Maternal mortality remains a significant public health concern in Nairobi slums, with community-based interventions potentially offering effective solutions. A mixed-methods approach was employed, including baseline surveys, a randomized controlled trial with CHW training, and post-intervention follow-up assessments. Data from 100 randomly selected households were analysed for changes in maternal health outcomes. CHW training resulted in an increased knowledge score of 25% among trained compared to non-trained CHWs (mean pre-training score: 48%, mean post-training score: 60%), with a significant reduction in home deliveries by 30% post-intervention. Training community health workers significantly improved their maternal health knowledge and practices, leading to a notable decrease in home deliveries and potential reductions in maternal mortality rates. Further training programmes should be expanded to include all CHWs in Nairobi slums to achieve broader impact. Longitudinal studies are recommended to monitor sustained effects over time. Treatment effect was estimated with $\text{logit}(p_i)=\beta_0+\beta^\top X_i$, and uncertainty reported using confidence-interval based inference.