Abstract
An increasing number of children are being born with the use of assisted reproduction techniques such as donor insemination, egg donation and surrogacy. There have been concerns that the use of these third-party reproduction techniques may have a negative effect on the quality of the relationship between the mother and father. Marital stability and quality was examined in a UK sample of donor insemination, egg donation and surrogacy families and families in which children were naturally conceived. Interview and questionnaire assessments of marital stability and quality were collected from mothers and fathers over five time points, when the children in the families were aged 1, 2, 3, 7 and 10. Of those families who participated when children were 10 years old, a minority of couples in each family type had divorced/separated and few differences emerged between the different family types in terms of mothers’ or fathers’ marital quality. Despite concerns, couples in families created by donor insemination, egg donation and surrogacy were found to be functioning well.
An increasing number of children are being born with the use of assisted reproduction techniques such as donor insemination, egg donation and surrogacy. There have been concerns that the use of these third-party reproduction techniques may have a negative effect on the quality of the relationship between the mother and father. Marital stability and quality was examined in a UK sample of donor insemination, egg donation, and surrogacy families and families in which children were naturally conceived. Interview and questionnaire assessments of marital stability and quality were collected from mothers and fathers over five time points, when the children in the families were aged 1, 2, 3, 7 and 10. Of those families who participated when children were 10 years old, a minority of couples in each family type had divorced/separated and few differences emerged between the different family types in terms of mothers’ or fathers’ marital quality. Despite concerns, couples in families created by donor insemination, egg donation and surrogacy were found to be functioning well.
Keywords
Introduction
Couples who are unable to conceive naturally and who wish to experience pregnancy, birth and raising their child from birth may turn to assisted reproduction. In cases where techniques such as IVF and intracytoplasmic sperm injection are inappropriate or unsuccessful, couples may use third-party reproduction techniques, such as donor insemination, egg donation and surrogacy. The involvement of a third party in reproduction has raised a number of concerns. Historically, this has been seen as undesirable due to fears that this would disturb the relationship between the mother and the father (
Department of Health and Social Security, 1984
). Additionally, there have been concerns that the imbalance in parents’ genetic relatedness to the child may have a negative effect on the marital relationship (Edelmann, 1989
). Unlike couples who conceive naturally, couples who have used third-party reproduction have had to accept that they are unable to experience the pregnancy and birth of a child who is their shared genetic offspring, which may have involved feelings of grief and loss (Hammer Burns and Covington, 2006
).The quality of the marital relationship has important implications for family functioning in general. Studies utilizing observational measures have found that higher levels of affection in the marital relationship are associated with higher levels of affection in the parent–child relationship (
Fauchier and Margolin, 2004
). These findings provide evidence for the ‘spillover hypothesis’, in which mood, affect or behaviour are considered to be transferred from one setting to another (Erel and Burman, 1995
). Marital quality has also been found to be related to child adjustment. For example, the frequency of marital conflict has been shown to affect children’s short-term coping skills as well as long-term adjustment (Cummings and Davies, 2002
, Parke and Buriel, 2006
). More specifically, high levels of marital conflict have been found to predict both internalizing and externalizing problems for girls and externalizing problems for boys (El-Sheikh and Whitson, 2006
).An early study of donor insemination families conducted in the USA found the rate of parental separation to be low compared with population norms (
Amuzu et al., 1990
). In a similar vein, early studies found the marriages of parents in donor insemination families to be stable and to be functioning within the normal range (Klock and Maier, 1991
, Klock et al., 1994
). In studies of both donor insemination and egg donation families conducted in the 1980s and 1990s, couples reported that the experience of infertility and undergoing fertility treatment had bought them closer together and had improved the quality of their relationships (Applegarth et al., 1995
, Leeton and Backwell, 1982
). The finding that fertility treatment results in an improvement in marital quality has also emerged in studies of couples who have undergone successful fertility treatment using their own gametes (Repokari et al., 2007
, Schmidt et al., 2005
).Cross-sectional studies can only reveal a snapshot of family functioning at a given time point (often when the long-awaited child has arrived). In order to examine marital stability over time, follow-up studies are required. In the European Study of Assisted Reproduction families, donor insemination, IVF, adoptive and natural conception families were recruited in the UK, the Netherlands, Italy and Spain. Family functioning was assessed when the children were aged between 4 and 8 years (
Golombok et al., 1996
), 12 years (Golombok et al., 2002
) and 18 years (Owen and Golombok, 2009
). By the time the offspring were 18 years old, 12% of couples had separated or divorced, with similar proportions of parents having separated or divorced in the different family types. In terms of the quality of the parents’ relationship, few differences were identified for mothers or fathers according to family type throughout the course of the study.In contrast, evidence of instability was found in a follow-up study of donor insemination families in New Zealand: of a sample of 44 families who were revisited 14 years after undergoing fertility treatment with donor spermatozoa, 46% of couples had either divorced or separated (
Daniels et al., 2009
). Although the rate of separation was not compared with population norms in New Zealand, the authors suggested that this high rate of separation may be related to the degree to which participants had been psychologically prepared for their donor insemination treatment.The present study explored marital stability and quality in a UK sample of families created through third-party donation (donor insemination, egg donation and surrogacy) and families in which parents conceived naturally over five time points, when the children in the families were aged 1, 2, 3, 7 and 10.
