Project EX-India: A classroom-based tobacco use prevention and cessation intervention program

Abstract

Tobacco use experimentation is most frequent between the ages of 15–24 in India. Therefore, programming to counteract tobacco use among adolescents is needed. There is a lack of evidence-based teen tobacco use prevention and cessation programs. The current study provides an outcome evaluation of the Project EX tobacco use prevention and cessation program among Indian adolescents (16–18 years). An eight-session classroom-based curriculum was adapted to the Indian context and translated from English to Hindi (local language). Next, it was tested using a quasi-experimental design with 624 Indian students at baseline, involving two program and two control schools, with a three-month post-program follow-up. Project EX involves motivation-enhancement (e.g., talk shows and games) and coping skills (e.g., complementary and alternative medicine) components. Program participants rated complementary and alternative medicine (CAM) activities like meditation, yoga and healthy breathing higher than talk shows and games. Compared to the standard care control condition, the program condition revealed a prevention effect, but not a cessation effect. Implications for prevention/cessation programming among Indian teens are discussed. This study was approved by the Independent Ethics Committee, Mumbai.

1. Introduction

The most susceptible time for experimenting with tobacco use in India is between the ages of 15 and 24 (Arora et al., 2010; GATS India, 2010; Mishra et al., 2005; Patel, 1999). According to the Global Youth Tobacco Survey (2009) of school going Indian students ages 13–15, 6.1% reported having ever smoked cigarettes, 4.4% reported current cigarette smoking, and 12.5% reported currently using other tobacco products. The prevalence rates of both smoking and smokeless tobacco use vary dramatically across different states in India (Rani et al., 2003). Smoking prevalence varies from 6.8% among 13–15 years old students in Bangalore (Muttappallymyalil et al., 2012; Shashidhar et al., 2012) to as low as 3% overall prevalence in Punjab (Muttappallymyalil et al., 2012; Siziya et al., 2008). One may speculate that youth in different states begin to use tobacco at different ages, possibly due to differences in urbanicity or culture.

The National Tobacco Control Cell (NTCC) at the Ministry of Health and Family Welfare (MoHFW) launched the National Tobacco Control Programme (NTCP) in the year 2007–08 with the aims to create awareness about harmful impact of tobacco consumption and to help people quit tobacco use through Tobacco Cessation Centers (TCCs) (National Tobacco Control Programme Report, 2014). Unfortunately, the NTCP does not address cessation component in schools.

Strategies to reduce tobacco use in India are challenged because of the use of multiple smoked and smokeless forms of tobacco. In a study of adolescents aged 10–19 years, qualitative analysis revealed that different forms of tobacco are easily available and accessible to teens at roadside stalls, general stores and pan (betel leaf) shops, as examples (Arora et al., 2010). Common forms of smokeless tobacco use/tobacco chewing include pan (piper betel leaf filled with slices of areca nut, lime, catechu, and other spices chewed with or without tobacco), gutkha (chewable tobacco mixed with areca nut) and mishri (a powdered tobacco rubbed on the gums as paste) (Rani et al., 2003).Out of 12.5% current users of “other” tobacco products, 9% of youth are current users of smokeless tobacco (Global Youth Tobacco Survey, 2009). Therefore, an intervention that addresses multiple forms of tobacco use is needed for adequate cessation and prevention effects. An evidence-based program, Project EX, was adapted culturally to address the multiple tobacco products use for India.

Project EX is an empirically tested and validated teen tobacco use cessation program (Espada et al., 2014; Sussman et al., 2004). Project EX has a school-based clinic version as well as a classroom version (see Sussman, 2012 for details). The clinic version has been effective in the United States of America (U.S.), China, and Russia (Idrisov et al., 2013; Sussman, 2012; Zheng et al., 2004). The classroom-based prevention/cessation version of Project EX has been effective for overall prevalence of 30-day smoking (applying across levels of use), as well as on cessation in California (Sussman et al., 2010; Sussman et al., 2007) and had some immediate impact in Spain (Espada et al., 2014). This paper focuses on reporting the three-month post-program impact of the Project EX prevention/cessation program with school going students (16–18 years) in Delhi, India. Program receptivity to and effects on prevention and cessation of tobacco use was examined in this pilot study using a two-group quasi-experimental study design, comparing Project EX with a standard care control condition.

