Background Globalisation can transform food systems and contribute to a nutrition transition from local diets to westernised diets high in ultra-processed foods. This is likely to influence malnutrition in all its forms, however there is no evidence on its association with the double burden of malnutrition (DBM), i.e. the co-existence of overnutrition and undernutrition within the same individual, household, or population. This analysis investigates the association between DBM, expressed as a stunted child and overweight mother in the same household, and economic, social, and political globalisation, and how these differ by country income and household wealth.

Methods We pooled anthropometric and demographic data for 1,131,069 children aged under 5 years and their mothers, aged 15–49 years, from 189 Demographic and Healthy Surveys (DHS) conducted between 1992 and 2018 in 55 low- and middle-income countries (LMICs). These were combined with country-level data on globalisation, using the KOF Globalisation Index de jure, and gross national income (GNI) from the World Bank. Multi-variate associations between the DBM and globalisation were tested using logistic regression with country and time fixed-effects and robust standard errors clustered by country. Interactions between globalisation, household wealth index and GNI were also included in the model.

Results The DBM was distributed unequally across groups of affluence in our sample, with higher probability of DBM among rich households of LMICs at low GNI levels, and among poor households of LMICs at high GNI levels. We identified that economic globalisation was associated with higher odds of children being stunted and having an overweight mother among the poorest but not the richest wealth quintiles (OR; 95% CI: 1.52; 1.18–1.97 for poorest quintile). These associations were attenuated as country income increased. Social globalisation was also associated with higher odds of children being stunted and having an overweight mother, although this did not vary significantly by household wealth or country income (OR; 95% CI: 1.43; 1.18–1.74). No associations were identified between political globalisation and the DBM.

Conclusion This analysis suggests that the DBM is affecting rich households of poorer LMICs and poor households of richer LMICs the most. It also indicates that economic and social globalisation are associated with higher DBM, although the impacts of economic globalisation are mostly realised by the world’s poorest.