Introduction Trends in social inequalities in smoking have been examined in a number of international publications; however these studies have rarely used multiple measures of health inequalities and did not compare simultaneously results in terms of relative vs absolute inequalities and measures of total impact vs measure of effect.

Methods Data from four successive waves of the Belgian Health Interview Survey (1997, 2001, 2004, 2008) were used to study trends in educational inequalities in daily smoking and calculate 4 measures of inequalities: the slope index of inequality (SII), the relative index of inequality (RII), both accounting for the size of each educational group, the population attributable risk (PAR), and the OR comparing the two extreme education groups.

Results All examined measures pointed towards significant inequalities in smoking. Time trends in social inequalities in smoking differed according to the indicator used. Looking at the relative measures of inequality, we found a significant increase between 1997 and 2008 especially for females for the OR (test of trend =+5% p=0.02 for females and trend=+8% p=0.08 for males) and a weaker slope for the RII (test of trend =+1% p=0.08 for females and trend=+1% p=0.07 for males). Looking at the absolute measures (PAR and SII), we found no significant increase.

Conclusion These results could be explained by the reduced prevalence of tobacco smoking in the country and the increase in the overall educational level of the population.