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