Factors influencing smoking cessation: Insights from Singapore’s nationwide health and lifestyle survey

ABSTRACT

Introduction: Singapore has implemented an evidence-based smoking cessation framework to support smokers in quitting. Our study investigated the prevalence and correlates of (1) quit attempts (QA) and quit intentions (QI) among current smokers, and (2) smoking cessation (SC) among ever-smokers in Singapore.

Method: Data was collected from a nationwide survey conducted between 2020 and 2022. QA was defined as attempting to stop smoking at least once in the past 12 months, while QI was defined as planning to quit smoking in the next 30 days or the next 6 months. SC referred to individuals who quit smoking over 6 months ago. Sociodemographic factors, doctor’s advice to quit and perceived harm from smoking were assessed using logistic regression among current smokers (n=1024) and ever-smokers (n=1457).

Results: Among current smokers, 31.3% and 41.2% reported QI and QA, respectively. Smokers with secondary or pre-tertiary education were less likely to report QI compared to those with a degree or higher. Doctor’s advice to quit was associated with a higher likelihood of QA. Among ever-smokers, 25.3% reported SC, and this was more likely when they perceived smoking one or more packs of cigarettes daily as posing a moderate or high health risk.

Conclusion: Educational campaigns should focus on simplifying messages for individuals with lower literacy levels. Smoking cessation training can be integrated into medical education, and graphic health warnings on cigarette packs can help effectively communicate the dangers of smoking.

CLINICAL IMPACT

What is New

• This study adds to global knowledge on smoking cessation by identifying key factors within a diverse, multi-ethnic population.
• Individuals were more likely to quit smoking if they had higher education attainment and perceived smoking as highly harmful.

Clinical Implications

• Smoking cessation campaigns should be designed to be easily understood, especially by those with lower literacy levels.
• Singapore’s current approach, including graphic health warning on cigarette packs, can serve as a model for other countries looking to boost smoking cessation rates.

The global prevalence of smoking has declined over the years. According to authors utilising data from the Global Burden of Disease, Injuries, and Risk Factors Study, from 1990 to 2020, the number of male smokers fell by 27.2%, whereas female smokers fell by 37.9%.¹ Moreover, the decline in smoking prevalence was higher in countries with a higher sociodemographic index than those with a lower sociodemographic index.¹ Despite the fall in smoking prevalence worldwide, challenges in smoking cessation remain. One prominent challenge is the emergence of the e-cigarette, the novelty of which has enticed younger individuals to start smoking.² Developing countries face additional challenges such as limited awareness and poor access to smoking cessation services.³

Similar to other countries, Singapore faces several challenges in improving its smoking cessation rates. A Singapore-based study on patients with substance use disorders identified several hurdles to cessation, including withdrawal symptoms, high cost of smoking cessation treatments, and lack of motivation.⁴ Nonetheless, smoking prevalence in Singapore has declined from 13.9% to 10.1% between 2010 to 2020.⁵ This decline can be attributed to Singapore’s multiprolonged approach to reducing smoking rates, which involves the enforcement of 2 primary legislative acts: the Tobacco (Control of Advertisement and Sale) Act and the Smoking (Prohibition in Certain Places) Act.⁶ The former act encompasses measures such as prohibiting the sale of tobacco products to individuals below 21 years old and banning advertisements of tobacco products. The latter act restricts smoking in specific public places, including public transport and hospitals. Additionally, there is the “I Quit” smoking cessation programme that tailors cessation plans for smokers according to their characteristics.⁷ These successful strategies to reduce smoking prevalence can be a blueprint for other countries with high smoking prevalence.

International studies, brought together through systematic reviews, have identified factors that influence smoking quit attempts and cessation.^8,9 They found that individuals were more likely to attempt quitting if they had made previous attempts to quit, had longer duration of abstinence and had motivation to quit.⁸ Moreover, individuals were more likely to cease smoking if they smoked more cigarettes per day and had a negative perception of smoking.⁹ However, a common limitation of these systematic reviews is methodological heterogeneity, whereby the study methods are so diverse that it is difficult to compare findings for several correlates.^8,9 With Singapore’s multicultural context and multiprolonged approach to smoking cessation, understanding the factors behind its success can provide a complementary perspective for other countries struggling with low cessation rates.

