Sage Journals: Discover world-class research

Abstract

Background:

The proportion of people exposed to second-hand smoke increases dramatically with a sharp increase in the prevalence of smoking. It is expected that awareness of the effect of second-hand smoke plays an important role in curbing smoking.

Objectives:

The objective of this study is to examine factors affecting awareness of second-hand smoke in Malaysia.

Methods:

Nationally representative data, covering a large sample size (4153 respondents) and collected based on multistage sampling, are used. Multiple logistic regression is employed to analyse the effects of demographic and lifestyle variables on the likelihood of being aware of second-hand smoke.

Results:

Results show that older people are less likely to be aware of second-hand smoke than youngsters and that less-educated individuals are less likely to be aware of second-hand smoke than well-educated individuals. Significant relationships are found between second-hand smoke awareness and wealth index, ethnicity and smoking behaviour.

Conclusion:

Demographic profiles should be considered when drawing up policies aimed at improving awareness of second-hand smoke among adults. The government needs to focus more on older people, wealthy individuals, less-educated individuals, Chinese individuals and smokers if a successful policy is to be implemented.

Keywords

Adult awareness education second-hand smoking

Introduction

Smoking-induced disease is one of the main killers in today’s society. Each year, more than 5 million deaths are associated with smoking.¹ At least one billion smokers reside in developing countries.¹ The negative health consequences of smoking have been well documented in the literature.² Lung cancer is highly associated with smoking. People who smoke have a 25% higher probability of developing lung cancer than non-smokers.² Smoking is also responsible for other cancers, such as liver cancer and colorectal cancer.² Furthermore, smoking is one of the main factors causing chronic obstructive pulmonary disease. Studies show that women who smoke are 22.4% more likely to have chronic obstructive pulmonary disease than their counterparts who do not smoke.³ Diabetes is another disease that is related to smoking. In general, compared with non-smokers, the probability of developing diabetes is 30–40% higher for smokers. There is also evidence that smoking has negative impacts on the immune system and reproductive health. Hence, smokers are more likely to acquire infectious diseases and experience the problem of birth defects than non-smokers.

Second-hand smoke is smoke produced by tobacco products, such as cigarettes, cigars and pipes. It is exhaled by persons who smoke and inhaled involuntarily by persons who do not smoke. Nowadays, second-hand smoke has become a serious public health concern. It is one of the main factors contributing to morbidity and mortality across globe. A report shows that, in the USA, second-hand smoke has caused approximately 2.5 million deaths among adults who are non-smokers since 1964.⁴ There are findings suggesting that individuals who are frequently exposed to second-hand smoke are 25–30% more likely to develop heart diseases than individuals who are not exposed to second-hand smoke.⁴ Furthermore, second-hand smoke can elevate the risk of acquiring stroke by at least 20–30%.⁴

In view of the undesirable realities of second-hand smoke, the Malaysian government has introduced an anti-smoking policy with the aim of reducing the number of people exposed to second-hand smoke. This policy includes increasing the tax on cigarettes and banning smoking in all the public areas. In addition, the Malaysian government has organized ‘tak nak’, which is a nationwide anti-smoking campaign. The purpose of this campaign is to create smoking awareness by educating people about the negative consequences of smoking. However, this campaign has not been very successful.⁵ In spite of policy makers’ efforts to curb second-hand smoke, people are still not well protected from tobacco smoke air pollution. A report shows that at least 40% of adults are exposed to second-hand smoke in their workplaces and homes.¹ Moreover, about 70% and 28% of adults breathe in tobacco smoke when they visit restaurants and use public transport, respectively.¹ A lack of awareness of the effects of second-hand smoke may be one of the reasons behind these alarming facts and figures. Although smokers understand the consequences of smoking on their own health, they may not be aware that tobacco smoke has serious negative impacts on non-smokers’ health.

There exist studies related to smoking behaviour in Malaysia.^5–8 Tan et al.⁵ used the Malaysian Household Expenditure Survey and found that well-educated and older individuals are less likely to smoke and also spend less on cigarettes than their less-educated and younger counterparts. Interestingly, they also found that being male increases the likelihood of smoking, as well as expenditure on cigarettes. Using the Third National Health and Morbidity Survey, Cheah and Naidu⁶ found positive relationships between smoking and young adults, poor individuals, men, unmarried individuals, Malay individuals, employed individuals, rural dwellers and those having only primary-level education. The findings of Tan et al.⁵ and Cheah and Naidu⁶ were corroborated by Tan⁷ and Lim et al.,⁸ who used different datasets.

The association between awareness of the effects of smoking and smoking behaviour has been demonstrated in previous studies. Researchers found that awareness of the effects of smoking can increase the likelihood of smoking cessation.⁹ Furthermore, other studies showed that awareness of the effects of smoking can influence smokers’ decisions to quit smoking.^{10, 11} Study conducted in China also suggested that awareness of the effects of smoking can determine participation decision of smoking among women.¹² The impact of pictorial health warnings on cigarette packs on smoking was also identified by researchers.¹³ Taken together, it can be concluded that awareness of the effects of smoking plays an important role in reducing smoking. Therefore, a population-based policy directed towards improving the awareness of second-hand smoke may help to reduce the proportion of people exposed to second-hand smoke. It is important to know which population groups are or are not aware of second-hand smoke if a more effective policy is to be introduced.

