Study des ign and participants

Participants were adult men aged ≥18 years, chosen because of the high prevalence of e-cigarette use in Malaysia^10,22. This exploratory online survey approached Malaysian EC users from the Malaysian Organization of Vape Entity (MOVE), a nationally representative entity that comprised 13485 registered EC users at the time of the study. EC users were approached through an official online Facebook social media page owned by MOVE. With permissions from the president and the administrator of the MOVE Facebook page, an announcement was created and hosted at the Facebook interface between January and April 2018. EC users who wished to participate in the study were required to fill in basic demographic details on a Google-form that consisted of name, age, gender, e-mail address and telephone contact number. A total of 4581 EC users volunteered to participate.

We estimated sample size with Stata 13.0 using the formula for one proportion, with the following parameters: P=1.4% which was the prevalence of EC use among the adult population in Kuantan²³; α=0.05; and Power=90%. To determine a difference of 1.8% in the prevalence of EC use in our study population, we estimated that a sample size n=642 was required. We accounted for a 10% non-response rate and obtained a total sample size of 707.

A sampling frame that consisted of a list of 4581 volunteered EC users was used to randomly select the required number of participants in our study. Of these, 931 participants were excluded as there was no e-mail address or telephone number provided on the basic demographic information form. From the remaining 3650 participants, selection was done through a systematic sampling method, with an interval of 5 participants, until the sample size of 707 was reached. In April 2018, an e-mail was sent to these participants that comprised consent forms, study information and a survey link. Participants were required to reply to the e-mail with the completed consent form within two weeks. Responses of the survey were verified when consent forms were received. Ethical approval was obtained from the Joint Penang Independent Ethics Committee (approval number: JPEC 17-0035).

The survey was answered using a Google-form link. To prevent redundancy of participation, respondents personal identity number and demographic information were checked and collected based on time-stamped validation by the site administrator. If this was unclear, response veracity was confirmed through a telephone call to the participant contact number. Ten responses were rejected as we were unable to ensure response veracity and no consent forms were returned. Three female participants were excluded. In total, we received a verified nationwide response from 694 e-cigarette users (response rate 98.1%) across all 13 states and 1 Federal Territory in Malaysia: Johor (81), Kedah (23), Kelantan (12), Federal Territory of Kuala Lumpur (120), Melaka (27), Negeri Sembilan (22), Pahang (31), Perak (33), Perlis (5), Pulau Pinang (28), Selangor (231), Terengganu (19), Sabah (23), and Sarawak (39).

Baseline definitions

Long-term EC usage had a threshold of ≥2 years, in accordance with recent prospective studies^15,16. A current smoker was defined as an adult who smoked 100 cigarettes in his lifetime and who currently smokes conventional cigarettes²⁴. Former smoker was defined as an adult who had smoked at least 100 cigarettes in his lifetime but who quit smoking at the time of the study²⁴.

Study instrument

We developed a structured, closed-ended, self-administered online questionnaire in local language based on current literature. The questionnaire was pilot tested among 10 EC users who were not included in the analysis. Detailed wording was moderated using feedback from these knowledgeable EC users. The final questionnaire consisted of 2 sections with 19 items. The first part assessed EC user profiles that included 9 items on sociodemographic characteristics: age, dichotomized as ≤30 years and >30 years; ethnicity categorised as Malays or non-Malays; education level categorised as tertiary education, or secondary education or less; employment status categorised as employed or unemployed; marital status categorised as married or single; monthly household income dichotomized as ≤4000 or >4000 MYR (1000 Malaysian Ringgit about 240 US$); smoking habits, current or former smoker with response options ‘yes’ or ‘no’; and long-term EC usage, with response options ‘yes’ or ‘no’, according to predefined criteria. The second part included 10 items adapted from literature that aimed to evaluate EC user perceptions^25–27. Items included: getting a definite nicotine hit from e-cigarettes; e-cigarettes are as satisfying as tobacco; e-cigarettes look and feel like a cigarette; e-cigarettes are healthier; e-cigarettes help to cut down smoking; e-cigarettes remove the urge to smoke; the crave for e-cigarettes is as much as for tobacco; use of e-cigarettes in places where smoking is banned; e-cigarettes help to stop smoking; and e-cigarettes allow the use of more nicotine. These items had response options ‘yes’ or ‘no.’

