Community-based model for preventing tobacco use among disadvantaged adolescents in urban slums of India.
ABSTRACT Tobacco consumption in multiple forms presents an emerging, significant and growing threat to the health of Indian adolescents, especially those from low socio-economic communities. Research in two phases was undertaken among economically disadvantaged adolescents in two urban slums of Delhi. In phase I, qualitative research methods such as focus group discussions and in-depth interviews were used to explore and understand the determinants influencing tobacco use among these adolescents. Prevalence of tobacco use was higher among boys than girls. Adolescents reported using tobacco in multiple forms, chewing tobacco being the most popular. Peer pressure, easy availability and affordability were important reasons associated with tobacco initiation and continued use. Though they had some knowledge about the harmful effects of tobacco, this was not sufficient to motivate them to abstain or quit. The community-based intervention model developed on the basis of the results of phase I was evaluated in phase II in a demonstration study with two slum communities. One was treated as the intervention and the other as control. A significant difference in current use of tobacco was observed between the study groups (p = 0.048), with the intervention group showing a reduction in use, compared with an increase in use among the control group. Post-intervention, the intervention group reported significantly lower fresh uptake (0.3%) of tobacco use compared with the control group (1.7%). No significant change was found for quit rate (p = 0.282) in the two groups. Community-based interventions can be effective in preventing adolescents from initiating tobacco use in a low-resource setting such as India.
- SourceAvailable from: Harshal Salve[Show abstract] [Hide abstract]
ABSTRACT: ijmedph Vol 4 | Issue 4 | Oct-Dec 2014 www.ijmedph.org SciBiolMed.Org -A non-profit private organization dedicated to Research in the field of Science, Biology and Medicine. It provides high quality, accurate and required information to enhance re-search and innovative concepts in scholarly publishing. Editor : Dr. Subhankar Chakraborty (USA) The journal is indexed with CAB Abstracts, An overview of qualitative research methodology for public health researchers Role of biomedical and behavioral interventions and their evidence in prevention of HIV infection: A literature review Indian public health standards for Ayurveda, Yoga and Naturopathy, Unani, Siddha and Homoeopathy facilities: An assessment Osteoporotic hip fractures in low-income group population, hospital based case control study from India A study of the drugs used in chronic obstructive pulmonary disease and their impact on quality-of-life more in this issue. Determinants of tobacco use and perception, attitude about an antitobacco act in rural Haryana, North India Background: Tobacco use is one of the most important preventable causes of morbidity and mortality in India. It is essential to study perception, practices and factors determining tobacco use to formulate the intervention for addressing this problem in the community. Materials and Methods: A cross-sectional study was carried out in 28 villages in Ballabgarh block of Faridabad, Haryana. This study aimed to understand factors determining tobacco use and to assess knowledge, attitude, practices about tobacco use and antitobacco act. Systematic random sampling was done to select study subjects. Total calculated sample size was 880. One eligible male and one female were selected randomly from household. Knowledge was assessed using semi-structured interview schedule while attitude using fi ve-point likert scale. In addition, focus group discussions and in-depth interviews were conducted among various stakeholders and opinion leaders to get insight about practices and factors determining tobacco use in the community. Results: Total 892 subjects were enrolled in the study, of which 51% were male. The mean age of the study subjects was 49.5 years (standard deviation: 17.5). Though awareness about harmful effects of tobacco use was reported, awareness about legislations under an antitobacco act was poor. Early initiation and continued use of all forms of tobacco were reported. Prevalent practice of tobacco use was attributed to pressure from peer groups, social customs and lack of de-addiction services at the community level by study subjects. Conclusion: Tobacco use in this rural community was mainly attributed to social and cultural factors. A multi-pronged public health approach is needed for addressing this complex problem the community.International J Medicine and Public Health. 10/2014; 4(4).
- [Show abstract] [Hide abstract]
ABSTRACT: Nicotine use and abuse is gaining increasing attention due to its negative and serious medical consequences. Multiple morbidities occur due to the intake of nicotine in various forms. To find the prevalence and type of nicotine use/abuse in females. House to house survey in a village in Purvanchal, i.e. eastern part of Uttar Pradesh and bordering Bihar. Semi-structured performa was used to collect data; a house to house visit was made to collect data. Simple percentages were calculated. Tooth powder form of nicotine use is common and the need to address this problem is urgent. Widespread dissemination of knowledge and legislative measures have to be undertaken to stop the problem.Industrial psychiatry journal 07/2010; 19(2):125-9.
