Article Text

Download PDFPDF

Tobacco smoke exposure and respiratory morbidity in young children
  1. A M Snodgrass1,
  2. P T Tan2,
  3. S E Soh3,4,
  4. A Goh1,
  5. L P Shek5,6,
  6. H P van Bever5,6,
  7. P D Gluckman3,7,
  8. K M Godfrey8,9,
  9. Y S Chong2,3,
  10. S M Saw10,
  11. K Kwek11,
  12. O H Teoh1,
  13. the GUSTO Study Group
    1. 1Department of Paediatric Medicine, KK Women's and Children's Hospital, Singapore, Singapore
    2. 2Department of Obstetrics & Gynaecology, Yong Loo Lin School of Medicine, National University of Singapore, National University Health System, Singapore, Singapore
    3. 3Singapore Institute for Clinical Sciences, Agency for Science and Technology Research (A*STAR), Brenner Centre for Molecular Medicine, Singapore, Singapore
    4. 4Department of Paediatrics, Yong Loo Lin School of Medicine, National University of Singapore, National University Health System, Singapore, Singapore
    5. 5Department of Paediatrics, Yong Loo Lin School of Medicine, National University of Singapore, Singapore, Singapore
    6. 6Khoo Teck Puat-National University Children's Medical Institute, National University Hospital, National University Health System, Singapore
    7. 7Liggins Institute, University of Auckland, Auckland, New Zealand
    8. 8Medical Research Council Lifecourse Epidemiology Unit, Southampton, UK
    9. 9NIHR Southampton Biomedical Research Centre, University of Southampton and University Hospital Southampton NHS Foundation Trust, Southampton, UK
    10. 10Saw Swee Hock School of Public Health, National University of Singapore, Singapore, Singapore
    11. 11Department of Maternal Fetal Medicine, KK Women's and Children's Hospital (KKH), Singapore, Singapore
    1. Correspondence to Dr AM Snodgrass, Department of Paediatric Medicine, KK Women's and Children's Hospital, Singapore, 100 Bukit Timah Road, Singapore 229899; amsnodgrass{at}yahoo.co.uk

    Abstract

    Objective Secondhand smoke exposure is a potentially preventable cause of significant respiratory morbidity in young children. Our study aimed to quantify respiratory morbidity in young children exposed to secondhand smoke to identify potentially modifiable factors.

    Materials and methods This study was embedded in a prospective birth cohort study of pregnant women and their children from fetal life onwards in Singapore (Growing Up in Singapore Towards healthy Outcomes, or GUSTO). Data on prenatal, antenatal and postnatal active and secondhand tobacco smoke exposure were obtained by an investigator-administered questionnaire for the periods before pregnancy, at 26–28 weeks’ gestation and 24 months after delivery. Data on respiratory morbidity (wheezing episodes, croupy cough, nebuliser use, snoring) and other morbidity (fever, hospitalisation, ear infection) of the child was collected at week 3 and at months 3, 6, 9, 12, 15, 18 and 24 after delivery. Information on parental atopy and potential confounders such as socioeconomic status and maternal educational level were also obtained. Statistical analysis of the data was performed to quantify any significant differences in incidence of respiratory morbidity in children exposed to tobacco smoke in utero and postdelivery, compared with those in smoke-free environments.

    Results Women who smoked regularly prior to pregnancy comprised 12.5% (n=155) of the study population; this number fell to 2.3% (n=29) during pregnancy. Mothers exposed to secondhand smoke in the household before pregnancy comprised 35.7% of the study population (n=441) and 31.5% (n=389) were exposed during pregnancy. Postnatally, the prevalence of secondhand tobacco smoke exposure from birth to 2 years of age was 29% (n=359). Participants of Malay ethnicity (p<0.001), mothers with no or primary level education (p<0.001) and mothers with low socioeconomic status (p<0.001) had the highest exposure to tobacco smoke. Offspring secondhand smoke exposure at home by 12 months and by 24 months of age was associated with an increase in hospital admissions due to respiratory disease (RR 1.89, 95% CI 1.02 to 3.50, p=0.04 by 12 months and RR 1.64, 95% CI 1.05 to 2.55, p=0.03 by 24 months) as well as all-cause hospitalisation (RR 1.57, 95% CI 1.14 to 2.17, p=0.01 by 12 months and RR 1.49, 95% CI 1.17 to 1.90, p=0.001 by 24 months), adjusting for parental atopy and child atopic dermatitis. Participants exposed to secondhand smoke by 12 months postdelivery had a significantly increased risk of having at least one wheezing episode (RR 1.71, 95% CI 1.38 to 2.11, p<0.001).

