Study setting, design and participants

The data used in this study were obtained from the tobacco and e-cigarette survey among Malaysian adolescents (TECMA) 2016¹⁵, a national school-based study that investigate cigarette smoking, e-cigarette and shisha use among school-going adolescents in Malaysia. The TECMA was a cross-sectional study that used a two-stage stratified cluster sampling design to select a representative sample of school-going adolescents aged 11–19 years in Malaysia. Data were collected from respondents using a self-administered approach (filling in answers on an OMR form). Prior to the study, respondents were given a detailed explanation about the survey and were informed that their participation was voluntary. Respondents were also assured that their information was confidential and would only be used for study purposes. Only respondents who gave their consent and also obtained the consent of their parents/guardians were allowed to participate in the study. The study protocol was approved the Ministry of Education Malaysia, whereas approval to recruit the students as participants was given by the individual schools and their respective districts, and the State Education Department. Ethical approval for the TECMA was obtained from the Medical Research & Ethics Committee, Ministry of Health Malaysia (NMRR-16-108-28789).

Sample size determination

Sampling frame used in the TECMA was provided by the Ministry of Education. Malaysia was stratified into 15 states, followed by the division of schools into urban and rural areas for each state, based on the classification of Ministry of Education. The school, as the primary sampling unit, was randomly selected from each state using systematic probability sampling proportional to the enrolment of students in each school. The second stage of sampling involved the selection of class at the selected school using the simple random sampling method. All students from the selected classes were invited to participated in the study. Sample size was determined using a prevalence of e-cigarette usage of 3%¹⁶, design effect of 1.5 and expected non-response rate of 20%. A total of 13980 students from 138 schools from 15 states were required for the survey. Detail description of the study can be found elsewhere¹⁵.

Instrument

The questionnaire used in this study was adapted from the Global Youth Tobacco Survey (GYTS) questionnaire. In order to suit the local situation, additional items were also included by the subject matter experts from the research institutes and Ministry of Health Malaysia.

Validity and reliability

Items adapted from the GYTS underwent forward and backward translation by content and language expert teams. The instrument was pretested among school-going adolescents in selected schools to establish face validity.

Measurement

The dependent variables in our study (exposure to SHS at home, outside the home and inside the school) were measured using the following items: a) ‘During the past seven days, on how many days has anyone smoked inside your home, in your presence?’ (responses: 0, 1–2, 3–4, 5–6, or 7 days); b) ‘During the past seven days, on how many days were you exposed to cigarette smoke from other people in places other than your home?’ (responses: 0, 1–2, 3–4, 5–6, or 7 days); and c) ‘During the past 30 days, have you ever seen anyone smoking inside the school compound?’ (responses: yes, no). Those who answered non-zero to the first two items were classified as being exposed to SHS at home and outside the home, respectively, while those who answered ‘yes’ to the third item were classified as being exposed to SHS inside the school^12,17.

The independent variables in this study were gender (male, female), age (≤12, 13–15, ≥16 years), ethnicity (Malay, Chinese, Indian, Bumiputra Sabah, Bumiputra Sarawak, other), school location (urban, rural), current smoking status (yes, no), and perceive SHS as harmful (yes, no).

The categorization of school location (urban, rural) was based on the classification of the Ministry of Education. Current smoker was defined as someone who smoked any tobacco product during the past 30 days. Perception of SHS as harmful was measured using the following item: ‘Do you think the smoke from other people’s cigarette smoking is harmful to you?’ (responses: yes, no). Those who answered ‘yes’ were considered to perceive SHS as harmful.

Statistical analyses

Data entry included the scanning of an OMR form into excel format, where the data were exported to SPSS statistical software for analysis. Data were cleaned and sample weight was calculated (taking into account the complex sample design and response rate) to ensure that estimates can be produced for the target population. Descriptive statistics were used to describe the characteristics of respondents. The prevalence of SHS exposure at home, outside the home, and inside the school were estimated. Simple logistic regression analysis was used to test the associations between all categorical independent variables with exposure to SHS. Variables with p<0.25 were included into the multiple logistic regression analysis to determine the influence of each variable on SHS exposure at home, outside the home and inside the school (except for gender and perception that SHS is harmful, as they were important variables with significant associations reported in other studies). A two-way interaction analysis between the independent variables in the models were carried out and no significant two-way interaction was detected (p>0.05) in all models. Diagnostic testing to assess the goodness-of-fit was conducted to ensure the fit of the logistic regression model for individual cases or covariates. SPSS version 21 software was used to carry out all statistical analysis, at 95% confident interval.

