Parental smoking and risk of childhood brain tumors

UWA Research Publication Milne, E., Greenop, K. R., Scott, R. J., Ashton, L. J., Cohn, R. J., de Klerk, N. H. and Armstrong, B. K. (2013), Parental smoking and risk of childhood brain tumors. Int. J. Cancer, 133: 253–259. doi: 10.1002/ijc.28004. Copyright © 2012 UICC This is the peer reviewed version of the following article: Milne, E., Greenop, K. R., Scott, R. J., Ashton, L. J., Cohn, R. J., de Klerk, N. H. and Armstrong, B. K. (2013), Parental smoking and risk of childhood brain tumors. Int. J. Cancer, 133: 253–259. doi: 10.1002/ijc.28004, which has been published in final form at http://dx.doi.org/10.1002/ijc.28004. This article may be used for non-commercial purposes in accordance with Wiley Terms and Conditions for self-archiving. This version was made available in the UWA Research Repository on 1 July 2014 in compliance with the publisher’s policies on archiving in institutional repositories. Use of the article is subject to copyright law. SHORT REPORT Parental smoking and risk of childhood brain tumours Short title: Parental smoking and CBT Authors: Elizabeth Milne1, Kathryn R.Greenop1, Rodney J. Scott2, 3, Lesley J. Ashton4, Richard J. Cohn5, 6, Nicholas H. de Klerk1, Bruce K.Armstrong7. 1 Telethon Institute for Child Health Research, Centre for Child Health Research, University of Western Australia, Perth, Western Australia, Australia. 2 Hunter Medical Research Institute, School of Biomedical Sciences, Faculty of Health, University of Newcastle, New South Wales, Australia. 3 Hunter Area Pathology Service, HNEHealth, Newcastle, New South Wales, Australia. 4 Children’s Cancer Institute Australia for Medical Research, Lowy Cancer Research Centre, University of New South Wales, Sydney, NSW, Australia. 5 Centre for Children's Cancer and Blood Disorders, Sydney Children's Hospital, Sydney, Australia; 6 School of Women's and Children's Health, Faculty of Medicine, University of New South Wales, NSW, Australia. 7 Sydney School of Public Health, University of Sydney, Sydney, New South Wales, Australia Author for correspondence: Dr Elizabeth Milne Telethon Institute for Child Health Research PO Box 855 West Perth, Western Australia 6872, Australia Telephone: +61 08 9489 7756 Facsimile: +61 08 9489 7700 Email: lizm@ichr.uwa.edu.au 1 Keywords: children, central nervous system neoplasms, tobacco, smoking, pregnancy, conception Abbreviations: CBT: childhood brain tumors, OR: odds ratio, CI: confidence interval, CD: collection district, IRSD: Index of relative socioeconomic disadvantage, RDD: random digit dialling, PNET: primitive neuroectodermal tumor. Current text word count: 2325 Novelty and impact of paper: In our national study of childhood brain tumors, we used a novel method to measure parental smoking. Parents were asked about annual smoking quantity, with recall anchored to occupational and residential history to reduce bias. Overall, parents’ smoking was not related to brain tumour risk, but the OR for maternal smoking during pregnancy was consistent with a four-fold increase in risk among children aged under 2 years, suggesting a possible association with early childhood tumours. 2 Abstract = 245 Childhood brain tumors (CBT) are the leading cause of cancer death in children, yet their etiology remains largely unknown. Tobacco smoke contains 61 known carcinogens and increases the risk of several adult cancers. This study investigated associations between parental smoking and risk of CBT in a population-based case-control study conducted between 2005 and 2010. Cases were identified through all 10 Australian paediatric oncology centres, controls via nationwide random-digit dialling, frequency matched to cases on age, sex and state of residence. Parental smoking information was obtained for 302 cases and 941 controls through mailed questionnaires that requested average daily cigarette use in each calendar year from age 15 to the child’s birth, linked to residential and occupational histories. Data were analysed using