Materials and methods
Participants
In the first phase of the study, 50 donor insemination families, 51 egg donation families, 42 surrogacy families and a comparison group of 80 natural conception families with a 1-year-old child participated. The egg donation and donor insemination families were recruited through nine fertility clinics in the UK. All two-parent heterosexual families with a child aged between 9 months and 1 year were asked to take part in the research. The exclusion criteria were severe congenital abnormalities and multiple births. The natural conception families were selected through maternity ward records on the basis of stratification to maximize comparability with the assisted reproduction samples. The selection criteria were that the child resulted from a singleton birth with a minimum of 30 weeks gestation, the child had no congenital abnormalities, the mother was at least 30 years of age, the child was the mother’s first or second child, the mother was married to or cohabiting with the child’s father and the pregnancy had been planned (
Golombok et al., 2004a
). A representative sample of surrogacy families was recruited through the UK Office of National Statistics and the surrogacy agency COTS (Golombok et al., 2004b
).These families were assessed when the children were aged 1, 2, 3, 7 and 10 years. Response rates for each phase of the study are presented in Table 1. By age 10, the study included 34 families with a child conceived by donor insemination, 30 families with a child conceived by egg donation, 33 families with a child born through surrogacy and 55 families with a naturally conceived child, representing 68% of the original sample, with no significant difference in retention rates between family types. Rather than actively withdrawing, the majority of those families lost to follow up had moved home and could not be traced.
Table 1Response rates for all family types at each phase of the study.
| DI | ED | SU | NC | |
|---|---|---|---|---|
| Phase 1 | 50 | 51 | 42 | 80 |
| Phase 2 | 46 (92) | 48 (94) | 37 (88) | 68 (85) |
| Phase 3 | 41 (82) | 41 (80) | 34 (81) | 67 (84) |
| Phase 4 | 36 (72) | 32 (63) | 32 (76) | 54 (68) |
| Phase 5 | 34 (68) | 30 (59) | 33 (79) | 55 (69) |
Values are n or n (% of original sample). Sample sizes do not always decrease over time, as in some cases families were unable to participate during one phase of the study (e.g. a family event, moving house) but were then able to participate at a later phase. DI = donor insemination; ED = egg donation; SU = surrogacy; NC = natural conception.
The demographic variables of those families who participated when the children were aged 10 (responders) were compared with those who did not participate at this phase of the study (non-responders). There was no association between whether families participated at age 10 and the following variables: method of conception (assisted reproduction versus non-assisted reproduction), mothers’ intention regarding whether to tell their child about the nature of their conception reported at age 1 (plan to disclose, uncertain, plan not to disclose), mothers’ age and whether the couple had male or female infertility. However, there was a significant association between socioeconomic status measured at age 1 and whether or not families took part at age 10 (chi-squared(2) = 6.76, P < 0.05). Families were categorized as: (i) professional/managerial; (ii) skilled non-manual; or (iii) skilled manual. Those families classified as professional/managerial were more likely to take part at age 10 (73%) and the families least likely to take part at age 10 were those classified as skilled non-manual (56%) and skilled manual (53%).
