Measure health equity by comparing health outcomes, access, experience, and financial protection across population groups, then use the gaps to guide action. Choose an indicator, disaggregate the data, quantify differences, and repeat the analysis. An average can hide who is being left behind.
This guide follows the WHO Health Inequality Monitor: use disaggregated data and summary measures together, then connect measurement to decisions.
On this page
- What health equity measurement means
- How to measure health equity step by step
- Which indicators should you measure
- How to choose the right comparison
- How to report results responsibly
- What measurement cannot tell you alone
- FAQ
What health equity measurement means
Health equity measurement asks whether avoidable and unfair differences exist between groups, and whether services and policies are reducing them. It examines who benefits and who faces barriers.
The terms matter. A health inequality is a measurable difference in health or in the factors that shape health. Health inequity adds a judgement about fairness and avoidability. Data show a gap. Context, evidence, and community knowledge help explain it.
The WHO Health Inequality Monitor data repository brings together disaggregated data across health topics and inequality dimensions, helping users examine patterns rather than averages.
Equity is not the same as equality
Equality means giving everyone the same thing. Equity focuses on fair opportunity to achieve health, which may require different support for groups facing different barriers. Measuring the same appointment offer for all patients does not show whether language, transport, disability access, cost, or digital access affects use.
A sound plan follows the path from resources to access, care, and outcomes.
How to measure health equity step by step
1. Define the decision
Begin with a decision, not a dashboard. You might assess rural reach, screening uptake, or financial protection.
Write the question in one sentence. Specify the population, setting, period, outcome, and possible action.
2. Choose the indicator and unit of analysis
Select an indicator that matches the question. Options include:
- Outcome: mortality, morbidity, recovery, quality of life, or patient-reported health
- Access: coverage, utilisation, waiting time, travel burden, or unmet need
- Quality and experience: safety, continuity, respectful treatment, or communication
- Financial protection: out-of-pocket costs, catastrophic spending, or forgone care because of cost
- Structural conditions: income, housing, education, employment, environment, or social protection
Define the numerator, denominator, period, and population before looking at results. Record whether the measure is crude or adjusted, and explain any adjustment in plain language.
3. Choose groups and dimensions
Disaggregate by dimensions relevant to the setting and feasible to collect safely. These may include age, sex, disability, place of residence, socioeconomic position, ethnicity, migration status, language, or other legally and ethically appropriate characteristics.
Do not treat these categories as interchangeable. Socioeconomic position might use income, education, occupation, or an area-based index. Geography might distinguish urban and rural residence, region, distance, or facility catchment.
Where barriers overlap, consider intersectional analysis. Show uncertainty and avoid exposing identifiable individuals.
4. Check data quality
Before interpretation, assess completeness, consistency, timeliness, and comparability. Ask whether the same definition was used across groups and periods. Check missingness by subgroup, since uneven missing data can distort apparent equity.
Review how the data were collected. Administrative records can reflect service use rather than all need. Surveys can capture unmet need but may exclude people without stable housing or the survey language. Registries may have clinical detail but incomplete demographic fields.
Document exclusions, coding choices, denominator changes, and any imputation. A transparent limitation is more useful than false precision.
5. Calculate absolute and relative differences
Use one absolute measure and one relative measure where appropriate. Absolute difference shows the gap in direct units. Relative difference shows the ratio between groups.
For example, coverage of 80% in one group and 60% in another is a 20 percentage-point gap. The ratio is 1.33. These measures answer different questions.
For ordered groups, such as wealth or education categories, compare the most advantaged and disadvantaged groups or use a summary measure that uses all groups. The WHO handbook covers pairwise comparisons and summary measures.
6. Add uncertainty and examine trends
An observed gap is an estimate. Report a confidence interval or another suitable uncertainty measure when the design and data support it.
Compare results across consistent time periods. A gap can narrow because one group improves, another worsens, or both change at different speeds. Report the underlying group values alongside the gap.
7. Link findings to action
Measurement has value when it changes a decision. For each priority gap, record the responsible team, intervention, and review date. Address the barrier suggested by the evidence, not merely the average.
Repeat the measure after implementation. Keep definitions stable and record changes that affect comparability. Equity monitoring is a cycle, not a one-time score.
