Social determinants of tobacco use among tribal communities in India: Evidence from the first wave of Longitudinal Ageing Study in India

Background Evidence on tobacco use among indigenous communities is scarce with available literature based either on a specific region or a particular tribe. Considering the large tribal population in India, it is pertinent to generate evidence on tobacco use among this community. Using nationally representative data, we aimed to estimate the prevalence of tobacco use and assess its determinants and regional variations among older tribal adults in India. Methods We analysed data from Longitudinal Ageing Study in India (LASI), wave-1 conducted in 2017–18. A sample of 11,365 tribal individuals aged ≥ 45 years was included in this study. Descriptive statistics was used to assess the prevalence of smokeless tobacco (SLT), smoking, and any form of tobacco use. Separate multivariable regression models were executed to assess the association of various socio-demographic variables with different forms of tobacco use, reported as adjusted odds ratio (AOR) with 95% confidence interval. Results The overall prevalence of tobacco use was around 46%, with 19% of smokers and nearly 32% smokeless tobacco (SLT) users. Participants from the lowest MPCE quintile group had a significantly higher risk of consuming (SLT) [AOR: 1.41 (95% CI: 1.04–1.92)]. Alcohol was found to be associated with both smoking [AOR: 2.09 (95% CI: 1.69–2.58)] and (SLT) [AOR: 3.05 (95% CI: 2.54–3.66)]. Relatively higher odds of consuming (SLT) were associated with the eastern region [AOR: 6.21 (95% CI: 3.91–9.88)]. Conclusion This study highlights the high burden of tobacco use and its social determinants among the tribal population in India, which can help tailor anti-tobacco messages for this vulnerable population to make tobacco control programs more effective.


Abstract:
ABSTRACT Background Evidence on tobacco use among indigenous communities is scarce; mainly based on either specific region or from particular tribe. Considering the large tribal population in India, it is pertinent to generate evidence on tobacco use among this community. We aimed to estimate the prevalence of tobacco use, assess its determinants and regional variations among older tribal adults in India using nationally representative data.

Methods
We analysed data from Longitudinal Ageing Study in India (LASI) wave-1 conducted in 2017-18. 11,365 tribal individuals aged ≥ 45 years were included in this study. Descriptive statistics was used to assess the socio-demographic profiles of the participants, expressed as frequency and proportions. Separate multivariable regression models were executed to find the association of various socio-demographic variables with different forms of tobacco, reported as adjusted odds ratio with 95% confidence interval.

Findings
The overall prevalence of tobacco use was found to be around 46%; with 19% of them smokers and nearly 32% of SLT (Smokeless tobacco) users. Participants from the most deprived group had a significantly higher risk of consuming smokeless tobacco users [Adjusted odds ratio (   Evidence on tobacco use among indigenous communities is scarce; mainly based on either 72 specific region or from particular tribe. Considering the large tribal population in India, it is 73 pertinent to generate evidence on tobacco use among this community. We aimed to estimate 74 the prevalence of tobacco use, assess its determinants and regional variations among older 75 tribal adults in India using nationally representative data.

77
We analysed data from Longitudinal Ageing Study in India (LASI) wave-1 conducted in 2017-78 18. 11,365 tribal individuals aged ≥ 45 years were included in this study. Descriptive statistics 79 was used to assess the socio-demographic profiles of the participants, expressed as frequency 80 and proportions. Separate multivariable regression models were executed to find the 81 association of various socio-demographic variables with different forms of tobacco, reported 82 as adjusted odds ratio with 95% confidence interval.

84
The overall prevalence of tobacco use was found to be around 46%; with 19% of them smokers This study highlights a high burden of tobacco use along with its social determinants which 93 can help in tailoring anti-tobacco messages for target population to make tobacco control 94 programs more effective. Globally, tobacco use is one of the greatest public health threats. The rampant use of various 116 tobacco products is a matter of concern in low-and middle-income countries (LMICs).