    Conclusions Secondhand smoke exposure during the prenatal and postnatal periods is associated with increased respiratory morbidity in children. Opportunistic screening and targeted smoking cessation counselling for parents at child hospital admissions and well-child outpatient visits, as well as preconception smoking cessation counselling for future pregnancies, may be beneficial to protect the child from negative health impacts.

    Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Introduction

    Exposure to tobacco smoke is detrimental to children and there is no known safe level of exposure.1 Prenatal tobacco exposure and secondhand smoke exposure in the home environment have been associated with a vast range of adverse health events2 ,3 and significant global disease burden.4 Burke et al5 noted in a recent systematic review in April 2012 that prenatal or postnatal secondhand smoke exposure was associated with a 30–70% increased risk of incident wheezing and a 21–85% increase in incident asthma (OR 1.85, 95% CI 1.35 to 2.53). A meta-analysis conducted by Jones et al,6 published in 2011, found that smoking by either parent or other household members significantly increased the risk of lower respiratory infections, particularly bronchiolitis (OR 2.51, 95% CI 1.96 to 3.21 for smoking by any household member). Secondhand smoke exposure has also been associated with reduced lung function in infants and children,7 ,8 increased risk of otitis media,9 ,10 poor growth,11 ,12 and behavioural and neurocognitive effects.13–16

    We sought to quantify any significant differences in the incidence of respiratory morbidity in children exposed to secondhand smoke (in utero and postdelivery), compared with those in smoke-free environments, in a local birth cohort.

    Materials and methods

    The GUSTO (Growing Up in Singapore Towards healthy Outcomes) study is a prospective birth cohort study of 1247 ethnic Chinese, Malay and Indian pregnant women and their offspring throughout pregnancy, birth and the early years of life, in Singapore, with participant recruitment spanning 2009–2010. Recruitment was aimed at pregnant women aged 18 years and above attending their first trimester antenatal dating ultrasound scan clinic at either of Singapore's two major public maternity units, namely National University Hospital (NUH), and KK Women's and Children's Hospital (KKH), between June 2009 and September 2010. The participants approached were Singapore citizens or permanent residents who were of Chinese, Malay or Indian ethnicity with homogeneous parental ethnic background, and who had the intention of eventually delivering in NUH or KKH and residing in Singapore for the next 5 years. In total, 3751 families were screened, of which 2034 met eligibility criteria; the response rate was 61.3%. Seventy participants (5.6%) dropped out during pregnancy and 88 participants (7.1%) withdrew voluntarily after delivery.17 The main aim of GUSTO is to evaluate the role of developmental factors in the early pathways to metabolic compromise; this provides an important opportunity to investigate the developmental pathways underlying disease risk in the three major ethnic groups in Singapore. Detailed data are collected for a range of environmental exposures in the parents and offspring including exposure to tobacco smoke, and respiratory morbidity is among a number of outcomes assessed in infancy and childhood.