Hence, our study aimed to examine the prevalence and correlates of different stages of smoking cessation in Singapore, which include intentions and attempts to quit and smoking cessation.

METHOD

This cross-sectional study is part of a larger study that aimed to examine drug consumption in Singapore (i.e. Health and Lifestyle Survey). In addition to drug consumption, the study captured comprehensive information on participants’ sociodemographic and lifestyle habits. The methodology is explained in a previous publication.¹⁰ Briefly, the study obtained its sample (n=6509) from an administrative database and adopted a disproportionate stratified sampling design. The analysis was weighted to account for disproportionate stratified sampling, non-response and post-stratification by age, sex and ethnicity. These processes are similar to those used in other population-based studies,^11,12 ensuring that the analysis is representative of Singapore’s population.

Individuals were included if they were (1) Singapore citizens or permanent residents living in Singapore, (2) aged 15‒65 years old, and (3) literate in English, Mandarin, Malay or Tamil. The exclusion criteria were (1) inability to do the interview and (2) hospitalised or institutionalised during the study period.

The study team employed a survey company to carry out the study between April 2021 and July 2022. The study team also trained lay interviewers to conduct the interviews. Due to COVID-19 restrictions, after obtaining consent from the participant face-to-face, a QR code or link was given to the participants, followed by a unique number for them to complete the online survey.

As the study involved asking sensitive questions such as illicit drug use, steps were taken to protect the participants’ identity. No identifiers were collected. The questionnaire was self-administered using tablets, and only verbal consent was taken to avoid documentation. Consent was not documented to minimise the risk of identifying the participants and to protect them from potential legal issues due to their responses. Moreover, parental consent was waived for participants who were minors. All questions had choices such as “I prefer not to answer” and “I don’t know” so that participants could skip questions they found personal. Three months after the survey’s completion, the survey company destroyed all contact information. The study protocol adhered to the ethical standards of the responsible committee on human experimentation (institutional and national) and with the principles of the Declaration of Helsinki. The study was approved by the Domain Specific Review Board from the National Healthcare Group.¹

Smoking-related outcomes

The outcome variables were intentions to quit, attempts at quitting, and smoking cessation. Participants were asked to indicate their smoking status using one of the following options: smoker (daily or sometimes a week), former smoker, and non-smoker. Among current smokers (i.e. participants who smoked daily or a few times a week), intentions to quit were assessed by asking them if they had planned to quit in the next 30 days, next 6 months or not at all in the next 6 months. Those who had planned to quit in the next 30 days or next 6 months were considered to have intentions to quit. Attempts at quitting were evaluated among the current smokers using the question: “How many times have you tried to stop smoking in the past 12 months?” Those who had tried to stop smoking at least once in the past 12 months were classified as having attempted to quit. In examining smoking cessation, the study used the term “ever-smoker” to denote participants who answered either “smoker (daily or sometimes a week)” or “former smoker”. Smoking cessation was assessed by asking former smokers whether they “quit more than 6 months ago” or “quit within the last 6 months”. Those who had quit more than 6 months ago were considered to have successfully achieved smoking cessation.

Independent variables

Our study examined the association between the outcome variables with the following variables: age, sex, ethnicity, education, marital status, having children, employment, personal income, number of chronic conditions, age of initiation of smoking, perceived risk of harm for occasional smoking and smoking 1 or more packs daily. The perceived risks of harm associated with smoking were self-reported using a 4-point Likert scale (i.e. no risk, little risk, moderate risk and great risk of harm). In this analysis, perceived risk of harm was dichotomised into no/little risk of harm and moderate/great risk of harm. For intentions and attempts to quit, additional variables were examined. These variables were doctor’s advice to quit smoking and nicotine dependence.

Chronic conditions were self-reported by individuals using a modified checklist of 18 chronic conditions. These conditions were asthma, arthritis, back problems, cancer, chronic inflammatory bowel disease, chronic lung diseases, congestive heart failure, diabetes, heart disease, hyperlipidaemia, hypertension, kidney failure, migraine, neurological conditions including Parkinson’s disease, stomach ulcer, stroke and thyroid disease. Our study classified the responses into 3 groups: no chronic condition, 1 chronic condition, and 2 or more chronic conditions.