Data analyses

Data collected were analysed using SPSS version 23.0. Normality check was conducted using statistical and graphical methods for numerical variables, and all numerical data were found to be normally distributed. Descriptive statistics were conducted for all variables. Chi-squared test was used to assess the association between long-term EC usage and categorical variables in this study. Significant variables in the chi-squared test were further examined using binary logistic regression analysis. Odds ratios (ORs) and 95% confidence intervals (CIs) were used to determine the strength of associations between long-term e-cigarette usage and independent variables. We fitted multivariate logistic regression models to examine these associations further. All significant variables at the bivariate level were entered into the multivariate logistic regression analyses to determine the predictors of long-term e-cigarette usage. Adjusted ORs (AORs) and 95% CIs are reported. Statistical significance was set at p<0.05. Multi-collinearity between independent variables was checked.

E-cigarette user profiles

Table 1 shows participant EC user profiles. The mean (±SD) age of EC users was 33 (±6.8) years, and the majority were aged >30 years (n=438; 63.1%). Most respondents were Malays (n=608; 87.6%), married (n=473; 68.2%), tertiary educated (n=493; 71%) and employed (n=646; 93.1%) with a monthly household income >4000 MYR (n=301; 43.4%). The mean (±SD) monthly household income was 5850 (±1013) MYR with an income range 100–15000 MYR. Almost all were ever smokers (n=690; 99.4%) while current and former smokers were 10.1% (n=70) and 84.1% (n=584), respectively. The prevalence of long-term EC usage was 82.3%.

E-cigarette user perceptions

Most users reported getting a definite nicotine hit from ECs (n=505; 72.8%). The majority perceived ECs to be as satisfying as tobacco (n=413; 59.5%) and healthier (n=650; 93.7%). Most users believed that ECs have helped them to cut down on tobacco smoking (n=658; 94.8%) and reduce the urge to smoke (n=613; 88.3%). Most respondents believed that ECs have helped them to stop smoking (n=605; 87.2%) (Table 2).

Association between long-term e-cigarette usage and user profiles

Table 3 shows the associations between long-term EC usage and user profiles. The odds of being long-term EC users was 2.5 times higher among those aged >30 years (OR=2.5; 95% CI: 1.7–3.3; p<0.001) compared with those aged ≤30 years, and this association was statistically significant. The odds of being long-term EC users was 2.0 times higher among those employed (OR=2.0; 95% CI: 1.1–3.9; p=0.034) compared with those unemployed, and this association was statistically significant. Similarly, the odds of being long-term EC users was 1.7 times higher among those who were married (OR=1.7; 95% CI: 1.1–2.5; p=0.007) compared with singles, and this association was statistically significant. The odds of being long-term EC users was 0.5 times lower among current smokers (OR=0.5; 95% CI: 0.3–0.8; p=0.005) compared with non-current smokers, and this association was statistically significant. In contrast, the odds of being long-term EC users was 2.4 times higher among former smokers (OR=2.4; 95% CI: 1.5–3.7; p<0.001).

Association between long-term e-cigarette usage and user perceptions

Table 4 shows the associations between long-term EC usage and user perceptions. The odds of being long-term EC users were higher among those who perceived to get a definite nicotine hit from ECs (OR=1.7; 95% CI: 1.1–2.5; p=0.020), felt that ECs were healthier than smoking (OR=2.0; 95% CI: 1.1–3.9; p=0.037), believed that ECs have helped them to cut down tobacco smoking (OR=2.5; 95% CI: 1.2–5.1; p=0.014), perceived that they did not have the urge to smoke as much since using ECs (OR=2.7; 95% CI: 1.6–4.4; p<0.001), and those who perceived that ECs have helped them to stop smoking (OR=3.1; 95% CI: 1.9–5.0; p<0.001) (Table 4). These associations were statistically significant.

Predictors of long-term e-cigarette usage

Multiple logistic regression analysis yielded two significant predictors of long-term EC usage. The most significant predictor of long-term EC usage in the model was the perception that ECs helped them to stop smoking (AOR=2.5; 95% CI: 1.6–4.4; p<0.001). This was followed by users aged >30 years who had about twice higher odds of being long-term EC users (AOR=2.4; 95% CI: 1.6–3.6; p<0.001). The final model, being the most parsimonious model was statistically significant (χ²=41.26; df=3; p<0.001) and predicted 82.3% of long-term EC usage correctly. There was no multi-collinearity between independent variables (Table 5).

Study population

The profile of EC users in this study was consistent with international and local trends, albeit with some notable differences. Vaping has become progressively popular among adolescents and young adults, worldwide, and the duration of EC use is believed to vary depending on whether usage is for experimental (e.g. curiosity) or goal-oriented (quit-smoking) reasons^18,28,29. The average age of EC users in the current study (33 years) was considerably higher than that reported in previous studies in Malaysia^19,23. This is probably due to differences in the design of the current and previous studies. While the current study targeted EC users across Malaysia, the earlier studies were state-focused and estimated prevalence of EC use among the general public. It is also possible that the self-selecting nature of participants in our sample reflects a higher participation by those who strongly identify as current regular users of e-cigarettes. To this extent, it seems reasonable to assume that this type of goal-oriented usage is concentrated in the older population, and this might have influenced the direction of our study findings.

