Prashant Kumar Singh1, Pankhuri Jain1, Nishikant Singh1, Lucky Singh2, Chandan Kumar3, Amit Yadav4, S V Subramanian5, Shalini Singh1. 1. Division of Preventive Oncology & Population Health, WHO FCTC Knowledge Hub on Smokeless Tobacco, ICMR-National Institute of Cancer Prevention and Research, Noida, Uttar Pradesh, India. 2. ICMR-National Institute of Medical Statistics, New Delhi, India. 3. Department of Policy and Management Studies, TERI School of Advanced Studies, New Delhi, India. 4. The International Union Against Tuberculosis and Lung Disease (The Union) South East Asia Office, New Delhi, India. 5. Harvard Center for Population and Development Studies, Department of Social and Behavioral Sciences, Harvard T.H. Chan School of Public Health, Boston, MA, United States.
Abstract
Introduction: This study hypothesises that the presence of a third person during the interaction between the survey investigator and the woman respondent leads to underreporting of smokeless tobacco (SLT) use by Indian women, including pregnant and breastfeeding women. Methods: Cross-sectional data from the National Family Health Survey conducted in 2015-16 was analysed for SLT use among women aged 15-49. Multivariate logistic regression examined the odds of SLT use reporting by women respondents in the presence of their husbands and other male or female adults. Results: SLT use reporting by women significantly varied by the presence of someone during the interview. The analysis shows that the odds of reporting SLT use among women who were neither pregnant nor lactating was 20.6% lower when they were interviewed in the presence of their husbands than when they were interviewed alone. Similarly, compared to those interviewed alone, the odds of women reporting SLT use was 16.5% lower among pregnant and breastfeeding women interviewed in the presence of any adult female. The odds of women under-reporting SLT use were higher in Central and Western India. Conclusions: This study argues that the current survey estimates misconstrue the authentic prevalence of tobacco use among women in India, including pregnant and lactating women. Due to social desirability or the presence of a third person during the survey interview, those respondents who do not report their tobacco use status are also more likely to forego essential support for successful tobacco cessation. Survey methodology must be strengthened to avert the presence of a third person during the interview to ensure better reporting and population health estimates.
Introduction: This study hypothesises that the presence of a third person during the interaction between the survey investigator and the woman respondent leads to underreporting of smokeless tobacco (SLT) use by Indian women, including pregnant and breastfeeding women. Methods: Cross-sectional data from the National Family Health Survey conducted in 2015-16 was analysed for SLT use among women aged 15-49. Multivariate logistic regression examined the odds of SLT use reporting by women respondents in the presence of their husbands and other male or female adults. Results: SLT use reporting by women significantly varied by the presence of someone during the interview. The analysis shows that the odds of reporting SLT use among women who were neither pregnant nor lactating was 20.6% lower when they were interviewed in the presence of their husbands than when they were interviewed alone. Similarly, compared to those interviewed alone, the odds of women reporting SLT use was 16.5% lower among pregnant and breastfeeding women interviewed in the presence of any adult female. The odds of women under-reporting SLT use were higher in Central and Western India. Conclusions: This study argues that the current survey estimates misconstrue the authentic prevalence of tobacco use among women in India, including pregnant and lactating women. Due to social desirability or the presence of a third person during the survey interview, those respondents who do not report their tobacco use status are also more likely to forego essential support for successful tobacco cessation. Survey methodology must be strengthened to avert the presence of a third person during the interview to ensure better reporting and population health estimates.