Evaluation of the Effectiveness of 5A-5R Counselling Services for Tobacco Cessation

Background and problem statement:

Tobacco use, encompassing both smoking and smokeless tobacco, constitutes a significant global health risk, contributing to eight million annual deaths and resulting in the loss of 229.77 million disability-adjusted life years (DALYs) worldwide. Despite concerted international efforts to mitigate tobacco consumption, approximately 80% of tobacco use is concentrated in low and middle-income countries, with Bangladesh standing out as a prominent example. The Global NCD Action Plan sets a target of a 30% reduction in tobacco prevalence by 2025; however, recent data indicates that 35.3% of Bangladeshi adults continue to use tobacco. In Bangladesh, the prevalence of tobacco use remains high, with 43.7% of adults reported as tobacco users in 2018, where smoking and smokeless tobacco have prevalence rates of 23.5% and 27.5%, respectively. Men exhibit a higher smoking prevalence than women, but both sexes have similar rates of smokeless tobacco use. Tobacco use is a major risk factor for non-communicable diseases (NCDs), including cardiovascular diseases (CVD), chronic lung disease (CRD), and cancer, contributing to substantial mortality rates in the country. In 2018, the estimated mortality from CVD, CRD, and cancer was 30%, 10%, and 12%, respectively. The Directorate General of Health Services (DGHS) in Bangladesh promotes brief intervention using the 5A-5R counselling technique, which is included in national guidelines for hypertension and diabetes management at the primary healthcare level. In this process, 5A (Ask, Advice, Assess, Assist, and Arrange) counselling approach applies to those people who want to quit tobacco, and 5R (Relevance, risk, reward, roadblock, repetition) applies to people who are not ready to change their behaviour. This WHO 5A-5R model is concise, comprehensive, and effective in improving attitudes toward smoking. Studies have shown that brief intervention counselling can lead to a 15% smoking cessation rate. Several cluster randomized trials, especially on 5A-based counselling have been conducted globally which have demonstrated the effectiveness of this approach on tobacco quitting, however, intervention studies evaluating the 5A-5R-based counselling approach in the South Asian context is scarce. Additionally, a systematic review demonstrated that mobile phone-based call play a crucial role in tobacco cessation, with text messages being more acceptable than application-based interventions. A comprehensive tobacco cessation counselling service using 5A-5R based approach coupled with mobile phone follow-up can be effective in reducing tobacco consumption prevalence, however, the investigators did not find any prior study evaluating such an approach, hence the investigators believe this study will not only generate evidence in Bangladesh context but also will be a global example. This study aims to assess whether phone-assisted culturally appropriate 5A-5R counselling services can enhance the tobacco quitting rate or reduce tobacco consumption among adults in union-level primary health care (PHC) settings in Bangladesh.

Research question/hypotheses and objectives: The 5A-5R-based counselling program following WHO PEN protocol has already been incorporated in the national protocol for diabetes and hypertension prevention in Bangladesh and aimed to be implemented at all care provider levels, however, in reality, implementation of this counselling activity for motivating tobacco user to quit tobacco is absent in primary health care settings. Therefore, the investigators want to assess whether with or without a mobile phone-assisted culturally appropriate 5A-5R counselling intervention program provided to any tobacco user will enhance the tobacco quitting rate among adults in union-level primary health care (PHC) settings in Bangladesh.

Methodology: The proposed study is a small-scale cluster randomized control trial using three arms (intervention arm, control arm), total duration is eight months (in total) (Preparatory phase: first one month, intervention: three months, follow-up: four months, data analysis and report writing phase: one month). The study will be conducted in three community clinics (CC) and their catchment communities in Three Unions (Sator, Mohonpur and Sotogram) in Dinajpur, a border-based district in Bangladesh. This small-scale intervention study will be conducted among men and women aged 18 and above. The investigators will purposefully select three unions of Birganj upazila from Dinajpur district; one union will be selected for providing conventional 5A-5R based counselling service according to the national protocol for prevention of diabetes and hypertension to quit tobacco among current tobacco users, one union will be selected for delivering mobile phone assisted enhanced 5A-5R Based counselling service and the other one union will be selected as control that will continue with usual care without introducing any structured 5A-5R based counselling service for tobacco cessation.

Type of Intervention: 5A-5R Based counselling service according to WHO PEN Protocol with mobile phone assisted remote support.

The primary outcome of this study is tobacco cessation in any form, the investigators will draw inferences of the efficacy of the intervention by comparing the intervention groups with the control group. For our sample size calculation, the investigators considered the tobacco quitting rate among Bangladeshi adults is 19.5%, at least a 10% reduction of tobacco consumption in any form. Considering 5% probability of type I error (95% two-sided confidence interval), 80% power, intervention to control sample ratio 1:1, and a minimum of 10% increase in tobacco quitting in any form, our estimated sample size is 213 per arm (Fleiss approach with continuity correction). Given that the investigators will enroll participants from the catchment areas of the community clinic (hence considered as a cluster), the investigators also considered the design effect (1.00514) and inflated the sample size and10% non-response rate, the estimated sample size is 238 respondents in each union, therefore the total respondents in three unions will be 714 respondents (in control and intervention arm).

