Should We Pay the Smokers? A Meta-Analysis of Financial Incentives for Smoking Cessation Among Smokers in Low Socioeconomic Group

IJPHCS Open Access: e-Journal International Journal of Public Health and Clinical Sciences e-ISSN : 2289-7577. Vol. 6:No. 2 March/April 2019 SHOULD WE PAY THE SMOKERS? A META-ANALYSIS OF FINANCIAL INCENTIVES FOR SMOKING CESSATION AMONG SMOKERS IN LOW SOCIOECONOMIC GROUP Fadzrul Hafiz Johani1, Muhammad Aklil Abd Rahiml,2, Zakiah Othman3, Shamsul Azhar Shah1*, Nazarudin Safian1 1 Department of Community Health, Faculty of Medicine, University Kebangsaan Malaysia, Kuala Lumpur, Malaysia. 2 Department of Community and Family Medicine, Faculty of Medicine, Universiti Malaysia Sabah, Malaysia 3 Department of Social and Preventive Medicine, Faculty of Medicine, University Malaya, Kuala Lumpur, Malaysia *Correspondence: drsham@ppukm.ukm.edu.my, Tel: +60195581454, Fax: +603-9145 6670 https://doi.org/10.32827/ijphcs.6.2.113 Abstract Today around 80% of smokers worldwide live in low- and middle-income countries, and in most countries, regardless of country income group, tobacco use is more concentrated in low socioeconomic status (SES) populations. This meta-analysis was conducted to review current available evidences to determine the effectiveness of financial incentive strategies on smoking cessation among low-SES smokers. Database search using PubMed, Science Direct and Cochrane Library were used to search financial incentive intervention prior to October 2018. Appraisal of methodological quality was assessed using Cochrane Collaboration’s tool. Six identified randomized control trials with 2450 and 2437 participants in intervention and control group respectively were included in the analysis. The random-effect model was used to combine results from individual studies. The pooled odds ratio (OR) was 2.16 (95% CI: 1.662.82) comparing financial incentive intervention with control. Heterogeneity was not significant across studies (Chi2 = 8.17, p = 0.15, I2 = 39%). Current evidences from the RCT researches suggest that financial incentives are promising potential strategy to encourage smoking cessation among low-SES smokers. Keyword: financial incentive, smoking cessation, low socioeconomic status, meta-analysis Fadzrul Hafiz Johani et.al. https://doi.org/10.32827/ijphcs.6.2.113 113 IJPHCS Open Access: e-Journal International Journal of Public Health and Clinical Sciences e-ISSN : 2289-7577. Vol. 6:No. 2 March/April 2019 1.0 Introduction Cigarette smoking is the only legal product that kills a large proportion of its consumers when used as intended by its manufacturer. The World Health Organization (WHO) has estimated that around six million people die each year from tobacco use and projected to increase to eight million by 2030 if no strong tobacco control measures are put in place 1. Today around 80% of smokers worldwide live in low- and middle-income countries, and in most countries, regardless of country income group, tobacco use is more concentrated in low socioeconomic status (SES) populations where the burden of tobacco-related illness and death is heaviest 2. In addition, smokers in the low-SES are more likely to be trapped in the vicious cycle of smoking and disadvantages 3. Some studies reported that smokers in low-SES are as likely to prioritise and attempt quitting as high-SES smokers, but they are less likely to be successful and confident in quitting 4,5. Smokers from low-SES face unique barriers to smoking cessation such as higher level of dependent on nicotine 6-9, higher level of stress in daily life 10,11 and living in the pro-smoking community norm 3,12 which increase the social stress and become a cue to smoke and prevent smoking cessation. There is a growing enthusiasm for incentive-based programmes to change unhealthy behaviours, including smoking 13, weight loss 14,15, alcohol consumption 16, vaccination uptake 17 and levels of physical activity 18. Incentives and rewards (terms used interchangeably) routinely feature in many smoking cessation programmes which can be used to encourage recruitment into the programme 19, to reward compliance with the process 20,21, and to reward cessation achieved at predefined stages 22. A variety of rewards have been used for these purposes, including cash payments, gift vouchers exchangeable for goods (excluding alcohol and cigarettes), food packages, certificates, promotional items such as clothes and others 23,24. The effectiveness of financial incentive strategies for smoking cessation were repeatedly evaluated in the general population 22,25. Thus, the aim of this study was to review current available evidences to determine the effectiveness of financial incentive strategies on smoking cessation among low socioeconomic smokers. 2.0 Methodology 2.1 Literature searches This meta-analysis was performed based on PRISMA statement 26. The literature search was performed on three databases which are PubMed, ScienceDirect and Cochrane Library. The search strategy utilized the PICO framework to improve searching for clinical question 27. The search term used were: ("poor" OR "poverty" OR "low income" OR "disadvantage*" OR “low socioeconom*”) AND ("financial incentive" OR "financial reward" OR “cash” OR “monetary” OR “payment” OR “lotter*” OR “gift card” OR “voucher”) AND ("smoking" OR “tobacco” OR "quit smoking" OR “stop smoking” OR "smoking cessation" OR “smoking abstinence”). The search was not restricted to any duration timeline. Fadzrul Hafiz Johani et.al. https://doi.org/10.32827/ijphcs.6.2.113 114 IJPHCS Open Access: e-Journal International Journal of Public Health and Clinical Sciences