Studies related to second-hand smoke are scarce, and research investigating demographic differences in awareness of second-hand smoke are even more scant. Although there is a lack of studies on the awareness of second-hand smoke, numerous studies have found significant relationships between demographic factors and awareness of diseases. For instance, Gelaw¹⁴ found that rural dwellers, youngsters, less wealthy individuals and less-educated individuals are more likely to have a low awareness of tuberculosis. Carpenter et al.¹⁵ found that older individuals are more aware of Alzheimer’s disease than are younger individuals. Other researchers identified that education level, socioeconomic status and being female are positively associated with an awareness of diabetes and colorectal cancer.^{16, 17} Similar findings were demonstrated by Islam et al.,¹⁸ in examining factors affecting the awareness of eye disease. Given these findings, we expect that demographic factors have a significant impact on awareness of second-hand smoke. In Malaysia, there is a study examining the factors affecting second-hand smoke exposure.¹⁹ However, the study did not investigate awareness of second-hand smoke. Moreover, the study included a limited range of demographic variables and did not focus specifically on the case of Malaysia.

In this study, we examine the factors associated with awareness of second-hand smoke. We use a logistic model to analyse the effects of demographic and health variables on the likelihood of being aware of the effects of second-hand smoke. Overall, this study offers several contributions to the literature. Firstly, the topic of the present study, that is, determinants of second-hand smoke awareness, has not been studied by researchers. Secondly, the present study includes a comprehensive list of demographic variables (i.e. age, sex, wealth index, education, ethnicity, marital status, house locality); thus, omission of variables is not an issue. Thirdly, the data used in the present study come from a large sample size. Hence, important findings can be generated, which can assist policy makers in developing more effective intervention measures. Fourthly, the country of interest is Malaysia, which has a high prevalence of smoking and lacks empirical studies related to second-hand smoke.

Methods

Sample

Analysis of the present study was based on a secondary dataset. Data from Global Adult Tobacco Survey (GATS) Malaysia were used. GATS Malaysia was jointly conducted by the Ministry of Health, Malaysia and the World Health Organization (WHO). The objective of the survey was to obtain information on the burden of smoking and anti-smoking policies. GATS Malaysia was a nationally representative survey, held in 2011. It is the latest smoking-related dataset in Malaysia and was made available to the public in 2014. GATS Malaysia comprised information related to smoking, such as tobacco use, cessation, second-hand smoke, economics, media and knowledge. Data were collected based on standardized methodology (e.g. questionnaire, sampling design, data collection) in order to generate globally comparable data. Therefore, GATS Malaysia was appropriate for the present study. The sampling of GATS was conducted in three stages designed by the Department of Statistics, Malaysia. In the first stage, a total of 426 enumeration blocks were selected from the total 74,756 enumeration blocks in Malaysia. Of the selected enumeration blocks, 222 were from urban areas and 204 were from rural areas. In the second stage, selection was based on living quarters in each selected enumeration block. Each enumeration block comprised 80 to 120 living quarters and only 12 living quarters were chosen. In the third stage, households in the selected living quarters were randomly selected, and members of the households were interviewed (face-to-face). Inclusion criteria were individuals aged 15 years and above, both sexes and all ethnic groups. Exclusion criteria were tourists and institutionalized individuals in hospitals, nursing homes, military bases or prisons.

Data collection

The Institute for Public Health Malaysia collaborated with the University of Malaya and the International Islamic University in carrying out GATS in Malaysia. The data collection process was outsourced to a private company that specialized in surveying. Nevertheless, the whole process, as well as the quality of the survey, was closely monitored by the Institute for Public Health Malaysia. Financial and technical support for this survey was provided by the WHO (Project HQTFI1003729).

To reduce errors in data collection, a pre-test was conducted in both urban and rural areas based on a sample size of 120 respondents. Respondents used in the pre-test were distributed based on age group, sex and smoking status. The pre-test was conducted by the WHO. In addition, all the interviewers and field supervisors attended a training campaign prior to the interview. The training included mock interviews and role-playing. To ensure that the survey procedures were consistent, three manuals were prepared and used – a field interviewer’s manual, a field supervisor’s manual and a question-by-question specification. The field interviewer’s manual covered interview techniques and methods of asking questions. The field supervisor’s manual comprised information on supervisors’ roles and data transfer procedures. Question-by-question specification offered instructions for each question in the questionnaire.

The questionnaires were used by field interviewers to interview respondents. Information obtained from the survey was then submitted to field supervisors. The field supervisors had to monitor the data collection done by the field interviewers and ensured that the interview was completed within the time frame. All the information obtained from the survey was transferred to a data management team at the Institute for Public Health Malaysia. The field-level data were analysed twice a week to ensure that there was no data collection error.

