Preterm birth (<37 weeks of gestation) occurs in up to 10% of pregnancies in developed countries and even more in developing countries.1,2 Preterm birth, especially very preterm birth, may have serious long-term effects, including cerebral palsy, respiratory disease, blindness, and deafness.3 Preterm birth is one of the largest health problems in reproductive health, because it is associated with more than 70% of morbidity and mortality in early life.4 Despite improvements of antenatal care, the incidence of preterm birth remains largely unchanged and may even have risen in the United States.5 The etiology of preterm birth is only partly understood.4
Advanced maternal age is associated with preterm birth,6 but a paternal age effect has not been documented.6–9 Advanced paternal age is associated with autosomal-dominant disorders and gene mutations,10 and several findings indicate that genetic factors play a role for the timing of the delivery.11 The placenta is largely derived from the father.12 If the timing of delivery is guided by placental or fetal genes, and if mutations in these genes are more common in the gametes of older men, paternal age may play a role.
We know of 4 studies evaluating the relation between paternal age and preterm birth, and none of them found an association.6–9 This lack of an apparent association may be the result of the definition of preterm birth being too broad or insufficient variation between paternal and maternal age. Studies on the effect of age itself are methodologically difficult.8
Using nationwide registry data in Denmark, we examined whether advanced paternal age increased the risk of preterm and very preterm birth.
From the Danish Fertility Database,13 we constructed a dataset of couples and their first child born in Denmark between 1980 and 1996 for whom parental age was either less than 30 years or more than 35 years at the birth of the child. We excluded the 30- to 34-year age group to reduce sample size, because we did not consider this group informative for the study. Only the first-born child of a couple was included, but either parent could be included in the cohort more than once with a different partner; thus, mothers were not necessarily primiparous. The Danish Fertility Database comprises information on all women and men of reproductive age in Denmark with annually updated socioeconomic conditions. Furthermore, the register contains information on sex, gestational age at birth, and birth weight of the children. Information on congenital malformations (International Classification of Diseases [ICD], 8th Revision: 740–759 before 1994 and ICD-10: Q00–Q99 since 1994) was obtained by linking the children to the National Hospital Register14 by means of the unique personal identification number (CPR number) assigned to all Danish residents.
To reduce the maternal age effect and limit confounding by subfecundity, we restricted our present analyses to couples in which the mothers were between 20 and 29 years of age. Because we focus here on the effect of advanced paternal age on preterm birth, we excluded fathers who were less than 20 years old (n = 503). An analysis done before excluding older mothers yielded the well-known association6 between maternal age and preterm delivery. We also excluded adopted children (n = 315), children with uncertain adoption status (n = 2263), children of multiple births (n = 761), and children with no information on gestational age at birth (n = 1590). A total of 70,347 singleton births remained available for analysis.
Paternal age was categorized into 6 age groups: 20–24, 25–29, 35–39, 40–44, 45–49, and 50+ years. Births before 37 weeks of gestation are preterm, and we defined very preterm birth as births before 32 weeks of gestation. Potential confounders included parity, education and income for both parents, calendar year, and sex of child. Parity (0, 1–2, 3+) was calculated based on previous live births and stillbirths of the mother as reported by the registry.13 We used information on parental education and income for the year before the birth of the child. Maternal and paternal education was classified as low (<9 years), medium (9–11 years), and high (>11 years). The average monthly income per parent was classified according to quartiles for each year and sex of the parent. If no information on education or income was available, an additional category indicating missing data was used in the analyses. The calendar years were grouped into 1980–1988 and 1989–1996.
We used logistic regression to evaluate the effect of paternal age on preterm and very preterm birth when adjusting for linear and quadratic terms of maternal age (in years) and potential confounders. We performed a test for trend regarding the effect of paternal age on preterm and very preterm birth. We also examined the paternal age effect by using single years of age in the logistic model. We furthermore presented results for combinations of 6 paternal age groups and 2 maternal age groups (20–24 and 25–29 years), with both parents age 20–24 years as the reference. We repeated the analyses excluding children with the following characteristics: diagnosis with congenital malformations within the first year of life, 1 or both parents without Danish citizenship, and parents without information on confounders. We also repeated the analyses separately among primiparous and multiparous women.
Table 1 shows that the incidence of very preterm birth increased with increasing paternal age. There were no important differences in sex ratio among paternal age groups. Older fathers tended to have a higher level of education, whereas their partners’ education levels were similar across paternal age groups.
Odds ratios (ORs) of very preterm birth increased with increasing paternal age (Table 2). This pattern was also apparent for all preterm birth, but to a lesser extent. Odds ratios for single year of paternal age were 1.006 (95% confidence interval [CI] = 1.001–1.012) for preterm birth and 1.017 (1.001–1.033) for very preterm birth. Births between 32 and 36 weeks showed ORs of 1.0 (0.9–1.1) for paternal age 25–29 years, 1.1 (1.0–1.2) for 35–39 years, 1.1 (0.9–1.3) for 40–44 years, 1.1 (0.9–1.4) for 45–49 years, and 1.0 (0.7–1.5) for 50+ years. This pattern demonstrated little evidence for trend (P = 0.13).
