Robust cross-country comparison of children meeting 24-HR movement guidelines: an odds solution for binary effect efficiency measures

Publication Name

Journal of Public Health (Germany)


Aim: International comparisons of public health measures relative to observed best practice are increasingly important for evaluating community health promotion strategies and policies such as meeting or not meeting public health guidelines. This study aimed to identify methods enabling robust evaluation with such binary effect measures at a population level. Subject and methods: Conventional efficiency comparisons of binary effect proportions are problematic due to a lack of consistency with alternate framing of resulting relative risks. In this paper, we illustrate such inconsistent efficiency measures comparing the proportion of school age children (9–11 years) meeting or not meeting integrated movement guidelines (IMGs) across the 12 countries from the International Study of Childhood Obesity, Lifestyle and the Environment (ISCOLE) study. IMGs jointly consider physical activity, sleep and sedentary behaviours. An odds method is developed to enable consistent efficiency comparison with alternative framing of binary effects. Results: A novel odds solution to relative risk problems arising with conventional efficiency comparison of binary effects with alternative framing is shown to provide consistent efficiency measures relative to best practice. Furthermore, this technical advancement is shown to extend to consistent indirect comparison and evidence translation. Conclusion: Robust methods for international cross-country comparison of binary effect measures such as meeting or not meeting guidelines are identified with a novel odds ratio method. This novel solution is particularly important for health promotion evaluation of IMGs given the need for consistent comparison in evaluating practice evidence of what works now and consistent evidence translation of treatment effects as and when they emerge.

Open Access Status

This publication may be available as open access



Link to publisher version (DOI)