Changes in dietary fat and declining coronary heart disease in Poland: population based study

and lipid profiles among men. 4 For Estonia, highly educated adults, compared with those with less education, gained a considerable advantage during the 1990s, in terms of mortality from circulatory diseases and total mortality. 5 According to Lawlor and colleagues, the children of these highly educated parents may paradoxically have the poorest insulin resistance profile.We cannot say how much of a paradox this really is, however, as we know nothing about whether their parents as children also had a poor insulin profile.Genetic, fetal, and early childhood factors should all be relevant in determining insulin resistance.In Lawlor and colleagues' study, parental education was important for insulin resistance among prepubertal and postpubertal children.Parental education can be taken as a measure of social circumstances when their children were born-that is, before the collapse of the Soviet system.Income was measured in 1997-2000; for Estonia this means when new food markets had opened up.In mutual adjustments, education but not income had an independent effect; thus it seems unlikely that it is consumption of "burgers, crisps, and processed food" that is creating the pattern of high insulin resistance among children of highly educated parents.We also noted that, in this study, children of highly educated fathers in Estonia had a 200 g lower birth weight than others, consistent with their higher insulin resistance.Anomalies such as those reported for Estonia and Portugal may be of special significance, as they point towards gaps in our understanding and warn against too simplistic a view of health inequalities.Correctly understanding the development of health and mortality in the formerly communist led countries of central and eastern Europe is likely to challenge (and has already challenged) many cherished epidemiological "truths."We thank Ilona Koupil for important comments to a previous version of this commentary.