e-ISSN : 2289-7577. Vol. 6:No. 2 March/April 2019 2.2 Study Selection The process of study selection was conducted in two phases by two reviewers after excluding duplicates studies. In the first phase, a pair of reviewers (F.H.J and Z.O) were independently screened the titles and abstracts for the potential article to be included in the study. During this phase, irrelevant studies were excluded, and any disagreement will be solved by the third reviewer (M.A.A.R). In the second phase, the full-text articles were retrieved for detailed evaluation for selection of articles. Study were included if they meet the following criteria: (i) Randomized controlled trials (RCTs) or clinical controlled trial (CCTs) that described the evaluation of financial incentive intervention with smoking cessation; (ii) among low-income or socially disadvantages smokers and (iii) adult smoker, published prior to October 2018. Studies that were not published in English, that were case reports or observational studies or full-texts not available were excluded. 2.3 Data Extraction The selected articles that met the inclusion and exclusion criteria by the two reviewers were retained for full review. Several authors were also contacted to get details of the eligible studies. The characteristic of each studies was examined including setting, study location, participants, intervention, follow-up period, primary outcome measures and results of the studies. 2.4 Assessment of methodological quality Studies included in the review were assessed for methodological quality and risk of bias using Cochrane Collaboration’s tool 28. The study quality was assessed by two authors (F.H.J and N.S) and any disagreement were resolved through discussion. The Cochrane Collaboration’s tool has been widely used for assessing risk of bias in randomized trials 29. Cochrane Collaboration’s tools evaluates six domains: random sequence generation (assessing selection bias on allocation to intervention due to inadequate generation of a randomized sequence), allocation concealment (assessing selection bias on allocation to intervention due to inadequate concealment allocation prior to assignment), blinding of participant and personnel (assessing performance bias due to knowledge of the allocated interventions by participants and personnel during the study), blinding of outcome assessment (assessing detection bias due to knowledge of the allocated intervention by outcome assessor) incomplete outcome data (assessing attrition bias due to amount, nature or handling of incomplete outcome data) and selective reporting (assessing reporting bias due to selective outcome reporting). Each study was given a rating of “Low Risk”, “High Risk” or “Unclear Risk” in the methodological quality for each domain according to pre-defined criteria 28 2.5 Meta-analysis Random-effect model was used to estimate the pooled effect size from included studies as suggested by 30. Odds ratio (OR) with 95% confidence interval (CI) and statistical measure of heterogeneity (Chi2 and I2) was calculated using Review Manager Ver5.3 31. All studies were included in the meta-analysis. Subgroup and sensitivity analysis were performed if p<0.10 and I2 value ≥ 50%. Fadzrul Hafiz Johani et.al. https://doi.org/10.32827/ijphcs.6.2.113 115 IJPHCS Open Access: e-Journal International Journal of Public Health and Clinical Sciences e-ISSN : 2289-7577. Vol. 6:No. 2 March/April 2019 2.6 Outcome measures The primary outcome measures were smoking abstinence at any point assessed after the start of the intervention. Smoking abstinence assess at 12-month duration or less after quit date is considered short-term abstinence while smoking abstinence assess at any point of time after 12 months is considered long-term abstinence 32. Biochemically validated evidence of quit rates was preferred over self-reported quit rates and biomarker of cotinine-confirmed measures were preferred over carbon monoxide (CO) measures. To ensure consistency in outcome measure, 7-day point prevalence abstinence rates were preferred although continuous abstinence rate were used if this was the only outcome measure reported. Analysis using intention-to-treat approach was used if possible. Where study had more than one type of intervention group, the most intensive condition was compared to the control group 33. 3.0 Results 3.1 Literature search The initial search strategy yielded a total of 119 literatures where 18 articles were excluded because of duplication. The remaining 101 articles were screened for title and abstracts which further excluded another 62 articles for irrelevant studies (n=53), language other than English (n=4), review articles (n=4) and protocol study (n=1). The retained 39 articles were assessed for their eligibility. Among them; ten studies using different intervention other than financial incentive such as psychosocial intervention; 17 studies have different outcome measure such as measuring quality of life and tobacco cessation service engagement; five studies had different target population other than low-income or socially disadvantages smoker; and one study measuring cost-effectiveness of financial incentive for smoking cessation among low income smoker. Six remaining articles 20,21,33-36 were included for meta-analysis. A flow diagram describing the article retrieval based on PRISMA flow diagram 26 is provided in Figure 1. Fadzrul Hafiz Johani et.al. https://doi.org/10.32827/ijphcs.6.2.113 116 IJPHCS International Journal of Public Health and Clinical Sciences e-ISSN : 2289-7577. Vol. 6:No. 2 March/April 2019 Identification Open Access: e-Journal PubMed (n = 49) Cochrane