Smoking has long been identified as a major global public health issue. It is the leading cause of preventable death worldwide and kills about 6 million people each year.1 Although Hong Kong has the lowest smoking prevalence among the major cities of China, at 11.1% as reported in 2010, it still accounts for about 5700 deaths annually, approximately one fifth of all deaths per year. In 1998, there were 1324 passive smoking–related deaths reported.2 3

According to the evidence-based MPOWER measures introduced by the World Health Organization4 to reduce the demand for tobacco, to provide smoking cessation services and cessation support in the public health care system, governments around the world have put more emphasis on smoking cessation programmes to reduce the tobacco-related health risks.5 On 1 January 2007, the Hong Kong Special Administrative Region (SAR) Government enacted the Smoking (Public Health) Ordinance and on 25 February 2009, tobacco tax was increased by 50%. In 2009, the Tung Wah Group of Hospitals (TWGHs) was commissioned by the Hong Kong SAR Government to provide a community-based smoking cessation service in Hong Kong.

The Integrated Centre on Smoking Cessation (ICSC) of the TWGHs was set up in different districts of Hong Kong, namely Shatin, Kwun Tong, Sheung Shui, Tuen Mun, Mongkok, Wanchai, Cheung Sha Wan, and Tsuen Wan to provide a free smoking cessation service to Hong Kong citizens. An integrated model of counselling and pharmacotherapy was adopted.6 7

Identification of predictors and determinants of success in smoking cessation is crucial for smoking cessation service.8 Over the last decade, health care professionals have endeavoured to identify the predictors and characteristics of successful quitters.9 Overseas studies have identified the following: old age, high socio-economic status,10 11 12 male gender, younger age at smoking initiation, previous quit attempts, being married, fewer depressive symptoms, fewer anxiety symptoms, lower prior tobacco consumption, lower score of Fagerström Test for Nicotine Dependence (FTND), no cohabitating smoker, and high cessation-related motivation/confidence.8 10 11 12 13 14 Nonetheless, many studies have shown that these predictors are not always consistent.11 15 This may be due to different methodologies and environments in different studies. Some studies were population surveys based on individual recall and did not include interventions. In this study, we analysed all potential variables and interventions. With a more comprehensive list of variables, we hoped to identify some robust independent predictors of successful quitters.

Clients who attended an ICSC in different districts in Hong Kong from 1 January 2010 to 31 December 2011 were recruited via smoking cessation hotlines, referral from health care professionals, or self-referral.

All clients received counselling, and pharmacotherapy was prescribed if the client agreed. An average of four face-to-face counselling sessions were conducted over the first 8-week intensive treatment phase by registered social workers who were all trained in tobacco cessation counselling. Phone follow-up and counselling were also offered during this treatment phase and between 9 and 12 weeks. The stage of change theory and motivational interviewing techniques were adopted.16 17 18 Clients were followed up by telephone at week 26 and week 52 to ascertain abstinence from smoking. The medications provided by ICSC included nicotine replacement therapy (NRT) and non-NRT. The former included nicotine patches, gum, lozenges, and inhalers. Oral medications included varenicline and bupropion. Medications were prescribed according to the clients’ personal preference and clinical conditions following a thorough explanation by counsellors or medical officers. For example, NRT gum would not be given to a client with dentures and a patch would not be given to a client with skin allergy.

This was a historical cohort study. All cases commenced treatment between 1 January 2010 and 31 December 2011. The inclusion criteria of the study were adults aged 18 years or above. Clients who were mentally unstable or cognitively impaired were excluded.

A structured questionnaire was used to collect the following information: (i) socio-demographic variables: age, gender, marital status, education, monthly household income, number of people living together; (ii) health-related variables: perceived health, cessation advice by nurse, cessation advice by doctor, cessation advice by any medical professional, severe/chronic illness, mental illness; (iii) smoking-related variables: age started smoking, years of smoking, cohabitation with another smoker, number of cigarettes smoked per day, FTND score,19 previous quit attempt, number of previous quit attempts, time of last attempt, reason to quit, high-risk situation; (iv) psychosocial variables: self-perceived stress, self-perceived depression, perceived importance, difficulty and confidence in quitting (from a scale of 0-100), perceived source of social support, Smoking Self-Efficacy Questionnaire (SEQ-12)20 21; and (v) intervention variables. Consent was obtained and confidentiality was assured. The questionnaire was self-administered and illiterate clients were given help as appropriate. Completed forms were validated by the counsellors.