Which indicators should you measure
A balanced scorecard needs more than one stage of the care pathway. Outcome data alone may show a gap too late. Access data alone may show use without quality or benefit.
| Measurement area | Example question | Useful indicators | Main caution |
|---|---|---|---|
| Need and risk | Who has the greatest need? | Disease prevalence, self-reported health, risk factors | Recorded need may be lower where diagnosis is limited |
| Access | Who can obtain care? | Coverage, unmet need, waiting time, distance | Utilisation is not the same as access |
| Quality and experience | Is care comparable and acceptable? | Safety, continuity, communication, patient experience | Averages can hide subgroup-specific experiences |
| Outcomes | Who benefits from care? | Recovery, complications, mortality, quality of life | Outcomes are shaped by factors outside the service |
| Financial protection | Who bears the cost? | Out-of-pocket spending, forgone care, financial hardship | Costs can be missed in incomplete household data |
The United Nations’ Sustainable Development Goal 3, target 3.8 frames universal health coverage around essential health-service coverage and financial risk protection. The UN metadata for service coverage, SDG 3.8.1 and financial protection, SDG 3.8.2 are references.
How to choose the right comparison
There is no single best equity statistic. Choose the comparison that fits the decision.
| Comparison | Best used for | Strength | Limitation |
|---|---|---|---|
| Group-to-group difference | A clear priority contrast | Easy to explain | Uses only two groups |
| Group-to-group ratio | Relative advantage or disadvantage | Shows proportional difference | Can look large when baseline values are small |
| Gradient across ordered groups | Wealth, education, or deprivation groups | Uses the pattern across categories | Needs a defensible ordering |
| Distribution across all groups | Multiple population categories | Avoids selecting only a pair | Can be harder to summarise |
| Trend by subgroup | Monitoring change over time | Shows whether progress is shared | Requires consistent definitions and repeated data |
Use the simplest measure that supports the decision. Complexity is not automatically more informative.
How to report results responsibly
Show averages and gaps together
Report the overall value with subgroup values and the chosen inequality measure. An average can improve while a subgroup falls behind, while a gap alone can hide poor results for everyone.
Name the population, period, and denominator
A statement such as “access is lower” is incomplete. Say which population, which service, which period, and compared with whom. Define terms such as “coverage” or “financial hardship.”
Avoid treating groups as causes
A difference by ethnicity, disability, or geography can reflect policy, discrimination, income, exposure, service design, or data access. Do not imply that group identity causes the gap.
Protect privacy and involve communities
Use suppression rules for small cells. Do not publish a breakdown that could identify people or expose a vulnerable community. Include affected people in decisions about measures and interpretation.
For evidence quality, definitions, and source handling, see the publication’s methodology. Related analysis is in Global Health Equity and Accessibility.
Rule: measure the gap, investigate the barrier, act with the affected community, and measure again.
FAQ
What is the first step in measuring health equity?
Define the decision and the population groups involved. Then choose an indicator that measures a relevant outcome, access barrier, care experience, or financial-protection issue.
Why is disaggregated data important?
Disaggregation shows how results differ between groups. Without it, an overall average can conceal unequal access, quality, outcomes, or financial burden.
Should I use absolute or relative inequality measures?
Use both when they add decision value. Absolute differences show the practical size of a gap, while relative measures show proportional contrast. Report the underlying group values as well.
How many indicators should an equity dashboard include?
Use enough indicators to cover the decision and the care pathway, but avoid a long list that no team can act on. A small set of well-defined measures is stronger than many poorly documented ones.
Can health equity be measured without income data?
Yes. Equity can be examined across dimensions such as geography, age, disability, sex, ethnicity, or education, subject to local law, ethics, and data quality. Income is one dimension, not the definition of equity.
How often should health equity be measured?
Measure it often enough to support the decision and detect change. The right interval depends on the indicator, sample size, intervention, and data cycle. Keep definitions consistent so trends remain interpretable.
Conclusion
Health equity measurement is a disciplined comparison, not a single score. Define the question, disaggregate credible data, report absolute and relative gaps, show uncertainty, and connect each finding to an accountable action.
Use the WHO Health Inequality Monitor and the UN’s SDG measurement guidance as technical references. For evidence-led health industry analysis, explore our topics and methodology.
Start with one priority gap, document the measure, and set a date to review whether the gap changed.