    Data on exposure to tobacco smoke were collected by investigator-administered questionnaires at weeks 26–28 of pregnancy (at antenatal clinic visit) as well as at month 24 postnatally (at study visit). Measures of both active and secondhand smoke exposure to tobacco smoke, in the household and at the workplace, prior to pregnancy, during pregnancy, from birth to before 6 months, from 6 months to less than 12 months and from 12 months to less than 24 months postdelivery, were obtained. Regular smoking was defined as smoking at least once a day for a year or more. Attempts to quantify the degree of tobacco smoke exposure were made by asking participants to report the age in years at which regular smoking started, the number of cigarettes smoked per day, the number of years during which a household member smoked daily at home and the proximity of the household smoker to the child (smoking in the presence of the child, in any part of the home, in the family car, directly outside the home or at a location distant from the home). The relationship of the household smoker to the child was also defined.

    Data on respiratory morbidity (wheezing episodes, croup or croupy cough, nebuliser use, snoring) and other morbidity (fever, hospitalisation, ear infection, allergy) of the children were collected by investigator-administered questionnaires at week 3 and at months 3, 6, 9, 12, 15, 18 and 24 of age. Croup was defined as a respiratory infection causing a barking (croupy) cough, hoarse voice, runny nose, fever, loud high-pitched hoarse noise when breathing in and breathing difficulty. A wheezing episode was defined as noisy breathing with a high-pitched, whistling sound heard from the chest and not the mouth; and episodes of bronchiolitis, defined as a respiratory infection causing wheeze, cough, fever, runny nose and breathing difficulty, were included in this category. Parents were asked whether their child had ever been prescribed nebulisers or inhaler treatment, and if their child snored during sleep. Familial predisposition towards asthma and atopy was sought by obtaining the presence or absence of family history of atopy. Demographical data obtained included parental ethnicity, highest maternal educational level, type of accommodation and monthly household income.

    Statistical analysis was performed using IBM SPSS V.21, and χ2 test or Fisher's exact test was used to assess the association between tobacco smoke and respiratory illness. Logistics regression and Poisson regression were used to perform multivariate analysis with binary outcome and to count data, respectively. Negative binomials were performed to overcome dispersion issues. Statistical significance was set at p<0.05. A total of 1236 participants were eligible for the analysis. Participants with twin pregnancies (n=10) were excluded from the analysis.

    Results

    Prenatal tobacco smoke exposure

    Active smoking by mother

    Women who smoked regularly prior to pregnancy comprised 12.5% (n=155) of the study population, and 2.3% (n=29) continued to smoke during pregnancy. Of these, 80.4% started smoking at the age of 18 years or below, with the mean age of smoking onset being 16.4 years (figure 1). Two participants had missing data on age of onset of smoking.

    Figure 1

    Age of onset of smoking in mothers.

    Demographic data for the active smokers in our study population are presented in table 1.

    Table 1

    Demographic characteristics of exposed (active smoking) and non-exposed participants

    There were 690 ethnic Chinese, 324 ethnic Malays and 223 ethnic Indians in the study population. Malay mothers were four times more likely to smoke compared with non-Malay mothers prior to pregnancy (OR 4.02, 95% CI 2.84 to 5.70, p<0.001), and were also more likely to continue smoking during pregnancy (OR 2.80, 95% CI 1.33 to 5.65, p=0.01). Most of these mothers had at least secondary level education (71.6% before pregnancy and 65.5%, during pregnancy) and came from the group with the lowest income bracket.

    Secondhand smoke exposure of mothers

    Demographic data for the participants exposed to secondhand smoke in the household are detailed in table 2.

    Table 2

    Demographic characteristics of exposed (secondhand smoke exposure) and non-exposed participants

    Before pregnancy, 35.7% (n=441 out of 1236) of women had a household member smoking at home on a daily basis for 6 months or more; this exposure fell to 31.5% (n=389) during pregnancy. One hundred and eighty-four women (14.9%) were exposed to secondhand smoke in the workplace prior to pregnancy; exposure also fell to 7.1% (n=88) during pregnancy. Within ethnic groups, the highest household exposure was noted in Malay women (63.2%, compared with 26.2% of Chinese women and 25.1% of Indian women, p<0.001). Univariate analysis showed a statistically significant difference in household exposure between Chinese and Malay women. Pregnant women with no or primary level education (prior to pregnancy—66.1%, n=37; during pregnancy—60.7%, n=34 out of 56) and within the lowest income bracket (S$0–S$999; prior to pregnancy—62.1%, n=18 out of 29; during pregnancy—58.6%, n=17 out of 29) had the highest chance of exposure to secondhand smoke in the household.