unconditional logistic regression, adjusting for frequency matching variables and potential confounders. Overall, parental smoking before or during pregnancy showed no association with CBT risk. The odds ratios for maternal smoking before and during pregnancy were 0.99 (95% CI: 0.70, 1.40) and 0.89 (95% CI: 0.61, 1.21) respectively, and those for paternal smoking before and during pregnancy were 0.99 (95% CI: 0.71, 1.38) and 1.04 (95% CI: 0.74, 1.46) respectively. In children under 24 months of age, the odds ratios for maternal smoking preconception and during pregnancy were 5.06 (95% CI 1.3519.00) and 4.61 (95% CI: 1.08, 19.63), although these results were based on modest numbers. Future studies should investigate the associations between maternal smoking and risk of CBT by the child’s age of diagnosis. 3 Introduction Childhood brain tumors (CBT) are the leading cause of cancer death in children, and apart from a few genetic syndromes and ionising radiation, their etiology is largely unknown. A range of environmental factors have been investigated as potential risk factors, but most of the findings have been inconclusive.1 Tobacco smoke contains 61 known carcinogens and is considered to increase the risk of adult cancers including lung, oro-pharyngeal, pancreatic and renal cancers.2 The results of previous studies of maternal smoking and CBT risk have been inconsistent. While most early studies, including a 1996 review3 and a 2002 meta-analysis,4 concluded there was no evidence of associations with maternal smoking before pregnancy and during pregnancy, moderately increased risks with smoking before5 pregnancy and during pregnancy6 have recently been reported.. In addition, Schuz and co-workers reported positive associations between maternal smoking during pregnancy and risk of ependymoma and medulloblastoma, but not astrocytoma7 but other relatively recent studies have reported null results.8-10 Previous studies of paternal smoking and risk of CBT have also produced inconsistent results; two studies published after the review by Norman and colleagues3 reported positive associations,8, 11 while six reported no association.5, 7, 12-15 These inconsistent findings could be partly due to differences in study design and the way parental smoking histories were obtained and quantified; and, perhaps to lack of investigation of disease subtypes in some studies.4 Here we present results from our Australian casecontrol study, in which detailed smoking histories were obtained from both parents, and associations within CBT subtypes were examined. Methods The Australian Study of Childhood Brain Tumours (Aus-CBT) was a national populationbased case-control study conducted between 2005 and 2010. The study design has been described previously.16 Briefly, incident cases were identified through all 10 Australian pediatric oncology centers where most CBTs are treated. Cases were eligible for inclusion if they were resident in Australia and had a biological parent available with adequate English skills to complete the questionnaires. Cases diagnosed in 2005 were recruited retrospectively, while 2006-2010 cases were recruited as soon as possible after diagnosis. Controls were recruited by national random digit dialing (RDD) between 2005 and 2010, and frequency matched to cases by age (within 1 year), sex and state of residence in a ratio of approximately 4 3:1. The RDD method has been described in detail elsewhere.17 Aus-CBT was approved by the Human Research Ethics Committees at all participating hospitals. Questionnaires mailed to parents asked about demographic characteristics, medical histories, engagement in activities involving potential exposure to carcinogens and diet. Parents were asked to indicate the average number of cigarettes smoked per day (CPD) in each calendar year from age 15 until the year after the index child’s birth along with their residential address and occupation in each year. Smoking data were referenced to the child’s birth year so that smoking during critical