BACKGROUND: Anaemia is a public health problem in Peru. In the Loreto region of the Amazon, ≥50% of children may be anaemic, although insufficient information exists for rural villages. METHODS: To generate more data about childhood anaemia in the Peruvian Amazon, haemoglobin was measured as part of a trachoma survey in 21 randomly selected villages. All children 1-9 y of age from 30 randomly selected households per village were recruited. Anaemia was classified according to the World Health Organization guidelines and a socio-economic status (SES) index was created for each household using principal component analysis. Spatial autocorrelation was determined using Moran's I and Ripley's K function. RESULTS: Of 678 children with complete haemoglobin data, 25.4% (95% confidence interval [CI] 21.2 to 30.1) had mild-or-worse anaemia and 22.1% (95% CI 15.6 to 30.3) had moderate-or-worse anaemia. Mild-or-worse anaemia was more common among children whose primary source of drinking water was surface water (prevalence ratio [PR] 1.26 [95% CI 1.14 to 1.40], p<0.001) and who were in the lowest SES tercile (PR 1.16 [95% CI 1.02 to 1.32], p=0.021). Moderate-or-worse anaemia was more common among boys (PR 1.32 [95% CI 1.09 to 1.60], p=0.005). No evidence of geospatial clustering was found. CONCLUSIONS: Remote villages of the Amazon would benefit from interventions for childhood anaemia and the poorest households would have the most to gain. Integrating anaemia screening into neglected tropical diseases surveys is an opportunity to use public health resources more efficiently.