Nicotine dependence was assessed using the Fagerstrom test for nicotine dependence.¹³ Individuals were classified as having very low nicotine dependence if the score was 0‒2, low nicotine dependence if the score was 3‒4, and moderate-to-high nicotine dependence if the score was 5 and above.¹³

Statistical analysis

Weighted percentages and unweighted counts were presented for all categorical variables. Multiple logistic regression was conducted to determine the correlates of each smoking-related outcome. For each outcome, the following steps were performed: (1) a univariate logistic regression model was generated to identify potential correlates, and (2) variables with P value <0.05 were included in the initial multivariable logistic regression using the enter method. Subsequently, variables with P value >0.05 were removed and another multivariable logistic regression analysis was performed to determine the final model.

The results of the models were presented in odds ratio (OR) and 95% confidence interval (CI). Standard error was estimated using Taylor linearisation. The analysis was generated using IBM SPSS version 23 (IBM Corp, Armonk, US) and Stata/MP 18.0 (StataCorp, College Station, US), with 2-tailed tests at a 5% significance level.

RESULTS

The analysis included 1457 ever-smokers, of which 1024 were current smokers. Among the 1024 current smokers, 1021 answered the question on intention to quit and 881 answered the question on attempts to quit.

Table 1. Characteristics of current smokers and ever-smokers.

Missing data: education (current smoker [n=12], ever-smoker [n=14]); marital status (current smoker [n=4], ever-smoker [n=7]); employment (current smoker [n=17], ever-smoker [n=23]); personal income (current smoker [n=163], ever-smoker [n=230]); number of chronic conditions (current smoker [n=12], ever-smoker [n=16]); age of initiation to smoking (current smoker [n=200], ever-smoker [n=280]); received advice from doctor to stop smoking (current smoker [n=3]); nicotine dependence (current smoker [n=3]); perceived risk of harm (smoke occasionally) (current smoker [n=169], ever-smoker [n=222]); perceived risk of harm (smoke 1 of more packs per day) (current smoker [n=164], ever-smoker [n=211])
^a Pre-tertiary education includes vocational institute/institute of technical education/pre-university/junior college/diploma/International Baccalaureate

Most participants in both groups were male (current smokers: 82.99%, ever-smoker: 80.54%), of Chinese ethnicity (current smoker: 63.48%, ever-smoker: 63.15%), married (current smoker: 59.24%, ever-smoker: 59.71%), had no children (current smoker: 65.94%, ever-smoker: 64.97%), and were either employed or self-employed (current smoker: 86.82%, ever-smoker: 86.26%) (Table 1). Regarding smoking-related variables, most participants in both groups started smoking at 18 years of age or later (current smoker: 51.55%, ever-smoker: 52.53%), and received advice from a doctor to quit smoking (current smoker: 84.55%, ever-smoker: 87.66%) (Table 1). Additionally, most participants in both groups believed that smoking occasionally or smoking at least 1 pack per day presented a moderate to great risk of harm (Table 1).

Fig. 1. Proportion of current smokers that had intentions or attempts to quit, and proportion of ever-smokers that achieved smoking cessation.

Among current smokers, the prevalence of intentions and attempts to quit were 31.27% (n=385) and 41.21% (n=427), respectively (Fig. 1). Among ever-smokers, the prevalence of smoking cessation was 25.25% (n=345).

Table 2. Logistic regression with intentions to quit as outcome.

Table 2 presents the logistic regression models for intentions to quit. In the multivariable model, current smokers were more likely to contemplate quitting if they were of Malay (versus [vs] Chinese, OR 1.62, 95% CI 1.004–2.61) or Indian ethnicity (vs Chinese, OR 2.59, 95% CI 1.61–4.18), and had attempted to quit previously (OR 8.15, 95% CI 5.02–13.22). Current smokers were less likely to have intentions to quit if they had secondary school education (vs degree and above, OR 0.32, 95% CI 0.14–0.69) or pre-tertiary education (vs degree and above, OR 0.37, 95% CI 0.17–0.80), separated/widowed/divorced (vs married, OR 0.34, 95% CI 0.17–0.69), and had low (vs very low, OR 0.53, 95% CI 0.31–0.89) or moderate-to-high nicotine dependence (vs very low, OR 0.53, 95% CI 0.29–0.97).