Library (n = 38) ScienceDirect (n = 32) Total (n = 119) Included Eligibility Screening Duplicate (n= 18) Abstract excluded: Irrelevant (n= 53) Not English (n= 4) Review (n= 4) Protocol study (n= 1) Abstract screened (n = 101) Full-text articles assessed for eligibility (n = 39) Studies included in analysis (n = 6) Full-text articles excluded (n =33) Reasons: -Different intervention (n = 10) -Different outcome (n = 17) -Different population (n = 5) -Cost effectiveness (n= 1) Figure 1: PRISMA flow diagram 3.2 Characteristics of included studies A detailed description of the included studies was listed in Table 1. The included studies were published between 2015 and 2018. All studies were RCTs study design. Five studies were conducted in United States 20,21,33,35,36 and one study from Switzerland 34. One study conducted among pregnant women 20 and one in mentally-ill adult 33 while others study participant were Fadzrul Hafiz Johani et.al. https://doi.org/10.32827/ijphcs.6.2.113 117 IJPHCS International Journal of Public Health and Clinical Sciences e-ISSN : 2289-7577. Vol. 6:No. 2 March/April 2019 Open Access: e-Journal recruited from general population or community-based health care clinic. Total financial incentive offered range from USD150 to approximate USD1557 (converted from CHf1500 on 28 June 2012 37) and the incentives were given in gift cards or monetary form. 3.3 Evaluation of Quality of Studies Individual rating for risk of bias in each study against the six domains from Cochrane Collaboration’s tool are reported in Table 2. Overall, most of the studies has low risk of bias across all six domains but blinding of participant and personnel was not possible to perform. Detail of assessments were provided in Supplemental Material. Table 2. Risk of bias assessment Random Study sequence generation 36 U 34 21 Blinding of Allocation participant concealment and personnel Blinding of Incomplete outcome outcome assessment data Selective reporting Others U H L L L L U L H L L L L L U H L L L H 33 L U H L L L L 35 L L H L L L H 20 L L H L L L H H: High Risk, L: Low Risk, U: Unclear Risk Fadzrul Hafiz Johani et.al. https://doi.org/10.32827/ijphcs.6.2.113 118 IJPHCS International Journal of Public Health and Clinical Sciences e-ISSN : 2289-7577. Vol. 6:No. 2 March/April 2019 Open Access: e-Journal Table 1. Characteristic of the studies Study, Design, Location 36 RCT Dallas, Texas Participant Characteristic Smoking Characteristic Intervention Incentive method; total; type Participant were socioeconomically disadvantage smoker who visit the Tobacco Cessation Clinic of safety net hospital. Cigarettes per day: Mean (SD) Intervention: 18.0(9.7) Control: 17.0(7.7) Intervention: 75 participants received usual care for smoking cessation plus financial incentive. $20 gift card for biochemically confirmed abstinence on the quit date then amount of the incentives increased by $5 with each weekly successive abstinent visit through 4 weeks after the quit date. Age: Mean (SD) year Intervention= 51.7(7.3) Control=52.6(7.4) HIS: Mean(SD) Intervention= 3.4(1.3) Control= 3.1 (1.2) Gender: (Male%) / (Female%) Intervention: (48%) / (52%) Control: (36.6%) / (63.4%) 34 RCT Geneva, Switzerland Participant were low-income smoker enrolled from the general population Age: Mean (SD) year Intervention=32(11) Control=32(11) Gender: (Male%) / (Female%) Intervention: (53%) / (47%) Control: (44%) / (56%) Cigarettes per day: Mean (SD) Intervention: 16(9) Control: 16(9) FTND: Mean(SD) Intervention: 4.1(2.3) Control: 3.9(2.4) Control: 71 participants received usual care for smoking cessation includes: 1. One initial educational session provided by a respiratory therapist 2. Weekly group support sessions facilitated by social workers. 3. Receive pharmacotherapy and individual follow-up weekly/as needed basis Intervention: 404 participant received financial incentives plus Internet-based support. Control: 401 participant received Internet-based support, but no financial incentives Internet-based support is a Stop-tabac.ch smoking cessation website, which offers fact sheets, discussion forums, testimonials, and an interactive “coach” that provides automatically written, personalized feedback reports. Primary Outcome Assessment; Follow-up 7-day PPA; 4 weeks follow-up verified by breath CO level. OR = 2.87 (1.42-5.77) AOR = 3.40 (1.61-7.16) Total incentive $150 Type: Gift card . Adjusted for pharmacological treatment, race, gender, age, years of education, cigarettes smoked per day Incentives given 6 times during 6 months from quit date: CHf 100 (1st week) CHf 150 (2nd week) CHf 200 (3rd week) CHf 300 (1st month) CHf 350 (3rd month) CHf 400 (6th month) Total incentive CHf 1500 Type: Gift card Result; Odds Ratio (Confidence Interval) Abstinence rates in intervention and control group were 49.3% versus 25.4%. 