Advertisement

Association of tobacco with hypertension in adult females: Evidence from National Family Health Survey-IV for an aspirational Indian state

Open AccessPublished:May 09, 2022DOI:https://doi.org/10.1016/j.cegh.2022.101061

      Abstract

      Background

      Elevated blood pressure and tobacco consumption have been the top two causes of preventable mortality worldwide. Females in the reproductive age carry a higher health impact for self and the offspring when exposed to one or both these risks. An in-depth assessment of the sociocultural determinants of hypertension and tobacco usage in an aspirational Indian state like Odisha is carried out through secondary analysis of the National Family Health Survey-IV (NFHS-4) dataset.

      Methods

      Data of 30,587 women aged between 15 and 49 years were extrapolated for analysis through frequency, percentage and logistic regression model to determine the predictors of tobacco use and hypertension and to correlate the two in the study population.

      Results

      In Odisha, 12.8% adult women were addicted to tobacco, mostly smokeless form (93.3%). High blood pressure was recorded in 7.3% of the respondent females. Increasing age, urban residence, extremes of wealth index, ever married status, obesity, diabetes, and tobacco usage (more for smokeless) were the predictors of hypertension. Education, caste and occupation had no significant associations.

      Conclusion

      Targeted interventions involving gender, marital status, obesity, residence and other socio-economic profiles may be adopted for risk reduction of non-communicable diseases like hypertension in the reproductive age females.

      Keywords

      1. Introduction

      The prevalence of non-communicable diseases (NCDs) has attained an epidemic proportion in developed and developing nations. NCDs share more than half of the disease burden and more than three-quarters of the deaths on the globe. Cardiovascular diseases (CVDs), cancers, respiratory diseases, and stroke account for 80% of premature NCD deaths and around 44% of CVD deaths.

      The Global Health Observatory: World Health Organization; Total NCD Mortality. Available from: https://www.who.int/data/gho/data/themes/topics/indicator-groups/indicator-group-details/GHO/total-ncd-mortality.

      Every fourth person on the earth is hypertensive. Hypertension is the third killer disease in the list of all age mortality. It accounts for 57% of stroke and 24% coronary heart disease deaths in India.
      • Kearney P.M.
      • Whelton M.
      • Reynolds K.
      • Muntner P.
      • Whelton P.K.
      • He J.
      Global burden of hypertension: analysis of worldwide data.
      Even though there is a declining trend (11%) in the prevalence from 2000 to 2010, the figures are still worrying at 28.4% in low-income countries and 17.7% in high-income countries as per 2015 data.
      World Health Organization
      Global Health Risks: Mortality and Burden of Disease Attributable to Selected Major Risks.
      By 2025, it is estimated to swell by 60%.
      • Kearney P.M.
      • Whelton M.
      • Reynolds K.
      • Muntner P.
      • Whelton P.K.
      • He J.
      Global burden of hypertension: analysis of worldwide data.
      Modifiable risk factors for NCDs include environmental and life style parameters. The socio-behavioral factors like food, exercise, stress, sleep, body mass index (BMI), drugs, alcohol, and tobacco are some known correlates. Hypertension and tobacco consumption have been the top two causes of preventable mortality worldwide.

      The Global Health Observatory: World Health Organization; Total NCD Mortality. Available from: https://www.who.int/data/gho/data/themes/topics/indicator-groups/indicator-group-details/GHO/total-ncd-mortality.

      Consumption of tobacco in smoke or smokeless forms has been a social malady for years across the globe. Sustainable Development Goal (SDG) aims at reduction in global burden of hypertension by one third by 2030. World Health Organization (WHO) has proposed a 25% reduction tobacco use by 2025.An estimate of the association between tobacco usage and hypertension and other interplaying risk factors will contribute towards risk reduction via guidelines and policies.
      India, a classic member of the low income nations faces the dual challenges of NCDs and Communicable diseases today. Odisha, an aspirational Indian state ranks third from the bottom in National Institute of Transforming India (NITI) Ayog health index list-2017-18. This Eastern Indian state boasts of a high tobacco usage as well. An in-depth assessment of the association between hypertension and tobacco usage among the female population of the state will be a helpful guide for the policy makers and health planners at large.

      2. Methods

      The study was designed to determine the predictors of hypertension and tobacco usage among females in Odisha aged 15–49 years and the associations between them.

      2.1 Study design and study population

      The current study used the data set of 33,721 adult females aged between 15 and 49 years from the thirty districts of the state of Odisha collated in National Family Health Survey (NFHS-4)
      NFHS 4 India fact sheet
      International Institute for Population Sciences.
      , a nationwide cross-sectional study conducted during 2015–16. The fourth round of the survey, for the very first time covered all the 640 districts of the country and expanded the domain of clinical, anthropometric, and biochemical testing to include measurement of blood glucose and blood pressure.
      The surveyors recorded three readings of blood pressure in each subject at a gap of 5 min and the average of the last two readings was reported discarding the first reading. Respondents with ‘observer recorded hypertension’ defined as ‘systolic blood pressure of more than 140 mmHg and/or diastolic blood pressure of more than 90 mm Hg’ were considered ‘hypertensive'.1953 subjects with ‘missing data’ were excluded. 1181 pregnant females were also excluded to avoid ‘pregnancy-associated hypertension’ as an effect modifier. A total of 30,587 samples were included for the analysis.

