We studied mean changes in birthweight from the first to the second delivery, and we found that the increase in birthweight was highest if the inter-pregnancy interval was less than six months. This finding was more prominent among women aged 35 years or older at the first delivery than among women less than 30 years. Also among women with first live born infant weighing less than 2500 g, we found the highest increase in birthweight at inter-pregnancy intervals less than six months.

In our study, all first-time singleton mothers in Norway during 1970–2010 were followed at least 10 years for a second delivery to occur. Thus, we limited potential biases due to overrepresentation of women with short inter-pregnancy intervals. To our knowledge, this study is the largest yet to report the relation of inter-pregnancy interval with pregnancy outcome, and we had statistical power for separate analyses of subgroups of women. We studied changes from the first to the second delivery in a follow-up study. Such data analytic approach, reduces biases that may occur if women with an underlying risk of low birthweight or other adverse pregnancy outcomes also have a short inter-pregnancy interval [12, 13, 16].

We made adjustments for factors that have been associated with birthweight and/or the interval between pregnancies, such as maternal age at the first delivery, hypertensive disorders in any of the two pregnancies, and a new father to the second pregnancy. In a subgroup of women, we made additional adjustment for smoking. However, adjustments did not change our results notably.

Confounding may remain. Women’s body mass index may vary by the length of the inter-pregnancy interval [13, 34], and high body mass index has been associated with high offspring birthweight [35, 36]. Unfortunately, information about body mass index was available for the last two years of our inclusion period only, and half of these women lacked information. Thus, the results among these women were unreliable and therefore not presented. Both short inter-pregnancy interval and low birthweight have been associated with low socioeconomic status [37,38,39].It cannot be ruled out that changes in socioeconomic status from the first to the second pregnancy have confounded our results. Unfortunately, we lacked individual information about income and socioeconomic status.

Our results may not be generalizable to all women in the world [39, 40]. In addition to the high standard of living in Norway, antenatal and obstetric health care is public, free of charge and used by almost all pregnant women [41, 42]. First trimester pregnancy termination on the woman’s request has been legally performed since 1979, and pregnancy termination is free of charge. The quality and accessibility of health care may influence maternal morbidity, the interval between deliveries, birthweight, and other pregnancy outcomes. A short inter-pregnancy interval has been reported to increase maternal morbidity [43, 44].

It is well known that mean birthweight increases from the first to the second delivery [45, 46]. Our results suggest that the increase in birthweight from the first to the second delivery is highest if the interval is short. Birthweight is associated with future health. Birthweight, also within normal ranges, has been inversely associated with mortality [9, 10]. The higher increase in birthweight after inter-pregnancy interval < 6 months in our study may partly be explained by higher increase in gestational age at delivery (mean increase two days). Both higher birthweight and higher gestational age may be advantageous for the infant, and these outcomes are strongly interrelated.

In most previous studies, women with stillbirth were excluded. Our study illustrates that if the first infant is stillborn or has low birthweight, birthweight is likely to be higher at the second delivery particularly if the inter-pregnancy interval is short.

In most previous studies the outcome variable was dichotomous, and a short inter-pregnancy interval has been associated adverse outcomes such as low birthweight, perinatal death, preterm birth, and poor school performance of the child [2,3,4,5,6,7,8,9, 47,48,49]. The strength of the associations, however, has varied considerably between studies. Studies that have applied within women analyses, questions whether the increased risk of adverse outcomes at short pregnancy intervals is causal [12, 13, 21, 25].

We estimated increased recurrence risk of preterm birth at inter-pregnancy intervals < 6 months.

Among the women with recurrent preterm birth and inter-pregnancy interval < 6 months, mean increase in gestational age from the first to the second delivery was 10 days, and mean increase in birthweight was 411 g. Such increase is favourable for the infant although delivery to term could not be reached. Previously, the risk of preterm delivery according to inter-pregnancy interval has shown inconsistent results [50]. We are not aware of previous studies of recurrent perinatal death risk according to inter-pregnancy interval.

The increase in birthweight at short inter-pregnancy intervals in our study could possibly be explained by underlying factors that are linked to both short inter-pregnancy interval and high birthweight. High fecundity could be one such factor. Women with high fecundity may get pregnant shortly after a delivery, have low risk of miscarriage, and the birthweight of their offspring may be higher than women with low fecundity [34]. A biological selection to a successful pregnancy after a short inter-pregnancy interval may be particularly pronounced among women at high reproductive age [7]. An alternative hypothesis is that at long inter-pregnancy intervals, the maternal age is increased and thereby the risk is increased of maternal complications that may affect fetal growth and duration of pregnancy.

A pregnancy enhances large changes in the maternal cardiovascular system, and these changes are essential for the provision of oxygen and nutrition to the fetus [51]. The maternal adaption to pregnancy includes increased cardiac output, new vessels, increased oxygen uptake, and higher levels of growth and pro-angiogenic factors [52, 53]. It is conceivable such changes remain after a pregnancy or may easily be reactivated, particularly shortly after a previous pregnancy. A biological memory of being pregnant is supported by the low risk of preeclampsia after a short inter-pregnancy interval [20, 43], and preeclampsia is closely linked to low maternal levels of pro-angiogenic factors [54]. Increasing evidence suggests that foetal cells that are transferred to the mother during pregnancy, have multilineage differentiation capacity [55, 56], and such cells may possibly contribute to a biological memory of being pregnant.