Norwegian women, like their counterparts in other developed countries, are increasingly changing partners between their first two pregnancies, and evidence from a population-based study suggests that this trend may have adverse implications for pregnancy outcomes and infant health.1 The rate of infant deaths within one year and the proportions of infants who were delivered preterm and were low-birth-weight declined between the first two pregnancies if the father was the same, but they increased if different men were involved. Moreover, for the second infant, the odds of each of these outcomes were roughly doubled if the mother had changed partners since her first birth.

The researchers used the national medical birth registry to identify women who had at least two births between 1967 and 1998; the study included 488,141 women, of whom 31,683 had changed partners between their first two pregnancies. Data from the registry and from Norway's vital statistics system permitted the investigators to examine the incidence of infant mortality (deaths by age one), preterm delivery (before 37 weeks' gestation) and low birth weight (less than 2,500 g), as well as maternal characteristics. In logistic regression analyses controlling for mother's age and education, the period in which the birth occurred (1967-1976, 1977-1986 or 1987-1998) and, in the case of second births, the interval between births, the researchers compared the likelihood of these outcomes for women who had changed partners and those who had not.

Over time, the proportion of women whose first two children had different fathers rose steadily from 3% to 10%. Women's level of education also increased, and while a change of partners was most common among those with the least education in each period, the gap grew over the 30 years of the study: In the earliest period, 5% of women with the lowest level of education (10 or fewer years of schooling) and 2% of those with the highest level (more than 14 years) changed partners between pregnancies; in the most recent period, the proportions were 19% and 6%, respectively.

First-born infants of women whose partner did not change had a higher rate of mortality within one year than did second-born infants (8.3 vs. 6.7 deaths per 1,000 births); in contrast, first- and second-born infants with different fathers had nearly identical mortality rates (8.3 and 8.7 per 1,000, respectively). Results of the logistic regression analysis revealed that first-born infants' risk of dying by age one was the same regardless of whether their mother changed partners before conceiving again; for second-born infants, however, the risk was significantly elevated if different men were involved in the pregnancies (relative risk, 1.8).

Similarly, the proportion of deliveries that were preterm declined between the first and second births (from 5% to 4%) among women whose infants had the same father, but it increased (from 6% to 8%) among those who changed partners. Women who changed partners had a slightly elevated risk of delivering their first infant before term (relative risk, 1.2), but the increase in risk was even greater in the second pregnancy (2.0).

This pattern was repeated with regard to low birth weight. Among women whose first two children had the same father, 4% had a low-birth-weight first infant, and 3% a low-birth-weight second infant; however, second infants born to women who changed partners had a higher incidence of low birth weight than first-borns (7% vs. 6%). As was the case for preterm delivery, the risk of this outcome was elevated among women who changed partners, and the difference was more modest with the first birth (relative risk, 1.3) than with the second (2.5).

For second-born infants, the mortality rate decreased as the level of maternal education increased, regardless of whether the mother had changed partners between pregnancies. The incidence of preterm birth and low birth weight declined with increasing maternal education only if both infants had the same father. At each level of education examined, women who changed partners had a greater risk of adverse outcomes than those who had the same partner; the differentials generally were largest among those with the most schooling.

The researchers conclude that "changes in lifestyle or social circumstances may accompany change of partner, and these changes could be important." They stress, however, that whether women who change partners between pregnancies also adopt a "less healthy lifestyle" than others still needs to be investigated.

—D. Hollander

1. Vatten LJ and Skjaerven R, Effects on pregnancy outcome of changing partner between first two births: prospective population study, British Medical Journal, 2003, 327(7424):1138-1141.