7-day PPA; 6- and 18-month follow-up both verified by breath CO level and either cotinine or thiocyanate measurements Rates of continuous abstinence between months 6 and 18 were 9.48% in the intervention group and 3.71% in the control group OR = 2.72 (1.47-5.02) AOR = 2.94 (1.57-5.50) Adjusted for sex and past quit attempts Fadzrul Hafiz Johani et.al. https://doi.org/10.32827/ijphcs.6.2.113 119 IJPHCS International Journal of Public Health and Clinical Sciences e-ISSN : 2289-7577. Vol. 6:No. 2 March/April 2019 Open Access: e-Journal 21 RCT Wisconsin, United States Participants were Medicaid recipients recruited from primary care patients and callers to the Wisconsin Tobacco Quit Line. Age: Mean (SD) year Intervention=45.0(11.2) Control=44.9(11.2) Cigarettes per day: Mean (SD) Intervention: 17.0(10.3) Control: 17.4(10.9) FTND (Item 1: % Smoking Within 30 Min) Intervention: 83.8% Control: 87.1% Gender: (Male%) / (Female%) Intervention: (40.0%) / (60.0%) Control: (39.4%) / (60.6%) 33 RCT New Hampshire, United States Participants were communitydwelling adult Medicaid beneficiaries with a mental illness diagnosis who were receiving services at community mental health centers (CMHC) Age: Mean (SD) year PV: (43.0 ± 10.8) PV+Q: (45.0 ± 10.7) PV+CBT: (46.0 ± 11.0) Gender: (Male%) / (Female%) PV: (44%) / (56%) PV+Q: (34%) / (66%) PV+CBT: (33%) / (67%) Cigarettes per day: Mean (SD) PV: 16.0(10.9) PV+Q: 18.0(10.2) PV+CBT: 17.0(10.5) FTND: Mean(SD) PV = 5.0(2.4) PV+Q = 5.0(2.1) PV+CBT = 6.0(2.4) Intervention 948 participants were offered five quitline cessation calls from Wisconsin Tobacco Quit Line (WTQL) and were encouraged to obtain cessation medication. Incentive were given in each counselling calls and during verified abstinence at 6month visit. Control 952 participants were offered five quitline cessation calls from WTQL and were encouraged to obtain cessation medication without financial incentive Intervention 75 participants in pharmacotherapy only (PV) + incentives 152 participants in pharmacotherapy and facilitated quitline (PV+Q) + incentives 108 participants in pharmacotherapy and telephone cognitive-behavioral therapy (PV+CBT) + incentives Control 71 participants in PV only 151 participants in PV+Q 104 participants in PV+CBT Participants in the incentive condition could receive payment of $30/call for up to five WTQL calls and $40 for producing biochemical evidence of abstinence at the 6-month follow-up visit. 7-day PPA; at 6months follow-up post study entry verified by breath CO level or cotinine or nicotine test Total incentive $ 190 Type: Monetary OR = 1.73 (1.36-2.20) AOR = 1.75 (1.37-2.23) Adjusted for type of entry into study, motivation to quit and Fagestrom-Test Nicotine Dependence (FTND) Participants in the incentive conditions received $50 in cash for verified abstinence on Mondays, Wednesdays, and Fridays in the first two weeks of the quit attempt and additional $75 for verified abstinence in third and fourth week. Total incentive $450 Type: Monetary Abstinence rates in incentive condition participants were 21.6% compared to 13.8% in controls 7-day PPA; at 12 months follow up confirmed with breath CO and urine cotinine test (or solely breath CO if a participant was using NRT) Abstinence rates at 12 months among participants receiving PV+Q with incentives is greater (14%) compared with PV+Q without incentives (4%) AOR = 3.94 (1.32-11.75) adjusted for gender, psychiatric diagnostic group, severity of nicotine dependence, and equipoise stratum. Fadzrul Hafiz Johani et.al. https://doi.org/10.32827/ijphcs.6.2.113 120 IJPHCS International Journal of Public Health and Clinical Sciences e-ISSN : 2289-7577. Vol. 6:No. 2 March/April 2019 Open Access: e-Journal 35 RCT Boston Massachusetts, United States Participants were low-SES and minority daily smokers with income ≤ $20,000/year receiving primary care at Boston Medical Center Cigarettes per day: Mean (SD) Intervention: 15.1(7) Control: 14.9(7) Age: Mean (SD) year Intervention= 49.9(11) Control= 50.1(10) FTND: Mean(SD) Intervention: 4.9(2) Control: 5.0(2) Gender: (Male%) / (Female%) Intervention: (43%) / (57%) Control: (49%), (51%) 20 RCT Wisconsin, United States Participant were low-income (Medicaid-registered) pregnant smokers receiving perinatal smoking cessation program at Wisconsin Women’s Health Foundation (WWHF). Age: Mean (SD) year Intervention= 26.7(5.4) Control= 26.1(5.1) Gender: All female Intervention 177 participants received the same materials in control; in addition, they received up to 4 hours of patient navigation delivered over 6 months, and financial incentives Control 175 participants received a low literacy smoking cessation brochure and a list of hospital and community resources for smoking cessation. Age first started smoking daily: Mean(SD) year Intervention: 16.3(3.4) Control: 16.4(3.3) Intervention 505 participants receiving smoking cessation counseling with monetary incentive FTND (Item 1: % Smoking Within 30 Min) Intervention: 54.7% Control: 58.4% Control 509 participants receiving smoking cessation counseling only Participants in incentive condition received $250 for smoking cessation 6 months after study enrollment; additional $500 for an additional 6 months after the initial cessation (12-month time point) 7-day PPA; at 12 months follow up confirmed with salivary cotinine Abstinence rates at 12 months follow up among participants in the navigation and incentives condition were 11.9% compared to 2.3% in control OR = 5.8 (1.9-17.1) AOR = 6.09 (2.01 – 18.40) Total incentive: $750 Type: Monetary Adjusted for age and sex Participants in incentive condition received $25 for each of 6 prebirth provider visits, $25–40 for each of 4 postbirth home visits at Weeks 1, 2, 4, and 6 (total = $130), $20 for each of 5 postbirth counseling calls, $40 for biochemically verified abstinence at the Week 1 and 6month visits. Total incentive: $460 Type: Monetary 7-day PPA; at 6month follow up verified by breath CO level. Incentive condition participants had a higher abstinence rate at 6-month post-birth than controls (14.65% vs. 9.23%) respectively OR = 1.69 (1.14-2.49) FTND: Fagerström Test for Nicotine Dependence, HSI: Heaviness of Smoking Index, CHf: Swiss franc, $: United States Dollar, SD: Standard deviation, Currency conversion 1 CHf = 1.038 USD (on 28 June 2012 during period of study) 37 Fadzrul Hafiz Johani et.al. https://doi.org/10.32827/ijphcs.6.2.113 121 IJPHCS Open Access: e-Journal International Journal of Public Health and Clinical Sciences e-ISSN : 2289-7577. Vol. 6:No. 2 March/April 2019 3.4 Publication Bias The shape and symmetry of the funnel plot of log ORs from the six studies as shown in Figure 2 suggest that there is minimal publication bias present. All of the studies have high precision except for one study 35. Figure 2. Funnel plot of the six studies included in the meta-analysis. 