Abstract

Background: Little is known about how population- attributable risks (PAR) for adverse birth outcomes due to smoking differ in adolescent and adult pregnancies. Methods: An analysis of community and hospital-based cross-sectional studies in Liverpool was undertaken to estimate the PAR values of low birthweight (LBW), preterm birth, and small for gestational age (SGA) births resulting from pregnancy smoking covering the period between 1983 and 2003. Maternal smoking status and pregnancy outcomes were available for a sample of 12 631 women. Results: The prevalence of maternal pregnancy smoking was 40 in the community sample and 33 in adults and 40 among adolescent pregnancies in the hospital sample. The PAR values (95 CI) associated with LBW, preterm birth and SGA outcomes due to maternal pregnancy smoking in the community sample were 27 (25-30), 13 (11-15) and 25 (23-27), respectively. The PAR values in adults in the hospital sample were 29 (27-31) for LBW, 16 (14-19) for preterm birth and 28 (26-31) for SGA. The corresponding PAR values in adolescents were 39 (34-43), 12 (7-18) and 31 (23-40). The LBW risk attributed to pregnancy smoking in adolescents was significantly higher than for adults (P = 0.05). Conclusion: About one-third of LBW, one-quarter of SGA and one-sixth of preterm births could be attributed to maternal smoking during pregnancy. The magnitude of the problem was greater among adolescent pregnancies, among whom a sub-group of mothers with very high risk for adverse birth outcomes due to pregnancy smoking was identified. (C) 2007 The Royal Institute of Public Health. Published by Elsevier Ltd. All rights reserved.