- [Show abstract] [Hide abstract]
ABSTRACT: Self-reported tobacco use among young people can underestimate the actual prevalence of tobacco use. Biochemical validation of self-reports is particularly recommended for intervention studies where cessation outcomes are to be measured. Literature on biochemical validation of self-reports of multiple forms of tobacco use in India is sparse, particularly among young people. The study was conducted during the baseline household survey of a community-based tobacco prevention and cessation intervention trial for youth (10-19 years old) residing in slum communities in Delhi, India in 2009. Salivary cotinine measurement on 1,224 samples showed that youth were under-reporting use of chewing and smoking tobacco. Self-reports had a low sensitivity (36.3%) and a positive predictive value of 72.6%. No statistically significant difference in under- reporting was found between youth in the control and intervention conditions of the trial, which will be taken into consideration in assessing intervention outcomes at a later time point. Biochemical validation of self-reported tobacco use should be considered during prevention and cessation studies among youth living in low-income settings in developing countries like India. Impact: The future results of biochemical validation from Project ACTIVITY (Advancing Cessation of Tobacco in Vulnerable Indian Tobacco Consuming Youth) will be useful to design validation studies in resource-poor settings.Asian Pacific journal of cancer prevention: APJCP 01/2011; 12(10):2551-4. · 1.50 Impact Factor
Community-based model for preventing tobacco
use among disadvantaged adolescents in urban
slums of India
MONIKA ARORA1,2*, ABHA TEWARI1, VIKAL TRIPATHY1, GAURANG
P. NAZAR1, NEERU S. JUNEJA1, LAKSHMY RAMAKRISHNAN3and
K. SRINATH REDDY2
1HRIDAY (Health Related Information Dissemination Amongst Youth), New Delhi, India2PHFI
(Public Health Foundation of India), New Delhi, India and3AIIMS (All India Institute of Medical
Sciences), New Delhi, India
*Corresponding author. E-mail: email@example.com
Tobacco consumption in multiple forms presents an
emerging, significant and growing threat to the health of
Indian adolescents, especially those from low socio-econ-
omic communities. Research in two phases was under-
taken among economically disadvantaged adolescents in
two urban slums of Delhi. In phase I, qualitative research
methods such as focus group discussions and in-depth
interviews were used to explore and understand the deter-
minants influencing tobacco use among these adolescents.
Prevalence of tobacco use was higher among boys than
girls. Adolescents reported using tobacco in multiple
forms, chewing tobacco being the most popular. Peer
pressure, easy availability and affordability were impor-
tant reasons associated with tobacco initiation and contin-
ued use. Though they had some knowledge about the
harmful effects of tobacco, this was not sufficient to
motivate them to abstain or quit. The community-based
intervention model developed on the basis of the results
of phase I was evaluated in phase II in a demonstration
study with two slum communities. One was treated as the
intervention and the other as control. A significant differ-
ence in current use of tobacco was observed between the
study groups (p ¼ 0.048), with the intervention group
showing a reduction in use, compared with an increase in
use among the control group. Post-intervention, the inter-
vention group reported significantly lower fresh uptake
(0.3%) of tobacco use compared with the control group
(1.7%). No significant change was found for quit rate
(p ¼ 0.282) in the two groups. Community-based inter-
ventions can be effective in preventing adolescents from
initiating tobacco use in a low-resource setting such as
Key words: tobacco control; adolescents; community intervention; low-SES
Socio-economic inequality is widely prevalent
in India, especially in urban areas (Perry, 2006).
linked to several behaviours that influence
health, and tobacco use is one such behaviour
(Jarvis and Wardle, 2006). The rate of smoking
is expected to rise by about 3% per year and
will cause around 1 million deaths annually in
2010 in India (Jha et al., 2008). Mortality due to
tobacco disproportionately affects young adults.
India, with more than 400 million adolescents
(Singh, 2005), has the highest number of adoles-
cents in the world (UNICEF, 2004). Sixty to
80% of the children in India live in low-
resource settings (PLAN, 2005) and consume
myriad varieties of tobacco products [Centers
Health Promotion International, Vol. 25 No. 2
Advance Access published 27 February, 2010
# The Author (2010). Published by Oxford University Press. All rights reserved.
For Permissions, please email: firstname.lastname@example.org
at Emory University on July 12, 2010
for Disease Control and Prevention (CDC)
Thus, adolescents are a highly vulnerable group
in a developing country such as India.
India is the second highest consumer of
tobacco in the world, after China. Currently,
about 230 million males and 11.9 million
females consume tobacco in India (Shafey et al.,
2009). Strategies designed to alleviate tobacco
use in India are challenged by the fact that
tobacco is consumed in multiple forms such as
smoked and smokeless forms. Owing to the
variety of tobacco products consumed in India,
varying health consequences are encountered,
e.g. higher rate of oral cancer (World Health
Organization, 1997) and increased incidence of
tuberculosis (TB) (Gajalakshmi et al., 2003).
