Psychological well-being and family functioning following identity-release gamete donation or standard IVF: follow-up of parents with adolescent children

Abstract

This study sought to investigate if heterosexual-couple parents with adolescent children following identity-release oocyte donation (OD), sperm donation (SD) or standard IVF differed with regard to psychological distress, family functioning, and parent-child relationships. The prospective longitudinal Swedish Study on Gamete Donation consists of couples recruited when starting treatment between 2005 and 2008 from seven Swedish university hospitals providing gamete donation. This study concerns the fifth wave of data collection and included a total of 205 mothers and fathers with adolescent children following OD (n = 73), SD (n = 67), or IVF with own gametes (n = 65). OD/SD parents had used identity-release donation and most had disclosed the donor conception to their child. Parents answered validated instruments measuring symptoms of anxiety and depression (HADS), family functioning (GF6+) and parent-child relationship. Results found that parents following OD or SD did not differ significantly from IVF-parents with regard to symptoms of anxiety and depression, family functioning, and perceived closeness and conflicts with their child. Irrespective of treatment group, most parents were within normal range on psychological distress and family functioning and reported positive parent-child relationships. However, SD mothers to a larger extent reported anxiety symptoms above cut-off compared to OD mothers (31% vs. 7.3%, p = 0.018). In conclusion, the present results add to previous research by including families with adolescent children following identity-release oocyte and sperm donation, most of whom were aware of their donor conception. Largely, our results confirm that the use of gamete donation does not interfere negatively with mothers’ and fathers’ psychological well-being and perceived family functioning.

Keywords:

• Donor conception
• family functioning
• identity-release
• psychological distress
• parent-child relations

Introduction

Couples who have trouble conceiving naturally can use medically assisted reproductive methods such as in-vitro fertilization (IVF) or use gametes donated from a third party. The donor has historically been anonymous but in recent decades an increasing number of countries have introduced identity-release donation that enables the donor-conceived person to obtain the identity of the donor (Golombok et al., 2016). For heterosexual couples, oocyte donation (OD) and sperm donation (SD) are the most common alternatives. With OD the mother will lack a genetic link to the child and in SD the father will lack a genetic link.

The potential importance of genetic relatedness between parent and child for family functioning has been debated among medical professionals and researchers and was investigated in a systemic review and meta-analysis by Zanchettin et al. (2022). The review included 45 studies of heterosexual and lesbian donor conception (DC) couples and single mothers, as well as families that had conceived through natural conception (NC), standard IVF (with the couples’ own gametes), surrogacy and adoption. The authors found that families who had conceived through DC were generally as well-functioning as IVF-families and, at times, better than NC-families. They discuss these findings as a possible consequence of the similarity in difficulties and experiences shared between DC- and IVF-parents as they attempt to become pregnant, which may in turn strengthen relational bonds both within the couple and between parents and the child.

Research comparing heterosexual-couple DC families and other family types with regard to family functioning and psychological distress has mainly concerned the period around pre-school and late childhood, i.e. when children are around 4–12 years old. Of these studies only one compared SD, OD, and IVF parents with each other (Murray et al., 2006), while the remaining studies included other family types such as NC parents and families with adopted children (Blake et al., 2014; Freeman & Golombok, 2012; Golombok, Brewaeys, et al., 2002; Golombok, MacCallum, et al., 2002). In the cases where significant differences between groups were found, there was a consistent pattern in favour of SD mothers compared to NC/adopted groups; both for more positive family functioning (Golombok, Brewaeys, et al., 2002), and more warmth and emotional involvement (Golombok, MacCallum, et al., 2002). Murray et al. (2006) found no differences between mothers following OD/SD and mothers following standard IVF. However, SD mothers were significantly better at responding sensitively to the child compared to OD mothers, while also being more emotionally overinvolved.

Identity development during adolescence includes active exploration of personal values and re-assessment of relational bonds, thus the importance ascribed to genetic connections may be emphasized more during adolescence compared to childhood. Consequently, discussions and concerns regarding identity issues may be more prevalent in DC families with adolescent children than in IVF or NC families, particularly in families that are open about the child’s genetic origins. To the best of our knowledge, only two studies investigating heterosexual-couple families with adolescent DC children have been conducted, and they show a similar pattern as families with pre-adolescent children. Golombok et al. (2017) compared SD, OD and NC mothers when the children were 14 years old. Overall, they found no group differences except for less positive parent-child relations for OD mothers compared to SD mothers. Owen and Golombok (2009) compared mothers and fathers with 18-year-old children following SD, IVF, NC or adoption. SD mothers reported lower anxiety compared to NC and adoption mothers and expressed more warmth towards their child compared to all the other groups of mothers. There were no significant differences between the fathers in any of the groups.

While previous studies indicate few differences in well-being between gamete donation and other types of families, the research base regarding fathers’ perspectives as well as families with older children is limited. In addition, most previous results concern parents who conceived with gametes from anonymous donors, and it is possible that parenting following identity-release donation entails specific challenges. Therefore, the aim of the present study was to investigate if heterosexual-couple parents with adolescent children conceived using oocyte donation, sperm donation, and in-vitro-fertilization, differ from each other with regards to psychological distress, family functioning, and parent-child relations.

