Community Based Group Intervention for Tobacco Cessation Essay
Objective: To determine the efficacy of community based group intervention for tobacco cessation and to identify the barriers for cessation intervention among rural men.
Methods: We recruited 400 men (20-40 years) currently using any form of tobacco from 20 villages of the Indian State of Tamil Nadu and randomized them equally into intervention and control groups. A physician offered two sessions of health education five weeks apart along with self-help material on tobacco cessation to the intervention group. The control group received only self-help material. The contents of the sessions included tobacco related health problems, benefits of quitting and coping strategies for withdrawal symptoms. The outcome measures at two months follow-up were self-reported point prevalence abstinence, quit attempts lasting > 24 hours and harm reduction of tobacco use > 50% of baseline. Follow-up data were available for 92%.
Findings: In the intention to treat analysis, self reported point prevalence abstinence of 12.5% in the intervention group was significantly higher than the 6.0% in the control group [Odds Ratio (OR) 2.39; 95% confidence interval 1.09-5.22]. Harm reduction was three times (OR 2.63, 95% CI 1.36-5.08) and quit attempts were two times (OR 1.66, 95% CI 0.97-2.84) higher in the intervention group than the control group. Time constraints, alcohol consumption and urban migration were found to be important barriers for group based tobacco cessation intervention.
Conclusion: Community based group intervention has the potential to increase the coverage of tobacco cessation services for men in rural Tamil Nadu.
(Clinical Trial Registration Number CTRI/2010/091/000221)
Tobacco has been projected to cause one billion preventable deaths in the twenty-first century, of which more than 70% will occur in low and middle income countries.1 The largest increase in tobacco related mortality has been projected to occur in India and China.2 The annual smoking-associated deaths in India were projected to be one million in this decade.3 As per the Global Adult Tobacco Survey (GATS) the current tobacco use in any form was 47.9% among adult males in India.4 The prevalence of tobacco use was significantly higher in rural, poorer, uneducated population compared to urban, wealthier and more educated population. 5 All forms of tobacco use were increasing in India and the largest increase was in the age group 15-24 years.6
Cessation by current smokers is the only way to avoid substantial proportion of tobacco related deaths worldwide before 2050. Tobacco cessation is rare in most low and middle income countries. For example the quit rate at population level in India was reported to be only two percent. Unless there is widespread cessation among smokers, 450 million deaths will occur in the world by 2050. Prevention of these deaths requires adult cessation.7 Seventy percent of the one million smoking related mortality in India was projected to occur in the middle age of 30-69 years.3 Half of the long term smokers will die during their productive middle age, losing 20-25 years of life.8 The benefits of smoking cessation were found to be largest in those who quit in the younger age group of 25 to 34 years.9 Therefore any cessation intervention needs to focus on the younger age group.
The evidence base for tobacco cessation interventions is limited in low and middle income countries.10 A few available studies reported quit rates although the primary objectives were not tobacco cessation intervention. A community based intervention study among rural tobacco users for primary prevention of oral cancer reported a quit rate of two percent in Ernakulam district (Kerala), one percent in Bhavnagar (Gujarat) and five percent in Srikakulam (Andhra Pradesh) at the end of one year.11 Another community based health education intervention program offered by health workers targeting the entire population of selected primary health centre areas in Karnataka reported a quit rate of 26.5% in the intervention group compared to 1.1% in the control group at the end of five years.12 The limitation of this study was that it was not randomized and mainly focussed on prevention of cancer.
The intervention was very intensive and delivered once a week by junior health workers and once a month by senior health workers for three years. It may be difficult to scale up such intensive interventions considering the limited human resources in the Indian rural health system.13 An unpublished randomized control trial in Kerala reported continuous smoking abstinence of 10.1% at six months follow-up in the intervention group that received brief intervention by a doctor and counselling by a non-doctor health professional compared to the control group that received only brief intervention by the doctor.14
Data from the World Health Organization supported 18 tobacco cessation clinics in India providing behavioural and pharmacological intervention reported a point prevalence abstinence of 14% at six weeks.15 In a vast country like India reaching every tobacco user with this clinic based intensive specialist treatment is not feasible.16 So there is a need to develop brief tobacco cessation techniques which focus on community and group settings.17 Since such studies are not available from India we decided to conduct this study to see the efficacy of such a study in Indian settings.