The economic cost of treating four major
tobacco-related diseases—cancer, cardiovascular
diseases, respiratory diseases and TB in India is
as high as US$1.7 billion (John et al., 2009),
which does not include the indirect and other
social costs associated with these diseases.
Tobacco use poses a major public health
threat particularly for adolescents in India, with
the current prevalence of tobacco use being
14% among 13–15-year-old school-going youth
as determined by the Global Youth Tobacco
Survey (GYTS) (CDC website: http://www.cdc.
India_factsheet.htm). The risk is particularly
high for adolescents belonging to the lower
socio-economic strata (SES), as suggested in
another school-based study, ‘Mobilising Youth
for Tobacco Related Initiatives’ (MYTRI). This
study, conducted among 10–16-year-old stu-
dents in Delhi and Chennai, found the preva-
lence of tobacco use to be twice as high among
government schools (low SES) when compared
with private schools (high SES) (Mathur et al.,
The higher rates of tobacco use reported
among adolescents living in urban, low-SES
areas in India (Nichter et al., 2004; Mishra
et al., 2005; Mohan et al., 2005; Stigler et al.,
2006) may be due to increased uptake by them
or less successful quit attempts (Bobak et al.,
2000; Jarvis and Wardle, 2006). Tobacco use
among adolescents is influenced by multiple
aetiological factors, including individual, socio-
cultural and environmental factors (Poland
et al., 2006). Further research is required to
delineate the complex aetiology of tobacco use
among the adolescents living in urban slums in
India to effectively prevent tobacco uptake and
Evidence from developed countries suggests
that programme- and policy-based interventions
have been successful in reducing the prevalence
of tobacco use, though most of the reductions
have been among the affluent class relative to
the poor (Jarvis and Wardle, 2006). As child-
hood and adolescence is the period when exper-
imentation with tobacco products is mostly
reported, interventions targeting adolescents,
both users and non-users, can be highly effec-
tive in tobacco control, though it is unclear how
best to intervene among this age group. There
are no published tobacco prevention and cessa-
tion studies among socio-economically disad-
vantaged adolescents in developing countries,
A study was thus undertaken to assess the
efficacy of a community-based intervention
model in a low-SES community in India, focus-
environmental factors contributing to tobacco
use. An evidence-based ‘best practices’ model
developed by the Canadian Tobacco Control
Research Initiative (CTCRI) with the CDC
(Milton et al., 2004) was adapted so that it
would be culturally appropriate for India. This
paper reports the results of the formative
research and a multi-component demonstration
study conducted with the following aims:
(1) To explore the aetiology of tobacco use in
multiple forms among adolescents from low
SES using qualitative research methods.
component community-based intervention
groups) for the prevention of uptake and
cessation of tobacco use among adolescents
from low SES, using quantitative research
Study subjects and design
The study was conducted in two phases. In
phase I (formative phase), extensive qualitative
work was conducted in two low-SES commu-
nities of Delhi in the year 2005. An attempt was
made to understand the determinants that
144 M. Arora et al.
at Emory University on July 12, 2010
trigger the onset of tobacco use, dynamics of
regular use of tobacco and factors that motivate
the user to quit tobacco. Ethics clearance for
this study was sought from Independent Ethics
Committee in Mumbai.
consent was taken from all the participants and
their parents or guardians.
Five focus group discussions (FGDs) were
conducted with 37 young persons (boys 31; girls
6), aged 10–19 years living in two urban slums
in Delhi. Out of these, two FGDs were con-
ducted with groups of migrant self-employed
boys living in night shelters, two with groups of
boys residing with family members and one
with girls residing with family members.
In-depth interviews were conducted with mul-
tiple stakeholder groups to assess the demand
for tobacco cessation in this age group. These
included three community leaders and health
professionals at two tobacco cessation clinics in
In phase II (intervention phase), two low-SES
communities in Delhi were identified, other than
the ones selected in phase I. Each of them had a
slum area and a low-SES housing colony (reset-
tlement colony) as a combined unit. This was a
demonstration study, with one community serving
as the intervention community and other as the
control community. These communities were
recruited on the basis of comparable socio-
demographic characteristics. Locally credible non-
government organizations (NGOs) working in
each of these communities were identified to
provide a channel for establishing rapport.
A questionnaire was developed on the basis of
the findings of the qualitative research in phase I
to assess knowledge, behaviour and attitudes
towards tobacco consumption and cessation. The
questionnaire at baseline was administered in
Hindi (local language) in September 2006 before
the implementation of the intervention and was
repeated after 1 year with the same sample at
endline to assess the efficacy of the intervention.
Since the sample comprised both literate and illit-
erate respondents, the questionnaire was adminis-
tered by a trained interviewer as per the
convenience of the respondents.