The outcome measure was self-reported 7-day point prevalence abstinence rate at week 26. Clients who were not able to be followed up or with an absent response for smoking status were considered to have not quitted.

Data management and analysis was performed using the Statistical Package for the Social Sciences (Windows version 22.0; SPSS Inc, Chicago [IL], US). Univariate logistic regression was used for all studied predictors. All predictors with a reported P value of <0.10 were then included in multiple logistic regression analysis. Backward elimination was used in the multivariate analysis to identify independent predictors of abstinence as well as to calculate the adjusted odds ratio (AOR) and 95% confidence interval. All statistical analyses were two-tailed tests and a P value of <0.05 was considered statistically significant.

A total of 4045 clients who attended the ICSC during 1 January 2010 to 31 December 2011 were reviewed and 3853 cases who met the inclusion criteria were analysed. The gender ratio of male-to-female was approximately 7:3. Their age ranged from 18 to 89 years with a mean of 42 years; mean duration of smoking of this cohort was 20 years, and mean cigarette consumption was 18 cigarettes per day (Table 1).

The abstinence rate at week 26 was 35.1% (1353/3853). Univariate analysis of basic demographic data revealed that successful quitting was related to older age, being married, and higher household income (Table 2). Mental illness was significantly related to failure to quit but chronic illness was not, for examples, hypertension, diabetes, and chronic obstructive pulmonary disease.

Analysis of smoking-related variables showed that successful quitting was related to longer years of smoking, not cohabiting with a smoker, lower daily cigarette consumption, and lower FTND score. Successful quitters were more likely to report “prove my ability to quit smoking” and “avoid discrimination as a smoker”. A higher number of high-risk situations in quitting were negatively related to quit rate. Significant individual high-risk situations included “under time pressure”, “arguing with others”, “depressed or frustrated”, “drinking alcohol or coffee”, “difficulty in sleeping”, and “bored” (Table 3).

The following psychosocial variables were correlated to quitting: not feeling stressed, not feeling depressed, high perceived importance of quitting, low perceived difficulty in quitting, high confidence in quitting, perceived support from spouse, and high SEQ-12 score (Table 4). All interventional variables were significant predictors of smoking abstinence: number of face-to-face counselling sessions, over-the-phone counselling, and use of medication.

All items reported P<0.10 in the univariate logistic regression analysis were included in the multiple logistic model with backward elimination. Only subjects with complete data in all fields of the included items were analysed (n=2714). As shown in Table 5, independent predictors of smoking abstinence at week 26 were older age, quitting based on “prove my ability to quit smoking”, high confidence in quitting, high external self-efficacy, more counselling sessions (both office and phone contact), and use of medication. The following characteristics were predictive of failure to quit: history of mental illness, high daily cigarette consumption, high FTND score, and feeling depressed.

This is the first comprehensive study of predictors of success for smoking cessation in a local smoking cessation service. Age, mental health, cigarette consumption, FTND score, reasons to quit, confidence in quitting, depressive mood, self-efficacy, sessions of office counselling, phone counselling, and medication treatment were identified as predictors among clients who volunteered to quit smoking.

In the univariate logistic analysis, most of the predictors were consistent with other studies. In many studies of predictors,15 22 results for gender, number of previous attempts, education level and social status, years of smoking, and history of depression have been inconsistent. In our study, a more comprehensive list of potential predictors from five domains (namely, demographics, health-related, smoking-related, psychosocial, and interventional variables) was included. After multiple logistic regression analysis, many commonly reported determinants/predictors were excluded. They included perceived health, marital status, cohabitation with a smoker, household income, gender, years of smoking, perceived importance of quitting or difficulty in quitting, feeling anxious, and internal self-efficacy in quitting.

