INTRODUCTION

Passive smoking, also called secondhand smoke (SHS), is defined as any tobacco smoke exposure as a result of other peoples’ active smoking1. More than one-third of the world’s population is exposed to SHS and the harmful effects of regular cigarette smoke2. In Indonesia, around 225720 people die annually from tobacco-related disease3. However, the number of smokers aged >15 years is about 89.6 million people or 33.8% of the population. Of these, 62.9% are male and 4.8% female smokers4, most smoke at home4,5.

Indonesia is the only country in Asia which is not a party to the World Health Organization Framework Convention on Tobacco Control (WHO FCTC). It has some of the weakest tobacco control measures in South-East Asia6 and although Indonesia implemented MPOWER measures to diminish tobacco use, none of the measures has worked7. Therefore, domestic studies on smoking behaviour and SHS exposure among Indonesians are needed as there exist few data on the prevalence of SHS in infants in Indonesia. However, it has been reported that about 57.3% of the children (13–15 years) and 78.4% of the adults are exposed to SHS at home5,8. Babies become vulnerable to cigarette smoke exposure because most of their time is spent at home9, and this has become a significant public health issue.

The adverse health effects of SHS may include the risk of sudden infant death syndrome (SIDS), birth defects, acute respiratory illnesses and chronic respiratory disease, cancer and other diseases1012, low birthweight11,13,14, and shorter body length15. Postnatal cigarette smoke exposure also can be a reason for the decrease in weight and length growth in the early months of life16. In children, up to eight years of age17, SHS has been associated with meningococcal carriage and disease, and medically attended accidents in children18. In addition, SHS exposure is associated with an increased risk of poor neurodevelopment19.

In the first 1000 days of life, socioeconomic factors may play a role in nutrition-sensitive causes of malnutrition20. Infants aged 6 months are introduced to complementary foods. Nutritious foods guarantee growth and health. For people with a lower middle income, setting aside a portion of their budget for their baby’s nutritional needs is necessary. For this reason, if parents smoke, the dietary needs of their children may be excluded or will not be a priority. Among the poor households in the urban slum areas of Indonesia, paternal smoking results in the diversion of household money from food to tobacco, thereby contributing to child malnutrition21.

Research on exposure to cigarette smoke among 6-month-old babies via paternal smoking in middle-low-income families is an area of regional importance. The aim of this study is to assess socioeconomic characteristics and paternal smoking in households and infants’ exposure to SHS in Indonesia.

METHODS

Study design, setting and participants

The data for this cross-sectional study were derived from an ongoing PEER (Partnerships for Enhanced Engagement in Research) Health prospective cohort study in Jakarta, Indonesia. The primary objective of the cohort study is to determine the effects of individual air pollution exposure in pregnancy and early infancy on maternal health (lung function, hypertension disorder in pregnancy, or preeclampsia) and neonatal/infant health (lung function and infection episodes).

The study population consists of pregnant women attending antenatal care and arranging to deliver the babies in seven community health centers in Jakarta, Indonesia. The study includes all pregnant women in the first antenatal visit, then followed up until the babies reach 6 months of age. Only women who have lived in their residential area for three months or more before the pregnancy are included in the study. Based on the sample size calculations, the study consists of 600 mother-infant pairs, starting from 2016 and finishing in 2020. After enrollment, pregnant mothers are followed up in line with routine antenatal care (ANC). After delivery and discharge, infants are invited for routine postnatal visits at the age of 40 days and at 2, 4 and 6 months. For those who refuse, a separate team of field workers visit the households.

To assess the socioeconomic characteristics and household risk factors related to paternal smoking and SHS exposure in 6-month-old infants, we limited participants to those with infants who reached the age of 6 months. We further included those with completed data for socioeconomic characteristics and smoking in households until July 2019. Thus, the present study includes 156 mother-infant pairs.

Measures

The questionnaire was developed by the PEER Health international team of experts. The questionnaire was pilot tested with 10 mothers resulting in no changes. Socioeconomic characteristics and smoking behaviour in the household were assessed by questionnaires during face-to-face interviews. Smokers were defined as individuals who smoked at least one cigarette per day. Regarding the father’s smoking behaviour, infants were considered exposed to SHS if the father smoked at home daily or weekly, and they were considered not exposed to SHS if the father had no smoking habit or never smoked or only smoked once a month at home.