Table 3. Logistic regression with attempts to quit as outcome.

For correlates for attempts to quit, current smokers were more likely to attempt quitting if they were of Malay (vs Chinese, OR 2.21, 95% CI 1.49–3.29) or Indian ethnicity (vs Chinese, OR 2.18, 95% CI 1.40–3.39), had received doctor’s advice to quit smoking (OR 2.36, 95% CI 1.29–4.32), and perceived occasional smoking to have moderate or great risk of harm (vs no/little harm, OR 1.91, 95% CI 1.24–2.94) (Table 3). They were less likely to attempt quitting if they had moderate-to-high nicotine dependence (vs very low dependence, OR 0.49, 95% CI 0.28–0.85).

Table 4. Logistic regression with smoking cessation as outcome.

In Table 4, ever-smokers were more likely to cease smoking if they had 2 or more chronic conditions (vs no chronic condition, OR 2.14, 95% CI 1.35–3.38), and perceived smoking 1 or more packs daily to have moderate or great risk of harm (vs no/little harm, OR 4.52, 95% CI 1.89–10.80). They were less likely to quit smoking if they were male (vs female, OR 0.64, 95% CI 0.43–0.94), and had primary school education and below (vs degree and above, OR 0.19, 95% CI 0.10–0.38), secondary school (vs degree and above, OR 0.39, 95% CI 0.23–0.65) or pre-tertiary education (vs degree and above, OR 0.54, 95% CI 0.34–0.84).

DISCUSSION

Our study examined the prevalence of intentions to quit, attempts at quitting and smoking cessation in Singapore. Our study has a higher proportion of current smokers who intended to quit (31.27%) and attempted to quit (41.21%), and the proportion of those who ceased smoking successfully lagged behind (25.25%). Consistent with this finding, a study from the US reported a higher proportion of intention to quit (77.1%) than smoking cessation (7.5%).¹⁴ A similar trend was also found in a Malaysian study, which found that while 49.1% of smokers attempted quitting, the success rate for smoking cessation was lower at 31.4%.¹⁵ These studies imply that globally, smokers face challenges in translating intention and attempts to quit into successful cessation.

A prominent finding in our study was that current smokers who were advised by a doctor to quit smoking were more likely to attempt quitting. Similarly, in a systematic review that examined 42 trials worldwide, brief advice from a physician increased the chance of smoking cessation.¹⁶ These findings suggest that integrating cessation advice into primary healthcare settings can be an important intervention for smoking cessation worldwide. Currently, in Singapore, a framework for addressing tobacco dependence exists within the primary healthcare setting.¹⁷ Moreover, smoking cessation clinics are available in polyclinics and restructured hospitals.¹⁷ Other countries can modify these measures by providing appropriate training to clinicians on offering advice and support based on their cultural context.

Our study also found that individuals with a higher perceived risk of harm from 1 or more packs of cigarettes were more likely to quit smoking successfully. Moreover, those with a higher perceived risk of harm from occasional smoking were more likely to attempt quitting. These findings align with studies in other countries, demonstrating that awareness of the harmful effects of smoking is associated with smoking cessation.^18,19 In a study among Korean adolescents, graphic health warnings increased the odds of attempts to quit by 1.72 times for boys and by 1.74 times for girls.¹⁸ An Australian market research reported that among ever-smokers, 61% of those who quit successfully said that graphic health warnings made them more concerned about their smoking habits, leading them to quit.¹⁹

Our study found that those with nicotine dependence were less likely to have the intention to quit or to attempt to quit smoking. This finding is not surprising, as they likely experienced stronger addiction or more severe withdrawal symptoms. This finding highlights the need for increased support, such as nicotine replacement therapy and counselling, for these individuals. Additionally, future research could explore interventions and communication strategies that might encourage them to develop the intention to quit or take steps towards quitting smoking.