The target sample size was 5112 respondents. However, only 4389 respondents completed the interview, of which 2160 lived in urban areas and 2229 lived in rural areas. The overall response rate was about 88.10%. For respondents aged 17 years and above, written consent from themselves had to be obtained before the interview. If respondents were aged 16 years and below, their parents’ or guardians’ written consent was necessary. All the protocols were approved by the Medical Research and Ethics Committee of the Ministry of Health, Malaysia. Full information of GATS Malaysia is described elsewhere.¹

Ethical considerations

GATS Malaysia is owned by the Ministry of Health, Malaysia, and the WHO. It was collected as part of a research project of the Institute for Public Health Malaysia. Because the full dataset was not freely available on the internet, we sought permission to use the data from the Ministry of Health, Malaysia. The data obtained was used only for the purpose of this study and is not intended for answering other studies’ research questions. Nonetheless, this study has been reviewed by the research committee of the Ministry of Health, Malaysia, prior to publication to ensure that there are no ethical issues.

Measures

The dependent variable is awareness of the effects of second-hand smoke and is formatted as a qualitative variable. It is obtained from the question, ‘Based on what you know or believe, does breathing other people’s smoke cause serious illness to a non-smoker?’ Those who answered ‘yes’ were coded as 1, while those who answered ‘no’ were coded as 0. The independent variables used in the analysis are grouped into two main categories: demography and lifestyle. The demography category consists of age, sex, wealth index, education, ethnicity, marital status and house locality. The lifestyle category comprises smoking.

Age was calculated from the respondents’ date of birth. To acquire a better understanding of the relationship between age and second-hand smoke awareness, age was divided into four categories: 15–24 years, 25–44 years, 45–64 years and ≥65 years. Wealth index was widely used in previous studies to measure a respondents’ economic background.^{20, 21} It was assessed based on the assets owned by the respondents. These assets included productive assets, non-productive assets and household amenities. For analytical purposes, the present study grouped the index into five quintiles: lowest, second, middle, fourth and highest. However, the survey did not ask about the respondents’ household income, which was difficult to measure owing to several factors. First, the respondents might not know their exact income. Second, the respondents might under-report their income in order to avoid tax. Third, some family members in a household did not reveal their income. Fourth, income was volatile as it varied with time. Last, income from non-labour activities, such as investment, was hard to measure. See the work of Rutstein and Johnson²² for a detailed description of wealth index.

To obtain information on education, the respondents were asked, ‘What is your highest level of education?’ The answers were collapsed to form a categorical variable: 1 for respondents who were less-educated, i.e. those who did not complete secondary school (<11 years of schooling), and 0 for respondents who were well-educated. Information on ethnicity was collected: Malay, Chinese, Indian and others. Because only a small number of the respondents reported ‘others’, this category was combined with ‘Indian’ to form a single category (Indian or others). The respondents’ marital statuses were also surveyed; these were categorized as married, widowed or divorced and single. House locality was assessed based on enumeration blocks. It was formatted as a categorical variable: 1 referred to urban; 0 represented rural. Smoking variable was assessed with the question, ‘Do you currently smoke tobacco every day, less than every day or not at all?’ To facilitate comparison, if the respondents answered ‘every day’ or ‘less than every day’, they were considered to be smokers, whereas if the respondents answered ‘not at all’, they were considered to be non-smokers.

Statistical analysis

Several statistical analyses were performed. First, descriptive statistics of all the independent variables were used to provide information on the characteristics of survey respondents. Next, the proportion of smoker by demographic factors was calculated to identify the prevalence of smoking. In addition, Pearson’s χ² test was calculated to assess the association between second-hand smoke awareness and the independent variables. Bivariate analysis might not offer very accurate estimates because it does not control for independent variables. Therefore, further analysis was conducted using multiple logistic regression. (See Wooldridge²³ for further discussion on logistic regression.) The effects of demographic and lifestyle variables on the likelihood of being aware of the effects of second-hand smoke were estimated from the multiple logistic regression. The overall significance of the regression model was assessed using likelihood ratio statistics. The significance level was based on p < 0.05. After removing data from respondents with incomplete information, data from only 4153 respondents were used for analysis. All the statistical analyses were carried out using STATA statistical software.²⁴

Results

Table 1 shows the characteristics of the survey respondents. A large proportion of the respondents are aged 25–44 years (41.70%), followed by those aged 45–64 (31.04%), 15–24 (17.65%) and ≥65 years (9.61%). The proportions of men (48.83%) and women (51.17%) in the sample are quite similar. The wealth index breakdown consists of 20.27% lowest, 20.23% second, 19.72% middle, 19.74% fourth and 20.03% highest. The proportion of well-educated individuals (51.89%) is slightly greater than that of less-educated individuals (48.11%). Of all the respondents, 59.67% are Malay, 15% are Chinese and 25.33% are Indian or others. In terms of marital status, the proportion of married individuals (63.47%) outweighs those who are widowed or divorced (11.82%) or single (24.71%). The majority of the respondents are rural dwellers (51.24%) and non-smokers (81.68%).