Demographic variables were compared between the different family types at each phase of the study. Mothers’ age differed between groups, with Helmert contrasts revealing mothers in assisted reproduction families as being significantly older than mothers in natural conception families. Helmert contrasts also revealed that mothers in egg donation families were significantly older than mothers in donor insemination families. In addition, there was a significant difference in family size, with a greater number of siblings in natural conception families as compared with assisted reproduction families. Socioeconomic status was also found to differ between family types. The majority of parents in natural conception families were categorized as professional/managerial, whereas socioeconomic status was more evenly spread in the donor insemination families.
Procedure
Ethical approval for the earlier phases of the study (when children were aged 1, 2 and 3) was obtained from the City University Ethics Committee and ethical approval for the latter phases (when children were aged 7 and 10) was gained from the Cambridge Psychology Research Ethics Committee, University of Cambridge (reference number 2000.36, approved 10th November 2006). When children were aged 1, 2, 3, 7 and 10, a research psychologist trained in the study techniques visited the families at home. A standardized interview and questionnaire relating to the quality of the marital relationship were administered to mothers and fathers individually.
Measures
Mothers and fathers were administered both interview and questionnaire measures to assess marital quality. Questionnaires were administered to mothers and fathers at each phase of the study, whereas interview ratings of mothers’ and fathers’ marital quality were only collected when children were aged 1, 7 and 10. Interview ratings of marital quality were not obtained when children were aged 2 because a more concise interview schedule was administered to parents that excluded some questions related to marital quality, and when children were aged 3 because interviews were only conducted with mothers.
Questionnaire measure of marital quality
Mothers and fathers completed the Golombok-Rust Inventory of Marital State (
Rust et al., 1990
), a 36-item questionnaire assessment of the overall quality of the relationship between couples who are either married or cohabiting. Scores range from 0 to 84, with higher scores indicating poorer marital quality. A score of 34 or more indicates martial dissatisfaction. Split-half reliability for this measure is 0.91 for men and 0.87 for women, and the questionnaire has been shown to discriminate between couples who are about to separate and those who are not.Interview measures of marital stability and quality
As part of a more general assessment of family functioning, a standardized interview designed to measure the quality of the marital relationship was administered to mothers and fathers separately. Information obtained during the interview was rated according to a standardized coding scheme (
Quinton and Rutter, 1988
). The following ratings were made.Marital stability
A rating was made of the family structure (married/cohabiting, divorced/separated).
Marital quality
Enjoyment of shared activities ranged from 1 (a great deal), 2 (quite a lot) to 3 (some). Confiding ranged from 1 (all important matters discussed adequately), 2 (the majority of important matters discussed adequately) to 3 (some/a minority of important matters adequately discussed). Quality of marriage was rated on a 3-point scale, ranging from 1 (marriage/cohabitation positive source of support and enjoyment), 2 (good marital/cohabitation history) to 3 (overall satisfactory history but some problems, or worse). At each time point, these three variables yielded a single factor and all factor loadings were moderate to substantial (ranging from 0.68 to 0.91). The scores from these three variables were combined to create an index of marital quality.
Results
Marital stability
At phase 1 of the study all couples were married/cohabiting. Of those couples who were still participating in the study when the children were 10 years old, 19 (13%) couples had divorced/separated (15% donor insemination, 13% egg donation, 12% surrogacy, 7% natural conception). At age 10, there was no difference in marital stability between family types (donor insemination versus egg donation versus surrogacy versus natural conception; chi-squared(3) = 1.37, not significant). Those parents who conceived a child using donated spermatozoa, donor eggs or those parents who had used a surrogate were just as likely to remain married/cohabiting after 10 years as those couples who conceived naturally. Likewise, marital status was not associated with the couples’ method of conception (assisted versus natural) (Fisher’s Exact test, not significant). Parents who conceived using assisted reproduction techniques were just as likely to be married after 10 years as those couples who conceived naturally.
Marital quality
Questionnaire measure
Mothers’ and fathers’ marital quality was compared between the different family types at each time point (Table 2). In terms of mothers’ marital quality, the only difference to emerge was when the children in the families were 2 years old (F(3) = 2.67, P < 0.05). Helmert contrasts revealed that mothers in natural conception families had higher levels of marital satisfaction (indicated by a lower mean score on the GRIMS) than mothers in assisted reproduction families.