periods relating to the pregnancy could be determined. Smoking during pregnancy was defined as smoking during the birth year (if the child was born in or after May) or during the year before the birth year (if the child was born before May). Preconception smoking was defined as smoking in the year before the birth (for births in or after May) or in the year two years before the birth (for births before May). We also investigated whether parental ever smoking, former smoking or pack-years of smoking were associated with CBT risk. In addition to parent-reported measures of socio-economic status (SES), each participant’s address was linked to an Australian Bureau of Statistics (ABS) Census Collection District (CD). The ABS assigns each CD a score for its area-based Index of Relative Socio-Economic Disadvantage (IRSD) after each quinquennial census [http://www.ausstats.abs.gov.au/Ausstats/subscriber.nsf/0/D729075E079F9FDECA25741700 11B088/$File/20390_2006.pdf]. Statistical analysis Unconditional logistic regression was used to estimate odds ratios (ORs) for the association between parental smoking in specific time periods and risk of CBT. All analyses included the frequency matching variables age, sex and State of residence, and variables that met the classical definition of confounding; that is, they were associated with case/control status, and with smoking among control parents. The variables included on this basis were ethnicity, parental age, child’s year of birth, household income and maternal alcohol consumption (for maternal smoking models). Interactions with child’s age and parental alcohol consumption were tested by fitting interaction terms in the models. Subgroup analyses were undertaken by CBT subtype where there were sufficient cases. Results We were notified of 794 CBT cases, of whom 64 were ineligible (36 with no Englishspeaking parent, 23 non-residents, five with no biological parent available). Of the 730 5 eligible cases, 568 (77.8%) were invited to take part by a physician, while a physician chose not to invite the other 162 for medical or psychosocial reasons. Parents of 374 cases consented (65.8% of invited, 51.2% of eligible). Information on smoking was available for 302 (85.1%) case mothers and 247 (68.5%) case fathers, while 29 families provided only demographic information and 39 provided no data. Between 2005 and 2010, 3,624 families eligible to be controls were identified by RDD, of whom 2,255 (62.2%) agreed to participate. In accordance with our age and sex frequency-matching quotas, we recruited 1,467 of these children to the study. Information on smoking was available for 941 control mothers (69.0% of recruited) and 801 control fathers (58.8% of recruited), while 413 families provided only demographic data and 104 families provided no data. Demographic and other characteristics of cases and controls who provided at least some data were similar, with some exceptions (Table 1). Controls were slightly more likely than cases to be female, have a mother aged over 35 years and have European ethnicity. A higher proportion of controls than cases were recruited in 2005-2006, as controls from our national leukemia study were frequency matched to CBT cases diagnosed in those years. The use of the 2005 and 2006 leukemia study controls also resulted in a higher percentage of controls than cases born between 1998 and 2003. Thus, the child’s age and year of recruitment were related. Both cases and controls lived in more socially advantaged CDs than the Australian population as a whole. The mean IRSD scores were 1025.3 for case CDs, 1030.5 for control CDs, and 1006.0 for all Australian CDs (t-test P-values: 0.33 for case vs control CDs and <0.001 for both case and control CDs vs all Australian CDs) (data not shown in tables). These findings are consistent with the observation that income and education were comparable among cases and controls (Table 1). Overall, 22% of mothers smoked preconception, 16.8% of mothers smoked during pregnancy, 29% of fathers smoked