The effect of age appeared to be consistent with the results of local23 24 and some international studies11 12 13 that older age was an independent predictor.25 Results for the predictive power of male gender have been controversial: some studies have reported it as a predictor of cessation success,8 10 26 while others have found it to have no significant effect or a negative effect.12 27 28 Our study could not confirm these findings. In addition, the role of marital status, education, household income, and number of cohabitants were shown not to be predictive, contrary to some overseas studies.29 30 Nonetheless, consistent with many studies, cigarette consumption and FTND score were negatively correlated with quit rate.8 27 31

Extensive research indicates that individual motivation, especially intrinsic motivation, is predictive of the long-term cessation result.8 In our study, two robust reasons to quit that could significantly predict abstinence were “prove my ability to quit smoking” and “avoid discrimination as a smoker”. This seemed to correspond to the “self-control” and “social influence” factors of Reasons for Quitting scale.32 In Hong Kong, smoking in some designated areas and public places is forbidden. This may precipitate the “avoid discrimination as a smoker” response. In service provision, operational initiatives and promotion strategies may be tailored to these two areas when motivating smokers to quit.

Perceived depressive mood (AOR=0.77) and history of mental illness (AOR=0.67) greatly enervated the success rate of quitting in our participants. Similar results have been reported in western countries as well as in Asia.8 33 34 35 This reinforces the importance of implementing appropriate mental health screening and referral in smoking cessation clinics. Presence of a chronic illness was not shown to be predictive although this may have been due to our relatively small sample size for this group of clients or because ours was a cohort of smokers who were motivated to quit. The effect of chronic illness may thus be attenuated. Studies have also shown that not all chronic diseases have the same impact on smoking cessation.36 37

The link between self-efficacy and successful quitting has long been established.22 38 Both external and internal self-efficacy in SEQ-12 scales have been found to be predictive in smoking cessation in western countries.21 In our study, after adjusting all potential predictors, a high degree of confidence and external self-efficacy were predictive of cessation, while the predictive ability of total and internal sub-score of SEQ-12 faded after adjustment. This is consistent with a previous Hong Kong study.20 Manifestation of cultural differences in self-efficacy during smoking cessation warranted further investigation. According to the results in the current study, smoking cessation counselling should focus more on helping clients to develop techniques to resist external temptation and to enhance external self-efficacy.

Consistent with overseas reviews of smoking cessation counselling,15 39 our study indicated that the number of sessions of face-to-face counselling or phone support were strong predictors (AOR=1.15 and 1.12, respectively). Both kinds of medication (NRT and non-NRT) were also associated with successful smoking cessation. Most previous predictor studies have not included these parameters, however.

There are some limitations in our study. Since this was a retrospective case review study of smokers who were motivated to quit, the results cannot be generalised to the whole smoking population. In addition, in the process of multiple logistic regression, only 2714 clients instead of all study subjects were analysed. Interventional variables such as office counselling, phone counselling, and medication modality were not randomly allocated. Patient compliance with medication was not evaluated, thus information on the use of medication may be biased. Another potential confounding factor was a small amount of missing data for some predictors. The effect of job nature and different chronic illnesses was not included in this study because of insufficient data; only chronic disease as a group was analysed. Self-reported 7-day point prevalence abstinence rate was not biochemically validated although previous study has shown that self-reported abstinence does not differ much to abstinence according to biochemical validation.40

This local study has identified a number of predictors of smoking abstinence at week 26 in clients who volunteered to seek treatment from a smoking cessation clinic. Most large-scale overseas studies have been based on a population survey. This was a large-scale comprehensive study performed in a real-life smoking cessation programme in Hong Kong. As such, it offers a better understanding of the determinants of successful quitting. Although some predictors have not been addressed and need further study, this study highlights the need for a holistic approach to the management of nicotine withdrawal, and to enhance external self-efficacy and motivation, to provide an adequate number of counselling sessions and to help smokers cope with mood problems.

