“Children of older fathers ‘more likely to die early’” is the headline in the Daily Mail today. Children of older fathers “are almost twice as likely to die before adulthood” warns the newspaper, reporting the results of a study in more than 100,000 children that showed that those born to fathers over 45 years old were less likely to live to be 19 than those born to men in their late 20s.
The newspaper story is based on a study of children born to parents of different ages. The study suggests a link between the age of the father and death due to some causes but not others, though the overall number of deaths recorded is small. As with all cohort studies, the question is whether the researchers have taken into account all other factors that could be responsible for any association seen. In this particular study, no adjustment was made for the health of the mother, and this could have had a large effect on child mortality.
Where did the story come from?
Dr Jin Liang Zhu and colleagues from the University of Aarhus in Denmark and the School of Public Health at the University of California in Los Angeles carried out this research. The study was funded by grants from the Danish National Research Foundation and was published in the (peer-reviewed) medical journal: European Journal of Epidemiology .
What kind of scientific study was this?
The study was a retrospective cohort study of couples with their first child, in which researchers were investigating the link between the age of the father and child mortality while adjusting for other factors that may have an effect, such as the age of the mother and socioeconomic factors.
The researchers used the Danish Fertility database (which holds data for all individuals in Denmark aged over 11 years old) to identify four different groups of couples with their first child. The first group consisted of all families recorded in the database where both members of the couple were aged over 35 at the time their child was born; the second group was all couples where fathers were over 35 years but with mothers younger than 30; the third group was of all mothers over 35 with fathers younger than 30; and the fourth group was a random sample from the database of parents who were both under 30 years old when their child was born.
The researchers collected data about death of the children by linking them to the Register of Causes of Death. This was possible because all children in Denmark are assigned a unique registration number when they are born. For the purposes of this study, causes of death were recorded as perinatal (around the time of birth), due to congenital malformation, ill-defined, due to injury or poisoning, or due to other diseases. The researchers assessed the link between paternal age category (15–24 years, 25–29 years, 35–39 years, 40–44 years, 45+ years) and risk of death using the parental age group 25–29 years as the reference point (i.e. comparing rates of death in other groups to this one). They also analysed the data according to age at death. They took into account other factors that may have played a part, including maternal age, how many other children there were, maternal and paternal education, income, country of origin and year. To investigate the cause of death after birth, the researchers excluded children with congenital malformations, and only included those who had a healthy birthweight (2,500g or more), or those that were born at or after 37 weeks.
What were the results of the study?
The researchers followed up 102,879 children for up to 18 years. During this time, 831 children died (601 of them under one year old). When taking into account other factors, children born to older fathers (over 45 years) had a higher rate of death than children born to fathers aged between 25 and 29 years old.
This pattern did not change depending on the age of the child’s death (i.e. before one year old or between one and 18 years old). When the researchers explored this association by cause of death, they found that paternal age was linked to death due to congenital malformations (physical abnormalities in children at birth), and death due to injury or poisoning.
What interpretations did the researchers draw from these results?
The researchers conclude that they have found an association between advanced paternal age and death due to congenital malformation, and due to injury or poisoning. They are cautious to point out that the results may be due to unmeasured factors that they have not adjusted for. They say that the findings should be weighed up against socioeconomic advantages for children born to older fathers.
What does the NHS Knowledge Service make of this study?
- The researchers conclude that the association between older age of fathers and death of offspring “may be related to differences in unadjusted lifestyle factors”. The researchers say that it is unlikely that the differences were related to different levels of healthcare, as the healthcare system is free for all residents. There may have been other factors to explain this pattern, that have not been considered.
- The researchers discuss that reduced fertility may be confounding the association seen, as older couples are generally less fertile than younger ones. No information was collected on fertility, so it is not possible to estimate the contribution of fertility.
- It is worth noting that deaths in children in this cohort were quite rare during the 18 years of follow up. Less than 1% of children died and the majority of those died when they were under the age of one year. This also means that when the researchers analysed deaths by the different paternal age groups, they were working with small sample sizes and outcome numbers (small numbers of deaths). This may have affected the accuracy of their results.
- For the majority of causes of death, the age of the father had no effect. The researchers say that the link between injury and poisoning and death may be due to social and behavioural factors that have not been controlled for, or it could indicate that older fathers may be more accident prone (due to impaired functioning or for behavioural reasons). More research would be needed to explore this.
As with all cohort studies, the problem comes when controlling for other factors that may affect the outcome. The researchers have taken into account some of the factors, but others have not been considered, such as maternal health, which may have a large effect on the risk of death, particularly before the age of one year. The researchers say that their findings add to a growing body of evidence that advanced paternal age has negative effects during reproduction.
Sir Muir Gray adds...
As an older father myself I wondered what would I have done had I known this but I can’t think what I would have done differently.
Analysis by Bazian
Edited by NHS Website
Links to the headlines
The Daily Telegraph, 2 June 2008
Daily Mail, 2 June 2008
Links to the science
Eur J Epidemiol 2008; Apr 28 [Epub ahead of print]