A two way view of gender bias in medicine.

"It isn't true, couldn't be; there must be other variables not taken into account that confound the results. " This was the remark of an incredulous psychiatrist colleague to the statement that there is reasoned evidence of inequalities in health care based on patient gender. His reaction aligns with that of health professionals' and researchers' longing for simplicity in outcomes and also with a contemporary style of data analysis. It is not unusual to hear statements like his when criticisms are voiced about health services for women compared to men. This editorial comment considers how health care contributes to the health inequalities associated with gender, a question addressed by many since the WHO European Report in 1990.16 There are two ways in which health service delivery and research can involve gender bias firstly, by assuming that women's and men's health situations and risks are similar, when in fact they are not, and secondly, by assuming differences where there are actually similarities. Actions and research that follow either approach may influence women's health outcomes in negative ways. An increasing number of scientific papers on gender bias in the past five years has intensified the debate because results often confirm the hypothesised presence of inequity. Yet there remain some conceptual, methodological, and empirical limitations that keep the picture hazy.