Changes in the social gradient in tobacco use in Argentina: Evidence from the 2005 and 2009 National Risk Factor Surveys

Introduction: Argentina's first national risk factor survey (NRFS) in 2005 showed a significant social gradient in tobacco use. Since then, important anti-tobacco policies have been enacted. Using data from the first and second rounds of the NRFS (2005 and 2009; N = 41,392 and 34,732), we evaluate if the gradient in smoking has changed. Methods: Logistic regression was used to analyze the prevalence of tobacco consumption. An interaction term was used in pooled models to test the significance of the change in the gradient. Results: Using income as the indicator of social position, the gradient in current smoking appears to be decreasing for low-income respondents (OR= 1.33, 95%CI = 1.15– 1.55 in 2005 and OR = 1.05, 95%CI = 0.92 – 1.19 in 2009; p = 0.07) and middle-income respondents (OR= 1.27, 95%CI = 1.10– 1.47 in 2005 and OR = 0.93, 95%CI = 0.83 – 1.05 in 2009; p < 0.001), relative to high-income respondents (adjusting for age and sex). However, if education (rather than income) is used, the gradient appears stable for most groups and widening for respondents with medium-low education (OR = 1.26, 95%CI = 1.09 – 1.46 in 2005 and OR = 1.43, 95% CI = 1.28 – 1.60 in 2009; p < 0.05). Predicted probabilities suggest that this is driven by decreases in smoking by respondents with high education levels. Conclusion: Changes in the social gradient in tobacco use are underway in Argentina. Surveillance programs should consider both income-based and education-based models, as these may reflect different patterns.