To assess cigarette smoking status and exposure, we examined: 1) whether the mother smoked, 2) whether the father smoked, 3) whether other family members smoked and lived in the same house, 4) number of family members who smoked, 5) number of cigarettes smoked by family members, 6) number of cigarettes smoked by the father, 7) number of cigarettes smoked by the mother, and 8) frequency of father smoking at home. The socioeconomic characteristics of the parents included: marital status; the educational level of the mother and father, whether elementary (low), high school (middle), or undergraduate and higher (high); family income, whether below or on/above the minimum wage (based on Jakarta Province 2018 minimum wage); the father’s and mother’s employment status; and number of children.

This study was approved by the ethics committee of the Faculty of Medicine, Universitas Indonesia, with the code of ethics No: 895/UN2.F1/ETIK/2015.

Statistical analysis

Descriptive statistics were used. A chi-squared test was used to compare the socioeconomic characteristics and household smoking behaviour of fathers who were smokers and those who were non-smokers and the relationship of infants’ SHS exposure and paternal smoking. The factors related to paternal smoking and SHS exposure in infants were evaluated using a multivariate logistic regression analysis.

RESULTS

Smoking behaviour in the household

Overall, the prevalence of smokers in the household was 71.2%, with fathers being the greatest contributors (59.6%). Maternal smoking was only 2.6% and smoking among other family members was 29.5% (Table 1). Together, the fathers and other household members smoked between 5–14 cigarettes per day, on average.

Table 1

Percentage of smoking fathers, mothers and other household members according to daily cigarette consumption, Jakarta, Indonesia, 2017–2019 (N=156)

Household membersNon-smokersCigarette consumption (cigarettes/day)

1–45–14≥15

n%n%n%n%
Fathers6340.41610.36541.7127.7
Mothers15398.121.310.600
Other11070.52012.82314.731.9

Socioeconomic characteristics, smoking behaviour in the family and smoking status of fathers

Participants were all married, with many having two (35.9%) and three or more (35.3%) children. More than half (54.4%) of the participants had family incomes under the regional minimum wage, and most mothers were unemployed (61.5%). Most of the fathers and mothers had middle-level education (75.6% and 71.8%, respectively). One-third of the children (32.7%) were exposed to paternal smoking at home, and 10.3% lived with two or more other household members who smoked.

In Table 2, the socioeconomic characteristics and smoking behaviour in the family with fathers who were smoking and not smoking were compared. Significant relationships between the smoking status of the father and the number of children and the father’s education and number of other household members who smoked were found in the bivariate analysis. Paternal smoking was more prevalent among poor families (with income under the minimum wage); however, this association was not significant. The relationships among the number of children, mother’s employment status, mother’s and father’s educational levels, other household members’ smoking status and number of other household members who smoked are shown in Table 3. Regarding the smoking status of the fathers, SHS exposure among children was likely to occur in poorer families, although the differences were not statistically significant in the bivariate analyses.

Table 2

Father’s smoking status in terms of socioeconomic characteristics and smoking in household, Jakarta, Indonesia, 2017–2019 (N=156)

CharacteristicsCategoriesFather’s smoking status

Non-smoker (N=63)Smoker (N=93)p*

n%n%
Number of childrenInfant with no siblings2555.62044.40.044
Infant with 1 sibling1832.13867.9
Infant with ≥2 siblings2036.43563.6
Marital statusMarried6340.49359.6-
Mother’s occupationUnemployed3637.56062.50.323
Employed2348.92451.1
Self-employed535.7964.3
Father’s occupationEmployed5840.88459.20.93
Self-employed964.3535.7
Mother’s educational levelLow114.3685.70.253
Middle4540.26759.8
High1745.92054.1
Father’s educational levelLow1253750.044
Middle4235.67664.4
High2058.81441.2
Family income status< Regional minimum wage2832.95767.10.056
≥ Regional minimum wage3549.33650.7
Mother’s smoking statusNon-smoker6341.18958.60.148
Smoker004100
Other household members’ smoking statusNon-smoker4440.46559.61
Smoker1839.12860.9
Number of other household members who smoke04540.96559.10.026
11653.31446.7
≥2212.51487.5

* Chi-squared test, significant at p<0.05.