1. WHO Report on the global tobacco epidemic. Geneva: World Health Organization; 2013.

2. Lam TH, Ho SY, Hedley AJ, Mak KH, Peto R. Mortality and smoking in Hong Kong: case-control study of all adult deaths in 1998. BMJ 2001;323:361. Crossref

3. Hong Kong Council on Smoking and Health: Annual Report 2011-2012. Hong Kong: Hong Kong Council on Smoking and Health; 2012.

4. WHO Report on the global tobacco epidemic, 2008: The MPOWER Package. Geneva: World Health Organization; 2008.

5. Wang L, Kong L, Wu F, Bai Y, Burton R. Preventing chronic diseases in China. Lancet 2005;366:1821-4. Crossref

6. 2008 PHS Guideline Update Panel, Liaisons, and Staff. Treating tobacco use and dependence: 2008 update U.S. Public Health Service Clinical Practice Guideline executive summary. Respir Care 2008;53:1217-22.

7. Reducing tobacco use: a report of the Surgeon General—executive summary. Nicotine Tob Res 2000;2:379-95. CrossRef

8. Caponnetto P, Polosa R. Common predictors of smoking cessation in clinical practice. Respir Med 2008;102:1182-92. Crossref

9. Prochaska JO, DiClemente CC, Velicer WF, Ginpil S, Norcross JC. Predicting change in smoking status for self-changers. Addict Behav 1985;10:395-406. Crossref

10. Hymowitz N, Cummings KM, Hyland A, Lynn WR, Pechacek TF, Hartwell TD. Predictors of smoking cessation in a cohort of adult smokers followed for five years. Tob Control 1997;6 Suppl 2:S57-62. Crossref

11. Monsó E, Campbell J, Tønnesen P, Gustavsson G, Morera J. Sociodemographic predictors of success in smoking intervention. Tob Control 2001;10:165-9. Crossref

12. Osler M, Prescott E. Psychosocial, behavioural, and health determinants of successful smoking cessation: a longitudinal study of Danish adults. Tob Control 1998;7:262-7. Crossref

13. Hyland A, Borland R, Li Q, et al. Individual-level predictors of cessation behaviours among participants in the International Tobacco Control (ITC) Four Country Survey. Tob Control 2006;15 Suppl 3:iii83-94. Crossref

14. Haas AL, Muñoz RF, Humfleet GL, Reus VI, Hall SM. Influences of mood, depression history, and treatment modality on outcomes in smoking cessation. J Consult Clin Psychol 2004;72:563-70. Crossref

15. Iliceto P, Fino E, Pasquariello S, D’Angelo Di Paola ME, Enea D. Predictors of success in smoking cessation among Italian adults motivated to quit. J Subst Abuse Treat 2013;44:534-40. Crossref

16. Miller WR, Rollnick S. Talking oneself into change: motivational interviewing, stages of change, and therapeutic process. J Cogn Psychother 2004;18:299-308. Crossref

17. DiClemente CC, Prochaska JO. Toward a comprehensive, transtheoretical model of change: stages of change and addictive behaviors. In: Miller WR, Heather N, editors. Treating addictive behaviors. 2nd ed. New York: Plenum Press; 1998: 3-24.

18. Lai DT, Cahill K, Qin Y, Tang JL. Motivational interviewing for smoking cessation. Cochrane Database Syst Rev 2010;(1):CD006936. Crossref

19. Heatherton TF, Kozlowski LT, Frecker RC, Fagerström KO. The Fagerström Test for Nicotine Dependence: a revision of the Fagerström Tolerance Questionnaire. Br J Addict 1991;86:1119-27. Crossref

20. Leung DY, Chan SS, Lau CP, Wong V, Lam TH. An evaluation of the psychometric properties of the Smoking Self-Efficacy Questionnaire (SEQ-12) among Chinese cardiac patients who smoke. Nicotine Tob Res 2008;10:1311-8. Crossref

21. Etter JF, Bergman MM, Humair JP, Perneger TV. Development and validation of a scale measuring self-efficacy of current and former smokers. Addiction 2000;95:901-13. Crossref