      2.2 Data variables

      A respondent who reported smoking or using any smokeless tobacco product on a daily or less-than-daily basis at the time of the survey is labeled as a ‘current tobacco user’. A subject who denied using smoke or smokeless tobacco in any form on a daily or less-than-daily basis at the time of the survey was identified as a ‘tobacco non-user’. A female who smoked cigarettes/pipe/cigars/bidis/hookah is labeled as ‘tobacco smoker’. A subject who chewed tobacco in the form of khaini/gutkha/paan masala-tobacco/paan-tobacco is called as ‘smokeless tobacco user’.
      Socio-demographic variables (age, wealth status, gender, caste, marital status, occupation, education and place of residence), physical factors like BMI and risk factors like diabetes were the other study variables included for analysis.

      2.3 Data collection methods

      Two of the researchers of our team went through the NFHS-4 data sheet provided to us and entered the requires variables in SPSS version.21. and after completion of data entry; data were cleaned up and the missing data were excluded and normality of test were checked and analysis was done.

      2.4 Data analysis

      Data were analyzed using SPSS software version-21. Descriptive statistics characterized the study population and the hypertension and tobacco use status. Application of Kolmogorov-Smirov test and Shaipro Wilks test confirmed that the data were having normal distribution with a significance level of 0.025. The Chi-Square test assessed the association between socio-demographic variables with tobacco usage and the hypertension prevalence. A Logistic regression analysis model was used to assess the determinants of tobacco usage and hypertension. Initially we conducted an uni-variate analysis by using all the determinants and crude odds ratio was calculated and then we have done the multivariate analysis by adjusting all the confounders and got the Adjusted odds ratio with 95% confidence interval. Hosmer and Lemeshow test and Cox & Snell R square tested the goodness-of-fit of the models. A p-value below 0.05 was considered significant to acknowledge the association.

      3. Results

      The mean age of the study population (n = 30,587) was 30.50 (±SD 9.92) years. ‘Observer recorded hypertension’ was noted in 2231 (7.3%) subjects who had a mean age of 37.44 (±SD 8.33) years. Normotensive females had a mean age of 29.95 (±SD 9.83) years. The mean systolic and diastolic blood pressure readings in the hypertensive group was 154.53 (±SD 24.254) and 96.66 (±S.D 16.554) mmHg respectively. For non-hypertensive subjects, the corresponding means were 112.53(±SD 11.142) and 76.71(±S.D 18.291 mmHg.
      Among the hypertensives, ‘45–49 years' constituted the largest age quintile (25.7%), and ‘15–19 years' was the smallest age quintile (2.9%). Majority of the respondents (76.8% hypertensives and 80.6% non-hypertensives) were from rural background and Hindus (93–95%). ‘Females with secondary education’ constituted the largest education quintile in either group. Majority of the women (84.4% & 68.9%) were ‘married'.
      Table 1 presents that the odds of tobacco usage were 29 times higher (aOR, 29.508[95% CI, 22.857–38.093]) in women of higher age (45–49 years) as compared to lower age (15–19 years). Rural subjects had 1.8 times higher odds (aOR, 1.818[95% CI, 1.641–2.016]) of using tobacco. Educational grades demonstrated no significant association with tobacco usage. Only agricultural workers were found with a significantly lower odds (aOR, 0.059 [95% CI, 0.375–0.881]) of addiction as compared to the unemployed or other professions. The wealth status of the respondents had an inverse association with tobacco usage. The poorest had 7 times higher odds (aOR, 7.240 [95% CI, 5.608–9.346]) of being addicted to tobacco. Tribal subjects had higher odds (1.538 ± 1.412–1.675) of tobacco addiction. The odds of tobacco usage increased from ‘unmarried’ to ‘currently married’ (6 times) to ‘widowed’ (12 times) to ‘divorcee’ (7 times) to ‘separated’ (10 times) women in that order.
      Table 1Multivariate logistic regression analysis for predictors of tobacco usage.
      Independent variables of tobacco usageAdjusted Odds

      Ratio (AOR)
      95% CIP-value
      Age (in yrs)
      15–191(Ref)
      20–242.9072.184–3.8690.001*
      25–295.8684.486–7.6760.001*
      30–3410.6568.206–13.8380.001*
      35–3916.36412.645–21.1750.001*
      40–4421.94616.976–28.3720.001*
      45–4929.50822.857–38.0930.001*
      Residence1 (Ref)

      1.818
      1.641–2.0160.001*
      Urban
      Rural
      Education

      Illiterate

      Primary

      Secondary

      Higher secondary
      1 (Ref)
      1.0470.938–1.1680.416
      0.9770.900–1.0600.573
      1.0310.894–1.1890.671
      Occupation
      Unemployed1 (Ref)
      Professional0.6480.298–1.4120.275
      Clerical1.9450.412–9.1910.401
      Salesman1.3530.726–2.5200.341
      Agricultural0.5790.375–0.8810.011*
      Services0.8550.629–1.1630.381
      Skilled/Unskilled0.8930.620–1.2860.542
      Others1.1840.668–2.0970.563
      Wealth index
      Richest1 (Ref)
      Poorest7.2405.608–9.3460.001*
      Poorer5.1593.983–6.6820.001*
      Middle3.5702.735–4.6600.001*
      Richer2.0991.570–2.8070.001*
      Caste
      Non-Tribal1 (Ref)
      Tribal1.5381.412–1.6750.001*
      Marital status
      Unmarried1 (Ref)
      Married6.1015.288–7.0380.001*
      Widow12.0429.861–14.7060.001*
      Divorcee7.3554.511–11,9900.001*
      Separated10.4077.570–14.3060.001*
      * Significant with p-value of below 0.05.
      Smokeless tobacco was found to be the dominant form (∼93%) of tobacco consumption in the respondents as presented in Table 2. They mostly used paan-tobacco, gutkha-paan masala and chewable tobacco.
      Table 2Forms of tobacco consumption (n = 30,587).
      Tobacco formTobacco users amongst hypertensives (n = 461)