For sociodemographic correlates, a noteworthy finding is that lower educational qualification is significantly associated with a lower likelihood of intention to quit and smoking cessation. These findings corroborate with findings from other population-based studies. A longitudinal Finland-based study found that higher educational levels were associated with higher odds of smoking cessation.²⁰ Similarly, a study that utilised data from Australia, Canada, the United Kingdom and the United States reported that smokers with lower education had 1.4 times higher odds of not intending to quit than those with higher education.²¹

The aforementioned findings can have broader significance to the development of global public health strategies. In Singapore, the perceived risk of harm from cigarette smoking is heightened by putting graphic health warnings on cigarette packs.²² Moreover, educational programmes were implemented to target students from a young age, raising awareness of the harmful effects of smoking.²³ These approaches can be valuable for other countries seeking to increase smoking cessation prevalence. However, successful adaption may require tailoring their approaches based on their population’s education level and awareness level.

One strength of our study is that the use of a structured questionnaire ensured that the methodology could be replicated in other countries. Although the specific questions related to smoking should be tailored according to the cultural context of the country, the overall structure of the questionnaire can be universally applied. By having a standardised questionnaire, the results from different countries can be compared. Another strength is that because the data is from a population-based study, the results can be generalised to the Singapore population. However, caution should be exercised when generalising to other countries. Although our results aligned with some studies from other countries, policies should be tailored according to the needs of the population.

A limitation of our study is that the study was done during the COVID-19 period. Studies have shown that COVID-19 can act as a hindrance and a catalyst for smoking cessation, due to reasons such as boredom and fear.^24,25 It is unclear whether the effect of COVID-19 on smoking cessation will be sustained over time. Another limitation of our quantitative study is that we are unable to examine the impact of smoking cessation interventions, as well as the reasons that people continue or quit smoking. Future studies can build on this study by examining the efficacy of different smoking cessation messages across countries, as well as the effect of clinician training on smoking cessation behaviours. Moreover, qualitative studies can explore the barriers and facilitators of smoking cessation.

CONCLUSION
In conclusion, our study contributes to the global knowledge of smoking cessation by identifying sociodemographics, perceived risk of harm from cigarette smoking, nicotine dependence, and clinicians’ advice associated with smoking cessation behaviour in a multi-ethnic population. As lower education is associated with poorer quit intention and smoking cessation, educational campaigns should be tailored for easy understanding by individuals with lower literacy levels. Moreover, smoking cessation training can be integrated into medical education. Singapore’s approach to smoking cessation behaviours, such as graphic health warnings on cigarette packs and smoking cessation clinics, can serve as a model for other countries aiming to improve smoking cessation behaviours.

This article was first published online on 15 October 2024 at annals.edu.sg.