    Postnatal tobacco smoke exposure

    Secondhand smoke exposure within the 24-month period following delivery was noted in 359 households (29%, missing data for 303 participants). Fathers comprised the majority of household smokers (61.9% at 6 months and 70.5% at 12 months postdelivery), followed by grandfathers (12.8% at 6 months and 11.5% at 12 months postdelivery). Mothers who smoked actively postdelivery made up 10.4% of smokers at 6 months at 8.8% at 12 months. Exposure to secondhand smoke was similar at months 12 and 24.

    Women smoking actively postdelivery comprised 3.9% of the study population (n=48). Among Malay women, 22 out of 323 (6.8%) were active smokers compared with 6 out of 223 Indian women (2.7%) and 20 out of 690 Chinese women (2.9%). The percentage of mothers smoking by month 12 was higher in households where fathers also smoked (76.9% vs 23.9%, p<0.001). All mothers who quit smoking during pregnancy but relapsed postdelivery had husbands who smoked.

    The majority of household smokers reported avoiding smoking in the presence of the child, but continued to smoke inside the home, family car or directly outside the home (table 3).

    Table 3

    Proximity of household smoker to child

    Exposure to tobacco smoke pre-pregnancy, during pregnancy and during the postnatal period was significantly associated with ethnicity, educational level and socioeconomic status. Malay women had the highest exposure while Indian women had the least (p<0.001). Mothers with no or primary level education had the highest exposure, while mothers with tertiary education had the least (p<0.001; with the exception of active smoking before pregnancy, where mothers with secondary level education had the highest exposure). Mothers with monthly household income S$0–S$999 had the highest exposure, while mothers with monthly household income ≥S$6000 had the least (p<0.001).

    Parental atopy

    A history of atopy in one or both parents was noted in 338 children (missing data for 337 participants or 27.3%). Parental atopy was defined as a history of asthma, eczema or rhinitis in either the mother, the father or both. Following logistic regression, univariate analysis showed that fathers who had ever had asthma were less likely to smoke during the postdelivery period (OR 0.51, 95% CI 0.30 to 0.89, p=0.018). There was no significant association between maternal asthma, atopic dermatitis or allergic rhinitis and smoking status, nor was there any association between paternal allergic rhinitis, atopic dermatitis and smoking.

    Respiratory morbidity

    By month 12 at postnatal follow-up, 200 children (16.2%) had experienced wheezing episodes, 70 (5.7%) had had croupy cough, 29 (2.3%) had suffered ear infections and 338 (27.3%) had presented a history of snoring. Fifty-four (4.4%) children had been hospitalised due to respiratory disease and 120 (9.7%) had been treated with nebulisers during this period.

    By month 24, 313 children (25.3%) had gone through wheezing episodes, 124 (10%) had had croupy cough, 52 (4.2%) had suffered ear infections and 363 (29.4%) had presented a history of snoring. Eighty-four (6.8%) children had been hospitalised due to respiratory disease and 197 (15.9%) had been treated with nebulisers during this period.

    The total number of wheezing episodes and hospital admissions for respiratory disease are presented in tables 4 and 5.

    Table 4

    Total number of wheezing episodes

    Table 5

    Hospital admissions due to respiratory illness

    Secondhand smoke exposure at home by 12 and 24 months of age was associated with an increase in hospital admissions due to respiratory disease (RR 1.89, 95% CI 1.02 to 3.50, p=0.04 by 12 months and RR 1.64, 95% CI 1.05 to 2.55, p=0.03 by 24 months as well as all-cause hospitalisation (RR 1.57, 95% CI 1.14 to 2.17, p=0.01 by 12 months and RR 1.49, 95% CI 1.17 to 1.90, p=0.001 by 24 months), after adjusting for parental atopy, child atopic dermatitis, ethnicity, education level and household income (tables 6 and 7). These results were consistent with multiple imputations.

