Act. AHRQ's job aid places it within an improvement approach that specifies an aim, measures and a change to test. Start small enough to learn from what happens before increasing the scale.
Use the same fictional program. In the prior week, six of 12 eligible clients completed follow-up within seven days: 50%. The team sets an illustrative aim of reaching 75% within four weeks while watching scheduling workload. Neither percentage is a recommended clinical benchmark.
| Step | The program's original teaching example |
|---|---|
| Plan | Test a shared scheduling checklist with one coordinator and the next week's eligible clients. Predict that completion will increase. Define the numerator, denominator, seven-day window and scheduling-time measure before starting. |
| Do | Use the checklist with 12 eligible clients, record visit completion and scheduling time, and note steps that were unclear or duplicated work. |
| Study | Compare the prediction with the findings: eight of 12 completed follow-up, or 66.7%, while average scheduling time increased from eight to 12 minutes. Review what happened and what remains uncertain. |
| Act | Decide whether to abandon, adapt or adopt the change. For example, revise a duplicated checklist field and test again before deciding to expand. |
“Do” means carrying out the test and recording what happened. “Study” means interpreting those observations. “Act” is the decision about the next change or cycle, not merely continuing the activity because it has started.
This pilot did not reach its illustrative 75% target, and it raised a workload concern. It still produced information. Explain what was learned before labeling the whole test a success or failure.
What can a better outcome establish, and what can it not?
The pilot's completion percentage increased from 50% to 66.7%. That describes the two groups under the stated definitions. It does not establish that the checklist caused the difference, and it does not show that clients' functioning improved.
AHRQ cautions that health outcomes have multiple influences, including factors outside providers' control. If the fictional program later finds better 30-day functioning ratings, interpretation still requires attention to who was measured, missing ratings and other differences between groups.
In this example, the pilot's 30-day outcome is not yet available when the seven-day completion results are reviewed. Calling an unavailable outcome “unchanged” would turn an absence of data into a finding.
A useful study explanation separates three statements: what changed, what might explain it, and what the data do not establish. The small test can inform the next cycle without proving a causal effect.
How should you practice reading a quality dashboard?
Use this editorial reading sequence for a fresh scenario:
- Name the aim before examining the most impressive number.
- Classify each measure by what it records and why it was selected.
- Check eligibility, numerator, denominator, time window and missing data.
- Identify the PDSA step described by the action in the scenario.
- State one finding and one limit on its interpretation.
For this example, the final note could read: “Follow-up completion increased in the pilot, scheduling workload also increased, and the functioning outcome is not yet available. The team can adapt and retest; the figures do not establish causation.” Your own practice data deserves the same reading, and planned CCM prep tracks accuracy in each domain.
Apply that reading method to relevant items in the 20 free CCM practice questions, then reserve a review session with the free CCM study plan. If you are still establishing whether you can apply, the separate CCM certification requirements guide covers eligibility. For your next quality scenario, identify the denominator before you interpret the percentage.
Trust the source
Official sources
Exam policies can change. Use these primary sources for the most current details.