COMMENTARY

Title, Date of Publication & Journal

Esben Agerbo, Preben Bo Mortensen, Trine Munk-Olsen.

‘Childlessness, parental mortality and psychiatric illness: a natural experiment based on in vitro fertility treatment and adoption.’

Journal of Epidemiology and Community Health

5 December 2012.

Claim supported by evidence?

The paper does not prove that childlessness causes premature death. It suggests that there is an association between involuntary childlessness and premature mortality, but stresses that the association may not be causal. Based on these data it is hard to determine the underlying cause of the results described.

Summary

• This research is interesting but only in relation to the very specific situation of people who are trying to have children. The findings cannot be generalised to the lifestyle choice of whether or not to have children, either in the wider Danish population or in the UK/rest of the world.

• The study only analyses data from couples who a) want a child and b) are seeking IVF treatment. It does not cover parents who are childless out of their own volition.

• Only relative risks are presented: twice to four-times a very low mortality rate is still a very low mortality rate.

• The data do not show any differences between biological parents and childless parents in terms of their rates of psychiatric mental illness.

Study Conclusions

• The study concludes that mortality is lower in biological parents than in the childless, but acknowledges that this conclusion only applies to parents who a) wish to have a child and b) are seeking IVF treatment.

• This association was found to be stronger in women than in men.

• The study does not conclude that the rate of psychiatric illness is lower in biological parents than for the childless. The data do suggest that the rate is lower among adopting parents than for the childless, but give several reasons as to why that is probably not a fair comparison. (See final point in Strengths/Limitations.)

Strengths/Limitations

• In restricting the study only to couples enrolled in IVF programmes, the authors account for many of the natural differences between childless couples and those who have children which would otherwise have biased the findings. This is what the authors mean by conducting a “natural experiment”. However other, unmeasured confounding factors cannot be ruled out (although many are accounted for).

• However, the disadvantage of such a “natural experiment” is that the people whose records have been analysed conform to a very particular set of circumstances. It is unlikely that these couples are representative of couples in general. Results are likely to be biased by the fact that having a child is probably a much more important goal in life for people actively seeking IVF treatment than for the wider population in general; i.e. these data might suggest that those who fulfil their life goals have better overall health than those who do not. The authors acknowledge this.

• Although over 26,000 couples’ records have been analysed (over the course of 12 years), only a small number of deaths were identified. Also, to put the figure in context, the numbers of couples studied are equivalent to about 5% of couples who got married over that period of time (Source: Statistics Denmark).

• The data suggest that incidence of psychiatric illness is lower in men and women who adopt, as compared with the childless. However, the authors point out that this may not be an entirely fair comparison: Danish regulations stipulate that parents must complete a course of fertility treatment before the adoption process can begin (so they will already have a “head-start” on their childless counterparts); parents who are at risk of developing a psychiatric illness are also less likely to have been approved for adoption in the first place – meaning that those who adopt are less likely to go to develop mental illness.

Glossary

Parity: The number of children born to a couple.

Nulliparous:  Having no children.

‘Before the headlines’ is a service provided to the SMC by volunteer statisticians: members of the Royal Statistical Society (RSS), Statisticians in the Pharmaceutical Industry (PSI) and experienced statisticians in academia and research.  A list of contributors, including affiliations, is available here.