Table 1.

Descriptive statistics of demographic and lifestyle variables (n = 4153).

Independent variable	Frequency	Percentage
Demographic
Age
15–24	733	17.65
25–44	1732	41.70
45–64	1289	31.04
≥65	399	9.61
Sex
Male	2028	48.83
Female	2125	51.17
Wealth index
Lowest	842	20.27
Second	840	20.23
Middle	819	19.72
Fourth	820	19.74
Highest	832	20.03
Education
Less-educated	1998	48.11
Well-educated	2155	51.89
Ethnicity
Malay	2478	59.67
Chinese	623	15.00
Indian or others	1052	25.33
Marital status
Married	2636	63.47
Widowed or divorced	491	11.82
Single	1026	24.71
House locality
Urban	2025	48.76
Rural	2128	51.24
Lifestyle
Smoking
Yes	761	18.32
No	3392	81.68

Source: Global Adult Tobacco Survey Malaysia.

The proportion of smokers by demographic factors is presented in Table 2. In terms of age, 23.27% of those aged 25–44 years are smokers, compared with 16.76%, 15.59% and 6.77% of those aged 45–64, 15–24 and ≥65 years, respectively. A higher proportion of men are smokers (36.34%), compared with 1.13% of women. Considering the wealth index, a higher proportion of those in the middle wealth index group (23.20%) compared with those in other wealth index groups (lowest (15.56%), second (16.43%), fourth (18.29%), highest (18.27%)) smoke. Close to 21% of those who are well-educated are smokers, while only 16.37% of those who are less-educated are smokers. Smoking is more common among Indians or others (21.10%) than among Chinese individuals (12.20%) or Malay individuals (18.68%). A larger proportion of single individuals are smokers (23.29%), compared with widowed or divorced (8.55%) and married (18.17%) individuals. The prevalence of smoking among urban dwellers (18.52%) and rural dwellers (18.14%) individuals is almost similar.

Table 2.

Proportion of smokers by demographic factors (n = 4153).

Independent variables	Smoker (percentage)
Demographic
Age
15–24	15.69
25–44	23.27
45–64	16.76
≥65	6.77
Sex
Male	36.34
Female	1.13
Wealth index
Lowest	15.56
Second	16.43
Middle	23.20
Fourth	18.29
Highest	18.27
Education
Less-educated	16.37
Well-educated	20.14
Ethnicity
Malay	18.68
Chinese	12.20
Indian or others	21.10
Marital status
Married	18.17
Widowed or divorced	8.55
Single	23.39
House locality
Urban	18.52
Rural	18.14

Source: Global Adult Tobacco Survey Malaysia.

The relationships between awareness of the effects of second-hand smoke and demographic, and lifestyle variables are presented in Table 3. Awareness of second-hand smoke is significantly related to age; 86.72% of those aged 25–44 years are aware of the effects of second-hand smoke, compared with 69.67% and 83.40% of those aged ≥65 and 45–64 years, respectively. A higher proportion of those in the lowest quintile of the wealth index are aware of the effects of second-hand smoke (90.02%), compared with those in the highest quintile of the wealth index (76.08%). However, there is not much differences between the second (88.81%) and middle quintiles (85.10%).

Table 3.

Second-hand smoke awareness by demographic and lifestyle variables (n = 4153).

Independent variables	Aware(n₁ = 3488)	Not aware(n₂ = 665)	χ ²	p
Demographic
Age
15–24	633 (86.36)	100 (13.64)	73.797	<0.001
25–44	1502 (86.72)	230 (13.28)
45–64	1075 (83.40)	214 (16.60)
≥65	278 (69.67)	121 (30.33)
Sex
Male	1682 (82.94)	346 (17.05)	3.241	0.072
Female	1806 (84.99)	319 (15.01)
Wealth index
Lowest	758 (90.02)	84 (9.98)	87.678	<0.001
Second	746 (88.81)	94 (11.19)
Middle	697 (85.10)	122 (14.90)
Fourth	654 (79.76)	166 (20.24)
Highest	633 (76.08)	199 (23.92)
Education
Less-educated	1574 (78.78)	424 (21.22)	77.677	<0.001
Well-educated	1914 (88.82)	241 (11.18)
Ethnicity
Malay	2128 (85.88)	350 (14.12)	16.569	<0.001
Chinese	502 (80.58)	121 (19.42)
Indian or others	858 (81.56)	194 (18.44)
Marital status
Married	2329 (84.94)	397 (15.06)	26.739	<0.001
Widowed or divorced	373 (75.97)	118 (24.03)
Single	876 (85.38)	150 (14.62)
House locality
Urban	1744 (86.12)	281 (13.88)	13.407	<0.001
Rural	1744 (81.95)	384 (18.05)
Lifestyle
Smoking
Yes	595 (78.19)	166 (21.81)	23.313	<0.001
No	2893 (85.29)	499 (14.71)

Percentages calculated for rows are in parentheses.