All adolescents aged 10–19 years (tobacco
users and non-users) living in the selected
households of the two communities were eli-
gible for the survey. Cluster sampling and
random sampling techniques were used to
identify the households in the communities. In
the first stage of sampling, blocks were selected
(based on the density of population in the
blocks) in each of the two communities to
provide the required sample size. Nine out of 21
blocks in the intervention community and 1 out
of 3 blocks in the control community were
selected. A random sampling technique was
used to select 772 and 834 households with 2154
and 2204 eligible adolescents, respectively, in
the intervention and control communities. Out
of these eligible adolescents, 1229 during base-
line and 1162 during endline were surveyed in
the intervention community, and 1152 at base-
line and 1083 at endline were surveyed in the
control community. Confidentiality
survey results was maintained.
A community-based multi-component interven-
tion model was developed, implemented and
evaluated in the intervention community. The
intervention model addressed numerous socio-
environmental and intra-personal factors that
trigger the onset of tobacco use and other
determinants that influence tobacco cessation.
Peer leaders, adult community leaders and
NGO personnel were identified and trained to
facilitate the intervention
intervention comprised interactive activities,
pre-tested posters, audio and video films, lec-
tures, street plays; in addition, pictorial hand-
outs, booklets and pamphlets were distributed
for knowledge enhancement by trained youth
peer leaders and community leaders. An aware-
ness rally was organized to reach and sensitize
the masses and positively influence tobacco use
norms in the community.
Multiple sessions (three to seven) of four
interactive activities were conducted at different
strategic locations in the community to dissemi-
nate information about the prevention of uptake
and cessation of tobacco use. On an average,
40–50 adolescents attended each session. Data
on process measures were collected to assess the
fidelity with which the intervention components
The two outcome measures used from the
‘current use’ of tobacco at baseline and endline.
‘Ever users’ were those who had consumed
tobacco in any form in their life time. ‘Current
were ‘everuse’ and
Community-based model for preventing tobacco use145
at Emory University on July 12, 2010
users’ were those who had consumed tobacco in
any form in the past 1 week. Current use at
baseline and endline were compared to calcu-
late the ‘quit rate’ and ‘rate of fresh uptake’.
Quit rate of tobacco use among the commu-
nities was estimated as the proportion of adoles-
cents who became non-users at endline from
among those who were current users in the
baseline survey. Rate of fresh uptake was esti-
mated as the proportion of adolescents who
reported themselves as ever users or current
users at endline from among those who were
never users at the baseline survey.
To validate self-reported tobacco consump-
tion or abstinence, salivary cotinine levels of a
subsample (25% of the sample at baseline and
endline) were also measured. Salivary cotinine
was biochemically analysed by enzyme-linked
immunosorbent assay (ELISA) using kits from
During phase I of the study, FGDs and in-depth
interviews were recorded and transcribed. The
transcripts were read repeatedly, and thematic
analysis was performed on the qualitative data.
With regard to phase II, which involved quanti-
tative data, comparisons between groups were
done using the x2test. Repeated measure logistic
regression was used to compare the control and
intervention communities over the study period.
An interaction model between year and commu-
nity was used to assess the difference between
the communities over the period of the study.
Logistic regression was used to calculate the odds
of fresh uptake and quitting. SAS 9.1 version was
used for the analysis of quantitative data.
Phase I (formative phase)
Socio-demographic description of participants
The participants were boys and girls in the age
group of 10–19 years. Most of them had migrated
from neighbouring states and villages. They had
left home for various reasons such as to escape
violence at home and to earn a livelihood. The
majority of participants were illiterate and
engaged in part-time/non-regular employment
such as rag picking, carrying lamps at wedding
processions and caterers. Their monthly income
ranged from INR 300–4000 (US$7–90) per
month. They usually obtained free food from reli-
gious institutions and used their earnings to
support their multiple addictions. The following
determinants were identified as being associated
with tobacco initiation, regular use and cessation.
Determinants that define initiation of tobacco
The participants reported that they started
using tobacco at a very early age as early as 6
years and used tobacco regularly. Some of them
reported initiation of tobacco use after they had
left alcohol, ganja (a form of cannabis) or other
I have been smoking bidi (made by rolling a dried,
rectangular piece of tenburni leaf [Diospyros mela-
noxylon] with sundried flaked tobacco) for the past
four years, a boy said that it feels good and after that
I got addicted to it (16-year-old boy).
When asked about the prevalence of the
types of addiction among participants, the
group responded unanimously and said,
We smoke bidi, cigarette, chew gutkha (a preparation of
crushed betel nut, tobacco, catechu, lime and sweet or
savory flavorings), drink alcohol and beer, bhang
(Indian hemp) and ganja and also smell solution (white
correction fluid containing toluene) (Group’s view).