Consistent with a previous report by Wong et al.¹⁹, EC users in the current study were predominantly Malays, married, and attained higher than secondary education. Compared to the earlier studies, a higher proportion of EC users in the current study were employed and earned a higher monthly household income.

The representation of participants by state in this study approximates the national population distribution, albeit with slight differences, with about one-third of respondents in the current study being from Selangor (including Putrajaya), which is the most populous and economically vibrant state in Malaysia as at 2019³⁰. Consistent with the national population distribution, respondents from Perlis were the least represented in our study.

E-cigarette user perceptions

Findings from the current study were in accordance with earlier reports regarding perceived benefits of EC use³¹. There is rich literature evidence supporting the use of ECs as a smoking cessation aid³², although the extent to which it helps smokers to completely quit smoking is still a subject of debate³³. Consistent with earlier studies, participants in the current study believed that ECs have helped them to cut down on tobacco smoking^34,35, reduced the urge to smoke^36–38 and ultimately helped them to stop smoking. While its benefit as a smoking cessation aid remains the most common reason for long-term EC usage among our study participants and vapers around the world, there are arguments regarding its role or efficacy in nicotine cessation^39,40, which, according to a recent systematic review, cast doubt on its overall efficacy in smoking cessation⁴¹. Nevertheless, the potential benefits that ECs offer over traditional cigarettes in preserving lung and other metabolic functions and reducing indoor and outdoor air pollution remain of interest to many stakeholders including public health practitioners and policy makers.

Predictors of long-term e-cigarette usage

We found in this study that those aged >30 years and the perception among vapers that ECs have helped them to quit smoking were significant predictors of long-term EC usage. This is perhaps not surprising, particularly amongst the predominantly goal-oriented EC users in our study cohort. Previous studies have shown that although nicotine addiction remains a key driver of EC use, other factors such as being more economical and environment-friendly, public policy acceptance, and being less harmful to health, were considered strong reasons for continuous and long-term EC usage among vapers in Malaysia and other parts of the world^6,19,42,43. Although we did not expressly collect data on the age at onset of EC use among participants in this study, the fact that most respondents engage in its use as a smoking cessation aid extends the argument that they possibly have been using it over years and may continue to do so for a long time. Our data on proportion of current smokers (10%) relative to former smokers (84%) among EC users in the current study further suggests that not only they may have been engaging in EC use for smoking cessation in the short-term, but also there is the possibility that they would continuously engage in its use over the long-term to avoid relapse and to attenuate cravings. This finding is consistent with that of a previous study that suggested that long-term EC usage was associated with a higher rate of smoking cessation¹⁵. Also, Zhuang et al.¹⁵ reported a higher likelihood of current EC users becoming long-term EC users, and that higher smoking cessation rates are associated with intensity of EC use (daily use)¹⁵. Combined, the findings of the current study that older, goal-oriented EC users were more likely to use ECs in the long-term have a sound basis in literature since the pharmacotherapeutic effect of smoking cessation aids in the real-world setting can only be detected if they are used for a sufficient long period of time^15,44.

Strengths and limitations

One of the major strengths of this study is the recruitment and participation of a nationwide sample of EC users across all states in Malaysia, improving the internal and external validity of the findings as well as its comparability with other similar studies conducted internationally. To the best of our knowledge, this is the first nationwide survey that explored the factors associated with long-term EC usage among vapers in Malaysia, and this provides a platform upon which future strategic interventions including public policies could built on.

The participants in this survey required access to online computer technology and thus our recruitment path may have been weighted not only towards those in contact with active long-term EC use peers, but also those of higher socioeconomic status. Participants were recruited from users registered in an association of EC users, and thus they may be expected to give ‘positive’ feedbacks about EC use as a smoking cessation aid, resulting in social desirability bias. According to the Malaysian Poisons Act 1952, the sale of ECs with nicotine content is illegal, unless prescribed by licensed health professionals⁴⁵. The lack of comparison perceptions among similar smokers who are not currently using ECs is another weakness and limits the full picture of the role that ECs may have in the lives of Malaysian smokers. From surveys elsewhere, one would assume that the perceptions of those smokers not currently using ECs are likely to be substantially different from those who are current users. They are more likely to believe that they are harmful and that they are not a useful substitute for tobacco smoking. This has further implications for harm reduction policy⁴⁶. Although user perceptions were well demonstrated in the current study, this online survey could not establish veracity of smoking abstinence through biochemical assessment. Thus, future studies should be directed towards verifying smoking abstinence through biochemical assessments in long-term EC users.