      No (%)
      Tobacco users amongst non-hypertensives (n = 3472)

      No (%)
      Total (n = 3933)

      No (%)
      N46134723933
      Smoke30 (6.51)237(6.82)304(7.61)
      Cigarette1 (0.22)10 (0.29)11 (0.28)
      Pipe0 (0.0)4 (0.12)4 (0.10)
      Cigar0 (0.0)6 (0.17)7 (0.15)
      Other smokes29 (6.3)217 (6.25)246 (6.25)
      Smokeless431 (93.49)3235(93.17)3629(92.39)
      Chewing tobacco58 (12.6)363 (10.5)421 (10.7)
      Snuff5 (1.1)20 (0.6)25 (0.6)
      Gutkha/paan masala162 (35.1)1406 (40.5)1568 (39.9)
      Paan-tobacco206 (44.7)1446 (41.6)1652 (42.0)
      The state had a 7.3% prevalence of hypertension in adult females The district reporting the maximum and minimum prevalence in the state were Deogarh (8.8%) and Nayagarh (4.16%) as shown in Fig. 1.
      Fig. 1
      Fig. 1Distric-Wise prevalence of hypertension.
      Table 3 presents that the odds of developing hypertension increased with increasing age. Older females (45–49 years) had 14 times higher odds (aOR, 14.515[95% CI, 11.206–18.801]) of being hypertensive than the young females (15–19 years). Urban females had higher odds (aOR, 1.25[95% CI, 1.13–1.38]) of being hypertensive. Urban subjects demonstrated a higher risk of having higher blood pressure (aOR, 1.254 [95% CI, 1.131–1.389]). Education and caste profile of the subject had no significant association with hypertension. ‘Unmarried’ women had the minimum risk for hypertension; ‘widows’ carried the highest risk (aOR, 6.33 [95%CI, 5.013–7.991]). Tobacco usage of both smoke and smokeless form had a higher risk (smoke: aOR, 1.769[95% CI, 1.206–2.594], smokeless: aOR, 1.804 [95% CI, 1.606–2.027]). Obesity had 2.5 times higher odds (aOR, 2.511[95% CI, 2.118–2.975]) of elevated blood pressure where as ‘Diabetes’ carried no significant risk. Tobacco consumption in any form was associated with higher risk of hypertension, more so with the smokeless form.
      Table 3Multivariate logistic regression for various predictors of hypertension.
      Independent variables

      of Hypertension
      Adjusted

      Odds Ratio (AOR)
      95% CIP-value
      Age (years)
      15–191(Ref)
      20–242.1671.612–2.9120.001*
      25–293.3622.539–4.4510.001*
      30–345.9394.539–7.7720.001*
      35–399.4227.250–12.2450.001*
      40–4411.9439.203–15.4990.001*
      45–4914.51511.206–18.8010.001*
      Residence1.131–1.3890.001*
      Rural1 (Ref)
      Urban1.254
      Education
      Illiterate1 (Ref)
      Primary1.0710.935–1.2270.321
      Secondary0.9800.885–1.0850.698
      Higher Secondary1.1700.987–1.3880.070
      Wealth index
      Poorest1 (Ref)
      Poorer0.8780.785–0.9820.023*
      Middle0.9580.846–1.0840.493
      Richer0.9820.848–1.1380.813
      Richest1.2741.089–1.4900.002*
      Caste
      Tribal1 (Ref)
      Non-Tribal1.0320.919–1.1580.595
      Marital status
      Unmarried1(Ref)
      Married3.8953.345–4.5360.001*
      Widow6.3305.013–7.9910.001*
      Divorcee3.1351.567–6.2720.001*
      Separated2.7151.627–4.5300.001*
      Diabetes
      No1(Ref)
      Yes1.1580.884–1.5170.287
      Obesity
      No1(Ref)
      Yes2.5112.118–2.9750.001*
      Tobacco use
      No1(Ref)
      Yes1.8021.609–2.0170.001*
      Tobacco type
      No Tobacco1(Ref)
      Smoke1.7691.206–2.5940.004*
      Smokeless1.8041.606–2.0270.001*
      * Significant with p-value of below 0.05.