Source: Global Adult Tobacco Survey Malaysia.

A higher proportion of well-educated individuals (88.82%) compared with less-educated individuals (78.78%) have second-hand smoke awareness. There is also an association between second-hand smoke awareness and ethnicity. Approximately 85.88% of Malay individuals are aware of the effects of second-hand smoke, compared with 81.56% and 80.58% of Indian or other individuals and Chinese individuals, respectively. Compared with widowed or divorced individuals (75.97%), a higher proportion of married (84.94%) and single individuals (85.38%) are aware of the effects of second-hand smoke. A higher proportion of urban dwellers are aware of the effects of second-hand smoke (86.12%) compared with rural dwellers (81.95%). In terms of smoking, 78.19% of smokers compared with 85.29% of non-smokers have second-hand smoke awareness. The results lead to the conclusion that elder people, individuals with a higher wealth index, the less-educated, Chinese individuals, widowed or divorced people, rural dwellers and smokers are associated with less awareness of the effects of second-hand smoke.

The multiple logistic regression that shows the association between the dependent and independent variables is illustrated in Table 4. The likelihood ratio is highly significant, thereby indicating that all the independent variables are jointly significant in affecting the dependent variable. In addition, a large proportion (84%) of the outcomes are correctly predicted by the regression model. Taken together, it can be concluded that the model is well specified.

Table 4.

Association between second-hand smoke awareness and demographic, and lifestyle variables (n = 4153).

Independent variables	Estimates	Standard error	Odds ratio	95% confidence interval	p
Constant	2.295	0.197	9.920	6.737, 14.608	<0.001
Demographic
Age
15–24	–	–	1.000	–	–
25–44	0.085	0.158	1.088	0.799, 1.482	0.590
45–64	−0.023	0.175	0.978	0.694, 1.378	0.897
≥65	−0.655	0.206	0.519	0.347, 0.778	0.001
Sex
Male	0.031	0.103	1.031	0.843, 1.262	0.764
Female	–	–	1.000	–	–
Wealth index
Lowest	–	–	1.000	–	–
Second	−0.131	0.161	0.877	0.639, 1.203	0.416
Middle	−0.324	0.156	0.723	0.532, 0.983	0.038
Fourth	−0.656	0.155	0.519	0.383, 0.703	<0.001
Highest	−0.733	0.162	0.481	0.350, 0.660	<0.001
Education
Less-educated	−0.351	0.106	0.704	0.572, 0.867	<0.001
Well-educated	–	–	1.000	–	–
Ethnicity
Malay	0.193	0.105	1.213	0.986, 1.491	0.067
Chinese	−0.326	0.143	0.722	0.545, 0.956	0.023
Indian or others	–	–	1.000	–	–
Marital status
Married	0.054	0.138	1.055	0.805, 1.383	0.696
Widowed or divorced	−0.074	0.186	0.929	0.645, 1.336	0.690
Single	–	–	1.000	–	–
House locality
Urban	0.121	0.095	1.128	0.937, 1.359	0.204
Rural	–	–	1.000	–	–
Lifestyle
Smoking
Yes	−0.642	0.121	0.526	0.415, 0.667	<0.001
No	–	–	1.000	–	–
Likelihood ratio	192.350
p	<0.001
Correct prediction (%)	0.840

Source: Global Adult Tobacco Survey Malaysia.

Only an age category is statistically significant (≥65 years). Individuals aged 65 years and above are less likely to be aware of the effects of second-hand smoke than individuals aged 15–24 years (odds ratio = 0.519; 95% confidence interval = 0.347, 0.778). Sex does not seem to have a significant effect on second-hand smoke awareness. Setting the lowest wealth index as the reference category, the odds of being aware of second-hand smoke reduces with the level of wealth index. Specifically, the middle, fourth and the highest wealth indexes have odds ratios of 0.723, 0.519 and 0.481, respectively. The relationship between education and second-hand smoke awareness is strong. Less-educated individuals have a lower likelihood of being aware of the effects of second-hand smoke than well-educated individuals (odds ratio = 0.704; 95% confidence interval = 0.572, 0.867).

Compared with Indian or other individuals, Chinese individuals are less likely to be aware of the effects of second-hand smoke (odds ratio = 0.722; 95% confidence interval = 0.545, 0.956). However, the difference in second-hand smoke awareness between Indians or other individuals and Malay individuals is not significant. It is surprising that marital status and house locality are not significantly associated with second-hand smoke awareness. The smoking variable also shows a strong relationship with second-hand smoke awareness. In particular, smokers display lower odds of being aware of the effects of second-hand smoke than non-smokers (odds ratio = 0.526; 95% confidence interval = 0.415, 0.667). Overall, the results of logistic regression are somewhat similar to those of Pearson’s χ² test. It should be noted, however, that marital status and house locality are only significant in bivariate analysis. This indicates that marital status and house locality do not have an impact on second-hand smoke awareness when other demographic variables are held constant.

