A pregnancy requires a reasonably good health and may have positive as well as negative health consequences for the woman. Part of these health effects may depend on the immune response to the exchange of fetal cells (microchimerism). The number of biological fathers to a woman's children may thus have a health effect beyond the parity effect. A possible design for studying this is to compare health effects for women with or without multiple partners but with the same parity. We compared total and cause specific mortality in these two groups in order to estimate their comparability and thus the problem of confounding. By using population registries we identified all women who had children with at least two different partners from 1973 to 1996 in Denmark (64,704 exposed women). Among all women who had at least two births in the above mentioned time period, we selected a random sample of 100,000 women to obtain information on women having one father for their children. We linked this cohort to a number of registries, including the Cause of Death Registry. We studied cause-specific mortality up to 1997. Altogether 1373 women died during follow-up. Women who had children with more than one partner had a higher relative mortality rate, which was even higher if she had more than two partners. This finding persisted after excluding unnatural deaths and did not depend on time from exposure. Although some of the findings were adjusted for parity, age and social factors, it is highly unlikely that these large differences are entirely related to microchimerism. The study shows that caution is needed when studying health effects of procreation with multiple partners.