in the preconception year and 26% of fathers smoked during the pregnancy. Paternal smoking in both periods was moderately correlated with maternal smoking during pregnancy: Spearman’s  = 0.36, P <0.001. Overall, there was little evidence that maternal or paternal smoking was associated with risk of CBT (Table 2). Although the ORs for paternal smoking 1-14 CPD in both periods were around 1.3, the estimates lacked precision due to a relatively small number of smokers, and the ORs for higher levels of smoking were below the null. The results were similar when parental smoking was mutually adjusted, when the analysis was restricted to children whose other parent did not smoke (data 6 not shown), and when the reference level was changed to ‘no smoking from 2 years before the birth year’ for mothers, and ‘no smoking ever before the birth’ for the fathers (data not shown). The OR for maternal smoking during pregnancy and paternal smoking in the preconception year was 0.87 (95% CI: 0.50, 1.50). No evidence was seen for associations with ever smoking, ex-smoking, or increasing pack-years for either parent (results not shown). When stratified by the child’s age at diagnosis or recruitment, the ORs for maternal smoking, both preconception and during pregnancy, were high for CBT diagnosed under 24 months of age, although this was based on small numbers (Table 3). There was no evidence of a similar age-interaction for paternal smoking (Table 3). There was no evidence of an interaction between parental smoking and alcohol consumption (data not shown). For low grade gliomas, the ORs for maternal smoking during pregnancy and paternal smoking preconception were 1.02 (95% CI 0.62, 1.67) and 1.02 (95% CI 0.66, 1.60) respectively (results not shown in tables). For PNET/medulloblastoma, the ORs were 0.82 (95% CI 0.39, 1.70) and 0.90 (95% CI 0.47, 1.70) for mother and father respectively. Other subtypes had insufficient numbers to analyse separately. There were insufficient numbers to investigate any interaction between smoking and child’s age within CBT subtypes. Discussion Overall, neither maternal nor paternal smoking before or during pregnancy was associated with an increased risk of CBT; these findings are consistent with most previous reports, as summarised in the introduction. There was an indication that maternal smoking before or during pregnancy was associated with an increased risk of CBT in infants (less than 24 months old), but this was based on a small number of cases and could be due to chance. Two previous studies also reported some variation in the association between maternal smoking during pregnancy and CBT risk by the child’s age. A Swedish cohort study6 reported an increased risk of CBT among children aged between 2 and 4 years (OR 1.64, 95% CI: 1.15, 2.33), but found little or no evidence of an association at other ages; and a Californian study18 reported ORs of 0.68 (95% CI: 0 43, 1.1) for CBT among children younger than 6 years, and 1.1 (95% CI: 0.73, 1.6) among older children. No formal assessment of the age by smoking interactions was reported in these papers. Three additional studies reported no differences in associations by age.7, 15, 19 Only one previous study investigated possible effect modification by alcohol use;20 no modification of the overall null association with smoking was observed. The reasons for the inconsistencies among previous studies of parental 7 smoking are not clear, but may be due to a combination of factors: different windows of exposure, dose categories and distributions of CBT subtypes; use of proxy respondents, measurement error, recall bias and lack of investigation of age-specific associations. An increased risk of CBT among infants associated with maternal smoking during pregnancy is biologically plausible. Chemicals found in tobacco smoke – including neurocarcinogens – can cross the placenta21 and cause chromosomal damage in the fetus.22 Animal studies have shown that the developing brain is much more likely to develop tumors as a