2. Method

2.1. School selection and experimental design

The target population for the EX prevention/cessation program was students of Class XI, both girls and boys in the age group 16–18 years, studying in four different schools (two government and two private schools) in Delhi, as a convenience sample. Schools with an active ongoing health education program with tobacco control as a major component were excluded at the time of selection procedure. The recruited schools were assigned to either a program or standard care control condition, blocked by school type; that is, there were two schools per condition, one government and one private school in each condition. School consent was sought prior implementation of the program.

2.2. Project EX-India curriculum contents, data collection, training, and implementation

The India-EX program consisted of eight prevention/cessation sessions. These are described in more detail elsewhere (Espada et al., 2014; Sun et al., 2007), and pertain to both tobacco users and non-users. The survey tools for this study were student questionnaires, which were adapted to the Indian context from the questionnaire used by collaborative partners of this study at University of Southern California, USA (e.g., Sussman et al., 2007). The student questionnaires along with parent and student information (tracking) sheets and consent forms were translated in Hindi for administration in government schools. For all three surveys, data was collected using self-administered surveys completed by the students in classrooms after obtaining active parental informed consent and active student assent, as mandated by the ethics board. The study was approved by the Independent Ethics Committee, Mumbai.

Researchers in India were trained through three two-hour Skype interactions. It was felt that in-person extensive training would not be necessary since they had been involved in extensive tobacco use prevention research previously (e.g., Arora et al., 2010). They were trained by the researchers in India (one of whom had translated the curriculum to Hindi) to deliver Project EX-India.

Project EX-India was implemented similarly to the parent program, twice a week the first two weeks followed by once a week the subsequent four weeks. The pretest was administered directly before beginning Session 1, the immediate posttest was administered directly after Session 8, and the follow-up questionnaire was administered in the classroom three months after the end of the program.

2.3. Adaptation of the Project EX prevention/cessation program to Indian youth

The Project EX classroom prevention/cessation curriculum was developed from the Project EX clinic program (Sun et al., 2007; Sussman et al., 2001). This program is constructed based on a motivation-coping skills-personal commitment model of teen tobacco prevention and cessation consisting of eight sessions, each 40 to 45-min in length.

The curriculum was adapted to the Indian context in five ways. Two sets of manuals [one for teacher and one for students] were prepared for the implementation of the intervention. The student manual was given to each participating student. The manuals were translated from English to Hindi (local language) by one of the researchers fluent in both languages. Second, the case studies and scenarios were made culturally relevant and the names were changed to Indian names so that the students could relate to the characters. For example, in Session 1 the name ‘Eddie’ was changed to ‘Sachin’, the term ‘girlfriend’ was changed to ‘best friend’ to comply with cultural norms, and the example of ‘grandmother’ dying from lung cancer was changed to the ‘grandfather’ dying of lung cancer. (Among the elderly, most smokers are males.) When introducing talk shows, the students were given the example of local talk shows like ‘Meri Baat’ or ‘Youth Express’ instead of ‘Oprah’ or ‘Dr. Phil’. For Session 3, “Sean Marsee's Smokeless Death” case study was replaced by Roshan Wankhede's story which is about a 17 year old boy who started consuming tobacco at the age of 13 due to peer pressure and developed oral cancer.

Third, the original curriculum targets use of tobacco in the form of pipes, cigars, and chewing tobacco. Various other forms of tobacco popular in India were included in this intervention. These forms include bidis (unfiltered cigarettes made by rolling a dried, rectangular piece of tendu leaf with sun-dried flaked tobacco), gutka, zarda, mishri, pan masala and khaini (chewing tobacco) (Reddy and Gupta, 2004). For example, in Session 3 students were asked a question about the most used form of chewing tobacco among children which is khaini followed by pan.

Fourth, the monetary amounts were changed from dollars to rupees. For instance, in Session 1, the original curriculum provides as a suggestion for the question “What are some reasons for quitting?” that one may “Save money. Over US $2555 per year is spent on smoking one pack per day”. The amount was changed to Rs 19,000 which indicates the cost per year of smoking and chewing one pack per day in India (chewing tobacco is much less expensive than cigarette smoking in India).

Finally, when Project EX was implemented as a research project in the United States, extrinsic motivators were provided as incentives. At the end of the program, students received certificate for participating in the program. In India, as in some other international settings, no incentives were provided to the students for attending the program sessions (e.g., Espada et al., 2014).

