Background: Obesity rates have been rising steeply across the globe in recent decades, posing a major threat to global human health. Despite this almost universal increase, differences between countries remain striking, even among equally developed societies.
Methods: We test if two cultural dimensions derived from a revised Hofstede model of culture from Minkov (2018), namely collectivism vs. individualism and monumentalism vs. flexibility, could help explain national variations in prevalence of obesity (BMI ≥ 30) among women and men around the world. We develop a theoretical framework that links these two cultural dimensions with obesity and then test their association empirically in analyses including 51 countries from all regions of the world as well as using imputed data for a total of 155 countries, representing 98% of the global population.
Results: In contrast to previous studies, we find that, adjusting for undernourishment and other potential confounds, individualism is associated with higher obesity prevalence in the male population, but not among the female population. We explain these findings by pointing to the different mechanisms through which individualism relates to health behavior, some of which are more gender-specific than others. A further novel finding is that flexibility, a national cultural trait that emphases humility, self-control, and restraint of desires, is a strong negative predictor of obesity in both genders beyond various potential confounds and is highly robust in specification curve analyses.
Conclusions: Our findings suggest that taking national culture into account can enhance our understanding of the obesity pandemic and should thus be considered by policy-makers in their design of interventions.
Bibliographical noteFunding Information:
The work of the second author was funded by the Estonian Research Council , Grant number: PRG380 , and by the National Research University - Higher School of Economics, Basic Research Program .
We are thankful to Takuro Furusawa, Boyd Swinburn, the participants of the Ronald F. Inglehart Laboratory for Comparative Social Research Regular Seminar, three anonymous reviewers, and the editor, Blair T. Johnson, for helpful feedback.
© 2022 Elsevier Ltd