Impact of rural-urban migration on indirect child mortality estimation in Kenya

Abstract

When estimating child mortality rates using the indirect method, it is assumed that all reported births and deaths occurred in the place where the mothers resided at the time of the survey. However, the migration of women can result in transferring data about deceased children from one place to another. In many developing countries, substantial migration happens between rural and urban regions, where child mortality disparities are significant. This migration creates challenges as child mortality estimates for rural and urban areas computed under the assumption of non-migration are likely to be erroneous. Our study pooled data from six Kenya Demographic and Health Surveys between 1989 and 2014. The study aimed to establish statistical evidence of the impact of rural-urban migration on indirect child mortality estimates. Our findings indicate that the inclusion of deceased children born to women who migrated from rural to urban regions led to a significant overestimation of infant, one-to-four, and under-five mortality rates in urban areas. On average, the overestimation of infant mortality rates ranged from 2.5% to 21.7%, while one-to-four mortality rate overestimation ranged from 4.0% to 41.2%. The average overestimation of under five mortality rate was between 3.0% and 26.8%. Based on these results, future indirect estimates of child mortality for rural and urban regions should be adjusted to account for the impacts of migration between these areas. Furthermore, it is essential to consider re-estimating trends of child mortality for rural and urban regions in Kenya to better understand the timing of mortality convergence between these regions.