2.4. Measures

Pretest and posttest measures were collected from students using a self-report, close-ended and fill-in-the-blank response questionnaire. The baseline questionnaire for this study included specifics on demographics and tobacco use behavior. Demographic items included age (in years), gender, and current living situation (with both parents, only with mother, only with father, with a relative/guardian or other). Tobacco use behavior was assessed with the Yes/No item asking “Have you ever smoked a cigarette/beedi in your life?”, “Have you ever chewed tobacco (e.g. gutkha, khaini, zarda, paan with tobacco etc.) in your life?, and “Have you ever used other smokeless tobacco products (e.g. snuf, nasswar, toothpaste with tobacco, gul, etc.) in your life? Open-ended as well as close-ended items to assess tobacco use in an average day, yesterday and the last month were also included.

In the process evaluation portion of the immediate posttest questionnaire, two measures of student responsiveness to the program were examined. The first measure was composed of eight items and assessed student's ratings of the program quality. These items asked the subject to rate how much they “liked” and “learned” from the sessions, how “interesting”, “informative” and “well-organized” the sessions were, how “enthusiastic” and “knowledgeable” the facilitator was, and how “helpful” the sessions were to not smoke in the future. Responses were on 10-point scales (“not at all” to “extremely”). As in previous work (e.g., Sussman et al., 2001), these adjectives were highly inter-correlated (Cronbach's α=.94 for the current study); thus ratings across the items were averaged to comprise a perceived program quality index.

In the second measure, participants rated how much they liked each of the key EX curriculum activities (8 total activities; from 1=“terrible” to 10 = “excellent”). The activities included 1) Talk Show: Family and Friends Confront Smokers about Their Habit, 2) Talk Show: Cigarettes May be Stressing You Out, 3) Healthy Breathing, 4) Game: Is Smoking on the Menu?, 5) Talk show: Quitting Smoking: I've Been There and It Does Get Better, 6) Yoga, 7) Meditation, and 8) Talk Show: WARNING! Waiting to Quit Smoking may Be Hazardous to Your Peace of Mind. These activity likeability items were averaged to get an overall impression of the program activities (Cronbach's α=.89 for the current study). The ratings of each of the eight activities were compared to get an impression of the relative liking of each of them (see Table 1).One other immediate forced choice posttest item, asked only of program participants, was: “Did taking the Project EX class help you to do any of the following?” Among eight forced-choice response categories included: “quit tobacco use completely”, “reduce the amount of tobacco you use, and you plan to quit completely”, “reduce the amount of tobacco you use, but you do not plan to quit completely”, “decide to quit smoking in the next two weeks”, “decide to quit smoking sometime in the future”, “strengthen your commitment to stay tobacco free (if you had already quit smoking)”, “strengthen your commitment to stay tobacco free (if you have never smoked)”, and “Other (please specify)”.

Table 1.

Likeability ratings of eight key activities in Project EX-India at immediate post-test.

Activity name	Mean	STD
Meditation	8.65	1.76a
Yoga	8.56	1.88a
Healthy Breathing	8.15	2.03c
Quitting Tobacco: I've Been There and It Does Get Better	7.97	2.19c
WARNING! Waiting to Quit Tobacco May Be Hazardous to Your Peace of Mind	7.82	2.46bc
Family and Friends Confront Smokers About Their Habit	7.57	2.23b
Tobacco May Be Stressing You Out	7.53	2.39b
Is Tobacco on the Menu?	7.43	2.38b

Note: The average likeability value of the program activities (response scale 1–10) was 7.96 (SD = 2.17). Same letter subscripts means not significantly different; different lettered subscripts means significantly different.

The third questionnaire for this study (implemented after three months) included the same tobacco use behavior items as in the pretest questionnaire. Tobacco use was examined at both pretest and follow-up. The yes-no items were combined to indicate any tobacco use. Like-wise, level of tobacco use was averaged across items to permit examination of a preventive effect.