Different social, environmental and personal
factors were cited for the onset of tobacco con-
sumption among participants. Some of these
included: imitating elders (particularly family
members), peer group influence and to have
‘fun’. Some of the respondents stated that they
had started using gutkha after their friends or
family members offered it to them. In many
cases, parents and family members also consumed
tobacco, both in chewed and smoked forms.
I started all these addictions when I was staying at
home. I started bidi first. I smoke two packets of ciga-
rettes in a day (16-year-old boy).
I started because my friends smoked. I didn’t know
how to smoke a bidi, my friend taught me and then
I learned how to smoke it. The taste of bidi is like
Pepsi (12-year-old boy).
Determinants that define regular use of tobacco
Accessibility and affordability. The majority of
participants mentioned that tobacco is easily
available everywhere, i.e. at roadside stalls,
general stores and paan (betel-leaf) shops.
146 M. Arora et al.
at Emory University on July 12, 2010
Since myriad varieties of tobacco products are
available, they are available over a wide price
range. Money is not a barrier for most of the
children as they earn a minimum of INR 50
(US$1.11) per day and can easily afford to buy
cheap tobacco products such as bidi and gutkha
[one packet of bidi/gutkha costs INR 1–7
(approximately US cents 2–14).
Consequences of tobacco use. Most of the
participants reported awareness of the negative
consequences of tobacco use on health, for
example tooth decay, TB and discoloration of
the teeth. They could not, however, distinguish
consequences. Almost all of them believed that
tobacco is injurious to health.
Cigarette has harmful ingredients and all. We also get
many diseases because of this. We feel burning sen-
sation after chewing gutkha. This gives us the feeling
of high (Group’s view).
Knowledge of tobacco control. Most of the
participants were aware of the Tobacco Control
Act in India.
The sale of bidi is banned below the age of 18 years. I
saw at India Gate that children less than 18 years do
not get any bidi or gutkha. It is written on the board
Social influences. The participants reported that
they get influenced by their peer group. Most of
them admitted that they find it difficult to resist
such peer group influence.
I take gutkha daily, I started because I saw villagers
taking this. When I smoked bidi for the first time,
I fainted. Gutkha is better than bidi and I take 4–5
packets of gutkha daily and it is also easily available at
the shop (13-year-old boy).
Determinants related to quitting behaviour
Lack of motivation to quit. The children were
not confident of having the skills to resist peer
group influences and they expressed a minimal
desire to quit. Some of them had attempted to
quit but were unsuccessful and started another
I tried to quit and I tried this continuously for 3 days.
One day I didn’t take but on second day I felt like
having it. I am not able to quit this addiction and
I started it again (10-year-old boy).
Perspectives of health professionals (tobacco
cessation experts) and community leaders
experts (clinical psychologist and pulmonologist)
revealed that no youth in the age group of 10–19
years visits tobacco cessation centres to avail the
facilities. Health professionals and community
leaders reported that tobacco consumption is
common among both adults and young persons
in the community and they all felt the need
for an extensive community development pro-
gramme to prevent tobacco uptake and promote
tobacco cessation among youth.
with tobacco cessation
Phase II (intervention phase)
The prevalence rates of ever use and current use
of tobacco at baseline and endline in the two
communities are presented in Table 1. Prevalence
of both current use and ever use increased in the
control community and decreased in the inter-
vention community (Fig. 1) at endline. At base-
line, both current use and ever use did not differ
However, at endline, the intervention group
showed significantly lower prevalence of tobacco
use among the males as well as the total sample.
Repeated measure logistic regression showed that
the intervention group showed significant decline
in current use for males and the total sample. The
prevalence for both ever use and current use was
significantly higher in boys compared with girls.
Classification of adolescents on the basis of
the salivary cotinine level among self-reported
users and non-users is presented in Table 2.
Adolescents were classified as tobacco user if
the amount of cotinine present in the saliva was
more than or equal to 10 ng/ml. Of the adoles-
cents in the intervention and control commu-
non-users, 13–17% were classified as users on
the basis of the cotinine level at baseline and
endline. False negative reporting of tobacco use
was not significantly different between the two
endline, p ¼ 0.354). In contrast, 0–21% of ado-
lescents were classified as non-users on the
basis of the cotinine level among those adoles-
cents who self-reported that they were users.
Rate of fresh uptake and quit rates for the two
communities are presented in Tables 3 and 4,
respectively. The rate of fresh uptake of tobacco
differed significantly between the intervention
p ¼ 0.551;for
Community-based model for preventing tobacco use147
at Emory University on July 12, 2010
and control communities. Out of 1055 never
users at baseline in the intervention community,
only 3 (0.28%) reported experimenting with
tobacco products during the intervention period.
In contrast, among the control community, 16
(1.67%) out of 958 never users at baseline
reported experimenting with tobacco at endline.