Prevalence of SHS at home, outside the home and inside the schools

This is the first study that describes SHS exposure at specific locations among a nationally representative sample of school-going adolescents in Malaysia. We found that more than one-third (37.8%) of the respondents were exposed to SHS at home. The prevalence of SHS exposure inside the home was higher than those reported in Saudi Arabia (32.7%)¹⁸ and among youth from 168 countries (30.4%)¹⁹. However, findings similar to ours were reported among youth in Gambia (38.2%)¹². The latest prevalence of SHS exposure reported in our study was significantly lower compared with previous studies conducted in Malaysia. For instance, the Malaysian Global Youth Tobacco Survey (GYTS 2009) reported a prevalence of 48.7%²⁰ while Lim et al.¹⁴ reported a prevalence of 56.4% among school-going adolescents in Malaysia. The reduction of SHS exposure was encouraging as the prevalence of smoking among Malaysian adults has plateaued since about a decade ago²¹. Higher level of awareness of smoking and SHS hazards among Malaysian adults might have caused them not smoke at home, leading to the reduction of SHS exposure at home among school-going adolescents. However, in-depth qualitative and quantitative studies are required to test our hypotheses. Enhancement of health promotion and smoke-free interventions among Malaysians are among the ways forward as both measures have shown promising efficacy in reducing SHS exposure inside the home²².

More than half (51.2%) of the respondents reported being exposed to SHS outside the home, which was similar to the prevalence reported by students in Saudi Arabia (49.3%)¹⁸. Another study conducted in 168 countries (44.2%)¹⁹ reported lower prevalence of SHS exposure outside the home. However, studies among youth from Gambia (61.4%)¹² and Malaysia (2009; 64.1%) reported higher prevalence of SHS exposure compared to our findings²³. Although the reduction of SHS exposure from 64.1% to 51.2% within 7 years is very encouraging, the prevalence of SHS exposure among Malaysian youth is still high. Therefore, more proactive measures such as expansion of smokefree public areas with comprehensive enforcement is needed. Adolescents should also be trained and encouraged to avoid SHS in public areas. The implementation of measures should involve all related stakeholders, parents, authorities, and community members, regardless of smoking status.

More than a quarter (27.3%) of school-going adolescents reported being exposed to SHS inside the school. The prevalence was higher than what had been reported by students in Gambia¹² (21.3%) but lower than the prevalence reported in the GYTS conducted in the Federated States of Micronesia (65.7%), Mongolia (56.4%) and China (54.5%)²⁰. The lower prevalence of SHS exposure in our study might be due to the introduction of smoke-free policies, which can create an anti-smoking norm inside the school. However, the prevalence of exposure in school was still high taking into consideration the prohibition of smoking in the school premises, which had been introduced two decades ago⁹. Our findings indicate the importance of cooperation among the stakeholders (school authority, health department and parent–teacher association) to tackle SHS exposure in the school. A study conducted in California reported lower risk of SHS exposure in school with stronger smoking regulations²⁴. Therefore, health promotion and a full enforcement of s smoking ban should be carried out in a collaboration between the school and health authorities, as the way forward to reduce SHS exposure inside the school.

Associated factors of SHS at home, outside the home and inside the schools

Our study found that male adolescents were at higher risk of SHS exposure outside the home and at school but less likely to be exposed to SHS in the home. Similar findings were reported among adolescents from Gambia, Saudi Arabia, and another 168 countries^12,18,19, while a study among adolescents in Cambodia reported a non-significant difference between genders¹⁷. Studies revealed that boys are more active and spend more time in school and outdoor areas compared to girls²⁵, which increases their risk for SHS exposure at these places. On the other hand, girls tend to be closer to their parents and stay at home more often, which increases their risk for SHS exposure at home rather than at outdoor areas²⁶.

The odds of being exposed to SHS were higher among older adolescents at home, outside the home and in the school compared with younger adolescents. Similar findings were reported in Cambodia and Saudi Arabia^17,18. A study among adults in Canada reported that those living with younger children tend to have smoke-free homes while those who were living with older children reported higher percentage of not having smoke-free rules in their homes²⁷, which may be the plausible explanation for higher risk of SHS exposure among older adolescents at home in our study. Along with that, older adolescents spent less time with the family at home and have higher chance to be outside home, which increases their risk of SHS exposure at places they visit²⁸.

