Back

Comparing physical activity policy evaluations across 27 EU member states: Diverse indicators, inconsistent findings

Messing, S.; Tcymbal, A.; Birkholz, L.; Bauman, A.; Whiting, S.; O'Hehir, C.; Woods, C.

2026-09-10 health policy
10.64898/2026.09.04.26361720 medRxiv
Show abstract

Background: Physical inactivity remains a major public health challenge, and policy action across sectors is needed to create environments that enable active lifestyles. Several tools have been developed to evaluate national physical activity policies, but it remains unclear whether they assess similar policy aspects or produce comparable results. This study compared evaluation results across the 27 EU member states and examined the indicators used by these tools. Methods: Eight physical activity policy evaluation tools were included that (a) assess national physical activity policies and (b) had been applied in multiple countries. For five tools covering at least half of the EU member states, aggregate scores were extracted or calculated and converted into country rankings. A total of 202 indicators were coded and mapped by category (policy content, process, structure), and data type (audit, content, implementation, impact assessment). Results: Country rankings varied considerably, with a mean difference of 13.0 positions between highest and lowest ranks. Some countries displayed consistent results (e.g., Austria, France, Germany), while others showed large discrepancies (e.g., Hungary, Spain). Indicator mapping revealed a focus on policy content (47.6%), followed by structure (21.6%) and process (13.7%) (6.2% other). Across tools, 20.3% of indicators audited policies, 26.0% assessed policy content, and 26.4% assessed policy implementation (16.3% other). Conclusion: The systematic comparison of indicators across physical activity policy evaluation tools provided insights into differences in country rankings. Future evaluations should clearly communicate their purpose, identify synergies between tools, address gaps in policy evaluation, and strengthen the reproducibility of evaluation processes.

Matching journals

The top 2 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.