Discussion

The objective of this study is to examine factors affecting awareness of the effect of second-hand smoke, in the light of a lack of studies on this topic. Overall, the present study’s results suggest that demographic factors, such as age, wealth index, education or ethnicity are significantly associated with awareness of the effect of second-hand smoke; these findings are quite similar to the findings of previous studies on awareness of diseases. In addition, the present study makes an attempt to investigate whether smoking is related to second-hand smoke awareness. The findings confirm the popular belief that smokers are less aware of the effects of second-hand smoke than are non-smokers.

With regard to the age variable, older individuals (≥65 years) are less likely to be aware of the effects of second-hand smoke than younger individuals (15–24 years). This finding can be explained by the fact that older people have poorer understanding than youngsters because of a lack of educational opportunities in the past.²⁵ Therefore, in general, older people tend to find it difficult to understand the negative impacts of smoking on non-smokers. If more health education programmes are made available to older people, it is plausible that knowledge about smoking among older people will be improved. These findings and arguments are, however, in contrast to those of previous studies.^14–16 Previous studies suggested that older individuals are more likely to have a better awareness of diseases than their younger counterparts because older individuals tend to have poorer health. It is worth noting that while individuals aged 15–24 years have a higher awareness of the effect of second-hand smoke than individuals aged ≥65 years, they are more likely to smoke. This implies that a policy directed towards reducing the prevalence of smoking needs to focus on youngsters as well.

It is noteworthy that although smoking is more prevalent among men than women, there is no significant difference in second-hand smoke awareness between the sexes. This implies that men have a higher tendency to smoke cigarette than women, even though they have the same level of second-hand smoke awareness as women. Therefore, one can conclude that there are factors other than second-hand smoke awareness that cause men to smoke. This finding calls for policy makers to put more emphasis on introducing intervention measures that can directly curb smoking among men rather than improving second-hand smoke awareness, if the objective of reducing smoking among men is to be met.

The effect of wealth index on second-hand smoke awareness is interesting. Holding other factors constant, individuals who are in the middle, fourth and highest quintiles of the wealth index are less likely to be aware of the effects of second-hand smoke than their counterparts who are in the lowest quintile. Based on this outcome, one can conclude that wealthier individuals are less aware of the effects of second-hand smoke than less wealthy individuals. This may be because wealthier individuals carry more job responsibilities and thus have less time on hand to learn about the negative consequences of smoking.²⁶ Another possible reason is that wealthier individuals are more able to purchase medical care and health supplements; thereby, they are unlikely to make efforts to prevent diseases by acquiring more health knowledge.²⁷ However, previous studies focusing on awareness of diseases found that socioeconomic status is positively associated with level of awareness.^{16–18, 28} It can, thus, be concluded that wealth index and socioeconomic status have different impacts on second-hand smoke awareness.

Education is strongly associated with second-hand smoke awareness, given that well-educated individuals are more likely to be aware of the effects of second-hand smoke than less-educated individuals. This association remains significant after other demographic factors are controlled for, thereby indicating that providing people with more education can help to improve awareness of second-hand smoke. Similar findings were evidenced by previous studies that focused on awareness of diseases.^{14, 17, 18, 28, 29} In particular, previous studies found that level of education is positively associated with the level of awareness of diseases. Three explanations are noteworthy. First, well-educated individuals are more able to understand the information on smoking than less-educated individuals because they have better comprehension skills. Second, compared with less-educated individuals, well-educated individuals tend to have more opportunities for exposure to health-related subjects and programmes. Third, education reduces the rate of time preference.³⁰ Individuals with a lower rate of time preference are more future-oriented than individuals with a higher rate of time preference. Therefore, in general, well-educated individuals are more devoted to make an effort to improve their health by acquiring more health knowledge than are less-educated individuals.

The results on ethnicity show that Chinese individuals are less likely to be aware of the effects of second-hand smoke than Indian or other individuals. However, there is no difference in the likelihood of having second-hand smoke awareness between Malay and Indian or other individuals. These findings may be partly explained by cultural and religious factors. It is plausible that different ethnicities may have different views on smoking. Because the present study is not an in-depth qualitative study, the actual reasons explaining the ethnic differences in second-hand smoke awareness are not well-identified. In terms of smoking likelihood, previous studies found that, of all the ethnic groups, Chinese has the lowest odds of smoking and spend the least amount of money on tobacco.^{5, 6, 8} Since Chinese individuals are unlikely to smoke, they are expected to be highly aware of the effects of second-hand smoke. Surprisingly, however, this expectation is not supported by the present study’s findings.

