Residual confounding explains the association between high parity and child mortality

Area of Science:

• Demography
• Public Health
• Epidemiology

Background:

• High parity is a demographic factor potentially influencing child survival.
• Previous studies suggest an association between high parity and increased child mortality.
• Selection bias and confounding factors may obscure the true relationship.

Purpose of the Study:

• To investigate the impact of high parity on under-five and neonatal mortality.
• To differentiate between physiological effects and confounding factors in the parity-mortality relationship.
• To analyze data from Demographic and Health Surveys (DHS) in low- and middle-income countries.

Main Methods:

• Analysis of DHS datasets from 47 low- and middle-income countries.
• Inclusion of women aged 35+ to measure completed fertility.
• Log-binomial regression controlling for wealth, education, residence, maternal age, time trends, and birth interval.
• Stratification by completed fertility to address confounding.

Main Results:

• A statistically significant association was found between high parity and increased child mortality.
• This association is largely attributed to background characteristics of mothers with high completed fertility, not physiological factors.
• Children of high completed fertility mothers showed increased mortality risk at every birth order, even after controlling for confounders.

Conclusions:

• The observed increased mortality risk associated with high parity is not physiologically driven.
• Maternal background characteristics, strongly correlated with high fertility, are the primary drivers of increased child mortality risk.
• Apparent increases in mortality with parity are due to population-level shifts towards mothers with adverse characteristics, not inherent risks of higher birth orders.