result of exposure to neurocarcinogens in utero than later in life,23 and this may also be the case for humans. Hence, it is plausible that tumors initiated through tobacco-induced damage to the fetal brain would manifest themselves relatively early in life. This is consistent with our finding that maternal smoking appeared to be associated with CBT risk only among very young children. Almost 78% of eligible cases were invited to participate by the treating clinician and 66% of invited parents consented, resulting in an overall participation fraction of 51%. Except for age and sex, where the distributions were similar to participating cases, information about eligible cases who did not participate was unavailable, so we could not determine whether our cases were representative of all eligible cases with respect to potential risk factors. Control families were recruited by national RDD using state-of-the-art methods and, according to the most recent data available, approximately 90% of Australian households had a landline telephone connection during the recruitment period [http://www.abs.gov.au/AUSSTATS/abs@.nsf/DetailsPage/1367.5Sep%202008]. Therefore, residences contacted are likely to be representative of the wider population. Telephone-based methods have been evaluated and shown to perform well in Australia for control subject recruitment, compared with other methods,24 and RDD is likely to be the best and most costeffective method available. Participation among eligible control families was 62% and, although no individual information was available for those who declined, area-based SES scores were higher among participating controls than among the wider Australian population. However, importantly, participating cases and controls had very similar SES distributions. Information about smoking was provided by approximately 81% case and 64% control mothers, and 66% case and 55% control fathers. Selection bias is possible since control parents who provided smoking data lived in areas with (on average) higher IRSD scores than those who did not; thus, they may have a lower prevalence of smoking. However, the 8 prevalence of smoking during pregnancy among our control mothers (16.8%) was similar to New South Wales Midwives data (population-based surveillance system) for birth years 19952007 (17.2%)25, suggesting that they are representative of the general population with respect to smoking. Further, the most likely consequence of selection bias due to the over- representation of higher SES controls would be inflated ORs for smoking; if this has occurred, then the true ORs for smoking would be even further below the null, which seems unlikely. All analyses were adjusted for household income and further adjustment for IRSD or education did not alter the effect estimates. Our questionnaires were designed to reduce measurement error by ‘anchoring’ parents’ recall of smoking to their residence and occupation in each year. Nonetheless, it is plausible that smoking was recalled less accurately by parents of older children (where the index pregnancy was further in the past). However, why the ORs among older children tended to be below unity is not clear, as parental smoking is unlikely to be protective against CBT among children of any age, and there is little reason to believe that recall of smoking among parents of older controls would be more complete than among parents of older cases. Given these results, it is reasonable to infer that the increased ORs observed for infants are not solely due to bias resulting from over-reporting of smoking among the parents of infant cases. In summary, this study provides little evidence that parental smoking in the preconception or pregnancy period is associated with an increased risk of CBT, except possibly among infant children of smoking mothers. We recommend that future studies investigate the associations between maternal smoking and risk of CBT by the child’s age of diagnosis. 