Socioeconomic inequalities in smoking in low and mid income countries: positive gradients among women?

Int J Equity Health. 2014 Feb 6:13:14. doi: 10.1186/1475-9276-13-14.

Abstract

Background: In Southern Europe, smoking among older women was more prevalent among the high educated than the lower educated, we call this a positive gradient. This is dominant in the early stages of the smoking epidemic model, later replaced by a negative gradient. The aim of this study is to assess if a positive gradient in smoking can also be observed in low and middle income countries in other regions of the world.

Methods: We used data of the World Health Survey from 49 countries and a total of 233,917 respondents. Multilevel logistic regression was used to model associations between individual level smoking and both individual level and country level determinants. We stratified results by education, occupation, sex and generation (younger vs. older than 45). Countries were grouped based on GDP and region.

Results: In Eastern Europe and the Eastern Mediterranean, we observed a positive gradient in smoking among older women and a negative gradient among younger women. In Sub-Saharan Africa and Latin America no clear gradient was observed: inequalities were relatively small. In South-East Asia and East Asia a strong negative gradient was observed. Among men, no positive gradients were observed, and like women the strongest negative gradients were seen in South-East Asia and East Asia.

Conclusions: A positive socio-economic gradient in smoking was found among older women in two regions, but not among younger women. But contrary to predictions derived from the smoking epidemic model, from a worldwide perspective the positive gradients are the exception rather than the rule.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adolescent
  • Adult
  • Age Distribution
  • Aged
  • Developing Countries / statistics & numerical data*
  • Educational Status
  • Female
  • Health Status Disparities
  • Humans
  • Logistic Models
  • Male
  • Middle Aged
  • Occupations / statistics & numerical data
  • Prevalence
  • Sex Distribution
  • Smoking / epidemiology*
  • Socioeconomic Factors
  • Young Adult