Outcomes Measurement
Core concepts
Outcomes measurement tells whether care produced the intended result. Understand the Donabedian categories: structure (resources), process (what is done), and outcome (patient health, experience, cost, safety). A useful measure has a clear operational definition: numerator, denominator, time window, data source, exclusions, and collection method. Reliability means repeated measurement is consistent; validity means the measure reflects the concept. Use baseline data, a benchmark, and balancing measures to detect unintended consequences. Risk adjustment matters when comparing groups with different acuity.
Traps
Common question traps include labeling a process measure (hand hygiene compliance) as an outcome, changing definitions mid-project, ignoring the denominator, and comparing unadjusted rates across units. Small numbers can produce unstable rates; a single month’s change may be variation, not improvement. Do not claim causation from a simple before-and-after comparison.
Worked example
A unit wants to reduce catheter-associated urinary tract infections. Define the outcome as infections per 1,000 catheter-days, with catheter-days as the denominator. Collect baseline for three months, report monthly, and track a balancing measure such as catheter use. If the rate falls after a removal protocol, interpret alongside process fidelity and run-chart trends.
Rule of thumb
Before collecting data, write the numerator, denominator, time period, exclusions, and data owner-then pilot the definition to ensure two reviewers count the same event.