      4. Discussion

      The prevalence of tobacco usage among the 15–49 years females of Odisha was estimated from NFHS dataset. 12.85% of the study population used tobacco. It is higher than the national average of 6.8% reported in NFHS-4(2016) for adult females.
      India-Key Indicators
      International Institute for Population Sciences.
      There has been a decreasing trend of tobacco use across the globe during the last two decades, a drop by 8.2% in females and 12.6% in males. Indian women reported 14.2% prevalence of tobacco addiction (esp. smokeless) as the national average in Global Adult Tobacco Survey (GATS)-2(2017), an encouraging drop of 6% since GATS-1 in 2010. Alarmingly, Odisha, alongside its neighboring states of West Bengal and Jharkhand reported the highest all-age-prevalence of 56.53% of smokeless tobacco users in GATS-II, a surge from 35.6% in 2010.
      • Ghosal S.
      • Sinha A.
      • Kanungo S.
      • Pati S.
      Declining trends in smokeless tobacco use among Indian women: findings from Global Adult Tobacco Survey I and II.
      Most (∼93%) tobacco users consumed smokeless forms; ‘paan-tobacco’ and ‘gutkha-paan masala’ topped the preference list. Different studies on Indian women have consistently reported smokeless form to be the predominant form of tobacco usage. GATS-I highlighted snuff/oral gul gudakhu (5.82%) and GATS-II reported about ‘betel quid with tobacco’(3.91%) to be the most used forms.
      • Ghosal S.
      • Sinha A.
      • Kanungo S.
      • Pati S.
      Declining trends in smokeless tobacco use among Indian women: findings from Global Adult Tobacco Survey I and II.
      In 2016, Kerala, Odisha, Tamil Nadu and Goa reported zero prevalence of smoking in females.
      • Abdulkader R.S.
      • Sinha D.N.
      • Jeyashree K.
      • et al.
      Trends in tobacco consumption in India 1987–2016: impact of the World health Organization framework convention on tobacco control.
      This pattern may be explained by postulations like more social and familial acceptance of smokeless tobacco in females as it is for smoking in males.
      The prevalence of tobacco use increased with increasing age; it was 29 times higher in the 45–49 years sub-group than the 15–19 years sub-group of females. A systematic review by Priya Mohan et al., in 2018 confirms this pattern of relationship of tobacco use and age.
      • Mohan P.
      • Lando H.A.
      • Panneer S.
      Assessment of tobacco consumption and control in India.
      A study on Nepalese women aged between 36 and 49 years reported 2.4 times higher usage of any form of tobacco than their younger (15–24 years) counterparts. A number of studies corroborate that prevalence of smokeless tobacco usage increase with age.
      India-Key Indicators
      International Institute for Population Sciences.
      ,
      • Sinha D.N.
      • Gupta P.C.
      • Ray C.S.
      • Singh P.K.
      Prevalence of smokeless tobacco use among adults in WHO South-East Asia.
      ,
      • Sreeramareddy C.T.
      • Ramakrishnareddy N.
      • Kumar H.H.
      • Sathian B.
      • Arokiasamy J.T.
      Prevalence, distribution and correlates of tobacco smoking and chewing in Nepal: a secondary data analysis of Nepal Demographic and Health Survey-2006.
      The rural women in Odisha in the reproductive age group had 1.8 times higher risk of tobacco addiction. The country average estimated from NFHS-4 data in some studies reports 8.1% rural prevalence,4.4% urban and 6.8% total prevalence.
      • Mohan P.
      • Lando H.A.
      • Panneer S.
      Assessment of tobacco consumption and control in India.
      GATS-II reports that most tobacco consuming women were from rural background with minimal education.
      • Ghosal S.
      • Sinha A.
      • Kanungo S.
      • Pati S.
      Declining trends in smokeless tobacco use among Indian women: findings from Global Adult Tobacco Survey I and II.
      Gupta et al. also highlighted about rural dominance in their study in North Indian community.
      • Gupta V.
      • Yadav K.
      • Anand K.
      Pattern of tobacco use across rural, urban, urban slum populations in a North Indian Community.
      Among the female subjects surveyed in the present study, the wealth status inversely correlated with tobacco usage in Odisha. Similarly, risk of tobacco addiction in the poorest quintile was more than the richest quintile across all states and UTs except Bihar, Nagaland, and Tripura as reported in GATS-II study.
      • Ghosal S.
      • Sinha A.
      • Kanungo S.
      • Pati S.
      Declining trends in smokeless tobacco use among Indian women: findings from Global Adult Tobacco Survey I and II.
      Educational status carried no association with tobacco usage in the study population covered. Some studies have highlighted educational level as an important contributing factor for tobacco usage with an inverse relationship especially for smokeless forms.
      • Ghosal S.
      • Sinha A.
      • Kanungo S.
      • Pati S.
      Declining trends in smokeless tobacco use among Indian women: findings from Global Adult Tobacco Survey I and II.
      Researchers report an inconsistent association between profession and tobacco addiction.
      • Ghosal S.
      • Sinha A.
      • Kanungo S.
      • Pati S.
      Declining trends in smokeless tobacco use among Indian women: findings from Global Adult Tobacco Survey I and II.
      ,
      • Mohan P.
      • Lando H.A.
      • Panneer S.
      Assessment of tobacco consumption and control in India.
      ,
      Odisha demonstrated a similar pattern. Surprisingly, women working in agricultural sector had less risk of being a tobacco user as compared to the unemployed or any other job. This aberration may be explained by the idea that, most probably the NFHS survey missed them during the day time when most of them were in the fields.
      The present study reports tribal women having higher prevalence of tobacco abuse in Odisha like any other Indian state.
      As per 2011 census, Odisha houses the third largest and the sixth densest (22.8%) tribal population in India.

      Census of India : Provisional Population Totals India : Paper 1 : Census 2011 (censusindia.gov.in).

