Protein-energy malnutrition (PEM) affects millions of children in the developing world. The relationship between malaria and PEM is controversial. The goal of this study was to evaluate whether undernutrition is associated with increased or decreased malaria attributable morbidity. Three cross-sectional surveys were conducted using insecticide-treated bed nets (ITNs) among children aged 0–36 months living in an area with intense malaria transmission. Data were collected on nutritional status, recent history of clinical illness, socioeconomic status, current malaria infection status, and hemoglobin. In multivariate models, stunted children had more malaria parasitemia (odds ratio [OR] 1.98, P< 0.0001), high-density parasitemia (OR 1.84; P< 0.0001), clinical malaria (OR 1.77; P< 0.06), and severe malarial anemia (OR 2.65; P< 0.0001) than nonstunted children. The association was evident in children with mild-to-moderate (− 3< height-for-age Z-score [HAZ]<− 2) and severe stunting (HAZ<− 3). The cross-sectional nature of the study limits the interpretation of causality, but the data provide further observational support that the presence of undernutrition, in particular chronic undernutrition, places children at higher, not lower risk of malaria-related morbidity.