What are the three main types of quality measures?

Structure measures describe the resources and organization of care, process measures describe whether appropriate care actions were taken, and outcome measures describe the resulting effect on patients' health.

Why do outcome measures need risk adjustment?

Because outcomes depend on patients' baseline characteristics as well as on care quality, fair comparison between providers requires statistical adjustment for differences in case-mix.

Quality Measurement and Metrics

Quality measurement is the practice of quantifying how well health care meets accepted standards, using defined indicators of the structure, process, and outcomes of care. Quality indicators translate the abstract idea of good care into countable measures that can be compared across providers, tracked over time, and used for accountability or improvement.

Definition

Quality measurement is the systematic use of defined indicators — quantitative descriptors of the structure, process, or outcome of care — to assess and compare the degree to which health services achieve desired outcomes consistent with current professional knowledge.

Scope

This topic covers the types of quality indicators (structure, process, outcome), the criteria a good measure must satisfy, the uses of measurement for internal improvement versus external accountability, and the well-known hazards of measurement such as gaming and risk-adjustment error. It is a methodological reference and does not prescribe which measures any organization should adopt.

Core questions

What can be measured to represent the quality of care?
How do structure, process, and outcome measures differ in what they capture?
What makes an indicator valid, reliable, and feasible?
How does measurement for improvement differ from measurement for accountability?
How should case-mix and risk be adjusted for fair comparison?

Key concepts

Structure, process, and outcome measures
Validity, reliability, and feasibility of indicators
Risk adjustment and case-mix
Process measures as actionable and attributable
Outcome measures as patient-relevant but multifactorial
Measurement for improvement vs. accountability
Gaming and unintended consequences of measurement

Key theories

Structure-process-outcome typology of indicators: Donabedian classified quality measures into three types: structural measures (characteristics of the care setting), process measures (whether appropriate actions were taken), and outcome measures (resulting health status); each offers different validity and attribution trade-offs.

Mechanisms

A quality measure links a numerator (events meeting a standard) to a denominator (the eligible population), often with explicit inclusion and exclusion criteria. Process measures ask whether evidence-based actions occurred and are directly actionable but only valuable when the process is genuinely linked to better outcomes. Outcome measures capture results that matter to patients but require risk adjustment because outcomes depend on patient case-mix as well as care quality. Structural measures are easiest to assess but most distal from outcomes.

Clinical relevance

Quality measures shape public reporting, accreditation, and pay-for-performance, and clinicians regularly encounter them as performance dashboards. This entry explains how such measures are constructed and interpreted; it is a reference on measurement and not guidance for treating individual patients.

Epidemiology

Measurement studies have repeatedly shown large gaps between recommended and delivered care; McGlynn and colleagues (2003) found US adults received about 55% of recommended care processes across a broad set of indicators, illustrating both the value and the burden of comprehensive process measurement.

Evidence & guidelines

Crossing the Quality Chasm (2001) framed measurement around six aims, and Leape and Berwick (2002) discussed how evidence should inform which practices and measures are prioritized. Indicator sets are maintained by national bodies and professional organizations, with attention to scientific acceptability and feasibility.

History

Quality measurement grew from early hospital standardization and Donabedian's 1966 conceptualization, expanding rapidly with public reporting and accountability initiatives from the 1990s onward. The proliferation of measures later prompted concern about measurement burden and a movement toward parsimonious, outcome-oriented metrics.

Debates

Process versus outcome measures: Process measures are actionable and fairly attributable to providers but valuable only when tightly linked to outcomes; outcome measures matter more to patients but require risk adjustment and may reflect factors outside providers' control.

Key figures

Avedis Donabedian
Elizabeth McGlynn
Lucian Leape
Donald Berwick

Seminal works

donabedian-1966
mcglynn-2003

Frequently asked questions

What are the three main types of quality measures?: Structure measures describe the resources and organization of care, process measures describe whether appropriate care actions were taken, and outcome measures describe the resulting effect on patients' health.
Why do outcome measures need risk adjustment?: Because outcomes depend on patients' baseline characteristics as well as on care quality, fair comparison between providers requires statistical adjustment for differences in case-mix.