Practice and Tools
Pharmacy performance metrics: define the measure before drawing the conclusion
- Summary
- A useful measure has an agreed definition and a clear relationship to the decision it supports. Activity, quality, experience, and outcomes answer different questions.
- Last reviewed
- September 29, 2026
Who is this metrics framework for?
A measure is useful only when its definition, numerator, denominator, data source, time period, and limitations are clear enough for another person to interpret it.
Pharmacy’s contribution becomes visible when it is measured well. This PharmD Hub™ framework helps pharmacy teams, managers, and service leaders define and interpret performance measures. Use it to explain what a number represents, where it came from, and which decision it can inform. It provides a method for choosing measures rather than a fixed set of required indicators.
Write a Measure Specification
Record the purpose, population, numerator, denominator where relevant, exclusions, data source, measurement period, owner, and review schedule. Explain missing data and known limitations. Keep definitions consistent when comparing periods.
Illustrative Measures
Access follow up completion: Eligible access cases with the required follow up documented within the locally defined interval divided by all eligible cases. Define the interval and what qualifies as follow up before reporting.
Service delivery: Completed eligible encounters during a defined period. Report this as activity; it does not establish clinical benefit by itself.
Patient experience: Results from a defined feedback question or instrument. State who was invited, who responded, and how nonresponse may affect interpretation.
Team development: Completion of a defined competency assessment. Distinguish attendance at training from demonstrated competency.
Illustrative calculation: If 18 of 24 eligible access cases meet your defined follow-up requirement, the completion rate is 75%. These are hypothetical figures, not PharmD Hub results or a benchmark. Report the counts and definition alongside the percentage; the measure does not establish that all 18 patients obtained medication.
Balance the View
An improvement in speed may coincide with additional workload or incomplete follow up. Consider measures that reveal unintended effects. Review results with the people doing the work before selecting a response.
Be Careful With Attribution
A change after an intervention does not, by itself, prove that the intervention caused it. Consider changes in population, staffing, policy, seasonality, and data capture. Describe associations honestly and reserve causal claims for evidence that supports them.
Use Measures to Guide a Decision
State what the result suggests, what remains uncertain, and what action or investigation follows. Do not label a number “industry standard” without a relevant source and comparable definition.
Apply the Framework to One Measure
Choose a measure your team already reports and write its full specification. Ask a colleague whether they could reproduce the result from that definition and data source. Resolve differences before using the measure in a proposal or an impact account.
Next in Practice and Tools
Technology and AI Review