      Among the tribals in India, 40.6% are below poverty line, 90% live in rural areas, 41% are illiterate. They are identified with poor health indicators, greater morbidity burden and limited access to health care services.
      ,
      Odisha had a prevalence of hypertension at a rate of 7.3% amongst the adult females studied. NFHS-IV data reflected a prevalence of 11.3% as national average in the year 2016. The all age and sex prevalence of hypertension in India was 30.7%; it was 23.7% in women. Eastern zone adults have the highest prevalence in men and women.
      • Gupta R.
      • Xavier R.
      Hypertension: the most important non communicable disease risk factor in India.
      A similar prevalence of 26.5% was reported by a study using health survey–based nationwide blood pressure data.
      • Ramakrishnan S.
      • Zachariah G.
      • Gupta K.
      • et al.
      Prevalence of hypertension among Indian adults: results from the great India blood pressure survey.
      Anchala et al. reported a similar prevalence of 29.8% in their meta-analysis from 142 hypertension prevalence studies from 1950 to 2013.
      • Anchala R.
      • Kannuri N.K.
      • Pant H.
      • et al.
      Hypertension in India: a systematic review and meta-analysis of prevalence, awareness, and control of hypertension.
      The lower prevalence reflected in NFHS-IV data may be due to overrepresentation of women (87.2%) and higher younger population (15–49 years).
      The present study population demonstrated a linear association between high blood pressure risk and increasing age. The risk increased by about 11–14 folds for the women in the forties as compared to the teens. J. Kishore et al. recognized age as non-modifiable risk factor, attributing to an increased risk of cardiovascular problems, such as hypertension.
      • Kishore J.
      • Gupta N.
      • Kohli C.
      • Kumar N.
      Prevalence of hypertension and determination of its risk factors in rural Delhi.
      Urban females carried a higher risk for hypertension in the present analysis. Gupta et al. (2012) had reported ∼17% higher prevalence in urban women aged between 35 and 70 years.
      • Gupta R.
      • Pandey R.M.
      • Misra A.
      • et al.
      High prevalence and low awareness, treatment and control of hypertension in Asian Indian women.
      A meta-analysis conducted in India in 2014 reports the gap to be 8%.
      • Anchala R.
      • Kannuri N.K.
      • Pant H.
      • et al.
      Hypertension in India: a systematic review and meta-analysis of prevalence, awareness, and control of hypertension.
      Epidemiological studies over the last 20 years have reported a stabilized urban prevalence around 25–30% with a rising rural trend from 15 to 25%. This urban-rural convergence of hypertension in India may be ascribed to changes in lifestyles of the rural folks.
      • Gupta R.
      Convergence in urban-rural prevalence of hypertension in India.
      In Odisha, females in the extremes of wealth strata (poorer & richest) were having higher risk of hypertension. Gupta et al. had reported a higher hypertension prevalence in Indian states with higher urbanization and social development.
      • Gupta R.
      • Gaur K.
      • Ram C.V.
      Emerging trends in hypertension epidemiology in India.
      A study from Maharashtra, based on DLHS-4 (2013) dataset had reports of higher blood pressure prevalence among the elderly, urbanites, rich and obese.
      • Bhise M.D.
      • Patra S.
      Prevalence and correlates of hypertension in Maharashtra, India: a multilevel analysis.
      In a study in 2020, higher prevalence was linked to older age, male sex, higher BMI, higher wealth status, and urban residence.
      • Gupta R.
      • Gaur K.
      • Ram C.V.
      Emerging trends in hypertension epidemiology in India.
      In the present study, education status of the respondents did not correlate with risk of hypertension. Most other studies have yielded variable results as regard association of education and hypertension. In an Indian study from Banaras, education, socioeconomic status, and alcohol use were not statistically linked to hypertension.
      • Shikha S.
      • Ravi S.
      • Gyan P.
      Prevalence and associated risk factors of hypertension: a cross-Sectional study in urban Varanasi.
      Higher educational status correlated with higher systolic blood pressure in a series of cross-sectional studies on middle-SES urban localities in Jaipur during 1992–2010.
      • Gupta R.
      • Guptha S.
      • Gupta V.P.
      • Agrawal A.
      • Gaur K.
      • Deedwania P.C.
      Twenty-year trends in cardiovascular risk factors in India and influence of educational status.
      Gupta et al. had found similar trends.
      • Gupta R.
      • Pandey R.M.
      • Misra A.
      • et al.
      High prevalence and low awareness, treatment and control of hypertension in Asian Indian women.
      Women in tribal community in Odisha were found to be having no protection from hypertension risk as reported in the present study. A study on tribal females in Bihar reported high (12.1%) prevalence of hypertension.
      • Anand N.
      • Hussain S.
      Prevalence of hypertension & associated risk factors among tribal population in a rural community of Katihar.
      Another secondary data analysis in 2020 highlighted scheduled tribes to be having higher risk than those from the upper caste.
      • Uddin J.
      • Acharya S.
      • Valles J.
      • Baker E.H.
      • Keith V.M.
      Caste differences in hypertension among women in India: diminishing health returns to socioeconomic status for lower caste groups.
      Kani tribe in Kerala had 48.3% prevalence of hypertension, higher than non-tribals in the state (Sajeev P et al.).
      • Sajeev P.
      • Soman B.
      Prevalence of non-communicable disease risk factors among the Kano tribe in Thiruvananthapuram district, Kerala.
      The traditional myth that tribals are naturally shielded form CVD risks is slowly melting away. Upliftment in socio-economic standards is exposing the thrifty phenotypes to the risks of abundance.
      After adjustment of other confounders, ‘ever married’ status was found to be a significant determinant (6–10 times) of hypertension risk in adult females. Similarly, in an Iranian longitudinal study ‘never married status’ had a lower risk of hypertension.
      • Ramezankhani Azra
      • et al.
      Associations of marital status with diabetes, hypertension, cardiovascular disease and all-cause mortality: a long-term follow-up study.
      Studies form Ghana (2014) and Bhutan (2021) demonstrate that marital status is an independent risk factor for hypertension in Ghana for women, rather than men.
      • Li K.
      • Ma X.
      • Yuan L.
      • et al.
      Age differences in the association between marital status and hypertension: a population-based study.
      • Kohori H.S.
      • Uematsu H.
      • Dorji N.
      • et al.
      Findings from the national survey for non-communicable disease risk factors and mental health using WHO STEPS in Bhutan.Published.
      • Tuoyire D.A.
      • Ayetey H.
      Gender differences in the association between marital status and hypertension in Ghana.
      Obesity carried a 2.5-fold higher risk for hypertension in the females enrolled from Odisha NFHS-4. However, diabetes had no significant associations with risk of hypertension. A study on nonpregnant females involving 319 subjects by NICPR (National Institute of Cancer Prevention and Research, India) in 2016 had published significant correlation of BMI with systolic and diastolic blood pressure. This association assumes significance since weight is a modifiable risk factor for hypertension and its complications.
      • Bhatt D.
      • Sharma S.
      • Gupta R.
      • Sinha D.N.
      • Mehrotra M.
      Predictors of hypertension among nonpregnant females attending health promotion clinic with special emphasis on smokeless tobacco: a cross-sectional study.
      The working definition of ‘diabetes’ in the survey did not have the ‘postprandial or fasting blood glucose criteria’, it only used the ‘random blood sugar’ instead and hence might have allowed for some missed cases. The lower number of diabetics in the relatively younger female subjects sampled in NFHS-4 may be the explanation. Multiple studies have favored obesity/overweight as a strong determinant of hypertension in adult females.
      • Bhinse M.D.
      • Patra S.
      Prevalence and correlates of hypertension in Maharashtra, India: a multilevel analysis. Population Studies.
      • Rajkumar E.
      • Romate J.
      Behavioural risk factors, hypertension knowledge, and hypertension in rural India.
      • Datta B.K.
      • Husain M.J.
      Uncontrolled hypertension among tobacco-users: women of prime childbearing age at risk in India.
      The present study reports tobacco abuse to be a significant (1.8-fold) determinant of hypertension in adult females even after adjustment of other confounders. This association has been strongly corroborated by various other studies in India and abroad. One publication has highlighted the dual risk of high blood pressure and tobacco consumption esp. in poor women in reproductive age group.
      • Datta B.K.
      • Husain M.J.
      Uncontrolled hypertension among tobacco-users: women of prime childbearing age at risk in India.
      Studies from Europe reported higher risk of hypertension for snuff users.
      • Bolinder G.M.
      • Ahlborg B.O.
      • Lindell J.H.
      Use of smokeless tobacco: blood pressure elevation and other health hazards found in a large-scale population survey.
      ,
      • Hergens M.P.
      • Alfredsson L.
      • Bolinder G.
      • Lambe M.
      • Pershagen G.
      • Ye W.
      Long-term use of Swedish moist snuff and the risk of myocardial infarction amongst men.
      Use of smokeless tobacco has been proved to be strongly associated with systolic hypertension in females in India.
      • Bhatt D.
      • Sharma S.
      • Gupta R.
      • Sinha D.N.
      • Mehrotra M.
      Predictors of hypertension among nonpregnant females attending health promotion clinic with special emphasis on smokeless tobacco: a cross-sectional study.
      Another study had emphasized about the overlapping risk of CVDs when both hypertension and tobacco use combined.
      India state level disease burden initiative CVD collaborators-the changing patterns of cardiovascular diseases and their risk factors in the states of India: the Global Burden of Disease Study 1990–2016.
      Chewing tobacco have been thought to allow for higher and longer blood levels of nicotine than smoking. The sodium consumed alongside smokeless tobacco also contributes to high blood pressure.