2.5. Data analysis

To assess the potential sampling bias due to attrition at the follow-up across conditions (external invalidity), we compared the participants that were surveyed at both pretest and follow-up time points (n = 444) and the participants that were only surveyed at pretest (n = 180). We used a logistic regression model, in which “attrition group” was the dependent variable and age, gender, and current living situation, ever tobacco use and monthly tobacco use at baseline were predictors. To assess the potential sampling bias due to attrition as a function of condition (internal invalidity), comparisons were made between the participants who were lost at follow-up and the rest that remained in the study at follow-up. The comparisons utilized separate logistic regression models for individual variables to indicate statistically significant differences (two-tailed p value at the .05 level). Again, all relevant demographic variables were examined, along with ever and monthly tobacco use.

Program receptivity was examined at immediate posttest in the program condition through a simple look at mean ratings on the receptivity measures, and the behavior commitment item. Also, all possible within-subjects mean comparison t-tests were conducted to provide a comparison on likeability among the eight activities. In an additional analysis, intent-to-treat (ITT) quit rates were calculated for those who said that they had ever-smoked at baseline. For prevention analysis, logistic regression was run using monthly tobacco use at follow-up among non-tobacco users at baseline(1=yes, 0=no) as the dependent variable and school type, condition, living condition, gender and age as independent variables. All the results were considered statistically significant at 5% level of significance. Analyses were done using statistical software SAS 9.4.

3. Results

3.1. Participants

The average class size was 40 and there were 4 sections per class in each school. A total of 654 students were eligible, out of those 624 students participated at the baseline study (303 in the control condition and 321 in the program condition) out of which 468 (75%) completed the immediate posttest and 444 (71%) completed the follow-up. At baseline, 58% of the sample was male, 66% of the sample attended the public schools, and only 16 or approximately 3% of the respondents were monthly smokers.

3.2. Assessment of attrition bias at three-month follow-up

First, regarding the analysis of external validity, there were no significant differences found between the respondents at baseline and those who followed–for age, gender, living conditions, ever tobacco use and monthly tobacco use at baseline. The comparison of demographic characteristics between pretest-only and follow-up (internal validity) revealed statistically significant differences as a function of condition (p < .01). Overall, 62% of males were followed up in the program condition (107 of 172), whereas 78% of males were followed up in the control condition (146 or 188); 77% of females were followed up in the program condition (115 of 149), whereas 66% were followed up in the control condition (76 of 115). Though the interaction effect between age and condition was significant (p < 0.04), the overall model was not (p = 0.09). The participants in the program condition were slightly younger in age in the group followed up (mean age was 15.96, SD = 0.73) as compared to the program condition baseline only group (mean age was 16.16, SD = 0.90), relative to the control condition (followed up mean=16.26, SD=0.69; baseline only mean=16.15, SD=0.61).

3.3. Receptivity analysis (process ratings at immediate post-test)

The average satisfaction of adolescents with the program was 7.77 (SD = 2.28). The average likeability rating of the program activities was 7.96 (SD = 2.17). The average likeability of the program activities ranged from 7.43 (SD = 2.38) for the game Is Tobacco on the Menu?, to 8.55 (SD = 1.88) for Meditation and 8.65 (SD = 1.76) for Yoga. Yoga and Meditation were rated more favorably than all other activities but did not differ from each other significantly, which is shown in Table 1. The least favorite activities were talk shows Family and Friend Confront Tobacco Users About Their Habit and Tobacco May be Stressing You Out and the game is Tobacco on the Menu?, which did not differ significantly from each other. In addition, 27.27% of adolescents in the program condition reported that Project EX helped to strengthen their commitment to stay tobacco free (if they had already quit using tobacco) or stay tobacco free (if they had never used tobacco before).

3.4. Cessation and prevention analysis

At baseline there were 16 monthly tobacco users out of which 9 were followed-up. All 9 of them quit. Out of these 9, 6 were in the control condition and 3 were in the program condition. Therefore, the compliance quit rate is 100%, regardless of condition. Intent-to-treat (ITT) quit rate slightly favors the control condition (not significantly though), therefore, no cessation effect was observed.

The overall logistic model for prevention analysis was significant (Likelihood Ratio=10.92, p=.05). Within the model, condition effect (Wald Chi-Square = 4.94, p = .03), living condition (Wald Chi-Square = 5.35, p = 0.02) and age (Wald Chi-Square = 4.06, p = 0.04) were statistically significant (see Table 2). Out of all respondents who were followed up, 32 started smoking at follow-up. Out of those 62.5% were in the control condition and 37.5% in the program condition.