117
Evidence suggests that almost 1.3 billion people use tobacco in the world, out of which 80% Tobacco is commonly used in three ways i.e., smokeless, smoking and dual form. Due to socio-124 cultural acceptability, smokeless tobacco is highly prevalent in India which includes chewing 125 tobacco such as khaini, gutkha, betel quid with tobacco, mishri, gul, and gudakhu.(6) Other 126 kinds include smoking (cigarette, hookah, and bidi), and use of any form of tobacco which 127 simultaneously predispose a risk of oral submucous fibrosis (OSMF), a premalignant disorder 128 with a potency to transform into oral cancer.(7) Evidence supports that smoking is responsible 129 for health ailments of the cardiovascular and respiratory systems. (6) 130 Tobacco not only results in loss of lives but also levies associated social and economic costs. in India and their unique health-related behaviours, it is pertinent to generate evidence on 161 tobacco use among this community so as to pave a way for future tobacco control programs 162 and policies targeting this marginalized group. Hence, we aimed to estimate the prevalence of 163 tobacco use, assess its determinants and regional variations among older tribal adults in India 164 using nationally representative data.

166
Overview of data 167 This study is based on the Longitudinal Aging Study of India (LASI), wave-1, which was a  Among them, participants who referred their caste to be "Scheduled tribes" (STs) were 183 included in this study. Following this, the conclusive sample size of 11,365 tribal individuals 184 aged ≥45 years was achieved as per the objective of this study. to 59 years, 60 to 74 years, and ≥75 years. Education was stratified into four categories as "no 192 formal education", "up to standard VII", "standard VIII-X" and "higher secondary and above".

193
Respondents who did not work more than three months in their lifetime and were not employed 194 during the survey were grouped as "currently not working" while the rest were "currently 195 working". Couples who were married or lived together (live-in) were grouped into "living with 196 partner" while the rest (widowed, divorced, separated, deserted and never married) were 197 clubbed into "living without partner". 29 States and 6 Union Territories (except Sikkim) of 198 India were arranged into six regions (north, central, east, north-east, west, and south) based on 199 their geographical location. Participants who were hardly or never involved in sports or 200 vigorous activities were grouped as "physically inactive", while others (daily to as less frequent 201 as "one to three times a month") were merged into "physically active" group. and smokeless), were clubbed as "smoking". Next, the users of smokeless tobacco and dual 210 users were grouped as "smokeless tobacco" users. Additionally, a third group of "any tobacco" all weighted proportions was reported as a measure of uncertainty. We considered p-value of 220 <0.05 to be significant. Univariate logistic regression was applied to determine the crude odds 221 ratio between distinct outcomes and attributes. Significantly associated variables were used in 222 separate multivariable models to obtain adjusted association between outcome variables and 223 various correlates, presented as adjusted odds ratio with 95% CI. Sampling weights were 224 considered in the analysis to adjust for multistage sampling design. permission to use the dataset was received and data is being acknowledged wherever entailed.

231
For this analysis, eventually, 11,365 schedule tribes aged 45 years and above, were included.

232
Their average age was 59.3 (±10·7) years. The majority of them belonged to the age group of 233 45-59 years (54.5%). Nearly equal number of male and female interviewees were included with 234 a little female predilection (53.7%). A greater part of the tribal population never went to school 235 (54.6%) and lived in a rural setup (77.4%). Most of the participants were physically inactive 236 (52.6%) and worked (56.2%) while interviewed (Table 1).  (Figure 2).

Reference
Univariate logistic regression models identified participants aged 75 years and above, males, 262 rural residents, lesser years of schooling, currently working, having partner and alcohol were 263 associated with tobacco use (Table 2). Table 3 shows the association of socio-demographic, 264 and behavioral factors with different forms of tobacco products among the study population, 265 after adjusting for all the covariates. There was a 74% greater likelihood of any tobacco use A higher risk of smoking was identified among participants from affluent tribal groups.

314
However, a review suggested smoking to be higher in medium wealth quintiles group of the Our study suggests that the prevalence of tobacco use was high among those who had no formal All methods were carried out following relevant guidelines and regulations.

365
Ethics approval and consent to participate 366 The present study utilizes de-identified data from a secondary source.

380
The authors declare that they have no competing interests.

382
No funds were received to conduct the present research.