      4.1 Strengths and limitations of the study

      The survey participants demonstrated a very high response rate of 97% among the eligible women. Respondents with missing data were eliminated during analysis. Confounders like pregnancy have been excluded from the study. The study bears some limitations as well. Overrepresentation of women in the survey may have reflected underreporting or overreporting for some variables Apart from anthropometric, blood pressure and blood sugar measurements, all other information provided was self-reported, which were likely to involve some bias. The blood pressure and blood sugar estimation was done only in single sitting leaving scope for error. Some cases of secondary hypertension might have been included in the surveyed data. The load of tobacco usage in terms of duration and volume has not been estimated in the survey. Second and third hand smoke exposure for the subjects have not been factored in to asses tobacco risks. Causality could not be derived from this cross-sectional study.

      4.2 Clinical implication for health managers and policy makers

      Health policies and planning need to be prioritized towards targeted approach instead of ‘one size fits all’ approach followed till date so far as prevention and control of risk factors of tobacco usage and non-communicable diseases like hypertension are concerned.

      4.3 Recommendation for future research

      Quasi experimental study designs like regression-discontinuity method and interrupted time series can yield more reliable insights towards causality assessment of the risk factors. Quantifying tobacco usage in terms of usage-years or dosing in the primary data series can be more useful.

      5. Conclusion

      Convergence of many risk factors (residence, wealth status, marital status, age, caste) for both tobacco usage and hypertension in adult females in the reproductive age pose as a multiplying impact on health outcomes in this representative study in a aspirational state of India like Odisha.

      Ethical approval

      The project was undertaken after allotment under ‘open call for writing manuscripts’ using NFHS-IV data advertised by the Resource Center for Cardiovascular Health, Department of Community Medicine and School of Public Health, Postgraduate Institute of Medical Education and Research (PGIMER), Chandigarh under the project ‘Strengthening management of hypertension services through capacity building, media and communication and stakeholder engagement in the state of Punjab’. The ethical approval for study was obtained from the Institute Ethics Committee, PGIMER, Chandigarh (PGI/IEC/2021/001139).

