Intended for healthcare professionals


Subfecundity and neonatal mortality: longitudinal study within the Danish national birth cohort

BMJ 2005; 330 doi: (Published 17 February 2005) Cite this as: BMJ 2005;330:393
  1. Olga Basso (bassoo2{at}, research associate professor1,
  2. J⊘rn Olsen, professor1
  1. 1 Danish Epidemiology Science Centre, Department of Epidemiology and Social Medicine, University of Århus, Århus, Denmark (DK8000)
  1. Correspondence to: O Basso
  • Accepted 1 December 2004


Treatment for infertility correlates with adverse outcomes in pregnancy, especially in singleton deliveries.1 A long time to pregnancy (subfecundity) also correlates with adverse outcomes,24 but few studies evaluating treatment take subfecundity into consideration. We explored the association between time to pregnancy and neonatal death in the Danish national birth cohort.

Participants, methods, and results

We identified 27 624 firstborn singleton babies, born alive from the 24th week of gestation between March 1998 and December 2001, whose mothers were enrolled in the Danish national birth cohort.4 5 Mothers had been interviewed during pregnancy (50% by the 16th completed week and 95% by the 25th) and asked about pregnancy planning and other factors.

Women who reported having planned or partly planned their pregnancy were asked how long it had taken them to conceive. If the answer was six months or longer, they were further asked whether they had received infertility treatment. We excluded 402 women reporting infertility treatment after less than one year of trying and 11 women with missing data on smoking. We analysed 27 329 births (with 66 deaths within 28 days of life (0.24%)). Age at death was recorded in the Danish birth registry.

We grouped women into five categories: up to two months of waiting time (reference); 3-12 months; > 12 months, no infertility treatment; > 12 months, treatment reported; and non-planners (including part planners). We examined the association between these categories and neonatal death through logistic regression, adjusting for mother's age, body mass index, smoking, and social class, the latter derived from the mother's job title.4

The risk of neonatal death increased with increasing time to pregnancy (table). Death between the 29th and the 365th day of life was not related to time to pregnancy (data not shown).

Neonatal deaths (per 1000 births) as a function of time to pregnancy and treatment and crude and adjusted logistic regressions

View this table:


Mothers waiting for longer than one year to conceive their first child gave birth to babies with a higher risk of neonatal death compared with children conceived sooner. We restricted the analysis to primiparae (73.5% of whom reported no previous pregnancies) because death of a previous baby may influence both the decision to conceive again and its outcome.

Infertility treatment was self reported and was only asked of women taking longer than six months to conceive, but there was little difference in risk between treated and untreated, although the causes of death may differ between the two groups.

Only about 35% of eligible women participated in the cohort,5 and this could cause bias if participants with a long time to pregnancy were at a different level of risk compared to the non-participants. Furthermore, we could not distinguish the length of infertility beyond one year, which limits our ability to identify a dose-response, if it exists.

We collected information on time to pregnancy and confounders before delivery, reducing the potential for other types of bias. The mother's job title may be a poor proxy for social class, but the adjustment appeared to have little effect on our estimates.

A long time to pregnancy per se is not commonly considered a marker of increased risk, and untreated women with a history of infertility may seek (or receive) inadequate prenatal care.

Even though neonatal death was a rare event in this population, it is a serious outcome and any potential risk marker should be considered. Our finding needs, however, to be corroborated elsewhere before it can be stated that a long time to pregnancy increases the risk of neonatal death.

If infertility itself is associated with adverse outcomes, an appropriate comparison group should be used when assessing effect of infertility treatment on pregnancy outcomes, lest adverse effects of treatment be overestimated.

What is already known on this topic

Infertility treatment is correlated to adverse pregnancy outcomes, and evidence indicates that subfecundity per se is also associated with adverse pregnancy outcomes

What this study adds

Subfecundity may be associated with an increased risk of neonatal death and should be included as a risk indicator in neonatal care

This article was posted on on 4 February 2005:


  • Contributors OB and JO had the idea for this study and formulated the hypothesis to be tested. OB analysed the data and both authors contributed to the interpretation of the findings. OB drafted the manuscript, and JO provided critical comments. JO is guarantor.

  • Funding OB had a grant from the Danish Medical Research Council (No 22-00-0008). The Danish Epidemiology Science Centre was established by a grant from the Danish National Research Foundation. The Danish National Birth Cohort is funded by the Danish National Research Foundation, the Pharmacy Foundation, the Egmont Foundation, the March of Dimes Birth Defects Foundation, the Augustinus Foundation, and the Health Foundation.

  • Competing interests None declared.

  • Ethical approval National Scientific Ethics Committee.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
View Abstract