The main objective of our study was to determine the efficacy of community based group intervention for tobacco cessation in rural Tamil Nadu. The secondary objective was to study the facilitators, barriers and acceptance for conducting group interventions.
The study was conducted in Tiruchirapalli district, Tamil Nadu, South India. The district was administratively divided into eight Taluks, 14 community development blocks and 479 villages with a total population of 2.42 million 52.9% residing in rural areas. The literacy rate in this district was 78% (rural 70%).18 One of the Taluks- Manaparai- with the largest number of villages was selected for the study.
We conducted a cluster randomized trial selecting 20 villages out of 97 based on probability proportion to the size of the population and randomized equally into intervention and control villages. (Figure 1) After randomization it was found that some of the control villages were contiguous with intervention villages. To avoid contamination of intervention, adjacent villages that were not contiguous with the intervention villages were selected as control. Each village was considered as a cluster and twenty participants were recruited from each cluster.
The estimated sample size (rounded) was 100 in each group and it was multiplied by two to account for design effect due to expected clustering. The study was designed to have 80% power at five percent significance level, assuming 15% abstinence in the intervention group and two percent abstinence in the control group.
Participants were recruited to the study based on the following inclusion criteria: (i) men (ii) aged 20-40 years (iii) resident of the village (iv) current user of any form of tobacco (v) willing to participate in the study and provide informed consent. Men who were planning to migrate in the next three months were excluded from the study to reduce the loss to follow up.
The baseline interview using a pre-tested structured interview schedule was done to collect socio-demographic details, tobacco use pattern and determinants related to outcome. Fagerstrom Test for Nicotine Dependence (FTND) was used to assess the dependence of smokers and smokeless tobacco users.19,20 For individuals who were using both forms of tobacco, both scales were administered and the highest score was taken as final score. The motivation to quit tobacco was assessed based on Prochaska’s five stages of change model.21
The outcomes were measured using structured interview schedule. The main trial outcome was self-reported point prevalence abstinence (No tobacco use in the past seven days) at two months after recruitment. The secondary outcomes were quit attempt (any attempt to quit tobacco which lasts more than 24 hours) and harm reduction (reduction of tobacco use more than 50% of baseline). Field notes were maintained to identify the potential facilitators, barriers and acceptance of this intervention in the community. This was an open label trial, the participants, the physician who delivered the intervention and the outcome assessor was not blind to the allocation status.
The intervention group received two sessions of health education delivered five weeks apart by a physician (SK) and self-help material. Each session lasted for 30 minutes. Two groups were formed in a cluster and each group comprised of 10 members. The control group received only self-help material. The content was developed based on the research done by the Project Quit Tobacco International.22-24
Components of first session covered topics like tobacco related facts, raising consciousness through photographs of diseased body parts using a flip chart, illness narratives, myths and facts, benefits of quitting tobacco and plan for quitting tobacco. The second session included discussion of their experience, withdrawal symptoms and measures to overcome them, coping strategies/prevention of relapse and hazards of second hand smoke. The contents of the booklet had information regarding harmful effects of tobacco, health benefits of quitting tobacco and advice on how to stop tobacco.
Ethical clearance was obtained from the Institutional Ethics Committee, Sree Chitra Tirunal Institute for Medical Sciences and Technology, Trivandrum that hosts India’s first Master of Public Health Program (MPH).25 Information sheet written in local language was distributed to the participants. The purpose of the study was explained to them and participation was voluntary. Written informed consent was obtained.
The data were analysed using SPSS Statistics version17 (Chicago, Illinois). The baseline characteristics of the intervention and control group were compared to look for any statistically significant difference. The mean and percentage values were compared using independent samples t-Test and Chi-square test respectively.