    Table 6

    Tobacco smoke exposure and child respiratory morbidity by 12 months of age

    Table 7

    Tobacco smoke exposure and child respiratory morbidity by 24 months of age

    Participants exposed to secondhand smoke by 12 months postdelivery had a significantly increased risk of having at least one wheezing episode (RR 1.71, 95% CI 1.38 to 2.11, p<0.001), but not by 24 months postdelivery. Statistical significance remained following adjustment for either maternal atopy (RR 1.73, 95% CI 1.40 to 2.14, p<0.001) or paternal atopy (RR 1.72, 95% CI 1.39 to 2.13, p<0.001) alone, after taking into account ethnicity, education level and socioeconomic status. The study was insufficiently powered to examine the independent effects of smoking at different time points before and during pregnancy and after delivery.

    Prenatal and postnatal secondhand smoke exposure

    Offspring of mothers exposed to secondhand smoke both prenatally and postnatally were noted to have a higher risk of wheezing episodes than offspring whose mothers were never exposed to secondhand smoke at both months 12 (RR 2.06, 95% CI 1.54 to 2.75, p<0.001) and 24 (RR 1.56, 95% CI 1.22 to 1.99, p<0.001). Risk of hospitalisation for respiratory disease was also significantly higher in offspring of mothers with secondhand smoke exposure during both prenatal and postnatal periods as opposed to those who were never exposed (RR 2.90, 95% CI 1.48 to 5.71, p=0.002 at month 12 and RR 2.68, 95% CI 1.60 to 4.49, p<0.001 at month 24). Risk of all-cause hospitalisation in offspring of mothers exposed to secondhand smoke both prenatally and postnatally was increased compared with offspring of mothers who were never exposed at month 12 (RR 1.88, 95% CI 1.34 to 2.63, p<0.001) and at month 24 (RR 1.94, 95% CI 1.47 to 2.55, p<0.001).

    Dropouts

    There were 156 dropouts (12.6%).17 No statistically significant difference in tobacco smoke exposure between dropouts and active participants was observed. Mothers who obtained no or primary level education were more likely to drop out compared with mothers with tertiary education (OR 2.95, 95% CI 1.56 to 5.56, p=0.001), and mothers in the lowest household income bracket (S$0–S$999) were more likely to drop out compared with mothers in the highest income bracket (≥S$6000; OR 5.09, 95% CI 2.17 to 11.92, p<0.001).

    Discussion

    Cigarette smoking is a significant problem in Singapore, although exposure reduction does occur with the advent of pregnancy and childbirth, as previously reported in other studies.18

    In our study, 12.6% of mothers were regular smokers prior to pregnancy, with a decrease noted in pregnancy (10.3% quit). The percentage of expectant mothers actively smoking during pregnancy (2.3%) was much lower than the 14% reported in previous studies.19 However, the epidemiology of female smokers in our study reflects a worrying global trend of progressive adoption of smoking by adolescent girls at young ages.20 Those women who continued smoking during pregnancy tended to be of Malay ethnicity, had a lower educational level and poorer socioeconomic status. (Median monthly household income per household member in Singapore was reported as S$2127 in 2012.21) Similar risk factors have been noted in previous studies; other predictors of continued smoking during pregnancy included increased maternal age and unmarried status.22 The benefits of smoking cessation counselling and active interventions to reduce tobacco use preconceptually and in the antenatal period might thus be maximised in this high-risk group.

    There was also significant secondhand smoke exposure prior to and during pregnancy, although a decreasing trend in secondhand smoke exposure postnatally was noted; 35.7% were exposed to secondhand smoke in the home prior to pregnancy, with exposure falling to 31.5% during pregnancy and then to 29% postdelivery. This continued reduction in exposure may be due to awareness of the deleterious effects of passive smoking on the pregnant mother and unborn child, and a desire to avoid adverse health consequences, or perhaps the wish to set a good example and preserve one's own health in order to provide a better future for the child.

