Abstract
OBJECTIVE: We sought to identify factors associated with linear growth among under-5 children in two urban informal settlements in Nairobi. DESIGN: We used longitudinal data for the period 2007-2012 from under-5 children recruited in the two sites between birth and 23 months and followed up until they reached 5 years of age. We fitted a generalized linear model on height-for-age Z-scores using the generalized estimating equations method to model linear growth trajectories among under-5 children. Known for its flexibility, the model provides strong parameter estimates and accounts for correlated observations on the same child. SETTING: Two urban informal settlements in Nairobi, Kenya.ParticipantsUnder-5 children (n 1917) and their mothers (n 1679). RESULTS: The findings show that child weight at birth, exclusive breast-feeding and immunization status were key determinants of linear growth among under-5 children. Additionally, maternal characteristics (mother's age, marital status) and household-level factors (socio-economic status, size of household) were significantly associated with child linear growth. There were biological differences in linear growth, as female children were more likely to grow faster than males. Finally, the model captured significant household-level effects to investigate further. CONCLUSIONS: Findings from the study point to the need to improve the targeting of child health programmes directed at the urban poor population in Nairobi. Specific modifiable determinants of child linear growth, particularly child weight at birth, exclusive breast-feeding, immunization status and mother's background characteristics, should be considered when designing interventions aiming at addressing child health inequities in these settings.