The objective of this study was to investigate the risk of stillbirth, neonatal death and lethal congenital anomaly in relation to proximity to hazardous industrial facilities at the time of birth in Cumbria, 1950–93. A retrospective cohort study was carried out using all 4325 stillbirths, 3430 neonatal deaths and 1569 deaths from congenital anomaly among the 287 993 births to mothers living in Cumbria between 1950 and 1993. Logistic regression was used to investigate the risk of each outcome in relation to proximity at birth to hazardous industrial sites, stratifying the analysis by time period and adjusting for social class, year of birth, birth order and multiple births. Continuous odds ratios for trend with proximity to sites were estimated. No significantly increased risk was found for stillbirth, lethal congenital anomaly or neonatal deaths in relation to proximity to hazardous industrial facilities, except for deaths from congenital heart defects in 1983–93. Overall, there was no evidence to suggest an increased risk of adverse lethal pregnancy outcome among babies whose mothers lived closer to hazardous industrial sites. The significantly increased risk of lethal congenital heart defects with proximity to these sites was likely to be a chance finding, given the large number of outcome groups and time periods analysed. However, effective environmental monitoring of industrial processes is required to enable studies to investigate the potential health risks of industrial pollution.