    Prenatal secondhand smoke exposure, in addition to postnatal exposure, was found to confer higher relative risk of total wheezing episodes, hospitalisation due to respiratory disease and all-cause hospitalisation at both months 12 and 24, compared with the risk of the said morbidities with any postnatal exposure versus no exposure at any time period. Despite the high correlation between exposures at preconception, antenatal and postnatal periods, statistical analysis did not show multicollinearity after all covariates were included.

    Postnatally, the majority of smokers were fathers. Smoking behaviour among fathers has received relatively little attention compared with maternal smoking, although the smoking behaviour of the woman's partner has consistently been cited as a risk factor for maternal postpartum smoking relapse,23 as borne out in our study; women who lived with another smoker were four times more likely to relapse than those who did not live with a smoker.22 Maternal relapses in smoking up to 24 months postdelivery may be influenced by the stresses of child rearing and smoking behaviour of other family members, and intensification of smoking cessation support services and counselling for all household smokers during this window period may be beneficial in maintaining behaviour change.

    Most household smokers in our study avoided smoking in the presence of the child, but continued to smoke inside the home or family car, or directly outside the home. Such attempts to reduce the child's secondhand smoke exposure are commendable and can be further improved. A cross-sectional survey conducted by Blackburn et al,24 in 2005, examining behavioural change in new fathers of infants aged 8–14 weeks old found that not smoking in the home appeared to be a more achievable goal as opposed to quitting. This has implications for the development of health promotion strategies to protect infants from passive smoking; avoiding smoking in the presence of infants or children in the household (a smoke-free policy at home) could be encouraged as an interim goal of behavioural change in the journey towards smoking cessation.

    Secondhand smoke exposure reduction is a relatively new area of scientific study. A Cochrane review of the secondhand smoke exposure reduction interventions in 2002 indicated that few programmes produced a statistically significant reduction.25 A further systematic review by Rosen et al,26 in 2011, which attempted to quantify the effects of interventions that encourage parental smoking cessation, reported quit rates of 23.1% in the intervention group versus 18.4% in the control group (overall relative risk 1.34, 95% CI 1.05 to 1.71, p=0.02). Combination approaches including both counselling and pharmacological therapy appear to be more effective than either alone. Two studies by Winickoff et al,27 ,28 reporting on a combination approach among smoking parents of children, noted increased numbers of quit attempts and decreased numbers of cigarettes smoked in the house and car following the intervention.

    Children exposed to secondhand smoke within the first 2 years of life comprised 29.1% of the study cohort. This was lower than the observed exposure of 40% worldwide (for ages 0–14 years) in 2004.4 In exposed children, the most significant global disease burdens were from lower respiratory tract infections in those younger than 5 years (5 939 000)—as corroborated by our study—and asthma (651 000).

    In our study, children exposed to secondhand smoke from birth to 24 months were found to have statistically significant increases in hospital admissions for respiratory illnesses as well as all-cause hospitalisation with exposure to secondhand smoke. Statistically significant increases in the total number of wheezing episodes after taking predisposition to wheeze (atopy) into account were also observed. This finding is similar to previous studies.5 ,6 Maternal atopy has been reported to have greater impact than paternal atopy on perinatal IgE production and asthma development in children;29 likely more complex mechanisms are involved in asthma than simply a genetic predisposition to wheeze. In our study, aside from likely influencing child predisposition to wheeze, atopy was also found to influence parental smoking behaviour. Further studies are needed to examine the relationship between parental atopy and smoking behaviour in order to quantify the overall effects on child respiratory morbidity.

    Wheezing episodes were common in our study population, with only 38.3% of participants remaining wheeze-free up to month 24, and 7.7% experiencing four or more wheezing episodes. Aside from economic implications in terms of healthcare utilisation and loss of productivity given parental time off from work during child illness episodes, the future risk of developing asthma in these children warrants further study and longer term follow-up.

