Two contrasting hypotheses have been presented to predict women's health variations. The Multiple burden hypothesis predicts that combining a paid job, being married, and having children is likely to be detrimental to women's health. The multiple attachment hypothesis predicts that multiple roles provide attachment to the community, which is likely to be beneficial to women's health. These hypotheses are examined in Britain and Finland, which have different patterns of women's employment participation. Lone mothers form a critical case, since they have fewer attachments and greater burdens, and therefore are expected to have poorer health. The socioeconomic position of lone mothers differs in Britain and Finland, but in both societies they are likely to have fewer attachments. We assess the extent to which health variations between women with different family and parental role combinations are because of the differences in their socioeconomic status and material circumstances. Comparable surveys from Britain and Finland from 1994 were used. Perceived general health and limiting long-standing illness were analysed for working age women (20-49 years) by family type and employment status. as well as other socioeconomic variables. In both countries, women living in two parent families and having children had better health than women living in other family types or on their own. Lone mothers form a disadvantaged group and showed overall worse health in both countries. Adjusting for employment status. education and household income weakened the association between family type and poor health. The findings are broadly in accordance with the multiple attachment hypothesis. Despite the more generous welfare state and high full-time employment among Finnish women, single lone mothers report poorer health than other women in Finland as well as in Britain. However, in Britain the disadvantaged social position of lone mothers accounts for a greater proportion of their poor health than in Finland.