22. Li L, Borland R, Yong HH, et al. Predictors of smoking cessation among adult smokers in Malaysia and Thailand: findings from the International Tobacco Control Southeast Asia Survey. Nicotine Tob Res 2010;12 Suppl:S34-44. Crossref

23. Yu DK, Wu KK, Abdullah AS, et al. Smoking cessation among Hong Kong Chinese smokers attending hospital as outpatients: impact of doctors’ advice, successful quitting and intention to quit. Asia Pac J Public Health 2004;16:115-20. Crossref

24. Abdullah AS, Yam HK. Intention to quit smoking, attempts to quit, and successful quitting among Hong Kong Chinese smokers: population prevalence and predictors. Am J Health Promot 2005;19:346-54. Crossref

25. Fiore MC, Novotny TE, Pierce JP, Hatziandreu EJ, Patel KM, Davis RM. Trends in cigarette smoking in the United States. The changing influence of gender and race. JAMA 1989;261:49-55. Crossref

26. Piper ME, Cook JW, Schlam TR, et al. Gender, race, and education differences in abstinence rates among participants in two randomized smoking cessation trials. Nicotine Tob Res 2010;12:647-57. Crossref

27. Vangeli E, Stapleton J, Smit ES, Borland R, West R. Predictors of attempts to stop smoking and their success in adult general population samples: a systematic review. Addiction 2011;106:2110-21. Crossref

28. Bjornson W, Rand C, Connett JE, et al. Gender differences in smoking cessation after 3 years in the Lung Health Study. Am J Public Health 1995;85:223-30. Crossref

29. Kim YJ. Predictors for successful smoking cessation in Korean adults. Asian Nurs Res (Korean Soc Nurs Sci) 2014;8:1-7. Crossref

30. Chandola T, Head J, Bartley M. Socio-demographic predictors of quitting smoking: how important are household factors? Addiction 2004;99:770-7. Crossref

31. Abdullah AS, Ho LM, Kwan YH, Cheung WL, McGhee SM, Chan WH. Promoting smoking cessation among the elderly: what are the predictors of intention to quit and successful quitting? J Aging Health 2006;18:552-64. Crossref

32. Curry S, Wagner EH, Grothaus LC. Intrinsic and extrinsic motivation for smoking cessation. J Consult Clin Psychol 1990;58:310-6. Crossref

33. Kim SK, Park JH, Lee JJ, et al. Smoking in elderly Koreans: prevalence and factors associated with smoking cessation. Arch Gerontol Geriatr 2013;56:214-9. Crossref

34. Lasser K, Boyd JW, Woolhandler S, Himmelstein DU, McCormick D, Bor DH. Smoking and mental illness: A population-based prevalence study. JAMA 2000;284:2606-10. Crossref

35. Covey LS, Glassman AH, Stetner F. Depression and depressive symptoms in smoking cessation. Compr Psychiatry 1990;31:350-4. Crossref

36. Salive ME, Cornoni-Huntley J, LaCroix AZ, Ostfeld AM, Wallace RB, Hennekens CH. Predictors of smoking cessation and relapse in older adults. Am J Public Health 1992;82:1268-71. Crossref

37. Freund KM, D’Agostino RB, Belanger AJ, Kannel WB, Stokes J 3rd. Predictors of smoking cessation: The Framingham Study. Am J Epidemiol 1992;15:957-64.

38. Gwaltney CJ, Metrik J, Kahler CW, Shiffman S. Self-efficacy and smoking cessation: a meta-analysis. Psychol Addict Behav 2009;23:56-66. Crossref

39. Clinical Practice Guideline Treating Tobacco Use and Dependence 2008 Update Panel, Liaisons, and Staff. A clinical practice guideline for treating tobacco use and dependence: 2008 update. A U.S. Public Health Service report. Am J Prev Med 2008;35:158-76.

40. Patrick DL, Cheadle A, Thompson DC, Diehr P, Koepsell T, Kinne S. The validity of self-reported smoking, a review and meta-analysis. Am J Public Health 1994;84:1086-93. Crossref