    Public health implications

    Environmental tobacco smoke exposure reduction does occur postdelivery and children may be the key motivators for parental abstinence. Opportunistic screening for secondhand smoke exposure in the household and targeted smoking cessation counselling in hospital when infants are admitted for respiratory illnesses should focus on both parents as well as on other family members such as grandparents. This strategy has previously been evaluated and found to be feasible with a high rate of acceptance, with 49% reporting quit attempts at 2-month follow-up.28 Other strategies to maintain abstinence might include reinforcement by paediatric providers and nurses during well-child visits for routine vaccination and developmental assessment. Smoking cessation counselling for future pregnancies is vital, and screening should ideally be performed preconception. Singapore has a multipronged approach to address tobacco use that involves tobacco control policies, tobacco taxation, public education and empowerment, partnerships and capacity building, and the provision of smoking cessation services in a supportive environment. Improving access of this high-risk group to the available services may improve quit rates and decrease relapse rates.

    Study limitations

    Quantification of tobacco smoke exposure after delivery was by retrospective recall and thus subject to recall bias. An increase in recall bias would falsely strengthen the positive association between secondhand smoke exposure and wheeze. In order to minimise this, data on secondhand smoke exposure were collected longitudinally throughout the preconception, antenatal and postnatal periods, with frequent follow-up postnatally. Exposures at months 12 and 24 were also found to be similar, although the relationship between SHS exposure and wheeze was only significant at month 12 and not month 24, perhaps indicating decreased susceptibility to wheeze with increasing age. Also, self-reported abstinence from smoking is not often a reliable measure with pregnant and postpartum women;30 due to reasons of wanting to maintain social desirability, many pregnant women under-report their smoking status31 and extent of secondhand smoke exposure.32 Biomarker validation would be a useful tool to confirm smoking and secondhand smoke exposure in this population. Diagnosis of respiratory morbidities, in particular croupy cough and snoring during sleep, was partly by parental report and thus possibly of uncertain accuracy. Within our study design, as participants sought medical attention in a community setting during the follow-up period, viral immunofluorescence studies on nasopharyngeal aspirates were not routinely performed, and the subsequent absence of virological data did not allow assessment of the relationship between respiratory syncytial virus (RSV) infection and later wheeze, and did not examine whether children with atopic asthma are also more prone to RSV infection. The considerable amount of missing data and exclusion of these participants from analysis may also have affected the statistical significance of our results, although the positive association between secondhand smoke exposure and total number of wheezing episodes as well as hospitalisation for respiratory disease was compelling. The study was also insufficiently powered to examine the independent effects of smoking at different time points before pregnancy, antenatally and postnatally.

    Possible confounders of the association between tobacco smoke exposure and wheeze, aside from the presence or absence of atopy, include socioeconomic status, changing disease susceptibility with increasing age (for secondhand smoke exposure postdelivery), and cumulative exposure owing to overlap between smoking exposure categories (eg, children of mothers actively smoking during pregnancy who are also exposed to secondhand smoke postdelivery) cannot be adequately accounted for.

    Conclusion

    Secondhand smoke exposure during both the prenatal and postnatal period is associated with increased respiratory morbidity in children. There were statistically significantly different patterns of tobacco smoke exposure based on ethnicity, educational level and socioeconomic status in this birth cohort, which can similarly be observed in other studies. Opportunistic screening and targeted smoking cessation counselling for parents at child hospital admissions and well-child outpatient visits, as well as preconception smoking cessation counselling for future pregnancies, may be beneficial. Identifying and addressing the social determinants of health can help to improve health equity and eventually aid in the recommendations for public health, to enhance efficacy of prenatal care and long-term child health.

    What this paper adds

    • Our study findings add to the growing body of evidence of increased respiratory morbidity in vulnerable populations (children) exposed to secondhand tobacco smoke. Our study also describes demographic characteristics of a group at high risk for secondhand smoke exposure in an ethnically diverse Asian population at currently relevant levels of secondhand smoke exposure. This is in spite of a robust national tobacco control programme that was first implemented in 1970 and has resulted in one of the lowest smoking rates in the world. It is hoped that this study will potentially inform future smoking cessation counselling and public health exposure reduction policies.