A strong relationship exists between smoking and second-hand smoke awareness. Both bivariate analysis and multiple regression show that smokers are less likely to be aware of the effects of second-hand smoke than non-smokers. The explanation is simple. It is clearly evident that smokers are more present-oriented and more impatient than non-smokers.^{31, 32} In other words, smokers are willing to trade-off their future health for the current satisfaction of smoking. Since acquiring information on smoking is a health investment, smokers are less likely to make an effort to learn about smoking effects than non-smokers. This indicates that poor awareness of second-hand smoke and smoking participation are closely related. If individuals have poor awareness of the effects of second-hand smoke, they are likely to engage in smoking behaviour. Therefore, intervention measures should increase smokers’ awareness of second-hand smoke so the likelihood of quitting smoking among smokers may increase. However, this conclusion needs to be supplemented by a study that includes smoking behaviour as the dependent variable and second-hand smoke awareness as the independent variable, so that the impact of awareness on smoking likelihood can be identified.

The findings of this present study must be interpreted carefully, as the analyses have four limitations. First, although the sample size is large, the data are extracted from a cross-sectional study. Hence, causality between second-hand smoke awareness and demographic and lifestyle variables cannot be identified. In addition, all the information is self-reported, which can result in reporting error. Some details, such as marital status and smoking behaviour are likely to be falsely reported, thereby leading to inaccurate estimates. Second, the regression model used in this study is developed based on the availability of data and findings of previous studies, instead of a specific theory. Furthermore, although the regression model comprises various independent variables, some important variables may be omitted from it. These include household size, awareness of anti-smoking media and insurance ownership.

Third, using a single question to assess the respondents’ awareness of the effects of second-hand smoke may be inadequate because it cannot identify the level of awareness. To improve the assessment, the respondents should be asked, ‘Based on your knowledge, does smoking cause the following illnesses to non-smokers?’ Then, several diseases should be listed in the question, such as heart disease, lung disease and cancers, and the respondents asked to reply ‘yes’ or ‘no’ for each disease, where ‘1’ is assigned for each ‘yes’ and ‘0’ for each ‘no’. Hence, the respondents who have better second-hand smoke awareness will score a higher value for this question than respondents with poorer second-hand smoke awareness. Fourth, the lifestyle variable consists only of smoking behaviour, which seems to be insufficient. Physical activity and alcohol consumption, for instance, are lifestyle variables that might also affect second-hand smoke awareness. Hence, they should also be included as independent variables, if the data allows.

Second-hand smoke is harmful. Policy makers need to learn about which groups of the population are unlikely to be aware of the effects of second-hand smoke if the goal of reducing the prevalence of smoking-induced disease is to be achieved. The findings discussed in this study will provided the government with directions for policy implementation. The government should make a concerted effort to improve second-hand smoke awareness among older people, aged 65 years and above. Providing older people with more smoking-related information may help to improve their awareness. However, the effectiveness of this intervention should be tested in a future study before it is widely introduced. Additionally, individuals who are in the middle, fourth and highest quintiles of the wealth index should be given special attention, as they are unlikely to be aware of the effects of second-hand smoke. The present study’s findings imply that besides increasing the tax on cigarettes, policies directed towards reducing the prevalence of smoking should also increase awareness of the effects of second-hand smoke among the wealthy. Moreover, less-educated individuals are the main focus of policy makers, as they are less aware about second-hand smoke. Thus, a policy aimed at educating them about the consequences of smoking might seem promising. To ensure that less-educated individuals learn about the effects of second-hand smoke, more public health awareness campaigns should be introduced.

Footnotes

Acknowledgements

The authors would like to thank the Director General of Health, Malaysia for his permission to use the data from the Global Adult Tobacco Survey Malaysia 2011 and to publish this paper.

Funding

This work was supported by the Population Studies Unit, University of Malaya (grant number IF002-2014).

Declaration of conflicting interests

The authors declare that there is no conflict of interest.

References

Institute for Public Health. Report of the Global Adult Tobacco Survey (GATS) Malaysia 2011. Report, Ministry of Health Malaysia, Putrajaya, 2012.

US Department of Health and Human Services. The health consequences of smoking – 50 years of progress: a report of the surgeon general. Report, US Department of Health and Human Services, Atlanta, 2014.

Thun

Lally

Calle

, et al. Cigarette smoking and changes in the histopathology of lung cancer. J Natl Cancer Inst 1997; 89: 1580–1586.

US Department of Health and Human Services. Let’s make the next generation tobacco-free: Your guide to the 50th anniversary surgeon general’s report on smoking and health. Report, US Department of Health and Human Services, Atlanta, 2014.

Tan

AKG

Yen

Nayga

RM.

Role of education in cigarette smoking: an analysis of Malaysian Household Survey data. Asian Econ J 2009; 23: 1–17.

Cheah

Naidu

BM.

Exploring factors influencing smoking behaviour in Malaysia. Asian Pac J Cancer Prev 2012; 13: 1125–1130.

Tan

AKG

. Distinguishing between non-smokers, casual smokers, and compulsive smokers: evidence from Malaysia. Atlantic Econ J 2012; 40: 173–184.