Table 3

Children’s SHS exposure considering father smoking at home by socioeconomic characteristics and household smoking, Jakarta, Indonesia, 2017–2019 (N=156)

CharacteristicsCategoriesSHS exposure

Not exposed (N=105)Exposed (N=51)p*

n%n%
Number of childrenInfant with no siblings3782.2817.80.036
Infant with 1 sibling3358.92341.1
Infant with ≥2 siblings3563.62036.4
Marital statusMarried10567.35132.7-
Mother’s occupationUnemployed6062.53637.50.009
Employed3983817
Self-employed646.2763.8
Father’ s occupationEmployed9667.64632.40.773
Self-employed964.3535.7
Mother’s educational levelLow228.6571.40.06
Middle75673733
High2875.7924.3
Father’s educational levelLow1253750.009
Middle7563.64336.4
High2985.3514.7
Family income status< Regional minimum wage5261.23338.80.094
≥ Regional minimum wage5374.51825.4
Mother’s smoking statusNon-smoker10468.44831.60.103
Smoker125375
Other household members’ smoking statusNon-smoker7972.53027.50.045
Smoker2554.32145.7
Number of other household members who smoke08072.73027.30.017
11963.31136.7
≥2637.51062.5

* Chi-squared test, significant at p<0.05.

Factors related to paternal smoking

To evaluate factors related to the smoking status of fathers, we examined the number of children, father’s educational level, family income, and number of household members who smoked in the logistic regression model. The multivariate logistic regression model showed that a larger number of children in the family increased the odds of fathers smoking (p=0.044), as fathers who had two children were more likely to smoke compared with those who had one child (OR=2.41; 95% CI: 1.02–5.67). Moreover, fathers with secondary education smoked more than those with higher education (OR=2.59; 95% CI: 1.19–5.63) (Table 4).

Table 4

Multivariate logistic regression analysis for factors related to father smoking, Jakarta, Indonesia, 2017–2019 (N=156)

FactorsCategoriesFather smokingp*

OR95% CI
Number of childrenInfant with no siblings1--
Infant with 1 sibling2.411.02–5.670.044
Infant with ≥2 siblings1.90.82–4.440.135
Father’s educational levelLow4.290.4–45.570.195
Middle2.591.19–5.630.045
High1--
Family income status< Regional minimum wage1.490.67–3.350.327
≥ Regional minimum wage1--
Number of other household members who smoke01--
10.550.24–1.290.171
≥24.590.96–21.990.057

* Significant at p<0.05.

Factors related to SHS exposure among infants

In the multivariate analysis (Table 5), children with one sibling (OR=3.09; 95% CI: 1.15–8.32) or who had fathers with low educational levels (OR=18.73; 95% CI: 1.54–227.93) and had two or more other household members who smoked (OR=4.54; 95% CI: 1.42–14.48) were more likely to be exposed to cigarette smoke at home from father’s smoking.

Table 5

Multivariate logistic regression analysis for factors related to SHS exposure among infants, Jakarta, Indonesia, 2017-2019 (N=156)

FactorsCategoriesFather smokingP*

OR95% CI
Number of childrenInfant with no siblings1--
Infant with 1 sibling3.091.15–8.320.026
Infant with >2 sibling2.530.93–6.940.071
Mother’s occupationUnemployed1--
Employed0.520.2–1.350.181
Self-employed1.50.42–5.420.526
Mother’s educational levelLow2.870.38–22.040.309
Middle0.790.28–2.220.650
High1--
Father’s educational levelLow18.731.54–227.930.022
Middle2.660.92–7.650.07
High1--
Family income status minimum wage< Regional minimum wage0.980.38–2.50.962
≥ Regional minimum wage1--
Mother’s smoking statusNon–smoker1--
Smoker4.780.46–49.330.19
Number of other household members who smoke01--
11.550.63–3.80.341
≥24.541.42–14.480.011

* Significant at p<0.05.

DISCUSSION

The prevalence of household smokers in this study was quite high, and it was primarily attributed to paternal smoking. The results of this study are in line with the study of Best et al.22 who examined 438336 households in rural areas in Indonesia where paternal smoking reached 73.7%. The study of Semba et al.21 showed that in 175583 urban poor households in Indonesia in 2007, the prevalence of paternal smoking reached 73.8%.