Of these experimenters, 15 were boys. The odds
Fig. 1: Current use and ever use among the control and intervention community at baseline and endline.
*p-Value for the difference between baseline and endline based on x2test; **p-Value for interaction effect
between community and year in a repeated measure logistic regression.
Table 1: Prevalence of current use and ever use of any tobacco product at baseline (2006) and endline (2007)
by study condition
TotalN % (CI)TotalN % (CI)
57 8.95 (6.73–11.16)
All comparisons between males and females significant at p , 0.0001.
Repeated measure logistic regressions (GEE coefficients for year and community interaction effect presented; estimates for
intervention community at endline with reference to the rest). Current use—male: estimate 20.307 (p-value 0.048); female:
estimate 20.306 (p-value 0.058). Total: estimate 20.318 (p-value 0.045).
Ever use—male: estimate 20.222 (p-value 0.074); female: estimate 20.210 (p-value 0.112). Total: estimate 20.231 (p-value
Bold p values indicate that though there was no difference between the intervention and control groups at baseline, there
was significant difference (p , 0.05) between these groups at the endline (This is true for males and Totals).
148 M. Arora et al.
at Emory University on July 12, 2010
(OR ¼ 5.96, p ¼ 0.005) of fresh uptake in the
control community was significant compared
with the intervention community.
Out of 32 current users at baseline in the
intervention community, 4 (11.11%) reported cur-
rently not using tobacco at endline. In the control
community, only 2 (4.55%) among the 42 tobacco
users at baseline reported not using tobacco at
endline. Quit rate was higher in the intervention
community but was not significantly different
from that of the control community (p ¼ 0.282).
Ever users were asked questions on the
different forms of tobacco products that they
had used from a list of 11 different forms. More
than 55% of ever users had used tobacco in
multiple forms (Table 5). Table 6 presents
tobacco use in smoking and chewable forms
among boys and girls in the two surveys. Girls
reported the use of only chewable forms of
tobacco products. Boys reported the use of both
smoking and chewable forms. Almost 50% of
ever user boys had used only chewable pro-
ducts, whereas one-third of the boys reported
having used both the products.
In both developed and developing countries,
tobaccocessation andcontrol have been
Table 3: Rate of fresh uptake among the intervention and control communities
Ever use of
Ever use of
Fisher’s exact test p-value ¼ 0.465
0.008 5.411 (1.56–18.81)
*Odds of fresh uptake in control community with reference to the intervention community.
Table 2: Salivary cotinine level and self-reported tobacco use in the two communities at baseline and endline
Reported current use Reported current use
NoYes TotalNo YesTotal
Cotinine levels Tobacco use status Tobacco use status (binary)
Cotinine level undetected (0 ng/ml)Non-user; absence of passive and active
Exposure to tobacco smoking; workout
Infrequent active smoking/tobacco use or
heavy exposure to passive smoking
,10 ng/ml Non-user
10–100 ng/mlTobacco user
.100 ng/ml Tobacco user
aComparing false negative reporting between control and intervention using x2test.
Bold values are false negative reporting rate between the intervention and control groups at baseline and endline.
Community-based model for preventing tobacco use149
at Emory University on July 12, 2010
identified as areas on which public health policy
should focus to bring down the huge burden of
diseases that tobacco causes globally. It is
important to focus efforts on both tobacco pre-
vention and cessation among adolescents, since
this is the age when tobacco initiation has
mostly been reported.
Successful community-based, peer-led, multi-
component tobacco intervention programmes
have been undertaken with adolescents in devel-
oped countries (Pentz et al., 1989; Perry et al.,
1989). In India, a community-based, multi-
component, peer-led, tobacco prevention inter-
vention has been implemented for the first time in
this demonstration study. School-based group ran-
domized trials such as the MYTRI in India have
reduced tobacco use in school students receiving
the intervention, compared with increased use in
control school students (Perry et al., 2009).
The qualitative research in phase I of this
study revealed a high usage of tobacco products
in multiple forms among adolescents in the two
study communities, which is in line with the
Indian component of the GYTS (http://www.cdc
The age at initiation of tobacco products in
this study is reported to be 6 years, which is
much lower than the 10 years reported earlier
in the Indian literature (Patel, 1999). This is
alarming, as starting tobacco use at such an
early age would further add to health inequality
among the disadvantaged population in India.
