Introduction This prospective, cohort study compares child protection outcomes over the first 5 years of life in a group of children born to self-declared drug-using mothers recruited during pregnancy (cases) and a group of children matched for gestational age, chronological age, maternal neighbourhood and place of delivery whose mothers made no such declaration of problematic drug use (controls). Methodology We monitored local child protection registers to identify cohort members who came to the attention of the local authority. Results Of the 71 original cases and 142 original controls, 55 (77%) and 96 (68%) remained in the area enrolled in local schools at 5 years of age. In total, 26 (47.3%) of the case children were subject to child protection procedures compared with 18 (18.8%) of the control children. This risk difference of 28.5% (95% CI 13.2% to 43.9%) has increased marginally since our previous report in this journal of child protection outcomes at 18 months of age (32% vs. 7%). However, the level of intervention deemed necessary to protect the child has increased significantly with six cases (compared with one control child) taken into the care of the local authority. ConclusionsDespite early maternal intentions and multiple supportive interventions, 27% of children born to women with significant substance abuse problems in our area required child protection during the pre-school years. Child protection risk assessment procedures need to weigh problematic maternal drug use heavily. Intervention studies with child welfare outcomes are needed to identify the most effective harm reduction strategies and inform public debate on how we can minimize child abuse related to substance misuse.