Background.There has been some concern that exposure to electromagnetic fields may cause birth defects. We studied risks of birth defects by residential exposure to 50-Hz magnetic fields from power lines.Methods.We estimated the distance between residence and power lines for 161,844 Norwegian residences, and their corresponding magnetic fields in the period 1980 to 1997. Risks of 24 categories of birth defects were compared across exposure levels, adjusting for social and demographic variables.Results.Among those living near power lines, the greatest reductions in risk were for cardiac defects (odds ratio = 0.5; 95% confidence interval = 0.3–0.9) and respiratory defects (0.4; 0.2–0.9). The largest increase in risk was for esophageal defects (2.5; 1.0–5.9). Other associations were weaker and had wide confidence intervals.Conclusions.There was little evidence that residence near power lines affected the risk of birth defects. The observed decreased risks of cardiac and respiratory defects and the increased risk of esophageal defects should be interpreted with caution given the number of endpoints, the imprecision in the calculations of the distance from the residence to the power line, and the limited information on pregnant women's change of residence.