Article Text

Download PDFPDF

Full term; an artificial concept
  1. Neil Marlow
  1. Correspondence to Neil Marlow, UCL EGA Institute for Women's Health, University College London, 74 Huntley Street, London WC1E 6AU, UK; n.marlow{at}

Statistics from

Babies born between 37 and 42 weeks of gestation are considered to have been born at the optimal time, and earlier or later delivery is known to be associated with increased risk of a range of outcomes from neonatal complications to impaired outcomes in terms of physical and intellectual ability and even death. Originally, there was an uncertainty about the length of gestation, because the basis for calculating gestational age (from the first day of the last menstrual period) was acknowledged as inaccurate due to conception at different points during the menstrual cycle and variation in an individual woman's cycle, thus a wide definition of full term was developed. With demand-led timing of delivery increasing around the world, many more babies are therefore delivered electively at 37–38 completed weeks of gestation, despite guidance from a range of official bodies.1

This rather artificial concept is further cemented in place with the use of ‘full term’ comparison groups in many studies as the reference population. Indeed there is currently much research interest in the moderate or late preterm babies (32–36 completed weeks) almost as a reaction to the increasing concentration on those babies born before 32 weeks, who need most resources to ensure survival and have a higher risk of adverse outcomes or intellectual deficits. Moderate or late preterm birth is known to be associated with a range of adverse outcomes.2

Therefore of interest, are several recent publications where gestational age has been treated correctly as a continuous variable. Indeed the inherent inaccuracy in the calculation of gestational age (even when confirmed or recalculated using a first trimester ultrasound scan) would simply act to dilute any associations with gestational age seen in such studies, emphasising the value of any statistically significant findings.

Mortality is closely related to gestation at birth in 40 weeks providing the nadir but rising either side of this. In a population study from Utah, Young et al derived mortalities for each gestational week and demonstrated that, even after the removal of deaths due to birth defects, the risk of death rose progressively and significantly as gestation fell below 39 weeks (figure 1).3

Figure 1

RR for neonatal death (<28d) after exclusion of deaths due to birth defects.3

More recent work well demonstrates that this risk extends out into childhood. Mackay et al combined the Scottish neonatal record with the Special Educational Needs database over the whole gestational range.4 Their pattern of findings is highly similar to those in the mortality study above with a significantly increased proportion of special needs at each gestational week below 40 weeks, albeit small (figure 2). These findings are important because most babies are born at term; the population attributable fraction for special needs due to gestational age at delivery is about the same for each gestational week from 37 to 39 weeks as for the whole population 33–36 weeks and twice as high as for all births at 28–32 weeks. Early term birth accounts for more than twice the attributable fraction from gestational age as all preterm births. A very similar pattern of effect is shown from a study in Denmark where the proportion of children not completing basic school was distributed similarly across gestation.5

Figure 2

Prevalence of special educational need by gestation at delivery.4

Quigley et al used the UK Millennium Study data to investigate performance in early school life using national attainment tests in relation to gestational age at birth using 39–41 weeks as reference and demonstrated increased risk of not working securely at key stage 1 (end of the first school year) at 37–38 weeks (RR 1.09 (95% CI 1.03 to 1.15)) which remained significantly elevated after correction for a range of confounders. This was a consistent finding across each of the component areas that constitute key stage 1.6

Although the increased risk associated with early term birth is relatively small, it is statistically significant and consistent across population studies. As perinatal illness in such babies is rarely severe enough to be responsible for neuromorbidity, it does remain to be seen whether changing elective delivery timing actually affects these short- and long-term risks.

Thus the use of a 5-week band of gestational age to define ‘full term’ as birth with an optimum prognosis seems an artificial construct – perinatal risk is a continuum for adverse outcomes that is minimal at 39–41 weeks of gestation.7 We have been aware for some time that elective delivery prior to 39 weeks of gestation may be associated with a doubling of RR of neonatal illness and admission for neonatal care.8 9 We now have good evidence for increased mortality and long-term morbidity associated with birth before 39 weeks. The sheer number of elective deliveries means that care should be taken not to generate extra morbidity by causing early term birth unless there are excellent maternal or fetal reasons for inducing delivery early. Counselling for elective delivery should be mindful of this and care has to be taken to avoid iatrogenic morbidity – it appears there are some things that the neonatologist cannot sort out!


Neil Marlow receives part funding from the Department of Health's NIHR Biomedical Research Centres funding scheme at UCLH/UCL.


View Abstract


  • Competing interests None.

  • Provenance and peer review Commissioned; internally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.