Tobacco Documents Online

338 Statement Smoking and Fetal Growth by Oliver Gilbert Brooke, M.D. FRCP Department of Child Health St. George's Hospital London SW 17 I am a graduate in Medicine of London University, attaining my basic medical degree in 1966. I received an MD degree (British medical equivalent of PhD) from London University in 1974. I am a pediatrician by training, with specialist interest in newborn medicine and nutrition. My present position is Reader in Pediatrics at St. George's Hospital (equivalent of full Professor in the United States) and I am head of Neonatology. Among other positions I have held, I have served as Scientific Officer of the Medical Research Counsel, Tropical Metabolism Research Unit at the University of the West Indies (1969-1972); Member of the Scientific Staff of the Division of Human Physiology, National Institute for Medical Research, Hampstead (1972-1973); Paediatric Registrar and Senior Registrar in Paediatrics at St. Mary's Hospital, London (1973-1976). I am a member of the European Advisory Committee on Feeding of Low Birth-weight Infants 339 which was organized by the Europea: Gastroenterology and Nutrition. Ii of the following scientific societ: Physiological Society, Neonatal Soc Association, European Society of Pf and Nutrition, and Paediatric ReseE an invited speaker and have present scientific meetings, and I am the.a scientific publications. - _ My interest in smoking and fetal gr 1970s when I carried out a study on size. This involved careful matchi for factors such as social status a accurate assessment of gestation. difficult to obtain in large scale but is very important in assessing When we analyzed our results we fou: for maternal age, height, gestation socio-economic status and race, and appeared to have little or no influ< other measures of fetal growth. (A: Archives of Disease in Childhood 53 Lancet 1, 1158, 1977). Since this i -2-

I 339 9 ty ute s an which was organized by the European Society of Paediatric Gastroenterology and Nutrition. In addition, I am a member of the following scientific societies: Nutrition Society, Physiological Society, Neonatal Society, British Paediatric Associition, European Society of Paediatric Gastroenterology and Nutrition, and Paediatric Research Society. I have been an invited speaker and have presented numerous papers at scientific meetings, and I am the.author of numerous scientific publications. My interest in smoking and fetal growth dates from the early 1970s when I carried out a study on ethnicity and birth size. This involved careful matching of pregnant women for factors such as social status and income, combined with accurate assessment of gestation. Such information is difficult to obtain in large scale epidemiological surveys but is very important in assessing the outcome of pregnancy. When we analyzed our results we found that, after controlling for maternal age, height, gestation of pregnancy, parity, socio-economic status and race, and infant sex, smoking appeared to have little or no influence on birth weight and other measures of fetal growth. (Alvear J & Brooke OG, Archives of Disease in Childhood 53, 27, 1978; Brooke O.G., Lancet 1, 1158, 1977). Since this unexpected finding was -2-

340 at variance with the prevailing views I became interested in carrying out further research in the subject. A review of the literature of the effects of smoking in pregnancy is contained in the U.S. Public Health Service report "Smoking and Health" (1979, 79-50066). This report marshals a large body of evidence to link smoking with low.birth weight. However there is a strong minority body of opinion which holds that the effects observed in humans are mainly mediated through social status or nutrition. The hypotheses here are as follows: 1. Women of low socio-economic status have been known for many years to have smaller babies than more privileged women. Smoking is a class linked practice. There are now far more smoking women in lower than in upper social class groups in Western societies. The link between smoking and low birth weight may therefore not be causal but mediated through other social•factors. 2. Smokers may be more likely than non-smokers to have small infants because of their personality or their genetic predisposition to respond unfavorably to stressful events. -3- 341 3. The effect of smoking c( nutrition; it is possib' or worse than non-smoker cause of the associatior reduced birth size. ' Evidence in favour of one or ano has been obtained by a number of 15-20 years: Yerushalmy produced evidence tha- to a particular type of reproducl J, Am. J. Obstet. Gynecol. 73, 8C Am J. Epidemiol. 93 443, 1971). Silverman's research supported th D.T. Am. J. Epidemiol. 105, 513, Davies et al, in a substantial an concluded that much of the effect was mediated through poor materna and was likely to have been of nu D P et al, Lancet 1, 385, 1976). -4-

341 i 3. The effect of smoking could be mediated through nutrition; it is possible that smokers eat less or worse than non-smokers, and that this is the cause of the association between smoking and reduced birth size. Evidence in favour of one or another of these hypotheses has been obtained by a number of workers over the past 15-20 years: Yerushalmy produced evidence that smoking was an "index to a particular type of reproductive outcome" (Yerushalmy J, Am. J. Obstet. Gynecol. 73, 808, 1957; Yerushalmy J, Am J. Epidemiol. 93 443, 1971). Silverman's research supported these conclusions (Silverman D.T. Am. J. Epidemiol. 105, 513, 1977). Davies et al, in a substantial and well documented study, concluded that much of the effect of smoking on fetal growth was mediated through poor maternal weight gain in pregnancy and was likely to have been of nutritional origin (Davies D P et al, Lancet 1, 385, 1976). -4- 0 tj ~ ~ CD ~ N

342 Rush reported similar'findings to Davies in an equally well- The• next 2 years will, it is hopf conducted study (Rush D, J. Obstet. Gynecol. Br. Cmmwlth. questions about the relationship: 81, 746, 1974). Rush also found that an effect of smoking intake and between psycho-social on birth weight was confined to women of lower social class This will lead directly to a lar€ (Rush D., Am J. Dis Children 129, 430, 1975). More recently survey of pregnancy outcome with these findings have been confirmed by Papoz et al, who found *.o data collection techniques, an no effect of smoking on birth weight in Parisian women of - of factors important in birth siz upper social class (Papoz L. et al, in: Maternal Nutrition The techniques used, particularly in Pregnancy, Ed. Dobbing J. London: Academic Press, 1981). collection of social and nutritio thoroughly developed and tested i I have been involved in research during the last two years to try to clarify these issues. This research has been - supported by the American tobacco industry. It involves a team approach (epidemiological, obstetric, pediatric, nutritional and psychological) to the investigation of pregnancy outcome in a hospital providing maternity services to a typical urban community with widely varying socio- economic conditions. The initial 18 month period has been involved with a pilot study of an unselected group of 160 women, examining the feasability of employing various study instruments to evaluate social status, psychological health and nutrition. The results are in the process of : analysis. -5- March 8, 1982 -6-

9 343 The next 2 years will, it is hoped, be devoted to answering questions about the relationships between smoking and food intake and between psycho-social stress and birth size. This will lead directly to a large scale epidemiological survey of pregnancy outcome with very careful attention to data collection techniques, and multivariate analysis of factors important in birth size and pregnancy outcome. The techniques used, particularly in relation to the collection of social and nutritional data, will have been thoroughly developed and tested in the initial studies. March 8, 1982 -6- I 0