3.5 Main Analysis The meta-analysis data using random-effect model 30 combining all six studies to explore the effect of financial incentive intervention for smoking cessation in low socioeconomic group of smokers shown in Figure 3. The forest plot illustrates the spread of the six studies risk estimates and their confidence intervals in relation to the summary OR of meta-analysis. There was no heterogeneity was found across studies (pheterogeneity = 0.15, I2 = 39%). Based on the six studies, the pooled OR estimates showed that financial incentive intervention was significantly associated with smoking cessation among low socioeconomic group of smokers. (pooled OR: 2.16; 95% CI: 1.66-2.82). When subgroup analysis done to explore between total amount of financial incentive received in the studies in relation to smoking cessation, the odds ratio of smokers who received financial incentive of more than USD500 was significantly magnified (pooled OR: 3.41; 95% CI: 1.67-6.97) compared to group that received financial incentive of less than USD500 (OR: 1.86; 95% CI: 1.53-2.26) as shown in Figure 4. Fadzrul Hafiz Johani et.al. https://doi.org/10.32827/ijphcs.6.2.113 122 IJPHCS Open Access: e-Journal International Journal of Public Health and Clinical Sciences e-ISSN : 2289-7577. Vol. 6:No. 2 March/April 2019 Figure 3. Forest plot of the six studies included in meta-analysis Figure 4. Subgroup analysis by amount of incentive offered 4.0 Discussion This study analyzed the evidence from a published RCTs identified from different databases to evaluate the effectiveness of financial incentive intervention on smoking cessation among low-SES. There was only six RCTs fulfilled the criteria that addressing this issue. The result of the metaanalysis that involved 2450 and 2437 participants in intervention and control group respectively suggest that financial incentive intervention have significantly higher smoking cessation rates Fadzrul Hafiz Johani et.al. https://doi.org/10.32827/ijphcs.6.2.113 123 IJPHCS Open Access: e-Journal International Journal of Public Health and Clinical Sciences e-ISSN : 2289-7577. Vol. 6:No. 2 March/April 2019 among low-SES (pooled OR: 2.16; 95% CI: 1.66-2.82) with homogeneity observed across studies (Chi2=8.17, p value=0.15, I2 = 39%). The small number of relevant studies precludes conclusions regarding optimal amount of financial incentive to achieve effect even though the subgroup analysis showed that financial incentive more than USD500 have higher effect estimate. Furthermore, to date, there have been no studies done at regional area other than United States and Switzerland to explore different effect of financial incentive according to ethnic group that have different smoking behavior and level of nicotine dependent 38-40 This analysis provides a useful synthesis of RCT evidence on the effect of financial incentive on smoking cessation among low-SES and facilitates identification of future research opportunities. However, there are several caveats which concern the validity of the assessment results that must be acknowledged and interpreted with caution. The homogeneity across studies in the meta-analysis could be due to small numbers of included studies 28. Besides, the pooled effect estimate does not differentiate duration of follow up and amount of financial incentive offered in the intervention group which are varies among studies. The six trials reviewed had some methodological limitation. One study 36 have a very short duration of 4-week follow up assessment thus it limits the ability of smoker to change behaviour and quit smoking because study by Phillippa Lally et al found that on average it takes 66 days for a habit to become ingrained 41 . The Transtheoretical Model of Change which is the famous theoretical model of behaviour change has been the basis for developing effective intervention to promote health behaviour change. The Transtheoretical Model 42-44 is the model of intentional change that describes how people modify a problem behaviour or acquire a positive behaviour. The model discovered that long term effect after 12 months of health habits, 43% failed to maintain their healthy behaviour but the risk of relapse dropped to 7% if the behaviour maintained until 5 years. Thus, a very much longer period of time was required in the research to see the true effect of smoking cessation behaviour as compared to the maximum intervention period of 18 months among the studies. All trials were conducted in United States except study by Etter&Schmid 2016 which conducted in the Switzerland. Thus, the finding may not be generalized to countries that have different index in measuring socioeconomic status, diverse sociocultural and political environment. Some of the trials have relatively small sample size. Study by Kendzor et al had only 75 participants in intervention group and 71 participants in control group and this may attenuate the power of study. Apart from that, trial by Lasser et al had incorporated additional intervention in incentive group which is differ from the control group. Thus, the effect of financial incentive may be masked by the effect of additional intervention and may not reflect the actual significant