Table 2Cross-sectional analysis: mothers’ and fathers’ questionnaire ratings of marital quality.
| Child’s age (years) | DI | ED | SU | NC | F | P-value | ||||
|---|---|---|---|---|---|---|---|---|---|---|
| n | Mean±SD | n | Mean±SD | n | Mean±SD | n | Mean±SD | |||
| Mothers’ marital quality | ||||||||||
| 1 | 46 | 24.07 ± 9.71 | 45 | 23.44 ± 10.44 | 37 | 21.62 ± 10.55 | 75 | 23.25 ± 10.88 | 0.39 | NS |
| 2 | 42 | 27.43 ± 9.61 | 36 | 25.19 ± 11.80 | 29 | 23.59 ± 9.89 | 50 | 21.42 ± 10.29 | 2.67 | <0.05 |
| 3 | 36 | 25.50 ± 10.12 | 32 | 26.16 ± 12.11 | 28 | 27.43 ± 10.11 | 56 | 23.82 ± 10.51 | 0.80 | NS |
| 7 | 30 | 25.73 ± 9.62 | 26 | 23.54 ± 11.55 | 23 | 24.78 ± 2.39 | 48 | 22.85 ± 10.74 | 0.49 | NS |
| 10 | 26 | 25.12 ± 11.54 | 25 | 23.56 ± 12.23 | 22 | 21.50 ± 7.81 | 45 | 22.04 ± 11.81 | 0.57 | NS |
| Fathers’ marital quality | ||||||||||
| 1 | 41 | 21.68 ± 9.11 | 40 | 23.67 ± 11.87 | 34 | 22.15 ± 10.37 | 60 | 22.33 ± 9.70 | 0.28 | NS |
| 2 | 34 | 21.47 ± 9.10 | 29 | 24.41 ± 11.17 | 22 | 23.50 ± 11.50 | 40 | 25.05 ± 8.05 | 0.91 | NS |
| 3 | 30 | 23.00 ± 10.01 | 26 | 25.73 ± 8.96 | 22 | 22.91 ± 9.81 | 42 | 23.31 ± 9.57 | 0.51 | NS |
| 7 | 25 | 21.80 ± 10.20 | 22 | 23.23 ± 9.87 | 18 | 21.61 ± 11.65 | 37 | 24.16 ± 9.46 | 0.39 | NS |
| 10 | 20 | 23.10 ± 11.89 | 17 | 25.35 ± 10.37 | 18 | 19.67 ± 9.95 | 35 | 24.86 ± 10.44 | 1.14 | NS |
DI = donor insemination; ED = egg donation; NC = natural conception; NS = not significant; SU = surrogacy.
Fathers’ marital satisfaction was compared between the different family types at each time point. No differences were found.
Interview measure
Mothers’ marital quality was compared between the different family types at thee time points, when children were aged 1, 7 and 10 (Table 3). A significant difference between groups was found for mothers when the children were aged 7 years old (F(3) = 2.62, P = 0.05). Helmert contrasts were not statistically significant. Mothers in the surrogacy families had the lowest levels of marital quality and those in egg donation families had the highest levels.
Table 3Cross-sectional analysis: mothers’ and fathers’ interview ratings of marital quality.
| Child’s age (years) | DI | ED | SU | NC | F | P-value | ||||
|---|---|---|---|---|---|---|---|---|---|---|
| n | Mean±SD | n | Mean±SD | n | Mean±SD | n | Mean±SD | |||
| Mothers’ marital quality | ||||||||||
| 1 | 49 | 5.43 ± 1.38 | 50 | 4.92 ± 1.48 | 39 | 5.18 ± 1.59 | 78 | 5.29 ± 1.64 | 1.01 | NS |
| 7 | 31 | 5.81 ± 1.60 | 27 | 5.11 ± 1.60 | 26 | 6.04 ± 1.61 | 50 | 5.18 ± 1.34 | 2.62 | <0.05 |
| 10 | 26 | 5.69 ± 1.72 | 23 | 5.65 ± 1.75 | 24 | 5.67 ± 1.40 | 48 | 5.33 ± 1.99 | 0.35 | NS |
| Fathers’ marital quality | ||||||||||
| 1 | 38 | 5.61 ± 1.55 | 36 | 5.03 ± 1.23 | 27 | 5.44 ± 1.58 | 49 | 5.22 ± 1.40 | 1.14 | NS |
| 7 | 24 | 5.38 ± 1.53 | 22 | 5.50 ± 1.71 | 22 | 5.82 ± 1.82 | 32 | 5.59 ± 1.37 | 0.32 | NS |
| 10 | 18 | 6.17 ± 1.47 | 19 | 5.47 ± 2.04 | 18 | 5.50 ± 1.43 | 27 | 5.70 ± 2.04 | 0.58 | NS |
DI = donor insemination; ED = egg donation; NC = natural conception; NS = not significant; SU = surrogacy.
