Childlessness “may increase likelihood of early death” BBC News has reported.
The story is based on a study of more than 21,000 Danish couples seeking IVF treatment. Researchers found that women who did not go on to have a child were four times as likely to die early compared with those who did have a child. Men who remained childless were twice as likely to die early.
Previous studies have found an association between childlessness and premature death. But this association is muddied by underlying factors – for example, obesity is a risk factor for both infertility and early death.
The researchers felt that using couples involved in IVF programmes would provide a more illuminating insight into the effects of parenthood on life-expectancy. They reasoned that people who had IVF treatment that was successful would have broadly the same risk factors for infertility as couples who had IVF that wasn't successful.
The researchers did find a statistically significant “survival benefit” to parenthood, but it should be viewed in context. The number of deaths among the 21,276 couples was extremely small – just 316 deaths in total.
Analyses based on such small numbers make the study’s results less reliable. It’s also worth bearing in mind that, in Denmark, the first three courses of IVF are given free-of-charge. Wealthier couples, who may be able to buy more IVF treatment sessions, are also more likely to live longer.
It may be that becoming a parent confers a survival advantage, possibly because new parents may become more health conscious themselves, but this study does not prove it.
The study was carried out by researchers from Aarhus University, the National Centre for Register-based Research and the Danish Clinical Registers. It was funded by the Stanley Medical Research Institute and the Danish Medical Research Council.
The study was published in the peer-reviewed Journal of Epidemiology and Community Health.
The Independent’s headline claiming that, “Parenthood is the secret to a longer life”, gave more credence to the results than was warranted and its short report of the study did not mention any limitations.
The Daily Mail’s headline was alarmist but, to its credit, the paper’s report pointed out that the study did not include couples who chose to be childless. It also mentioned other factors that might have affected results, such as substance abuse, depression, psychiatric illness and physical illness linked to infertility.
The paper claimed that adopting children reduces the chances of early death, but this result only applied to men, not women, in the study.
The BBC included discussion of the study’s limitations from an independent expert.
The researchers say that, so far, studies have suggested that childlessness increases the risk of premature death and psychiatric illness.
However, they point out that these results might be affected by confounding factors that affect both the risk of infertility and of early death or illness, such as obesity, smoking or alcohol misuse.
They also say that previous research has not separated those who choose not to have children from the involuntarily childless.
This was a cohort study that followed up 21,276 childless couples seeking IVF treatment, using a number of national and social registers.
Researchers say their study is based on a “natural experiment” – the event of becoming a parent – and therefore the results are more likely to be reliable.
“Natural experiments” do not have an experimental study design but are, in fact, observational studies. They do not benefit from randomisation and so can’t be used to infer causation, just association. This point is made very clear by the researchers at the start of the study.
The researchers used data from various registers, linking this information to the personal identification number assigned to all Danes. These registers were:
Their cohort consisted of women from the IVF register who, in the previous year, were living with their partner. Women or partners who were not childless or who had had a psychiatric illness were excluded. The couples were followed from their entry on the IVF register until the date of their death, first psychiatric illness, emigration or until the study ended in 2008.
They analysed the data using standard statistical methods, adjusting their results for factors that might affect the results. These were:
Other factors included marital break-up, psychiatric disorders and number of IVF treatments.
In the period from 1994 to 2005, 21,276 childless couples joined the IVF register. A total of 96 women and 220 men died during the follow-up period and 710 women and 553 men were diagnosed with a psychiatric illness.
It’s worth noting that the researchers present adjusted and unadjusted results, emphasising the unadjusted results in the summary of their published research report. This means that the main results below do not take into account other factors in personal health that influence mortality. They found (unadjusted) that:
The researchers point out that the findings were slightly altered once adjusted for confounding factors.
The researchers say their results suggest that mortality rates are higher in the childless, in particular childless women. They say that since their study is based on a “natural experiment” the results are less likely to be due to reverse causation or confounding factors.
This study only looked at couples seeking IVF treatment. As a result its findings cannot be generalised to other couples who choose not to have children or to couples who may be involuntarily childless but who choose not to have IVF.
This is important because the headlines suggest that parenthood itself leads to a longer life, while this study simply shows that women who remain childless despite IVF have some extra risk.
Couples who have IVF may differ in several ways from these other groups, not least because having IVF usually indicates that parenthood is an important goal in life. While there may be a negative effect on health if IVF treatment proves unsuccessful, people who are happy to remain childless may experience no adverse effects as the result of their decision.
Although researchers tried to adjust for confounding factors that might have affected the results, it is still possible that certain things altered the chance of having children through IVF and of premature death. These might include income and as yet unidentified medical factors.
It is possible – and it would be nice to think – that parenthood confers a survival advantage, but the findings from this study should be viewed with caution.