different between the group. Finally, among all trials, none described evidence of dose-response relationship according to amount of financial incentive offered with smoking cessation rates. Thus, amount of financial incentive that would yield optimal smoking cessation rate is unknown. The longest period of follow-up assessment was done at 18-month 34. Long term abstinence beyond this point doubt to be sustained as the trend of abstinence rates reduce by time 33,34,36. The incentive might have the short-term desired effect but still weaken the intrinsic motivation. Thus, once the incentives have been removed, the desired outcome will be pursued less eagerly, suggesting challenges for long-term sustainability 45 . This similar pattern of behaviour changes was also seen in other incentive-based intervention 16,46-48. Fadzrul Hafiz Johani et.al. https://doi.org/10.32827/ijphcs.6.2.113 124 IJPHCS Open Access: e-Journal International Journal of Public Health and Clinical Sciences e-ISSN : 2289-7577. Vol. 6:No. 2 March/April 2019 Generally, the key finding of this analysis support those similar reviews of financial incentive interventions across wider range of smoker population 13 in which they are effective in encouraging smoker to quit smoking. 5.0 Policy Implication Smoking cessation among low-SES smokers through financial incentive strategies are promising potential component. Nonetheless, there are challenges that need to be addressed in the effort to implement and sustain incentive-based policies such as feasibility, sustainability of funding mechanism through multifaceted approach and substantial commitment from government and nongovernment organization. Regulation on eligibility for low-SES to be a non-smoker as a prerequisite to receive financial assistance from social welfare or other financial institution should be a mandatory implementation which is similar of what some country did for vaccination issues 49. 6.0 Future research Future research in RCTs design to measure the effectiveness of financial incentive on smoking cessation in low-SES among diverse sociocultural may be required as smoking behaviour and nicotine dependence are genetically related 40,50. Furthermore, future RCTs should explicitly assess optimal amount of monetary offered to yield most desired effect of smoking cessation. Trial should be reported based on CONSORT statement 51 to facilitate quality assessment and ensure data availability for meta-analysis. 7.0 Conclusion Current evidences from the RCT researches suggest that financial incentives are promising potential strategy to encourage smoking cessation among low-SES smokers. Acknowledgements The authors would like to thank the University Kebangsaan Malaysia for sponsoring this work to be published in an open access journal. Author Contributions Fadzrul Hafiz Johani et.al. https://doi.org/10.32827/ijphcs.6.2.113 125 IJPHCS Open Access: e-Journal International Journal of Public Health and Clinical Sciences e-ISSN : 2289-7577. Vol. 6:No. 2 March/April 2019 F.H.J designed the study, F.H.J and Z.O performed data collection. F.H.J and M.A.A.R performed data analysis and wrote the manuscript. F.H.J and N.S assessed studies’ quality. M.R.A.M edited the manuscript. S.A.S provided suggestion for analysis and review the manuscript. Conflict of Interest The authors declare no conflict of interest. Supplemental Material Supplemental Table 1: Detail assessment on risk of bias References 1 2 3 4 5 6 7 8 9 10 WHO. WHO global report on trends in prevalence of tobacco smoking 2015. (World Health Organization, 2015). NCI & WHO. The Economics of Tobacco and Tobacco Control. Report No. NIH Publication No. 16-CA-8029A, (U.S National Cancer Institute and World Health Organization, Tobacco Control Monograph, 2016). Eriksen, M., Mackay, J., Schluger, N., Islami, F. & Drope, J. The Tobacco Atlas. (American Cancer Society, Atlanta, Georgia, 2015). Cancer, C. f. B. R. i. Quitting strategies used by current smokers and recent quitters: Findings from the 2015 Victorian smoking and health survey. Melbourne: Cancer Council Victoria, (2016). Giskes, K., van Lenthe, F. J., Turrell, G., Brug, J. & Mackenbach, J. P. Smokers living in deprived areas are less likely to quit: a longitudinal follow-up. Tobacco control 15, 485488 (2006). Hyland, A. et al. Individual-level predictors of cessation behaviours among participants in the International Tobacco Control (ITC) Four Country Survey. Tobacco control 15, iii83iii94 (2006). Siahpush, M., McNeill, A., Borland, R. & Fong, G. Socioeconomic variations in nicotine dependence, self-efficacy, and intention to quit across four countries: findings from the International Tobacco Control (ITC) Four Country Survey. Tobacco control 15, iii71-iii75 (2006). Kotz, D. & West, R. Explaining the social gradient in smoking cessation: it’s not in the trying, but in the succeeding. Tobacco control 18, 43-46 (2009). Picco, L., Subramaniam, M., Abdin, E., Vaingankar, J. A. & Chong, S. A. Smoking and nicotine dependence in Singapore: findings from a cross-sectional epidemiological study. (2017). Adler, N. E. et al. Socioeconomic status and health: the challenge of the gradient. American psychologist 49, 15 (1994). Fadzrul Hafiz Johani et.al. https://doi.org/10.32827/ijphcs.6.2.113 126 IJPHCS Open Access: e-Journal 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 International Journal of Public Health and Clinical Sciences e-ISSN : 2289-7577. Vol. 6:No. 2 March/April 2019 Aneshensel, C. S. Social stress: Theory and research. Annual review of sociology 18, 15-38 (1992). Echeverría, S., Diez-Roux, A. V., Shea, S., Borrell, L. N. & Jackson, S. Associations