Fathers’ marital quality was compared between family types at each phase of the study. Once again, no differences were found between the different family types.
The relationship between the demographic variables that differed between groups (mothers’ age, family size and socioeconomic status) and the outcome variables that differed significantly different between groups were examined. No significant relationships were found.
Discussion
Marital stability and marital quality was examined in a longitudinal study of families created by donor insemination, egg donation, surrogacy and families in which parents conceived naturally. Comparisons were conducted at five time points, when the children in the families were aged 1, 2, 3, 7 and 10. Few differences emerged between the different family types in terms of mothers’ or fathers’ marital quality as assessed by self-report questionnaire and investigator-based interviews. The couples in all of the different family types appeared to be functioning well.
Of the families who were participating when children were 10 years old, a minority of couples in each family type had divorced/separated. This appears to be marginally lower than the average rate of divorce in the UK: statistics from 2005 indicate that approximately 45% of marriages will end in divorce and that almost half these divorces will occur before couples reach their tenth anniversary (
Wilson and Smallwood, 2008
). The findings in this study are contrary to the findings of Daniels et al., 2009
, in which approximately 45% of donor insemination couples had separated when assessed 14 years after their original assessment. This disparity could be explained by the differences in sampling techniques, the time that had lapsed between follow-up studies (10 years in the present compared with 14 years in the New Zealand study) or the cultural and social contexts in which the studies were conducted. In addition to the relatively low rate of separation/divorce, mothers’ and fathers’ scores on the standardized questionnaire of marital state indicated that their levels of marital satisfaction were typically above average.Of interest to future researchers in this field will be the relationship between marital quality and parents’ decisions as to whether to tell the child about their donor conception. In adoption research, fathers who are committed to maintaining a good relationship with their partner have been found to be more likely to be involved in communication about the adoption (
Freeark et al., 2008
). Mothers’ and fathers’ level of involvement in adoption communication were also found to be significantly correlated: either couples engaged collaboratively in communication about adoption with their children or both parents avoided the topic.The findings presented in this paper have a number of limitations. Firstly, those families in which couples were experiencing many difficulties may have been the most likely to have dropped out over the 10-year span of the study. It is likely that a selection effect has been in play from the time couples started their fertility treatment, as those couples who did not have a strong relationship may have been the least likely to last the course of fertility treatment and go on to have a child. However, it should be noted that in the present study, mothers and fathers’ marital quality at age 1 was found to be unrelated to whether or not families participated at age 10. Secondly, this study under-represents families of a lower economic status, as these families were the most likely to drop out over time. Thirdly, in order to examine marital stability and quality over time and make comparisons between the different family types, a longitudinal statistical approach is desirable. Due to small and diminishing sample sizes, this was not considered to be an appropriate statistical approach in the present study. Although the difficulties of recruiting a sample of this nature cannot be underestimated, future studies would benefit from larger sample sizes. The two differences that did emerge in the analysis (in mothers’ marital quality at age 2 as assessed by questionnaire and age 7 as assessed by interview) may be chance effects resulting from multiple comparisons being conducted. Larger samples would allow for more stringent statistical comparisons to be conducted and for small differences between groups to be detected over time, were they to exist.
Despite its limitations, this study is unique in presenting data from donor insemination, egg donation and surrogacy families over five time points. That few differences emerged between groups contributes to the growing body of research in this field which demonstrates that families created by donor insemination, egg donation and surrogacy are functioning well.