Chewing of tobacco products is more prevalent
among adolescents. Boys reported the use of
both smoked and smokeless forms of tobacco
reported both in the qualitative and quantitative
phases of this study. Girls self-reported the use
of only chewing forms. Chewing tobacco is not
overtly visible, and the social stigma attached to
smoking, especially among girls, can be avoided
if tobacco is chewed. Moreover, in India,
chewing forms of tobacco are available in small
packs which can be easily hidden from teachers
andparents or other adults.
tobacco use in the survey questionnaire also
Table 5: Multiple forms of tobacco use among the
ever users of tobacco products
Frequency Per cent Frequency Per cent
aTotal of 11 forms of tobacco products were listed on the
survey. These included cigarette, bidi, chilum, hukka,
zarda, khaini, gutkha, paan with tobacco and snuff.
A maximum of 9 out of 11 were reported to be used by
Table 6: Use of smoking and chewable forms of
tobacco among boys and girls
Baseline Chewable and smoking
Chewing forms only
Smoking forms only
Chewable and smoking
Chewing forms only
Smoking forms only
Table 4: Quit rate among the intervention and control communities
p-value *Odds ratio
use at endline
use at endline
*Odds of quitting in control community with reference to the intervention community.
150 M. Arora et al.
at Emory University on July 12, 2010
highlighted that tobacco is consumed in mul-
tiple forms by adolescents. Almost 50% of
respondents reported using tobacco in more
than one form as noted in phase II of this study.
Qualitative data from phase I of this study indi-
cated a very high prevalence of tobacco use
(90%), which differed from the results of the
survey questionnaire in phase II (self-reported
tobacco use 5.7% at the baseline survey).
Self-reported tobacco use among these adoles-
cents was validated by salivary cotinine levels,
and highlighted high levels of under-reporting
among adolescents during the survey. About
one in seven self-reported non-users was classi-
fied as a user on the basis of the cotinine level.
earlier in a large study conducted in the USA,
UK and Poland (West et al., 2007), which forms
the basis for the biochemical validation of self-
reported smoking status.
The present study highlights the fact that ado-
lescents belonging to low SES are more prone to
early initiation of tobacco use because of several
limited knowledge of the negative consequences
of tobacco use, peer pressure and acceptance of
tobacco consumption by adults and elders, a
finding which is consistent with another study
conducted in Delhi and Chennai in India
(Mathur et al., 2008). This leads to unchecked
tobacco consumption among adolescents resid-
ing in low-resource settings, further adding to
health inequality. These determinants provide
adolescents with an enabling environment that
aids conversion of experimentation to regular
use of tobacco. As several of these adolescents
start working at an early age, easy accessibility
and affordability of tobacco products encourages
their tobacco habit. Access to disposable income
has been found to be a risk factor for tobacco
use in other studies in India (Nichter et al.,
2004). Lack of cessation facilities was also found
to be a determinant that prevented them from
quitting. Few of them thought of quitting; some
gave up one addiction, but started another one
soon, highlighting the vulnerable environment in
which these adolescents live.
Comparison of quit rates and rates of fresh
uptake between the intervention and control
communities clearly underscores the benefits of
the intervention in controlling tobacco initiation
and enhancing tobacco cessation. Though quit
rates did not differ significantly between the two
has been highlighted
communities attempted cessation and a few of
them were successful in achieving it. No interven-
tion was administered in the control community
but wider influences such as tobacco control
policy changes and the mass media cannot be
ruled out. These results point towards the need to
set up community-based cessation services apart
from hospital-based services currently available in
Section25_952.htm) to facilitate their use by
adolescents and the common people. The two
hospital-based cessation centres in Delhi are
insufficient to cater to the entire population of
The limitations of this study included a small
number of communities (two) and loss to follow
up. For such multi-component interventions to be
effective and promote cessation, the period of
intervention has to be for a longer duration than
the 1 year of intervention in phase II of this study.
The results demonstrating the efficacy of this
intervention model have been encouraging, con-
sidering its short duration. With these experiences
and results, a group randomized trial has been
plannedin the third
ACTIVITY (Advancing Cessation of Tobacco in
Vulnerable Indian Tobacco Consuming Youth),
funded by the Fogarty International Center at the
National Institutes of Health. This large-scale
randomized intervention trial is currently being
conducted in 14 low-SES communities of Delhi.
Community-based interventions that are sustain-
able and cost-effective can be a promising
approach for the prevention and cessation of
tobacco use, particularly for protecting vulnerable
adolescents in developing countries such as India.
CARENIDHI, HOPE Project and ARPANA
Trust, who coordinated and supported us in the
implementation of the project. We are also
thankful to our field staff and the community
leaders who made this project successful. This
research would not have been possible without
the participation of youth and their parents.
are thankfulto ourpartner NGOs:
This work was supported by the Proposal
Development Grant and Start up Grant from
Community-based model for preventing tobacco use 151
at Emory University on July 12, 2010
Research in Developing Countries’, New Delhi,
India (Project: Tobacco use among young
groups in urban India: dimensions, determinants
and dynamics of behaviour change through a
Bobak, M., Jha, P., Nguyen, S. and Jarvis, M. (2000)
Poverty and smoking. In Jha, P. and Chaloupka, F. J.