of neighborhood problems and neighborhood social cohesion with mental health and health behaviors: the Multi-Ethnic Study of Atherosclerosis. Health & place 14, 853-865 (2008). Cahill, K., Hartmann‐Boyce, J. & Perera, R. Incentives for smoking cessation. Cochrane Database of Systematic Reviews (2015). Paul‐Ebhohimhen, V. & Avenell, A. Systematic review of the use of financial incentives in treatments for obesity and overweight. Obesity Reviews 9, 355-367 (2008). Volpp, K. G. et al. Financial incentive–based approaches for weight loss: a randomized trial. Jama 300, 2631-2637 (2008). Lussier, J. P., Heil, S. H., Mongeon, J. A., Badger, G. J. & Higgins, S. T. A meta‐analysis of voucher‐based reinforcement therapy for substance use disorders. Addiction 101, 192203 (2006). Wigham, S. et al. Parental financial incentives for increasing preschool vaccination uptake: systematic review. Pediatrics 134, e1117-e1128 (2014). Mitchell, M. S. et al. Financial incentives for exercise adherence in adults: systematic review and meta-analysis. American journal of preventive medicine 45, 658-667 (2013). Lagarde, M., Haines, A. & Palmer, N. Conditional cash transfers for improving uptake of health interventions in low-and middle-income countries: a systematic review. Jama 298, 1900-1910 (2007). Baker, T. B. et al. A randomized controlled trial of financial incentives to low income pregnant women to engage in smoking cessation treatment: Effects on post-birth abstinence. (2018). Fraser, D. L. et al. A Randomized Trial of Incentives for Smoking Treatment in Medicaid Members. American journal of preventive medicine 53, 754-763 (2017). Cahill, K. & Perera, R. Competitions and incentives for smoking cessation. Cochrane database of systematic reviews (2011). Crossland, N., Thomson, G., Morgan, H., Dombrowski, S. U. & Hoddinott, P. Incentives for breastfeeding and for smoking cessation in pregnancy: An exploration of types and meanings. Social Science & Medicine 128, 10-17 (2015). Adams, J., Giles, E. L., McColl, E. & Sniehotta, F. F. Carrots, sticks and health behaviours: a framework for documenting the complexity of financial incentive interventions to change health behaviours. Health psychology review 8, 286-295 (2014). Giles, E. L., Robalino, S., McColl, E., Sniehotta, F. F. & Adams, J. The effectiveness of financial incentives for health behaviour change: systematic review and meta-analysis. PloS one 9, e90347 (2014). Moher, D., Liberati, A., Tetzlaff, J. & Altman, D. G. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Annals of internal medicine 151, 264-269 (2009). Schardt, C., Adams, M. B., Owens, T., Keitz, S. & Fontelo, P. Utilization of the PICO framework to improve searching PubMed for clinical questions. BMC medical informatics and decision making 7, 16 (2007). Green, S. & Higgins, J. (Version, 2005). Higgins, J. P. et al. The Cochrane Collaboration’s tool for assessing risk of bias in randomised trials. Bmj 343, d5928 (2011). Fadzrul Hafiz Johani et.al. https://doi.org/10.32827/ijphcs.6.2.113 127 IJPHCS Open Access: e-Journal 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 International Journal of Public Health and Clinical Sciences e-ISSN : 2289-7577. Vol. 6:No. 2 March/April 2019 Borenstein, M., Hedges, L. V., Higgins, J. P. & Rothstein, H. R. A basic introduction to fixed‐effect and random‐effects models for meta‐analysis. Research synthesis methods 1, 97-111 (2010). Review Manager Version 5.3 v. Version 5.3 ( The Cochrane Collaboration, Copenhagen: The Nordic Cochrane Centre, 2014). Ockene, J. K. et al. The physician-delivered smoking intervention project: can short-term interventions produce long-term effects for a general outpatient population? Health Psychology 13, 278 (1994). Brunette, M. F. et al. Randomized Trial of Interventions for Smoking Cessation Among Medicaid Beneficiaries With Mental Illness. Psychiatric Services 69, 274-280 (2017). Etter, J.-F. & Schmid, F. Effects of large financial incentives for long-term smoking cessation: a randomized trial. Journal of the American College of Cardiology 68, 777-785 (2016). Lasser, K. E. et al. Effect of patient navigation and financial incentives on smoking cessation among primary care patients at an urban safety-net hospital: a randomized clinical trial. JAMA internal medicine 177, 1798-1807 (2017). Kendzor, D. E. et al. Financial incentives for abstinence among socioeconomically disadvantaged individuals in smoking cessation treatment. American journal of public health 105, 1198-1205 (2015). XE. XE Currency Converter: CHf to USD, <https://www.xe.com/currencyconverter/convert/?Amount=1&From=CHF&To=USD> (2012). Munafò, M. R., Clark, T. G., Johnstone, E. C., Murphy, M. F. G. & Walton, R. T. The genetic basis for smoking behavior: A systematic review and meta-analysis. Nicotine & Tobacco Research 6, 583-597, doi:10.1080/14622200410001734030 (2004). Ohmoto, M. et al. Association between dopamine receptor 2 TaqIA polymorphisms and smoking behavior with an influence of ethnicity: a systematic review and meta-analysis update. Nicotine & Tobacco Research 15, 633-642 (2012). Kubota, T. & Yokoyama, A. in Clinical Relevance of Genetic Factors in Pulmonary Diseases 77-91 (Springer, 2018). Lally, P., Van Jaarsveld, C. H., Potts, H. W. & Wardle, J. How are habits formed: Modelling habit formation in the real world. European journal of social psychology 40, 998-1009 (2010). Prochaska, J. O. & DiClemente, C. C. in Treating addictive behaviors 3-27 (Springer, 1986). Prochaska, J. O., DiClemente, C. C. & Norcross, J. C. In search of how people change: applications to addictive behaviors. American psychologist 