An efficient method to conduct large population survey in a low resource setting: Tamil Nadu tobacco survey

Abstract

Background: Resource allocation for tobacco surveillance in low-income settings like India is a challenge. The current study describes an efficient method to conduct a large population-based tobacco survey in an Indian state. Methods: Tamil Nadu Tobacco Survey (TNTS) was conducted in Tamil Nadu, the sixth most populous state in India, between March and November, 2015. About 100,000 subjects aged 15 years and above, representing both urban and rural populations within 32 districts, were included. The Global Adult Tobacco Survey (GATS) questionnaire was modified and translated into the local language (Tamil) to develop the survey questionnaire, which was pre-tested in 1,690 participants in three districts in November 2014. The survey was conducted through research collaboration between 31 educational institutions and three NGOs. Once collected, data were double-entered using an open access tool (Epidata). Results: The quality and the accuracy of the data was ensured at every level and the data was double entered to minimise the entry error. Among a total number of 32,945 participating households, 111,363 eligible individuals were identified, of which 99,825 individuals completed the survey. The overall Household Response Rate (HRR) was 91.23% (range within districts: 72-99%). The overall Individual Response Rate was 89.24% (range within districts: 73-99%). The unweighted population almost equally represented the weighted population in the selected demographic variables such as age, gender, and type of residence. Conclusions: TNTS was conducted in an efficient manner utilizing local resources, without compromising on quality. This method can be replicated in any setting with the low or limited resource.