(eds), Tobacco Control in Developing Countries. Oxford
University Press Inc., New York, pp. 41–61.
Gajalakshmi, V., Peto, R., Kanaka, T. S. and Jha, P. (2003)
Smoking and mortality from tuberculosis and other
diseases in India: retrospective study of 43000 adult
male deaths and 35000 controls. The Lancet, 362,
Jarvis, M. J. and Wardle, J. (2006) Social patterning of
individual health behaviours: the case of cigarette
smoking. In Marmot, M. and Wilkinson, R. G. (eds),
Social Determinants of Health, 2nd edition. Oxford
University Press, Oxford.
Jha, P., Jacob, B., Gajalakshmi, V., Gupta, P. C., Dhingra,
N., Kumar, R. et al. (2008) A nationally representative
case–control study of smoking and death in India. The
New England Journal of Medicine, 358, 1137–1147.
John, R. M., Sung, H.-Y. and Max, W. (2009) Economic
cost of tobacco use in India, 2004. Tobacco Control, 18,
Mathur, C., Stigler, M. S., Perry, C. L., Arora, M. and
Reddy, K. S. (2008) Differences in prevalence of
tobacco use among Indian urban youth: the role of
socioeconomic status. Nicotine and Tobacco Research,
Milton, M. H., Maule, C. O., Lee, S. L., Backinger, C.,
Malarcher, A. N. and Husten, C. G. (2004) Youth
Tobacco Cessation: A Guide for Making Informed
Decisions. Centers for Disease Control and Prevention,
Mishra, A., Arora, M., Stigler, M. H., Komro, K. A., Lytle,
L. A., Reddy, K. S. et al. (2005) Indian youth speak about
tobacco: results of focus group discussions with school
students. Health Education & Behavior, 32, 363–379.
Mohan, S., Sarma, P. S. and Thankappan, K. R. (2005)
performance predict tobacco use among adolescent
and low educational
Nichter, M., Nichter, M. and Van Sickle, D. (2004)
Popular perceptions of tobacco products and patterns of
use among male college students in India. Social Science
and Medicine, 59, 415–431.
Patel, D. R. (1999) Smoking and children. Indian Journal
of Pediatrics, 66, 817–824.
Pentz, M. A., MacKinnon, D. P., Dwyer, J. H., Wang,
E. Y. I., Hansen, W. B., Flay, B. R. et al. (1989)
Longitudinal effects of the Midwestern Prevention
Project on regular and experimental smoking in adoles-
cents. Preventive Medicine, 18, 304–321.
Perry, A. (2006) India Inc.: Bombay’s Boom. Time, June
26, pp. 40–44.
Perry, C. L., Klepp,K.
Community-wide strategies for cardiovascular health:
the Minnesota Heart Health Program youth program.
Health Education Research, 4, 87–101.
Perry, C. L., Stigler, M. H., Arora, M. and Reddy, K. S.
(2009) Preventing tobacco use among young people in
India: Project MYTRI. American Journal of Public
Health, DOI 10.2105/AJPH.2008.145433.
PLAN (2005) Growing up in Asia: Plan’s Strategic
Framework for Fighting Child Poverty in Asia (2005–
2015). Plan Limited, Bankok, Thailand.
Poland, B., Frohlich, K., Haines, R. J., Mykhalovskiy, E.,
Rock, M. and Sparks, R. J. (2006) The social context of
smoking: the next frontier in tobacco control? Tobacco
Control, 15, 59–63.
Shafey, O., Eriksen, M., Ross, H. and Mackay, J. (eds)
(2009) The Tobacco Atlas, 3rd edition. The American
Cancer Society, Atlanta.
Singh, N. (ed.) (2005) If I Were Rain: Celebrating the Spirit
of India’s Disadvantaged Urban Child. Youthreach, New
Stigler, M. H., Perry, C. L., Arora, M. and Reddy, K. S.
(2006) Why are urban Indian 6th graders using more
MYTRI. Tobacco Control, 15, i54–i60.
UNICEF (2004) Mapping India’s Children: UNICEF in
Action. UNICEF, New Delhi.
West, R., Zatonski, W., Przewozniak, K. and Jarvis, M. J.
(2007) Can we trust national smoking prevalence
figures? Discrepancies between biochemically assessed
and self-reported smoking rates in three countries.
Cancer Epidemiology Biomarkers and Prevention, 16,
World Health Organization (1997) Tobacco or Health: A
Global Status Report, Country Profiles by Region. World
Health Organization, Geneva.
in Kerala,India. PreventiveMedicine, 41,
I. andSillers, C. (1989)
152M. Arora et al.
at Emory University on July 12, 2010