9 Consortium statement The Aus-CBT consortium conducted the study and the Telethon Institute for Child Health Research (TICHR), University of Western Australia, was the coordinating centre. Bruce Armstrong (Sydney School of Public Health, University of Sydney), Elizabeth Milne, Nicholas de Klerk, Caroline Bower, Peter Dallas (TICHR), Frank van Bockxmeer (Royal Perth Hospital, University of WA), Rodney Scott and John Attia (University of Newcastle), Lin Fritschi (WA Institute for Medical Research), Lesley Ashton, Michelle Haber and Murray Norris (Children’s Cancer Institute Australia for Medical Research, Lowy Cancer Research Centre, UNSW), Margaret Miller (Edith Cowan University) and Judith Thompson (WA Cancer Registry) were the research investigators. The authors acknowledge the contribution made by our clinical co-investigators who recruited and cared for study patients at each participating hospital: Nicholas Gottardo (Princess Margaret Hospital, TICHR); John Heath and Elizabeth Smibert (Royal Children’s Hospital, Melbourne); Peter Downie (Monash Medical Centre, Melbourne); Tim Hassell and Ross Pinkerton (Royal Children’s Hospital Brisbane); Maria Kirby (Women’s and Children’s Hospital, Adelaide); Stewart Kellie and Luciano dalla Pozza (Children’s Hospital at Westmead); Frank Alvaro (John Hunter Hospital, Newcastle); Richard Cohn (Sydney Children’s Hospital) and John Daubenton (Royal Hobart Hospital). The authors also acknowledge the Clinical Research Associates at each hospital, and the study coordinators: Jackie Mansour, Somer Dawson, Tamika Heiden, and Helen Bailey. Funding: The National Health and Medical Research Council (NHMRC) funded Aus-ALL (Grant number: 254539) and Aus-CBT (Grant number: 404089). Elizabeth Milne was supported by an NHMRC Fellowship. Support for Rodney Scott was in part from NBN Children's Cancer Research Fund. 10 References 1. Baldwin RT, Preston-Martin S. Epidemiology of brain tumors in childhood--a review. Toxicol Appl Pharmacol 2004;199:118-31. 2. IARC. Tobacco smoke and involuntary smoking, IARC Monographs Evaluating the Carcinogenic Risks in Humans Volume 83. Geneva: World Health Organisation, 2004. 3. Norman MA, Holly EA, Preston-Martin S. Childhood brain tumors and exposure to tobacco smoke. Cancer Epidemiol Biomarkers Prev 1996;5:85-91. 4. Huncharek M, Kupelnick B, Klassen H. Maternal smoking during pregnancy and the risk of childhood brain tumors: a meta-analysis of 6566 subjects from twelve epidemiological studies. J Neurooncol 2002;57:51-7. 5. Sorahan T, McKinney PA, Mann JR, Lancashire RJ, Stiller CA, Birch JM, Dodd HE, Cartwright RA. Childhood cancer and parental use of tobacco: findings from the interregional epidemiological study of childhood cancer (IRESCC). Br J Cancer 2001;84:141-6. 6. Brooks DR, Mucci LA, Hatch EE, Cnattingius S. Maternal smoking during pregnancy and risk of brain tumors in the offspring. A prospective study of 1.4 million Swedish births. Cancer Causes Control 2004;15:997-1005. 7. Schuz J, Kaletsch U, Kaatsch P, Meinert R, Michaelis J. Risk factors for pediatric tumors of the central nervous system: results from a German population-based case-control study. Med Pediatr Oncol 2001;36:274-82. 8. Plichart M, Menegaux F, Lacour B, Hartmann O, Frappaz D, Doz F, Bertozzi A-I, Defaschelles A-S, Pierre-Kahn A, Icher C, Chastagner P, Plantaz D, et al. Parental smoking, maternal alcohol, coffee and tea consumption during pregnancy and childhood malignant central nervous system tumours: the ESCALE study (SFCE). Eur J Cancer Prev 2008;17:376-83. 9. Filippini G, Maisonneuve P, McCredie M, Peris-Bonet R, Modan B, Preston-Martin S, Mueller BA, Holly EA, Cordier S, Choi NW, Little J, Arslan A, et al. Relation of childhood brain tumors to exposure of parents and children to tobacco smoke: the SEARCH international case-control study. Surveillance of Environmental Aspects Related to Cancer in Humans. Int J Cancer 2002;100:206-13. 11 10. Stavrou EP, Baker DF, Bishop JF. Maternal smoking during pregnancy and childhood cancer in New South Wales: a record linkage investigation. Cancer Causes Control 2009;20:1551-8. 