This study showed that lower levels of education were associated with paternal smoking and SHS exposure among infants. Several studies showed that parents’ low educational level was associated with SHS exposure in children15,2125. In this study, a significant relationship was found between the number of family members who smoked and exposure to cigarette smoking among infants, and there was a tendency that the greater the number of family members who smoked, the greater the risk of fathers becoming smokers. This occurred if the baby’s parents lived at home with their extended families. These results are in line with the study by Mwaniki and Gray26. They are also consistent with a meta-analysis study by Leonardi-Bee et al.27 and systematic literature review by Kusel et al.18, which found that the use of cigarettes significantly increased if at least one parent smoked, if mothers smoked, fathers smoked, both parents smoked, siblings smoked and other family members smoked. In another study, tobacco use was influenced by the number of close friends who smoked28.

In the present study, we found that a large number of children in the family was associated with higher odds of smoking in fathers and SHS exposure in children. Hawkins and Berkman29 also found that babies with siblings were at greater risk of SHS exposure. This study suggests that parents in Indonesia should change their smoking behaviours and rules at home, especially if they have more children. Further studies are needed to better understand these decision-making processes and help recognise critical periods for appropriate interventions29. Moreover, health promotion programmes should place more stress on tackling smoking in fathers who have infants with one or more siblings. Indeed, fathers’ and other household members’ smoking behaviours have received far less attention or health promotion consideration than mothers’ smoking behaviours. Protecting babies from smoking parents and other smokers in the family is crucial in diminishing SHS exposure in early infancy, when the risks of asthma attacks, respiratory infections, ear infections, and SIDS are more common30,31.

In this study, no relationship was found between family income and father’s smoking status. However, higher percentages of paternal smoking and SHS exposure among infants occurred in low-income families than in high-income families. Previous studies showed that lower social class increased the risk of exposure to cigarette smoke in women and children at home23,25,32. Furthermore, paternal smoking was associated with an increased risk of stunting and wasting in children from urban poor families21, and underweight and stunting in rural children in Indonesia22. According to Semba et al.21, the risk of undernutrition in early childhood was due to the fact that a large proportion of family income was used for cigarettes and a small proportion for buying food.

Other than the risk of undernutrition, the high prevalence of smoking in households and exposure to SHS in infants, the current study raises concerns about the risk of cardiovascular diseases (CVDs) at a younger age in Indonesia. Forty-five per cent of CVD deaths in those aged 30–44 years were more likely to be caused by tobacco use. CVDs are the number one cause of death with 558736 each year (36.3% of all deaths). CVD deaths caused by tobacco usage were 147510 or 26% of all CVD deaths each year3. Therefore, it is crucial to control tobacco to prevent and control disability and death caused by CVDs.

Getting smokers to quit would be problematic as smoking is part of the Indonesian culture. To address this, the most effective approach is indirectly asking whether father’s responsibility for their partners and children is of larger cultural value33. Regarding this, community-based smoke-free home initiatives in Yogyakarta, Indonesia, could redefine smoking cessation as a health issue for women and children34. This policy will go hand-in-hand with government regulations prohibiting smoking on public transport and public places, and providing smoking cessation services in hospitals and health centres etc.

Limitations

A limitation of this study was the small sample size, which led in some cases to wide confidence intervals. Further, we used a questionnaire to measure self-reported SHS and did not objectively measure nicotine or cotinine levels. However, several studies that have relied on self-reports on smoking behaviours found good agreement between cotinine/nicotine levels and self-reports, indicating the suitability of questionnaires as a means of measuring smoking behaviour and exposure3537. The current study also did not include e-cigarettes or heat-not-burn tobacco users in Indonesia. These limitations will need to be addressed in future studies.

CONCLUSIONS

The findings of the present study highlight that infants who are only a few months old and living with smokers are exposed to SHS at home. The father’s educational level, infants having siblings, and/or the number of other household members who smoke affect father’s smoking status and SHS exposure among infants at home. Designing an appropriate anti-smoking education program to target families is recommended. Furthermore, comprehensive tobacco control programmes to increase the adoption of smoke-free homes are likely to be an effective way to reduce SHS exposure and promote decreased cigarette smoking in families with children.