OBJECTIVE: Fatigue is a common complaint worldwide and associated with disability and high health service use costs. We tested the hypothesis that maternal fatigue would be associated independently with maternal common mental disorder ('maternal CMD') in a rural, low-income country setting. METHODS: The analysis was conducted using data from a population-based cohort located in the Butajira demographic surveillance site, Ethiopia. A total of 1065 women were recruited in pregnancy and followed up to 2.5 (n = 1009; 94.7%) and 3.5 years post-partum (n = 989; 92.9%). Maternal CMD symptoms were measured using a locally validated version of the Self-Reporting Questionnaire and fatigue was measured using a dichotomised item from the Patient Health Questionnaire-15. Physical health indicators included haemoglobin level, body mass index and illness episodes. Generalised estimating equations were used to conduct hypothesis-driven and exploratory multivariable analyses in the panel at 2.5 and 3.5 years. RESULTS: The prevalence of maternal fatigue was 8.3% at 2.5 years and 5.5% at 3.5 years post-partum. Psychological symptoms of maternal CMD were associated independently with complaints of fatigue after adjusting for anaemia, body mass index, physical ill health, poverty and other confounding variables: adjusted odds ratio (aOR), 1.46; 95% confidence interval (CI), 1.28-1.66 for each one point increase in SRQ score. In the multivariable model, only psychosocial factors (CMD and stressful life events) and self-reported physical ill health were associated significantly with complaints of fatigue. CONCLUSION: Complaints of fatigue are associated strongly with maternal CMD and other psychosocial risk factors in this rural, low-income country setting with a high burden of undernutrition and infectious disease. Fatigue should be understood as a potential indicator of CMD in primary care to improve detection and treatment.