Background Defining and measuring population health in places is fundamental for local and national planning and conducting cross-geographic health comparisons. Yet availability and comparability of place-level health data is unknown.

Methods A scoping review was performed to identify how Organisation for Economic Co-operation and Development (OECD) countries measure overall health for sub-national geographies within each country. The search was conducted across MEDLINE, Scopus and Google Scholar, supplemented by searching all 38 OECD countries statistical agency and public health institute websites. For all three electronic databases, three concepts were created to identify studies where health indicators would have been used to assess health at a population-level: (1) health indicator, (2) population assessment and (3) OECD countries. Only at the full article assessment stage were studies excluded for not having health indicator data at a sub-country geography.

Results Out of a total of 1,157 non-duplicate titles and abstracts screened, 210 full texts were reviewed and sixty publications selected; plus extracted information from 37 of 38 OECD countries statistical agency and/or public health institute websites. Twelve health indicators were identified where data was available at a population level for sub-national geographies. Data sources varied by categorisation into mortality (all-cause, cause-specific, life expectancy at birth, life expectancy at 65 years, preventable, excess or amenable) or morbidity (self-rated health, long-standing illness, disability, activity limitations or healthy life expectancy) health indicators: the former mostly from national statistical agencies and the latter from population-level surveys. In all cases, geographic boundaries used administrative definitions. Region, or equivalent large subnational entities, was the predominant geographic level for both mortality and morbidity indicators. All-cause mortality, and some cause-specific mortality indicators, were available at regional level for all 38 OECD countries. All other mortality indicators were frequently available at this level, with the exception of life expectancy at 65 years (5 countries only). Similar but slightly fewer indicators were available for urban areas (max countries per most frequent indicator = 24), followed by municipality (range of 1–14 countries per indicator). Other geographies, particularly those at smaller granularity, were infrequently available across health indicators and countries.

Conclusion Health indicator data at sub-national geographies are generally only available for a limited number of indicators at large administrative boundaries. Relative uniformity of health indicator question format allows cross-national comparisons. However, wider availability of health indicators at smaller, and non-administrative, geographies is needed to explore the best way to measure population health in local areas.