Introduction Discussions on inequalities in health and trends in these often concentrate explicitly or implicitly on relative inequality. This paper explores practical and ethical implications of that tendency.

Methods Trends in relative and absolute inequalities in coronary heart disease in Scotland between 1991 and 2006 were assessed using existing data. A hypothetical alternative scenario, featuring different trends, was devised and its implications compared with those of the actual scenario.

Results The ‘headline’ actual trend was an increase in relative inequality between the most and least deprived population groups. However, there was also a reduction in absolute inequality, and both groups benefited from substantially reduced mortality rates. In the hypothetical scenario relative inequality would have lessened, but at the expense of higher numbers of deaths in both groups than in the actual scenario, and a lesser improvement in absolute inequality. The differences between the two scenarios raised ethical questions.

Conclusion When talking about health inequalities, defining desirable reductions in them, assessing trends and judging success and failure, it is important, on social justice and other grounds, to consider both absolute and relative inequality.