      Availability of data and material

      This research work was performed based on secondary data which is freely available in public domain for all registered users from the Demographic and Health Surveys Program (DHS) at https://dhsprogram.com.

      Funding

      This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

      Declaration of competing interest

      The authors declared no potential conflicts of interest concerning the research, authorship and/or publication of this article.

      Acknowledgments

      The authors would like to acknowledge and thank the Resource Center for Cardiovascular Health, PGIMER, Chandigarh for providing technical support for the project. Acknowledgement is also extended to Global Health Advocacy Incubator (GHAI) for supporting the study and the demographic and Health Surveys ( DHS ) Program for providing the data set (survey ref no. 155509 downloaded on June 3, 2021).

      References

      1. The Global Health Observatory: World Health Organization; Total NCD Mortality. Available from: https://www.who.int/data/gho/data/themes/topics/indicator-groups/indicator-group-details/GHO/total-ncd-mortality.

        • Kearney P.M.
        • Whelton M.
        • Reynolds K.
        • Muntner P.
        • Whelton P.K.
        • He J.
        Global burden of hypertension: analysis of worldwide data.
        Lancet. 2005; 365 (Jan 15-21): 217-223https://doi.org/10.1016/S0140-6736(05)17741-1
        • World Health Organization
        Global Health Risks: Mortality and Burden of Disease Attributable to Selected Major Risks.
        World Health Organization, 2009
        • NFHS 4 India fact sheet
        International Institute for Population Sciences.
        (Available from)
        • India-Key Indicators
        International Institute for Population Sciences.
        2021 (Available from:)
        • Ghosal S.
        • Sinha A.
        • Kanungo S.
        • Pati S.
        Declining trends in smokeless tobacco use among Indian women: findings from Global Adult Tobacco Survey I and II.
        BMC Publ Health. 2021; 21 (Nov 9): 2047https://doi.org/10.1186/s12889-021-12089-6
        • Abdulkader R.S.
        • Sinha D.N.
        • Jeyashree K.
        • et al.
        Trends in tobacco consumption in India 1987–2016: impact of the World health Organization framework convention on tobacco control.
        Int J Publ Health. 2019; 64 (Jul): 841-851https://doi.org/10.1007/s00038-019-01252-x
        • Mohan P.
        • Lando H.A.
        • Panneer S.
        Assessment of tobacco consumption and control in India.
        Indian J Clin Med. 2018; 9https://doi.org/10.1177/1179916118759289
        • Sinha D.N.
        • Gupta P.C.
        • Ray C.S.
        • Singh P.K.
        Prevalence of smokeless tobacco use among adults in WHO South-East Asia.
        Indian J Cancer. 2012; 49 (Oct-Dec): 342-346https://doi.org/10.4103/0019-509X.107726
        • Sreeramareddy C.T.
        • Ramakrishnareddy N.
        • Kumar H.H.
        • Sathian B.
        • Arokiasamy J.T.
        Prevalence, distribution and correlates of tobacco smoking and chewing in Nepal: a secondary data analysis of Nepal Demographic and Health Survey-2006.
        Subst Abuse Treat Prev Pol. 2011; 6 (Dec 20) (Available from:): 1-9
        • Gupta V.
        • Yadav K.
        • Anand K.
        Pattern of tobacco use across rural, urban, urban slum populations in a North Indian Community.
        Indian J Community Med. 2010; 35 (April): 245-251
      2. Report of the Expert Committee on Tribal Health, Tribal Health in India: Bridging the Gap and Roadmap for the Future-Policy Brief. 2018 (Ministry of Health & Family Welfare and Ministry of Tribal Affairs. Available from:)
      3. Census of India : Provisional Population Totals India : Paper 1 : Census 2011 (censusindia.gov.in).

