# A secondary multivariate decomposition analysis of factors influencing smokeless tobacco cessation across two Global Adult Tobacco Survey waves in India and Bangladesh

## TL;DR

This study examines factors influencing attempts to quit smokeless tobacco in India and Bangladesh using data from two global surveys.

## Contribution

The study uses multivariate decomposition to analyze changes in smokeless tobacco cessation attempts across two time periods in two countries.

## Key findings

- In India, exposure to health warnings and quitting advice had positive effects on quit attempts.
- In Bangladesh, print media warnings had positive effects, while reduced pro-SLT ads had negative effects.
- Overall, changes in quit attempt prevalence were not statistically significant in either country.

## Abstract

This study explores patterns and determinants of attempted quitting of smokeless tobacco (AQSLT) in India and Bangladesh, using data from two waves of the Global Adult Tobacco Survey (GATS).

A secondary analysis was conducted using nationally representative data from GATS Wave 1 (2009–2010) and Wave 2 (2016–2017) in India and Bangladesh. Adults who reported using SLT currently or within 12 months at the time of each wave were included. Changes in AQSLT prevalence, associated factors, and contribution of independent variables were assessed using descriptive statistics, logistic regression, and multivariate decomposition analysis.

While smokeless tobacco (SLT) use declined from Wave 1 to Wave 2 in both countries, changes in AQSLT prevalence were not statistically significant. In India, multivariate decomposition revealed that increases in the proportion of individuals who received quitting advice from a doctor or healthcare professional, noticed health warnings on SLT products, saw SLT warnings in print media, and were exposed to pro-SLT advertisements, had positive endowment effects on AQSLT. Behavioral changes among those exposed to SLT product warnings had the strongest negative composition effect. In Bangladesh, increased exposure to warnings in print media had positive, while decreased exposure to pro-SLT advertisements had a negative endowment effect on AQSLT. No significant composition effects were observed in Bangladesh.

Interventions such as providing advice and health warnings show inconsistent effects on quit behavior. Findings warrant further evaluation of effectiveness of interventions and exploring tested culturally sensitive cessation strategies, that effectively motivate quit attempts among SLT users.