Adolescents of Malay and Bumiputra Sarawak ethnicity were more likely to be exposed to SHS at home, outside the home and in the school compared with Chinese adolescents. Another study among Malaysian adolescents reported similar findings¹⁴. Our findings can be explained by the higher prevalence of smoking among adults of Malay, and Bumiputra Sarawak, ethnicity compared with Chinese, as adolescents living with smoking adults have a higher risk of being exposed to SHS²¹. In addition, a previous study among Malaysian adults reported that Malay adults were less likely to adopt a smoke-free home policy, which explains the higher likelihood of SHS exposure at home among Malay adolescents²⁹.

Adolescents who were from rural areas were more likely to be exposed to SHS at home, outside the home and in the school. Other studies among adolescents from Gambia and Korea reported similar findings^12,26. The Malaysian National Health and Morbidity Survey (2019) reported a significantly higher prevalence of smoking among adults residing in rural areas, which might have contributed to an increased risk of SHS exposure among adolescents from rural areas²¹. In addition, adolescents from rural areas had higher risk of SHS exposure possibly due to the higher prevalence of smoking among adolescents from rural areas¹⁵. It has been shown that having friends who smoke increases the risk of SHS exposure among adolescents¹⁸. Furthermore, the higher likelihood of SHS exposure in rural areas might be due to rural dwellers having less exposure to health promotion or anti-smoking campaigns³⁰. This calls for more anti-tobacco programs and strict enforcement of tobacco control laws in the country, particularly in rural areas.

Our study showed that adolescents who were current smokers had higher odds of being exposed to SHS at home, outside the home and in the school. Our findings are consistent with those reported in studies among adolescents in Saudi Arabia, Gambia, and Malaysia^12,14,18. These findings might have been expected in view of the fact that adolescents who smoke are less likely to perceive SHS as harmful and thus less likely to avoid SHS³¹. In addition, evidence suggested that smoking adolescents were more likely to have friends who smoked, compared to nonsmokers, and therefore have a higher risk of being exposed to SHS³².

It might have been expected that adolescents who had higher awareness of the harmful effects of SHS would have a lower risk of SHS exposure and vice versa. However, in our study, adolescents who did not perceive SHS as harmful had lower risk of being exposed to SHS outside the home and in the school, in contrast to what had been reported among adolescents in Saudi Arabia¹⁸. Nevertheless, our findings are in line with studies conducted among adolescents from South Africa, which reported that the perception of SHS as being harmful was associated with higher likelihood of SHS exposure³³. Another recent study among adolescents from the United States similarly reported that those who thought more about the harmful effects of SHS had higher risk of SHS exposure, which is attributable to the fact that those who were exposed to SHS thought more about the harmful effects of SHS³⁴. This phenomenon could also be due to the lack of SHS avoidance, as highlighted in a study in Jordan that avoidance behavior towards SHS can be low even with high awareness of the hazards of SHS³⁵.

The Malaysian government is committed towards achieving the target of becoming a tobacco-free nation with all Malaysians being protected from SHS exposure by the year 2045 (endgame for tobacco)⁵. In achieving this target, several strategies have been implemented, including strengthening of existing tobacco control activities, strengthening of enforcement of national tobacco control legislation, empowerment of community, strengthening of multisectoral collaboration and strengthening of tobacco control activities based on the MPOWER strategy (Monitor tobacco use and prevention policies; Protect people from tobacco smoke; Offer help to quit tobacco use; Warn about the dangers of tobacco; Enforce bans on tobacco advertising, promotion and sponsorship; Raise taxes on tobacco). The implementation of such interrelated tobacco control measures has contributed to the reduction of SHS exposure at home and public places in Malaysia⁵.

Strengths and limitations

This study is not without limitations. First, our study utilized cross sectional data and therefore causal relationships could not be established. Second, SHS exposure was based on self-reporting and thus there is the possibility of under-reporting or over-reporting due to recall bias. Despite these limitations, another study has proven that the self-report method can be reliable to measure SHS exposure³⁶. In addition, our study is a nationwide study involving a representative sample, which enables generalization of the findings to adolescents in Malaysia.