11. Cordier S, Monfort C, Filippini G, Preston-Martin S, Lubin F, Mueller BA, Holly EA, Peris-Bonet R, McCredie M, Choi W, Little J, Arslan A. Parental exposure to polycyclic aromatic hydrocarbons and the risk of childhood brain tumors: The SEARCH International Childhood Brain Tumor Study. Am J Epidemiol 2004;159:1109-16. 12. Sorahan T, Prior P, Lancashire RJ, Faux SP, Hulten MA, Peck IM, Stewart AM. Childhood cancer and parental use of tobacco: deaths from 1971 to 1976. Br J Cancer 1997;76:1525-31. 13. Hu J, Mao Y, Ugnat AM. Parental cigarette smoking, hard liquor consumption and the risk of childhood brain tumors--a case-control study in northeast China. Acta Oncol 2000;39:979-84. 14. Filippini G, Farinotti M, Ferrarini M. Active and passive smoking during pregnancy and risk of central nervous system tumours in children. Paediatr Perinat Epidemiol 2000;14:78-84. 15. Pang D, McNally R, Birch JM. Parental smoking and childhood cancer: results from the United Kingdom Childhood Cancer Study. Br J Cancer 2003;88:373-81. 16. Milne E, Greenop KR, Bower C, Miller M, van Bockxmeer FM, Scott RJ, de Klerk NH, Ashton LJ, Gottardo NG, Armstrong BK. Maternal use of Folic Acid and Other Supplements and Risk of Childhood Brain Tumors Cancer Epidemiol Biomarkers Prev 2012;21:1933-41. 17. Bailey H, Milne E, de Klerk N, Fritschi L, Bower C, Attia J, Armstrong B. Representativeness of child controls recruited by random digit dialing. Paediatr Perinat Epidemiol 2010;24:293-302. 18. Norman MA, Holly EA, Ahn DK, Preston-Martin S, Mueller BA, Bracci PM. Prenatal exposure to tobacco smoke and childhood brain tumors: results from the United States West Coast childhood brain tumor study. Cancer Epidemiol Biomarkers Prev 1996;5:127-33. 19. John EM, Savitz DA, Sandler DP. Prenatal exposure to parents' smoking and childhood cancer. Am J Epidemiol 1991;133:123-32. 12 20. Gold EB, Leviton A, Lopez R, Gilles FH, Hedley-Whyte ET, Kolonel LN, Lyon JL, Swanson GM, Weiss NS, West D, et al. Parental smoking and risk of childhood brain tumors. Am J Epidemiol 1993;137:620-8. 21. Jauniaux E, Burton GJ. Morphological and biological effects of maternal exposure to tobacco smoke on the feto-placental unit. Early Hum Dev 2007;83:699-706. 22. de la Chica RA, Ribas I, Giraldo J, Egozcue J, Fuster C. Chromosomal instability in amniocytes from fetuses of mothers who smoke. JAMA 2005;293:1212-22. 23. Rice JM, Ward JM. Age dependence of susceptibility to carcinogenesis in the nervous system. Ann N Y Acad Sci 1982;381:274-89. 24. Valery PC, Williams G, McWhirter W, Sleigh A, Scott D, Bain C. Electronic telephone directory listings: preferred sampling frame for a population-based study of childhood cancer in Australia. Ann Epidemiol 2000;10:504-8. 25. Mohsin M, Bauman AE, Forero R. Socioeconomic correlates and trends in smoking in pregnancy in New South Wales, Australia. J Epidemiol Community Health 2011;65:727-32. 13 Table 1: Distribution of demographic and birth characteristics in the Australian Study of Childhood Brain Tumors, 2005-2010. Child gender Female Male 0-1 2-4 5-9 10-14 NSW/ACT Victoria/Tasmania SA/NT Western Australia Queensland 1990-1997 1998-2003 2004-2010 2005-2006 122 180 30 85 89 98 102 85 19 42 54 84 125 93 107 40.4 59.6 9.9 28.1 29.5 32.5 33.8 28.1 6.3 13.9 17.9 27.1 41.4 30.8 35.4 445 496 110 303 293 235 283 251 77 112 218 223 469 249 415 47.3 52.7 11.7 32.2 31.1 25.0 30.1 26.7 8.2 11.9 23.2 23.7 49.8 26.5 44.1 2007-2008 2009-2010 ≤24 25-34 35+ <24 25-34 35+ Didn’t complete secondary school Complete secondary school and/or trade certificate University/College Up to $40, 000 $40, 001-$70,000 $70,001-$100,000 >$100, 000 1 2 3+ European At least 50% 99 96 45 187 70 15 151 100 42 32.8 31.8 14.9 61.9 23.2 5.6 56.8 37.6 13.9 268 258 87 593 261 26 434 335 91 28.5 27.4 9.2 63.0 27.7 3.3 54.6 42.1 9.7 99 32.8 301 32.0 161 49 78 80 93 138 98 66 185 73 53.3 16.3 26.0 26.7 31.0 45.7 32.5 21.9 61.3 24.2 549 127 262 249 299 399 326 216 679 170 58.3 13.6 28.0 26.6 31.9 42.4 34.6 23.0 