Early childhood exposure to non-violent discipline and physical and psychological aggression in low- and middle-income countries: National, regional, and global prevalence estimates

Abstract

Background: Advocates for children's rights have recommended the elimination of all forms of violent discipline given its detrimental effects on children's development. Yet, little is known about the global prevalence of various forms of discipline, including physical and psychological aggression, as well as alternative forms of non-violent discipline, especially in low- and middle-income countries (LMICs).

Objective: This study aims to obtain national, regional, and global prevalence estimates of the percentage and number of 2- to- 4-y-olds in LMICs exposed to these disciplinary practices by their caregivers.

Participants and setting: We use data collected between 2010 and 2016 from 107,063 2- to- 4-y-old children living in 49 LMICs as part of the Multiple Indicator Cluster Survey (MICS).

Methods: Using the best-fitting model based on cross-validation techniques, we performed predictive modeling to generate country-level prevalence estimates for 131 LMICs in 2013, as well as 95% confidence intervals around these estimates.

Results: We estimate that 296.2 million 2- to- 4-y-olds (95% CI 256.9, 300.9) were exposed to non-violent discipline in 2013, which corresponds to 83.9% of the population. Furthermore, 220.4 million (95% CI 138.1, 283.7) and 230.7 million (95% CI 128.4, 300.6) children were exposed to aggressive physical and psychological discipline, respectively, which corresponds to prevalence of 62.5% and 65.4%. We also identify a high heterogeneity in the estimates across and within regions, finding a higher prevalence of both violent disciplinary methods in South Asia and Sub-Saharan Africa.

Conclusions: These results suggest the need for new policies and programs to minimize violent discipline around the world.

Keywords: Child discipline; Low- and middle-income countries; Non-violent discipline; Physical punishment; Predictive modeling; Psychological aggression.