Table 2.

Logistic regression for prevention analysis for monthly tobacco use from baseline to follow-up.

Variable	Results of the logistic regression
	Odds ratio	95% CI	p-Value
Condition
Program	0.39	0.17–0.89	0.03*
Control	Reference
Living Condition
Living with parents	0.19	0.05–0.77	0.02*
Not living with parents	Reference
School type
Public	1.73	0.72–4.18	0.22
Private	Reference
Gender
Female	1.09	0.52–2.31	0.82
Male	Reference
Age	1.86	1.02–3.41	0.04*

^*Significant at 0.05 level.

4. Discussion

The spiraling rates of tobacco use among adolescents in India (Patel, 1999) suggest a need to develop cessation and prevention programs that target multiple forms of tobacco use. Project EX is an evidence-based program designed to produce tobacco prevention and cessation effects. Motivation enhancement, social skills, and life skills material, are important components of Project EX. The program was culturally adapted and targeted various forms of tobacco use for successful implementation in schools in northern India.

This study tested the effects of the prevention/cessation program utilizing a quasi-experimental design. The program included complementary and alternative medicine activities, games and talk shows in addition to curriculum, to engage students. Program receptivity analysis suggests that alternate medicine activities such as yoga, meditation and healthy breathing were rated as most likeable among all activities. During the program, the smokers make a personal commitment to quit and to subsequently review the commitment. Non-smokers make personal commitments to stay tobacco free and serve as a “listening ear” to those who may be trying to quit (Espada et al., 2014; Sun et al., 2007). About 27% reported that Project EX helped to strengthen their commitment to stay tobacco free or never use tobacco.

Although there was no significant cessation effect of the program, we did find a significant prevention effect. The monthly smoking at follow up was significantly associated with the condition effect (program vs. control), living condition of the respondent and age. A total of 62.5% (20 out of 32) of the smokers at follow-up were in the control condition, with an even percentage of nonsmokers and sample sizes at baseline. Therefore, this program shows promising results for prevention of tobacco use among adolescents.

As it has been shown in this study, Project EX can have a significant prevention effect. Still, there are some notable limitations. First, fidelity of implementation was not assessed, although we know that all sessions were implemented in all program classes. The only data collected on the facilitators was from the ratings of the participants. In these ratings the facilitators were moderately favorably rated and these measures were included in the perceived program quality receptivity measures. Second, for such interventions to be successful and effective, the period of intervention and evaluation has to be longer to observe a long term impact (e.g., one-year follow-up). Third, the rate of self-reported ever tobacco use was really low at baseline. This could be due to the fear of consequences of reporting tobacco use, or it can be due to the region in which the data were collected of the specific schools involved. Saliva cotinine testing can be done to validate self-reports. Fourth, Project EX has not been tested in other locations in India and it would be interesting to replicate the program in other States. Also, a much larger sample size and more schools are needed to be able to accurately assess the cessation impact of Project EX-India. Last, the study did not use a cluster randomized controlled study design and was under-powered due to financial limitations. Based on the data extracted from previous work in other locations (e.g. Sussman et al., 2007), to achieve a medium effect size, we would have needed to double our sample size to achieve adequate statistical power (80%) for the prevention effects Our sample size would have needed to be increased 10 times to have a sufficient sample size to detect cessation effects. However, it represented one of the first and few of such efforts in India.

However, as a pilot study, the implementation was successful. All sessions were implemented, attrition was not higher than in several U.S. studies (see Sussman et al., 2007), and the overall program and activities were rated favorably by participants. The current study suggested promise as a prevention program but did not include sufficient numbers of regular tobacco users to be able to engage in an appropriate cessation analysis (there was insufficient statistical power). Further replications in this cultural context might be required. It appears important to engage in a needs assessment prior to implementing the Project EX prevention/cessation program. States in which regular tobacco use is most prevalent might be a better target population to consider. Also, perhaps, the program might be adapted for emerging adults who are older and may have a more entrenched tobacco problem.

HIGHLIGHTS.

• Outcome evaluation of school based tobacco control program Project EX in India

• Quasi-experimental two-condition study design (program, standard care control)

• Four schools, two in each condition, conditions comparable at baseline

• Program condition revealed a significant prevention effect among adolescents

• Alternative medicine components were rated more favorably than other activities