      4. Statistical Profile of Scheduled Tribes in India 2013. Government of India, 2014 (Ministry of Tribal Affairs. Available from:)
        • Gupta R.
        • Xavier R.
        Hypertension: the most important non communicable disease risk factor in India.
        Indian Heart J. 2018; 70 (Jul-Aug): 565-572https://doi.org/10.1016/j.ihj.2018.02.003
        • Ramakrishnan S.
        • Zachariah G.
        • Gupta K.
        • et al.
        Prevalence of hypertension among Indian adults: results from the great India blood pressure survey.
        Indian Heart J. 2019; 71 (Jul-Aug): 309-313https://doi.org/10.1016/j.ihj.2019.09.012
        • Anchala R.
        • Kannuri N.K.
        • Pant H.
        • et al.
        Hypertension in India: a systematic review and meta-analysis of prevalence, awareness, and control of hypertension.
        J Hypertens. 2014; 32 (Jun) (PMID: 24621804; PMCID: PMC4011565): 1170-1177https://doi.org/10.1097/HJH.0000000000000146
        • Kishore J.
        • Gupta N.
        • Kohli C.
        • Kumar N.
        Prevalence of hypertension and determination of its risk factors in rural Delhi.
        Int J Hypertens. 2016; 7962595https://doi.org/10.1155/2016/7962595
        • Gupta R.
        • Pandey R.M.
        • Misra A.
        • et al.
        High prevalence and low awareness, treatment and control of hypertension in Asian Indian women.
        J Hum Hypertens. 2012; 26 (Oct) (Epub 2011 Sep 1. PMID: 21881598): 585-593https://doi.org/10.1038/jhh.2011.79
        • Gupta R.
        Convergence in urban-rural prevalence of hypertension in India.
        J Hum Hypertens. 2016; 30 (Feb) (Epub 2015 Jun 25. PMID: 26108364): 79-82https://doi.org/10.1038/jhh.2015.48
        • Gupta R.
        • Gaur K.
        • Ram C.V.
        Emerging trends in hypertension epidemiology in India.
        J Hum Hypertens. 2019; 33 (Aug) (Epub 2018 Sep. 25): 575-587https://doi.org/10.1038/s41371-018-0117-3
        • Bhise M.D.
        • Patra S.
        Prevalence and correlates of hypertension in Maharashtra, India: a multilevel analysis.
        PLoS One. 2018; 13 (Feb 5) (PMID: 29401464; PMCID: PMC5798824)e0191948https://doi.org/10.1371/journal.pone.0191948
        • Shikha S.
        • Ravi S.
        • Gyan P.
        Prevalence and associated risk factors of hypertension: a cross-Sectional study in urban Varanasi.
        Int J Hypertens. 2017; (2017)5491838https://doi.org/10.1155/2017/5491838
        • Gupta R.
        • Guptha S.
        • Gupta V.P.
        • Agrawal A.
        • Gaur K.
        • Deedwania P.C.
        Twenty-year trends in cardiovascular risk factors in India and influence of educational status.
        Eur J Prev Cardiol. 2012; 19 (Dec) (Epub 2011 Sep 26. PMID: 21947630): 1258-1271https://doi.org/10.1177/1741826711424567
        • Anand N.
        • Hussain S.
        Prevalence of hypertension & associated risk factors among tribal population in a rural community of Katihar.
        J Evol Med Dent Sci. 2020; 9 (june): 1725-1729
        • Uddin J.
        • Acharya S.
        • Valles J.
        • Baker E.H.
        • Keith V.M.
        Caste differences in hypertension among women in India: diminishing health returns to socioeconomic status for lower caste groups.
        J Racial Ethn Health Disparities. 2020; 7 (Oct) (Epub 2020 Feb 20. PMID: 32078742): 987-995https://doi.org/10.1007/s40615-020-00723-9
        • Sajeev P.
        • Soman B.
        Prevalence of non-communicable disease risk factors among the Kano tribe in Thiruvananthapuram district, Kerala.
        Indian Heart J. 2018; 70: 598-603
        • Ramezankhani Azra
        • et al.
        Associations of marital status with diabetes, hypertension, cardiovascular disease and all-cause mortality: a long-term follow-up study.
        PLoS One. 2019; 14 (22 Apr)e0215593https://doi.org/10.1371/journal.pone.0215593
        • Li K.
        • Ma X.
        • Yuan L.
        • et al.
        Age differences in the association between marital status and hypertension: a population-based study.
        J Hum Hypertens. 2021; https://doi.org/10.1038/s41371-021-00558-9
        • Kohori H.S.
        • Uematsu H.
        • Dorji N.
        • et al.
        Findings from the national survey for non-communicable disease risk factors and mental health using WHO STEPS in Bhutan.Published.
        (August 31,)
        • Tuoyire D.A.
        • Ayetey H.
        Gender differences in the association between marital status and hypertension in Ghana.
        J Biosoc Sci. 2018; 3 (May 51): 1-22https://doi.org/10.1017/S0021932018000147
        • Bhatt D.
        • Sharma S.
        • Gupta R.
        • Sinha D.N.
        • Mehrotra M.
        Predictors of hypertension among nonpregnant females attending health promotion clinic with special emphasis on smokeless tobacco: a cross-sectional study.
        BioMed Res Int. 2017; 8765217https://doi.org/10.1155/2017/8765217
      5. National family health survey (NFHS-4).
        (India, 16: Odisha [Internet]. [cited 2021 Jul 2]. Available from:)
        • Bhinse M.D.
        • Patra S.
        Prevalence and correlates of hypertension in Maharashtra, India: a multilevel analysis. Population Studies.
        PLoS One. 2018; 13 (Feb 5)e0191948https://doi.org/10.1371/journal.pone.0191948
        • Rajkumar E.
        • Romate J.
        Behavioural risk factors, hypertension knowledge, and hypertension in rural India.
        Int J Hypertens. 2020; (Mar 9;2020)8108202https://doi.org/10.1155/2020/8108202
        • Datta B.K.
        • Husain M.J.
        Uncontrolled hypertension among tobacco-users: women of prime childbearing age at risk in India.
        BMC Wom Health. 2021; 21: 146https://doi.org/10.1186/s12905-021-01280-x
        • Bolinder G.M.
        • Ahlborg B.O.
        • Lindell J.H.
        Use of smokeless tobacco: blood pressure elevation and other health hazards found in a large-scale population survey.
        J Intern Med. 1992; 232 (Oct): 327-334
        • Hergens M.P.
        • Alfredsson L.
        • Bolinder G.
        • Lambe M.
        • Pershagen G.
        • Ye W.
        Long-term use of Swedish moist snuff and the risk of myocardial infarction amongst men.
        J Intern Med. 2007; 262 (Sep): 351-359https://doi.org/10.1111/j.1365-2796.2007.01816.x
      6. India state level disease burden initiative CVD collaborators-the changing patterns of cardiovascular diseases and their risk factors in the states of India: the Global Burden of Disease Study 1990–2016.
        Lancet Global Health. 2018; 6 (Dec)e1339-e1351https://doi.org/10.1016/s2214-109x(18)30407-8