The prevalence of congenital malformation diagnosed within the first year of life was 4% for term birth, 9% for preterm birth, and 22% for very preterm birth. When children with congenital malformations were excluded (n = 2698), the associations of paternal age with preterm and very preterm birth decreased slightly (Table 3).
We obtained similar results when we performed the analyses among primiparous women (n = 62,141) (Table 3) and multiparous women (n = 8206) (data not shown).
When children of parents without Danish citizenship (n = 6989) or children of parents with missing information on confounders (n = 5609) were excluded, the association between paternal age and preterm and very preterm birth remained unchanged (data not shown).
We found that the risk of preterm birth increased with increasing paternal age. This association was almost entirely the result of the increased risk of very preterm birth, which may explain why previous studies did not find any association.6–9
The germline mutation rate is much higher in men than in women, and the difference increases with age.10,15 Paternal age effect on preterm birth may be explained by a reduced capacity of the DNA repair system, a consequence of cumulative environmental exposures, or a selective advantage of the gametes. Congenital malformations may be part of the causal chain. Our results suggest, however, that congenital malformations do not explain all the association between paternal age and very preterm birth.
Genomic imprinting is a recently recognized phenomenon that plays a role in fetal and placental growth and development, and the expression of specific genes on the paternal genome are crucial for placental development.12 Mutations of these imprinted genes may cause dysfunction of the placenta (such as maternal hypertension and placental abruption), which could result in preterm birth.16
Age of procreation is determined both by fecundity and planning. Some couples have children at a later age because they could not conceive at a younger age. Because a long time-to-pregnancy is associated with preterm birth,17 we restricted our analyses to women in an age group in which low fecundity is not expected. Paternal age has been linked to male subfecundity,18 but we are not aware of any study evaluating the association between male fecundity and preterm birth. Because we have no data on male fecundity, we are unable to evaluate whether male fecundity confounds our results.
The magnitude of effect was markedly larger in couples in which the man was 50+ years old and the woman was 20 to 24 years old. It is possible that social factors associated with this marked age disparity between partners partially explain this finding.
Our study had several advantages: the large age variation among the fathers and the narrow age range for the mothers, pregnancy outcomes obtained by linkage to validated medical registers, and the possibility of adjustment for parental socioeconomic status. On the other hand, our database did not have information on other risk factors for preterm birth such as infertility, infections, smoking, or a history of preterm birth, which may confound our results. However, there is little evidence that paternal history of preterm birth increases preterm birth in subsequent births.19
Our study is the first to show an association between advanced paternal age and very preterm birth. If corroborated, this finding may lead to new insights into the paternal etiology of preterm birth.
1. Mattison DR, Damus K, Fiore E, et al. Preterm delivery: a public health perspective. Paediatr Perinat Epidemiol
. 2001;15(suppl 2):7–16.
2. Kelly A, Kevany J, de Onis M, et al. A WHO collaborative study of maternal anthropometry and pregnancy outcomes. Int J Gynaecol Obstet
3. Marlow N. Neurocognitive outcome after very preterm birth. Arch Dis Child Fetal Neonatal Ed
4. Slattery MM, Morrison JJ. Preterm delivery. Lancet
5. Branum AM, Schoendorf KC. Changing patterns of low birthweight and preterm birth in the United States, 1981–98. Paediatr Perinat Epidemiol
6. Abel EL, Kruger M, Burd L. Effects of maternal and paternal age on Caucasian and Native American preterm births and birth weights. Am J Perinatol
7. Olshan AF, Ananth CV, Savitz DA. Intrauterine growth retardation as an endpoint in mutation epidemiology: an evaluation based on paternal age. Mutat Res
8. Kinzler WL, Ananth CV, Smulian JC, et al. Parental age difference and adverse perinatal outcomes in the United States. Paediatr Perinat Epidemiol
9. Tough SC, Faber AJ, Svenson LW, et al. Is paternal age associated with an increased risk of low birthweight, preterm delivery, and multiple birth? Can J Public Health
10. Crow JF. The origins, patterns and implications of human spontaneous mutation. Nat Rev Genet
11. Ward K. Genetic factors in preterm birth. BJOG
. 2003;110(suppl 20):117.
12. Miozzo M, Simoni G. The role of imprinted genes in fetal growth. Biol Neonate
13. Knudsen LB. The Danish Fertility Database. Dan Med Bull
14. Andersen TF, Madsen M, Jørgensen J, et al. The Danish National Hospital Register. A valuable source of data for modern health sciences. Dan Med Bull
15. Goriely A, McVean GA, Röjmyr M, et al. Evidence for selective advantage of pathogenic FGFR2 mutations in the male germ line. Science
16. Hladky K, Yankowitz J, Hansen WF. Placental abruption. Obstet Gynecol Surv
17. Basso O, Baird DD. Infertility and preterm delivery, birthweight, and Caesarean section: a study within the Danish National Birth Cohort. Hum Reprod
18. Kidd SA, Eskenazi B, Wyrobek AJ. Effects of male age on semen quality and fertility: a review of the literature. Fertil Steril
19. Basso O, Olsen J, Christensen K. Study of environmental, social, and paternal factors in preterm delivery using sibs and half sibs. A population-based study in Denmark. J Epidemiol Community Health