Pilot Randomised Controlled Trial of ToQuit Intervention in India (ToQuit)

ToQuit study aims to overcome the accessibility and feasibility hurdles by developing and evaluating a tobacco-cessation intervention that can be delivered using low cost and easily available mobile text messaging. India has 1.2 billion mobile phone subscriptions (tele-density 92%) and this provides a unique opportunity to significantly increase the penetration and coverage of tobacco cessation interventions in a low health-resource setting.

This pilot Randomised Control Trial (RCT) will aim to empirically evaluate the feasibility and acceptability of the mobile-based brief intervention (BI) for tobacco use, generate preliminary estimates of the impact of the BI, and fine-tune the procedures for a definitive RCT.

This protocol will serve as a research framework for the implementation of the proposed pilot RCT.

Objectives:

1. To test and refine procedures for a definitive RCT.
2. To assess the feasibility and acceptability of delivering the mobile-based BI for tobacco use.
3. To refine the mobile-based BI package for a definitive RCT.
4. To assess the impact of the mobile-based BI on tobacco use.
5. To inform sample size calculations for the definitive RCT.

Study setting:

The study recruitment will be conducted online and hence participants will include any eligible individual residing in India.

Sample size:

We will include a sample of 100 participants, 50 each in the intervention and control groups. This sample size estimates are based on our previous experiences of pilot RCTs and deemed sufficient to achieve the goals of understanding the acceptability, feasibility, and preliminary impact of an intervention. The sample size of 80 enrolled tobacco users randomly allocated with a 1:1 allocation ratio, with 5% type I error and 6% abstinence in the control arm at follow up would provide 80% statistical power to detect an effect of 5.3 (relative risk) for abstinence in favour of ToQuit intervention.11 Considering a 20% loss rate during the follow-up period as per our experience in case series, the total sample of 100 participants would be needed.

Recruitment sources:

• The recruitment advertisements will be shared over the social media accounts of Sangath India (an organization where all the research staff is based) and contacts of the researchers. We will also display advertisement posters in the public health facilities and dental clinics in Goa state and at public places such as village panchayats, supermarkets etc. The individuals willing to quit tobacco will be asked to send 'ToQuit' message on the designated mobile number.

• Gatekeepers such as Anganwadi (community-based integrated child development services delivery centres in India) workers , local panchayat leaders, Accredited social health activist (ASHA) workers: The researchers will build awareness about the project in the community and seek assistance from community gatekeepers. Gatekeepers would:

  1. Refer patients to the researchers.
  2. Inform people in the community about ToQuit, and ask them to contact ToQuit researchers
• Medical officer (MO) at primary health centres (PHC)/dentists at district hospital (DH) and private dental clinics: The PHC MO and dentists at DH and private clinics will be informed about the study and will be requested to refer patients who are using tobacco to the researchers for screening. Project team will leave calling cards with them to handout to the patients.

Screening:

When a 'ToQuit' message is received from a potential participant, one of the researchers will call on the mobile number and inform them about the study. If the respondent is willing to participate, researcher will proceed with initial screening for eligibility.

Tobacco users who do not fulfil the eligibility criteria will be thanked for their time, and will be suggested to seek help from a general practitioner, primary health centre, or a tobacco cessation clinic (TCC) in their area.

Consenting:

Potential participants who fulfill all the eligibility criteria will be given information about their participation in the study. Each potential participant will also be taken through the informed consent process and will be requested to provide written consent to participate in the study. The study researcher will ensure that any questions/queries the participant may have about the intervention are resolved before the participant provides their consent. All consent procedures will be audio-recorded with the permission of the participant.

When done remotely, if accessible to the potential participant, a researcher will send the information sheet and consent form via email/WhatsApp messaging app. If the potential participant does not have access to these, then the researcher will read out the information sheet before the screening.

Respondents who do not consent to participate in the study will be requested to give minimal sociodemographic data (age, gender, educational status, employment status, marital status), to examine whether there is a systematic difference between those who consent and those who do not. Reason for refusal will be recorded. They will be thanked for their time and will be suggested to seek help from a general practitioner, primary health centre, or a TCC in their area.