72.2 18.1 Child state residenceb Birth year Year of diagnosis/ Recruitment Maternal age group Paternal age group Best parental education Household income Birth order Ethnic groupc 14 Control n 1363 Control %a Category Child age group Case n 335 Case %a Variable Provided demographic data Mother returned exposure questionnaire Father returned exposure questionnaire 302 941 247 801 European At least 50% non12 4.0 30 3.2 European Indeterminate 32 10.6 62 6.6 Tumor diagnosis Low grade 144 47.7 gliomas High grade 26 8.6 gliomas Embryonal 71 23.5 tumorsd Germ cell tumors 20 6.6 Ependymomas 22 7.3 e Others 19 6.3 a Percentages are of participants whose mother returned the exposure questionnaire. b ACT: Australian Capital Territory; NSW, New South Wales; NT, Northern Territory; SA, South Australia. c European, at least 3 European grandparents; 50% European, 2 European grandparents; at least 50% non-European, 2 non-European grandparents and ethnicity of 2 other grandparents non-European or unknown; indeterminate, no 2 grandparents of same ethnicity (i.e European or non-European) and 2+ grandparents of unknown ethnicity. d Includes 46 medulloblastomas, 22 primitive neuroectodermal tumors, 3 atypical teratoid rhabdoid tumors. e Includes 6 meningiomas, 9 choroid plexus tumors, 4 unclassified. 15 Table 2: Association between parental smoking before and during the pregnancy and risk of childhood brain tumors Case n Maternal smoking preconceptiona None 228 1-14 CPD 41 15+ CPD 31 Any 72 P-value for dose trend % Control n % OR 95% CI 76.0 13.7 10.3 24.0 731 101 105 206 78.0 10.8 11.2 22.0 1.00 1.16 0.82 0.99 Referent 0.76, 1.78 0.52, 1.31 0.70, 1.40 0.61 780 85 72 157 83.2 9.1 7.7 16.8 1.00 0.91 0.88 0.89 Referent 0.56, 1.47 0.52, 1.49 0.61, 1.31 0.57 Maternal smoking during pregnancya None 249 1-14 CPD 28 15+ CPD 23 Any 51 P-value for dose trend 83.0 9.3 7.7 17.0 Paternal smoking preconceptionb None 168 69.4 568 71.9 1.00 Referent 1-14 CPD 32 13.2 71 9.0 1.31 0.82, 2.11 15+ CPD 42 17.4 151 19.1 0.83 0.55, 1.24 Any 74 30.6 222 28.1 0.99 0.71, 1.38 P-value for dose trend 0.54 b Paternal smoking during pregnancy None 171 70.7 588 74.4 1.00 Referent 1-14 CPD 29 12.0 65 8.2 1.30 0.79, 2.13 15+ CPD 42 17.3 137 17.4 0.92 0.61, 1.38 Any 71 29.3 202 25.6 1.04 0.74, 1.46 P-value for dose trend 0.85 a Adjusted for matching variables, child’s ethnicity, year of birth group, mother’s age group, alcohol consumption during pregnancy, household income. b Adjusted for matching variables, child’s ethnicity, year of birth group, father’s age group, household income. 16 Table 3: Parental smoking stratified by child’s age at diagnosis or recruitment Age Smoking n Case/Controls OR 95% CI P-value of Interaction a Maternal smoking preconception 0-1 None 20/93 1.00 Referent 0.09 Any 10/17 5.06 1.35, 19.00 2-4 None 65/241 1.00 Referent Any 20/60 1.34 0.69, 2.60 5-9 None 71/226 1.00 Referent Any 18/67 0.78 0.39, 1.54 10-14 None 72/171 1.00 Referent Any 24/62 0.61 0.32, 1.16 Maternal smoking during pregnancy a 0-1 None 23/98 1.00 Referent 0.10 Any 7/12 4.61 1.08, 19.63 2-4 None 71/251 1.00 Referent Any 14/50 1.00 0.48, 2.07 5-9 None 79/244 1.00 Referent Any 10/49 0.60 0.26, 1.35 10-14 None 76/187 1.00 Referent Any 20/46 0.73 0.36, 1.47 Paternal smoking preconception b 0-1 None 20/80 1.00 Referent 0.25 Any 6/24 1.11 0.29, 4.20 2-4 None 47/190 1.00 Referent Any 28/68 1.45 0.79, 2.68 5-9 None 54/176 1.00 Referent Any 19/69 0.74 0.38, 1.44 10-14 None 47/122 1.00 Referent Any 21/61 0.75 0.39, 1.44 Paternal smoking during pregnancy b 0-1 None 20/82 1.00 Referent 0.26 Any 6/22 1.48 0.39, 5.65 2-4 None 49/198 1.00 Referent Any 26/60 1.61 0.86, 3.01 5-9 None 55/182 1.00 Referent Any 18/63 0.75 0.38, 1.49 10-14 None 47/126 1.00 Referent Any 21/57 0.82 0.43, 1.60 a Adjusted for matching variables, child’s ethnicity, year of birth group, mother’s age group, alcohol consumption during pregnancy, household income. b Adjusted for matching variables, child’s ethnicity, year of birth group, father’s age group, household income. 17