The impact of ART on union dissolution: a register-based study in Denmark 1994-2010.

STUDY QUESTION

Are couples initiating ART treatment at higher risk for future union dissolution compared to other couples?

SUMMARY ANSWER

There is no effect of ART treatments in future marital dissolution over a period of 16 years when adjusting for all confounders.

WHAT IS KNOWN ALREADY

Findings regarding marital stability and infertility treatments have been sparse and controversial. While there is data showing higher divorce rates among women who go through infertility treatments, there is also some evidence of this experience bringing couples closer by forcing them to communicate more and to deal with the surrounding stigma. Using a population-based study and couple-level data, we investigated the extent to which ART treatment increases the risk for divorce/marital dissolution during up to 16 years of follow-up.

STUDY DESIGN SIZE, DURATION: Register-based national cohort study including all women registered with ART treatment in Denmark between 1 January 1994 and 30 September 2009 (n = 42 845). Marital/cohabiting status was confirmed by matching these women to partners who they were married to or shared an address with. To account for having a significant relationship at baseline (2 years), marital/cohabiting status was confirmed by accessing this variable before the establishment of the cohort back to 1 January 1992.

PARTICIPANTS/MATERIALS, SETTING, METHODS: A comparison group from the background population including five controls per case and matched to female age at baseline was prospectively sampled. Participants could change status during follow-up if they entered ART. The final sample had 148 972 couples, followed until marital dissolution, death of self/spouse, migration or until 31 December 2010. We used Cox regression models adjusting for female and male age, education, marriage, common child at baseline and live-born child during follow-up.

MAIN RESULTS AND THE ROLE OF CHANCE

At baseline, the majority of couples were married (69%). More non-ART couples opted for marriage (70% versus 64%; P < 0.0001) and already had common children at study entry (43% versus 9%; P < 0.0001). During the 16 years of follow-up the majority of couples had children with their baseline partners (56% non-ART versus 65% ART), and 22% ended up separated or divorced (20% ART versus 22% non-ART). Findings revealed a lower risk of break-up among ART couples (crude HR 0.84, 95% CI 0.82-0.86), even after adjusting for both partners' age, education, partnership status and having a common child at baseline (adj HR 0.83, 95% CI 0.80-0.86). However, when subsequent common children (time-dependent) was added to the model, no difference in the risk of dissolution was found (adj HR 1.00, 95% CI 0.99-1.01). A significant interaction between ART status and common children showed that the risk of break-up was attributed to childlessness regardless of having gone through ART treatment.

LIMITATIONS REASON FOR CAUTION

This study did not control for involuntary childlessness, non-ART fertility care (ovulation induction, IUI) and biological parenthood. Additionally, there are important predictors of divorce that were not considered. We were unable to adjust for religion, existence of previous marital relationships, income, employment, health status of parents and child(ren), and quality of relationship.

WIDER IMPLICATION OF FINDINGS

The finding that going through ART does not increase the risk of break up per se is reassuring for couples who underwent ART and have children or are contemplating to start ART.

STUDY FUNDING/COMPETING INTEREST(S): This work was supported by FCT (Portuguese Foundation for Science and Technology), grant ref. SFRH/BPD/85789/2012. The authors have no conflicts of interest.

TRIAL REGISTRATION NUMBER

N/A.