Background: Community health workers in Dangme-West district, Ghana, treated children aged 2-59 months with fever with either artesunate-amodiaquine (AAQ) or AAQ plus amoxicillin (AAQ + AMX) within a cluster-randomized controlled trial (registration no. TDR/UNDP Trial registration A: 20189). The intervention was introduced in a stepped-wedge manner. The aim of the study was reduction of mortality. This paper reports on the reduction of morbidity, notably anaemia, severe anaemia and severe illness. Clusters of 100 children were randomized in to AAQ, AAQ + AMX and pre-intervention arms. Six months later the pre-intervention clusters were randomized in to the AAQ and AAQ + AMX arms.
Methods: Data were collected in eight cross-sectional surveys. Using stratified sampling, 10 clusters were randomly selected per survey. Blood samples were taken to assess haemoglobin. Caregivers were interviewed about diseases (signs and symptoms) among their children in the preceding 14 days. Multivariate logistic regression analysis was used to determine the impact on anaemia, severe anaemia and severe illness.
Results: Compared with the pre-intervention clusters, anaemia was reduced in the AAQ (OR = 0.20, 95% CI 0.12-0.33) and AAQ + AMX (OR = 0.23, 95% CI 0.15-0.36) clusters, severe anaemia was reduced in the AAQ (OR = 0.20, 95% CI 0.09-0.45) and AAQ + AMX (OR = 0.12, 95% CI 0.04-0.31) clusters and severe illness was reduced in the AAQ (OR = 0.46, 95% CI 0.26-0.80) and AAQ + AMX (OR = 0.38, 95% CI 0.22-0.63) clusters. No significant differences were found in outcome variables between the AAQ and AAQ + AMX clusters.
Conclusions: Treating fever with antimalarials significantly reduced the prevalence of anaemia, severe anaemia and severe illness. We found no significant reduction in outcomes when the AAQ and AAQ+AMX clusters were compared.
Keywords: Amoxicillin; Anaemia; Artesunate; Children; Community; Fever.