References
- Pregnancy outcome, health of children, and family adjustment after donor insemination.Obstet. Gynecol. 1990; 75: 899-905
- Families created through ovum donation: a preliminary investigation of obstetrical outcome and psychosocial adjustment.J. Assist. Reprod. Gen. 1995; 12: 574-580
- Effects of marital conflict on children: recent advances and emerging themes in process-oriented research.J. Child Psychol. Psychiatry. 2002; 43: 31-63
- Parental information sharing with donor insemination conceived offspring: a follow-up study.Hum. Rep. 2009; 24: 1099-1105
Department of Health and Social Security, 1984. Report of the committee of inquiry into human fertilisation and embryology (Warnock Report). Cmnd. 9314. HMSO, London, England.
- Review. Psychological aspects of artificial insemination by donor.J. Psychosom. Obst. Gyn. 1989; 10: 3-13
- Longitudinal relations between marital conflict and child adjustment: vagal regulation as a protective factor.J. Fam. Psychol. 2006; 20: 30-39
- Interrelatedness of marital relations and parent–child relations: a meta-analytic review.Psychol. Bull. 1995; 118: 108-132
- Affection and conflict in marital and parent–child relationships.J. Marital Fam. Ther. 2004; 30: 197-211
- Fathers, mothers and marriages: what shapes adoption conversation in families with young adopted children?.Adopt. Q. 2008; 11: 1-23
- The European study of assisted reproduction families: family functioning and child development.Hum. Rep. 1996; 11: 2324-2331
- The European study of assisted reproduction families: the transition to adolescence.Hum. Rep. 2002; 17: 830-840
- Parenting infants conceived by gamete donation.J. Fam. Psychol. 2004; 18: 443-452
- Families created through surrogacy arrangements: parent-child relationships in the 1st year of life.Dev. Psychol. 2004; 40: 400-411
- Psychology of infertility.in: Covington S.N. Hammer Burns L. Infertility Couselling. A Comprehensive Handbook for Clinicians. Cambridge University press, New York2006
- Psychological factors related to donor insemination.Fertil. Steril. 1991; 56: 489-495
- A prospective study of donor insemination recipients: secrecy, privacy, and disclosure.Fertil. Steril. 1994; 62: 477-484
- A preliminary psychosocial follow-up of parents and their children conceived by artificial insemination by donor (AID).Clin. Reprod. Fertil. 1982; 1: 307-310
- Families created by assisted reproduction: parent child relationships in late adolescence.J. Adolesc. 2009; 32: 835-848
- Socialization in the family: ethnic and ecological perspectives.in: Renninger K. Irving E. Handbook of Child Psychology. John Wiley and Sons, Inc., New Jersey2006: 429-504
- Parenting Breakdown: The Making and Breaking of inter-Generational Links.Avebury Gower Publishing, Aldershot, England1988
- Infertility treatment and marital relationships: a 1-year prospective study among successfully treated ART couples and their controls.Hum. Rep. 2007; 22: 1481-1491
- The GRIMS. A psychometric instrument for the assessment of marital discord.J. Fam. Ther. 1990; 12: 45-57
- Does infertility cause marital benefit? An epidemiological study of 2250 women and men in fertility treatment.Patient Educ. Couns. 2005; 59: 244-251
- The proportion of marriages ending in divorce.Popul. Trends. 2008; 131: 28-36
Article info
Publication history
Published online: September 19, 2012
Accepted:
September 5,
2012
Received in revised form:
July 30,
2012
Received:
June 15,
2012
Declaration: The authors report no financial or commercial conflicts of interest.Footnotes
Lucy Blake is a post-doctoral research associate at the Centre for Family Research, University of Cambridge. Her main areas of interest are family relationships and psychological wellbeing in non-traditional families. Her PhD examined family functioning in donor insemination and egg donation families in the UK, with particular reference to the issue of disclosure. She is currently investigating family functioning in families in which gay fathers have had a child using a surrogate and an egg donor. Her most recent publications have explored family relationships and donor conception from the perspective of the child.
Identification
Copyright
© 2012 Reproductive Healthcare Ltd. Published by Elsevier Inc. All rights reserved.