    References

    Footnotes

    • Collaborators The GUSTO study group includes Pratibha Agarwal, Arijit Biswas, Choon Looi Bong, Birit FP Broekman, Shirong Cai, Jerry Kok Yen Chan, Yiong Huak Chan, Cornelia Yin Ing Chee, Helen Y H Chen, Yin Bun Cheung, Audrey Chia, Amutha Chinnadurai, Chai Kiat Chng, Mary Foong-Fong Chong, YSC, Shang Chee Chong, Mei Chien Chua, Chun Ming Ding, Eric Andrew Finkelstein, Doris Fok, Marielle V Fortier, PDG, KMG, AG, Yam Thiam Daniel Goh, Joshua J Gooley, Wee Meng Han, Mark Hanson, Christiani Jeyakumar Henry, Joanna D Holbrook, Chin-Ying Hsu, Hazel Inskip, Jeevesh Kapur, KK, Ivy Yee-Man Lau, Bee Wah Lee, Yung Seng Lee, Ngee Lek, Sok Bee Lim, Yen-Ling Low, Iliana Magiati, Lourdes Mary Daniel, Michael Meaney, Cheryl Ngo, Krishnamoorthy Naiduvaje, Wei Wei Pang, Anqi Qiu, Boon Long Quah, Victor Samuel Rajadurai, Mary Rauff, Salome A Rebello, Jenny L Richmond, Anne Rifkin-Graboi, SMS, LPS, Allan Sheppard, Borys Shuter, Leher Singh, SES, Walter Stunkel, Lin Lin Su, Kok Hian Tan, OHT, Mya Thway Tint, HPvB, Rob M van Dam, Inez Bik Yun Wong, P C Wong, Fabian Yap and George Seow Heong Yeo.

    • Contributors All the authors had full access to all data in the study, and take responsibility for the integrity of the data and accuracy of the data analysis. AMS had the primary role in data analysis, data interpretation and writing of the report. PTT contributed to data collection, analysis, interpretation and report writing. SES, AG, LPS, HPvB, PDG, KMG, YSC, SMS and OHT contributed to the study design, data analysis, and interpretation and report writing. KK contributed to the study design.

    • Funding This research is supported by the Singapore National Research Foundation under its Translational and Clinical Research (TCR) Flagship Programme and administered by the Singapore Ministry of Health's National Medical Research Council (NMRC), Singapore, NMRC/TCR/004-NUS/2008; NMRC/TCR/012-NUHS/2014. Additional funding is provided by the Singapore Institute for Clinical Sciences, Agency for Science Technology and Research (A*STAR), Singapore. KMG is supported by the National Institute for Health Research through the NIHR Southampton Biomedical Research Centre and by the European Union’s Seventh Framework Programme (FP7/2007-2013), projects EarlyNutrition and ODIN under grant agreements numbers 289346 and 613977.

    • Competing interests None declared.

    • Patient consent Obtained.

    • Ethics approval This study received ethical approval from Singhealth Centralised Institutional Review Board (CIRB) and National Healthcare Group DSRB.

    • Provenance and peer review Not commissioned; externally peer reviewed.

    • Data sharing statement Investigators interested in exploring the possibility of collaborations should contact lead principal investigator YSC (yap_seng_xhong@nuhs.edu.sg), principal investigators SMS (seang_mei_saw@nuhs.edu.sg), KK (Kenneth.Kwek.YC@kkh.com.sg) and PDG (pd.gluckman@auckland.ac.nz). GUSTO has a website, mainly focused on information for the participants, at http://www.gusto.sg/. More information is also available on the website of the Translational Clinical Research (TCR) Flagship Programme on Developmental Pathways to Metabolic Disease, which is commonly known as Developmental Origins: Singapore (DeVOS), website http://devos.sg/