Lim

Ghazali

Kee

, et al. Epidemiology of smoking among Malaysian adult males: prevalence and associated factors. BMC Public Health 2013; 13: 8.

Rutten

LJF

Augustson

Moser

, et al. Smoking knowledge and behaviour in the United States: sociodemographic, smoking status, and geographic patterns. Nicotine Tob Res 2008; 10: 1559–1570.

10.

Nourjah

Wagener

Eberhardt

, et al. Knowledge of risk factors and risk behaviours related to coronary heart disease among blue and white collar males. J Public Health Policy 1994; 15: 443–459.

11.

Siahpush

McNeill

Hammond

, et al. Socioeconomic and country variations in knowledge of health risks of tobacco smoking and toxic constituents of smoke: results from the 2002 International Tobacco Control (ITC) Four Country Survey. Tob Control 2006; 15: iii65–iii70.

12.

Finch

Novotny

, et al. Smoking knowledge, attitudes, and behaviors among rural-to-urban migrant women in Beijing, China. Asia Pac J Public Health 2010; 22: 342–353.

13.

Borland

Hill

Initial impact of the new Australian tobacco health warnings on knowledge and beliefs. Tob Control 1997; 6: 317–325.

14.

Gelaw

SM.

Socioeconomic factors associated with knowledge on tuberculosis among adults in Ethiopia. Tuberc Res Treat 2016; 2016: 6207457.

15.

Carpenter

Zoller

Balsis

, et al. Demographic and contextual factors related to knowledge about Alzheimer’s disease. Am J Alzheimers Dis Other Demen 2011; 26: 121–126.

16.

Yardley

Glover

Allen-Mersh

TG.

Demographic factors associated with knowledge of colorectal cancer symptoms in a UK population-based survey. Ann R Coll Surg Engl 2000; 82: 205–209.

17.

Angeles-Llerenas

Carbajal-Sanchez

Allen

, et al. Gender, body mass index and socio-demographic variables associated with knowledge about type 2 diabetes mellitus among 13293 Mexican students. Acta Diabetol 2005; 42: 36–45.

18.

Islam

FMA

Chakrabarti

Islam

, et al. Factors associated with awareness, attitudes and practices regarding common eye diseases in the general population in a rural district in Bangladesh: the Bangladesh Population-based Diabetes and Eye Study (BPDES). PLoS One 2015; 10: e0133043.

19.

Lee

Lim

Lee

, et al. Secondhand smoke exposures in indoor public places in seven Asian countries. Int J Hyg Envir Health 2010; 213: 348–351.

20.

Dovern

Quaas

Rickels

A comprehensive wealth index for cities in Germany. Ecol Indic 2014; 41: 79–86.

21.

Gupta

Kaur

Islam

, et al. Association of household wealth index, educational status, and social capital with hypertension awareness, treatment, and control in South Asia. Am J Hypertens 2017; 30: 373–381.

22.

Rutstein

Johnson

The DHS wealth index. DHS comparative report no. 6. ORC Macro, Calverton, MD, 2004.

23.

Wooldridge

JM.

Econometric analysis of cross section and panel data. 2nd ed. Cambridge, MA: MIT Press, 2010, p. 1064.

24.

StataCorp. Stata statistical software: Release 13.1. Texas: Stata Corporation, 2013.

25.

Cheah

Goh

KL.

Blood glucose screening among elderly Malaysians: who to target?

J Diabetes 2017; 9: 85–92.

26.

Cheah

Azahadi

Phang

, et al. Factors affecting participation decision and amount of physical activity among urban dwellers in Malaysia. Public Health 2017; 146: 84–91.

27.

Cheah

Tang

CF.

Factors influencing the use of preventive medical care in Malaysia: evidence from National Health and Morbidity Survey data. Asian Econ J 2017; 31: 119–137.

28.

Parmenter

Waller

Wardle

Demographic variation in nutrition knowledge in England. Health Educ Res 2000; 15: 163–174.

29.

Amin

Mostafa

Sarriff

Factors associated with the general public knowledge and awareness of cardiovascular diseases and its risk factors in Penang – Malaysia. IOSR J Pharm 2014; 4: 21–27.

30.

Van der Pol

. Health, education and time preference. Health Econ 2011; 20: 917–929.

31.

Khwaja

Silverman

Sloan

Time preference, time discounting, and smoking decisions. J Health Econ 2007; 5: 927–949.

32.

Ida

Goto

Simultaneous measurement of time and risk preferences: stated preference discrete choice modelling analysis depending on smoking behaviour. Int Econ Rev 2009; 50: 1169–1182.

Factors associated with second-hand smoke awareness in Malaysia

Abstract

Background:

Objectives:

Methods:

Results:

Conclusion:

Keywords

Introduction

Methods

Sample

Data collection

Ethical considerations

Measures

Statistical analysis

Results

Discussion

Footnotes

Acknowledgements

Funding

Declaration of conflicting interests

References