The selection process of the respondents for the study:

the investigators will deploy three Community Health Workers (CHWs) in three selected unions to make home visits within the catchment areas to reach all adults, irrespective of sex, who are current tobacco consumers. They will initially give some information about the harmful effects of using consuming tobacco products and will introduce the 5A-5R-based counselling service at respective CCs and also invite them to attend the 5A-5R-based counselling session in the selected CCs (in the intervention arm). Interested participants will come to the CCs and with informed consent, will be enrolled in the study to receive the intervention designed for each CC for each selected union.

Randomization procedure: The investigators will use a computer-based approach for selecting the control and intervention union and also to identify one designated CC from 3 CCs in each arm (2 for the intervention arm and 1 for the control arm). Neither the respondents nor the investigators will be blind about the CCs allocation, however, the outcome assessors (data analysts) will be unaware of the arm allocation.

Details of intervention: For intervention delivery, the investigators will provide 5A-5R-based counselling service with BCC material (leaflet) delivery with added weekly mobile phone-based calls and texts will be delivered. The investigators will provide weekly one mobile phone call and one behavioural change text message to facilitate tobacco quitting to ensure remote ongoing support in the intervention arm. While implementing this intervention, the investigators will ensure to exclude other effects and will select a study site where no other behavioural or any related national program is going on.

5A-5R based counselling approach: The 5As (Ask, Advise, Assess, Assist, Arrange) summarize all the activities that a healthcare provider can do to help a tobacco user (both smoking within 3-5 minutes in a primary care setting. This model can guide participants through the right process to talk to respondents who are ready to quit tobacco use and deliver advice.

Feasibility of the intervention by the CHCPs: According to the national protocol the 5A-5R-based counselling approach for tobacco quitting will not take more than 3-5 minutes, the investigators believe the CHCPs can be able to deliver the intervention with their designated daily work schedule.

Training of the CHCPs: The investigators will follow the the 5A-5R-based counselling protocol by the WHO Package of Essential Non-communicable Diseases Intervention for primary health care and the national protocol for the management of diabetes and hypertension. This training will be conducted at a local primary health care facility and master trainers trained on WHO PEN protocol will facilitate the sessions. There will be pre and post-training evaluations of the participants to ensure proper understanding of the training materials. The investigators will conduct supportive supervision and on-the-job training during their visit to selected CCs.

Behavioural Change Communication (BCC) Materials and Mobile phone-based remote support: The investigators will ensure that the BCC developed will be acceptable culturally and according to the Bangladesh Government's Tobacco Control Policy and World Health Organization. The investigators will follow the guidelines of the National Tobacco Control Cell (NTCC), and WHO tobacco control program and also consult with other organizations working with tobacco cessation, such as the National Heart Foundation, Cancer Society, and so on. The investigators will also collect examples from neighboring countries such as India, Sri Lanka, and Thailand. BCC material will contain information related to types of tobacco, health hazards of tobacco use, the importance of quitting tobacco, and how to quit tobacco. Additionally, to have other health benefits the investigators will include some motivational messages for increasing physical activity, and fruit and vegetable intake. Before delivering the final BCC material, the investigators will share this with the Bangladesh Tobacco Control Research Network and take their recommendations to finalize the one-page document. the investigators will incorporate pictorial messages and bullet point health advice regarding tobacco quitting. The text messages will also be aligned with BCC materials and delivered in Bangla and English language.

Control arm:

The investigators will select one CC from the particular union as a control and will not introduce any 5A-5R-based tobacco counselling services or any BCC material or any phone call-based remote service in all 8 months of the study period. The investigators will select such a CC as control where there will be a geographical boundary and a certain distance; so, there is less chance of information contamination (or information sharing). However, after completing the study (after 2 months of the study completion) by estimating the efficacy of 5A-5R based tobacco counselling service with or without mobile phone assisted remote service to enhance tobacco quitting, considering the ethical concern the investigators will deliver all counselling materials/leaflets to the Community Health Care Providers (CHCPs) of the control area to deliver these educational materials to the selected participants. Our aim is to ensure that they are not deprived of the intervention materials after the study completion.

Statistical Analysis:

Baseline characteristics of participants, including sociodemographic characteristics, and lifestyle factors will be compared between the two intervention and one control subdistricts . A simple difference-in-difference (DiD) estimates for tobacco use status will be implemented with multivariate logistic regression analyses, which take the following form.

Log(p/1-p) = β0 + β1*Time + β2*Interventions + β3*[Time*Intervention] + β4*Covariates + ε The β3 coefficient captures the difference in change over time. Clustered standard errors at union/community level will be specified to allow for intragroup correlation, relaxing the usual requirement that the observations within a cluster are independent. Marginal effects will be calculated to illustrate how the predicted probability changes over time among the three groups.