Baseline assessments:

Baseline measures administered to consenting participants will include:

1. Sociodemographic data:

   Data about potential confounders such as age, gender, employment status, educational status, and marital status will be collected through a socio-demographic form.

2. Tobacco use:

Type of tobacco used to understand if there are any differences in outcomes in smokers and smokeless tobacco users.

2. The Alcohol, Smoking and Substance Involvement Screening Test (ASSIST): Participants will then be administered the 6 questions about tobacco use from the ASSIST tool. The ASSIST is the screening test which covers all psychoactive substances including alcohol, tobacco and illicit drugs, and can help identify individuals who may have hazardous, harmful or dependent use of one or more substance.12 ASSIST has shown good concurrent, construct, predictive and discriminative validity to be used in India. It has good reliability and feasibility in culturally diverse settings and with different substance use patterns.13 3. Information required for delivery of intervention The participant will be asked the questions related to intervention delivery such as preferred days and time, and preferred language to send the intervention messages. The researcher will inform the participant about tentative schedule of receiving messages and end the call.

Sampling:

The participants will be stratified in two groups, smokeless tobacco users and smokers. Consenting eligible participants in both these strata will be randomly allocated to intervention or control group, in 1:1 ratio. An independent data manager will generate a randomisation list and participants will be randomised using Sequentially Numbered Opaque Sealed Envelopes to maximise allocation concealment. Consenting eligible participants' contact numbers will be added to the intervention database.

Intervention:

Participants from the ToQuit intervention arm will receive the messages on their mobile phone for 8 weeks. The intervention messages will consist of specific content areas derived from the formative research of the study. Some of the components of the intervention include: Psychoeducation about consequences of tobacco use and benefits of quitting; goal setting; managing goals and self-monitoring of behaviour; avoidance strategies; self-awareness and reflection messages; social and pharmacological support, help-seeking; identifying and managing cravings; and relapse prevention strategies. The messages will be sent for 3-4 days a week.

Participants from the control group will be sent the details of other functional tobacco helplines in India.

Nested qualitative study:

Qualitative outcome interviews will be conducted with a sub-sample of participants from the intervention group to gain insight into the impact of the mobile-based BI on the tobacco use. We will purposively select those who report abstinence and those who do not report abstinence. An interview guide will be developed focused on understanding the acceptability, feasibility, and perceived impact of the intervention.

Data analysis:

- Quantitative- Quantitative outcome data will be analysed using STATA 16, consistent with CONSORT guidelines for parallel randomized controlled trials. The process indicators will be presented using a trial flow chart.14 Baseline characteristics of participants with respect to consented and not consented, outcome assessments completed and not completed, and differences between intervention and the control arms, will be presented as proportions or means as appropriate. For primary analysis, complete case analysis will be conducted, adjusted for baseline variables with imbalance, if any. Secondary analysis will be intention-to-treat at the 3 month end-point regardless of treatment adherence. Regression analysis will be conducted for the treatment outcomes. Effect size will be reported as odds ratio with 95% confidence interval for the primary outcome, between intervention and control group. Standardized mean differences with 95% confidence intervals will be calculated for ASSIST scores of the two arms, and changes in scores from baseline and follow up.

Missing outcome data will be imputed using mean imputation or multiple imputation. Sensitivity analysis will be conducted assuming data missing at random, and outcome groups will be compared by replacing missing values with 'abstinence achieved' as 'best case scenario' and by failure to achieve the abstinence as 'worst case scenario'.

- Qualitative- Thematic analysis will be used to analyse the qualitative data from the outcome interviews, using N-vivo version 12.15 Analysis will involve generation of codes from raw data. Themes will be derived by retrieving pieces of data pertaining to codes and by examining their meaning in relation to the research questions (acceptability of the content and delivery of the mobile-based intervention, barriers to engagement, impact on tobacco use).