REFERENCES

1. Dai X, Gakidou E, Lopez AD. Evolution of the global smoking epidemic over the past half century: strengthening the evidence base for policy action. Tob Control 2022;31:129-37.
2. Kim JY. Directions and Challenges in Smoking Cessation Treatment. Tuberc Respir Dis 2020;83:S1-5.
3. Al-Worafi YM. Smoking Cessation in Developing Countries: Challenges and Recommendations. In: Al-Worafi YM (Ed.), Handbook of Medical and Health Sciences in Developing Countries [Internet] (Cham: Springer International Publishing; 2024).
4. Mareen ZSH, Harjit Singh KK, Ng CWL, et al. Challenges to smoking cessation in patients with substance use disorders. Ann Acad Med Singap 2024;53:124-6.
5. Ministry of Health Singapore. Smoking Prevalence in Singapore Population dropped from 13.9% in 2010 to 10.1% in 2020. https://www.moh.gov.sg/news-highlights/details/smoking-prevalence-in-s’pore-population-dropped-from-13.9-in-2010-to-10.1-in-2020. Accessed 7 October 2024.
6. sg. 2013. Smoking Control Programmes. https://www.healthhub.sg/live-healthy/tobacco%20and%20tax. Accessed 7 October 2024.
7. sg. 2018. IQuit. https://www.healthhub.sg/programmes/iquit#Coping. Accessed 7 October 2024.
8. Vangeli E, Stapleton J, Smit ES, et al. Predictors of attempts to stop smoking and their success in adult general population samples: a systematic review. Addiction 2011;106:2110-21.
9. Cengelli S, O’Loughlin J, Lauzon B, et al. A systematic review of longitudinal population-based studies on the predictors of smoking cessation in adolescent and young adult smokers. Tob Control 2012;21:355-62.
10. Subramaniam M, Koh YS, Sambasivam R, et al. Prevalence of consumption of illicit drugs and associated factors from a nationwide epidemiological survey: The Singapore Health and Lifestyle Survey. Ann Acad Med Singap 2024;53:222-32.
11. Subramaniam M, Abdin E, Vaingankar JA, et al. Tracking the mental health of a nation: prevalence and correlates of mental disorders in the second Singapore mental health study. Epidemiol Psychiatr Sci 2020;29:e29.
12. AshaRani P, Abdin E, Kumarasan R, et al. Study protocol for a nationwide Knowledge, Attitudes and Practices (KAP) survey on diabetes in Singapore’s general population. BMJ Open 2020;10:e037125.
13. Heatherton TF, Kozlowski LT, Frecker RC, et al. The Fagerström Test for Nicotine Dependence: a revision of the Fagerstrom Tolerance Questionnaire. British Journal of Addiction 1991;86:1119-27.
14. Leventhal AM, Dai H, Higgins ST. Smoking Cessation Prevalence and Inequalities in the United States: 2014-2019. J Natl Cancer Inst 2022;114:381-90.
15. Lim KH, Heng PP, Nik Mohamed MH, et al. Prevalence and Factors Associated With Attempts to Quit and Smoking Cessation in Malaysia. Asia Pac J Public Health 2019;31:22S-31S.
16. Stead LF, Buitrago D, Preciado N, et al. Physician advice for smoking cessation. Cochrane Tobacco Addiction Group, editor. Cochrane Database of Systematic Reviews [Internet]. 31 May 2013. http://doi.wiley.com/10.1002/14651858.CD000165.pub4. Accessed 24 July 2024.
17. Lua YHA, How CH, Ng CWM. Smoking cessation in primary care. Singapore Med J 2024;65:38-44.
18. Joo HJ, Joo JH, Kim SH, et al. Association Between Graphic Health Warning Labels on Cigarette Packs and Smoking Cessation Attempts in Korean Adolescent Smokers: A Cross-Sectional Study. Front Public Health 2022;10:789707.
19. Australian Government Department of Health and Aged Care. Evaluation of effectiveness of graphic health warnings on tobacco product packaging [Internet]. August 2018. https://www.health.gov.au/resources/publications/evaluation-of-effectiveness-of-graphic-health-warnings-on-tobacco-product-packaging. Accessed 24 July 2024.
20. Ruokolainen O, Härkänen T, Lahti J, et al. Association between educational level and smoking cessation in an 11-year follow-up study of a national health survey. Scand J Public Health 2021;49:951-60.
21. Siahpush M, McNeill A, Borland R, et al. Socioeconomic variations in nicotine dependence, self-efficacy, and intention to quit across four countries: findings from the International Tobacco Control (ITC) Four Country Survey. Tob Control 2006;15:iii71-5.
22. Ministry of Health Singapore. Standardised packaging & enlarged graphic health warnings mandatory for all tobacco products in Singapore from 1 July 2020 [Internet], 29 June 2020. https://www.moh.gov.sg/news-highlights/details/standardised-packaging-enlarged-graphic-health-warnings-mandatory-for-all-tobacco-products-in-singapore-from-1-july-2020. Accessed 24 July 2024.
23. Ministry of Education Singapore. Educating Students on the Harmful Effects of Smoking, 21 January 2021. https://www.moe.gov.sg/news/forum-letter-replies/20210121-educating-students-on-the-harmful-effects-of-smoking. Accessed 24 July 2024.
24. Johnston E, Bains M, Hunter A, et al. The Impact of the COVID-19 Pandemic on Smoking, Vaping, and Smoking Cessation Services in the United Kingdom: A Qualitative Study. Nicotine Tob Res 2023;25:339-44.
25. Guignard R, Quatremère G, Pasquereau A, et al. Barriers Against and Motivations for Quitting Smoking during the COVID-19 Health Crisis: Results of a Qualitative Study in France. Int J Environ Res Public Health 2022;19:13051.