47, 1102 (1992). Prochaska, J. O. & Velicer, W. F. The transtheoretical model of health behavior change. American journal of health promotion 12, 38-48 (1997). Gneezy, U., Meier, S. & Rey-Biel, P. When and why incentives (don't) work to modify behavior. Journal of Economic Perspectives 25, 191-210 (2011). DeFulio, A. & Silverman, K. The use of incentives to reinforce medication adherence. Preventive medicine 55, S86-S94 (2012). Operario, D., Kuo, C., Sosa-Rubí, S. G. & Gálarraga, O. Conditional economic incentives for reducing HIV risk behaviors: integration of psychology and behavioral economics. Health Psychology 32, 932 (2013). Fadzrul Hafiz Johani et.al. https://doi.org/10.32827/ijphcs.6.2.113 128 IJPHCS Open Access: e-Journal 48 49 50 51 International Journal of Public Health and Clinical Sciences e-ISSN : 2289-7577. Vol. 6:No. 2 March/April 2019 Haff, N. et al. The role of behavioral economic incentive design and demographic characteristics in financial incentive-based approaches to changing health behaviors: a meta-analysis. American Journal of Health Promotion 29, 314-323 (2015). Klapdor, M. & Grove, A. ‘No Jab No Pay’ and other immunisation measures, <https://www.aph.gov.au/About_Parliament/Parliamentary_Departments/Parliamentary_Li brary/pubs/rp/BudgetReview201516/Vaccination> (2015). Davies, G. & J Soundy, T. The genetics of smoking and nicotine addiction. Vol. Spec No (2009). Begg, C. et al. Improving the quality of reporting of randomized controlled trials: the CONSORT statement. Jama 276, 637-639 (1996). Fadzrul Hafiz Johani et.al. https://doi.org/10.32827/ijphcs.6.2.113 129 IJPHCS International Journal of Public Health and Clinical Sciences e-ISSN : 2289-7577. Vol. 6:No. 2 March/April 2019 Open Access: e-Journal Supplemental Table 1: Detail assessment on risk of bias Study Kendzor et al. 2015 Random sequence generation Allocation concealment Blinding of participants and personnel Blinding of outcome assessment Incomplete outcome data Selective reporting Other bias Unclear Risk Unclear Risk High Risk Low Risk Low Risk Low Risk Low Risk Participants recruited during their visit to the Tobacco Cessation Clinic at the Dallas County, Texas Eligible participants were randomly assigned Both participant and personnel were not blinded primary outcome assessed using biochemically confirmed CO level No loss of follow up for 4 weeks intervention period outcomes that are of interest in the review have been reported in the pre-specified way. none Etter&Schmid Unclear Risk Procedure of 2016 Low Risk High Risk Low Risk Low Risk Low Risk Low Risk Participants could not be blinded. Researchers were not blinded primary outcome was verified by carbon monoxide and either cotinine or thiocyanate measurements No Participants were excluded from analysis outcomes that are of interest in the review have been reported in the pre-specified way. none randomization was not explained allocation using sealed opaque envelope drawn by participants. Fraser et al. 2017 Low Risk Unclear Risk High Risk Low Risk Low Risk Low Risk High Risk Randomization occurred via computer- generated lists Allocation of participant to treatment or control group not explained Counsellors at the WTQL were not blinded. Participant in incentive group were not blinded primary outcome was verified by biochemical evidence using the intent-to-treat principle outcomes that are of interest in the review have been reported in the pre-specified way. Clinics choose different cut-scores although testing method and cut-score was the same for initial screening and follow-up Brunette et al. 2017 Low Risk Unclear Risk High Risk Low risk Low risk Low Risk Low Risk Computer-generated tables for each stratum within each site were used for random assignment Participants were randomly assigned to receive incentives for biologically verified abstinence or no incentives Both participant and personnel were not blinded biologically confirmed with expired breath carbon monoxide and urine cotinine Missing observations were imputed as smoking outcomes that are of interest in the review have been reported in the pre-specified way none Fadzrul Hafiz Johani et.al. https://doi.org/10.32827/ijphcs.6.2.113 130 IJPHCS International Journal of Public Health and Clinical Sciences e-ISSN : 2289-7577. Vol. 6:No. 2 March/April 2019 Open Access: e-Journal Study Random sequence generation Allocation concealment Blinding of participants and personnel Blinding of outcome assessment Incomplete outcome data Selective reporting Other bias Lasser et al. 2017 Low Risk Low Risk High Risk Low Risk Low Risk Low Risk High Risk randomized participants using a random number generator using sealed envelopes Unblinded study (participant and researcher were unblinded) assessment using biochemically confirmed with saliva cotinine or urine anabasine test Using intention-to-treat analysis outcomes that are of interest in the review have been reported in the pre-specified way. Baseline smoking status was not confirmed biochemically Baker et al. 2018 Low Risk Low Risk High Risk Low Risk Low Risk Low Risk High Risk First Breath staff used randomization tables prepared by the UW-CTRI to randomize women upon consent. Separate computer determined randomization tables were used Both participant and personnel were not blinded assessment using biochemically confirmed CO level Participants with missing data for the primary outcome were counted as smoking outcomes that are of interest in the review have been reported in the pre-specified way. Smoking status was not confirmed biochemically at baseline Fadzrul Hafiz Johani et.al. https://doi.org/10.32827/ijphcs.6.2.113 131