The Maasai of northern Tanzania, a semi-nomadic ethnic group predominantly reliant on pastoralism, face a number of
challenges anticipated to have negative impacts on child health, including marginalisation, vulnerabilities to drought,
substandard service provision and on-going land grabbing conflicts. Yet, stemming from a lack of appropriate national
survey data, no large-scale comparative study of Maasai child health has been conducted. Savannas Forever Tanzania
surveyed the health of over 3500 children from 56 villages in northern Tanzania between 2009 and 2011. The major ethnic
groups sampled were the Maasai, Sukuma, Rangi, and the Meru. Using multilevel regression we compare each ethnic group
on the basis of (i) measurements of child health, including anthropometric indicators of nutritional status and self-reported
incidence of disease; and (ii) important proximate determinants of child health, including food insecurity, diet,
breastfeeding behaviour and vaccination coverage. We then (iii) contrast households among the Maasai by the extent to
which subsistence is reliant on livestock herding. Measures of both child nutritional status and disease confirm that the
Maasai are substantially disadvantaged compared to neighbouring ethnic groups, Meru are relatively advantaged, and
Rangi and Sukuma intermediate in most comparisons. However, Maasai children were less likely to report malaria and worm
infections. Food insecurity was high throughout the study site, but particularly severe for the Maasai, and reflected in lower
dietary intake of carbohydrate-rich staple foods, and fruits and vegetables. Breastfeeding was extended in the Maasai,
despite higher reported consumption of cow’s milk, a potential weaning food. Vaccination coverage was lowest in Maasai
and Sukuma. Maasai who rely primarily on livestock herding showed signs of further disadvantage compared to Maasai
relying primarily on agriculture. We discuss the potential ecological, socioeconomic, demographic and cultural factors
responsible for these differences and the implications for population health research and policy.