Material and methods

Setting

The study was conducted in the context of Swedish legislation, making publicly funded donor insemination available to heterosexual couples in 1985, and IVF-treatment with donor oocytes or sperm available to heterosexual couples in 2003. All recipient couples undergo a psychosocial evaluation to ensure beneficial circumstances for the future child and to encourage early disclosure of the donor conception to the child. The Swedish legislation on identity-release donations further stipulates that any offspring following gamete donation is entitled to obtain identifying information about their donor when sufficiently mature (Stoll, 2008).

Participants

The present study is part of the prospective longitudinal Swedish Study on Gamete Donation (SSGD), where consecutive groups of couples starting treatment were recruited between 2005 and 2008 from all seven University hospitals providing gamete donation in Sweden. Inclusion criteria for the SSGD were being able to read Swedish and undergoing one round of treatment. Of eligible heterosexual couples approached at project start, a total of 866 individuals starting treatment with oocyte donation, sperm donation or standard IVF (with own gametes) were included in SSGD, with response rates between 71% and 81%. Four waves of data collection have been conducted previously; at the start of treatment (T1), two months after treatment (T2), when the child was 1–4 years old (T3), and when the child was 7–8 years old (T4).

The present study concerned participants at the fifth wave of data collection of the SSGD, conducted 2022–2023, i.e. 14–18 years after treatment start. Inclusion criteria for the present study were being part of a heterosexual couple at treatment start and having an adolescent child (age 13–17) conceived following gamete donation or standard IVF. SSGD participants who had actively dropped out or who were non-responders at both of the latest two waves of data collection (T3 and T4) were not approached at T5. In the SSGD, the first child born to a participating couple following conception (via one or more treatment cycles) in 2005–2010 is considered the target child, and parents were instructed to complete specific items (e.g. parent-child relationship) in relation to this child (indicated by the child’s year and month of birth, which was manually written on each questionnaire).

Individual questionnaires were distributed via mail with a cover letter informing about the purpose of the study, and a pre-stamped return envelope. Non-responders were sent two reminders. No compensation was provided for participation.

Of the 309 heterosexual-couple parents approached at T5, a total of 205 responded and were included in the final study sample: 73 parents following oocyte donation (response rate 70%), 67 parents following sperm donation (response rate 60%), and 65 parents following IVF with own gametes (response rate 69%). The 205 participants represented a total of 131 couples, of which 74 couples were represented by both parents and 57 couples by one of the parents.

Data collection

Participant characteristics

Participant characteristics included family type (OD, SD, IVF), age, gender, relationship with co-parent of the adolescent child (married/cohabiting, divorced/separated), and main occupation (employed, unemployed, other). Recipients of donor gametes were also asked to report donor type (identity-release, known/directed) and if they had started talking with their child about his/her donor conception (dichotomized into yes/no).

Psychological distress

The Hospital Anxiety and Depression Scale (HADS) was used to assess symptoms of anxiety and depression (Zigmond & Snaith, 1983). The scale consists of two subscales measuring symptoms of anxiety (HADS-A) and depression (HADS-D), with a score ranging from 0 to 21 for each subscale. A higher score indicates more severity and a subscale score ≥8 indicates possible or probable caseness and has been suggested as an optimal balance between sensitivity and specificity of anxiety disorder and depression, respectively (Bjelland et al., 2002). Missing values were replaced with the participant’s mean score for each subscale respectively. The instrument has demonstrated good internal consistency and concurrent validity (Bjelland et al., 2002). In the present study, internal consistency was good for the HADS-A (α = 0.81) and the HADS-D (α = 0.79).

Family functioning

A short version of the General Functioning Subscale based on the McMaster Family Assessment Device (FAD) (Epstein et al., 1983) was used to assess family functioning. The short version (GF6+) includes the six positive items from the GF12 subscale, e.g. ‘In times of crisis, we can turn to each other for support’ and ‘We can express feelings to each other’. Responses are indicated on a four-point Likert-scale and form a mean total score ranging from 1 (best functioning) to 4 (worse functioning), with cut-off scores greater than 2 indicating unhealthy family functioning (Boterhoven de Haan et al., 2015). Missing values were replaced with the participant’s mean score of the completed items. The GF6+ is a reliable and valid instrument to assess overall family functioning and has demonstrated equivalent psychometric properties to the GF12 (Boterhoven de Haan et al., 2015). The Swedish version of the GF12 has demonstrated satisfactory reliability and acceptable validity in a bariatric sample (Bylund et al., 2016). In the present study, internal consistency was excellent (α = 0.91).

Parent-child relationship

Two items developed by Goisis and Palma (2021) were used to assess aspects of parent-child relationship covering closeness and conflict frequency. Participants were asked ‘Overall, how close would you say you are to your child?’, and ‘Most parents have occasional quarrels with their children. How often do you quarrel with your child?’. Answers were indicated on a four-point Likert-scale ranging from ‘not at all close’ (1) to ‘very close’ (4), and ‘most days’ (1) to ‘almost never’ (4), respectively.

Ethical considerations

The study was approved by the Swedish Ethical Review Authority (Dnr 2022-03739-01, date of approval 10 October 2022). Participation in the study was voluntary and all participants provided written informed consent to participate. All methods were carried out in accordance with guidelines, regulations, and adhered to the principles outlined in the Declaration of Helsinki.

Statistical analysis

Non-parametric tests were used for all analyses due to the data not being normally distributed. Participants were divided by gender, comparing mothers and fathers separately, as to avoid dependence between observations. Kruskal-Wallis test was conducted to compare the outcome measures (HADS, GF6+, Closeness, and Conflict) based on family types (OD/SD/IVF). The potential impact of having used a known/directed donor (N = 9) was investigated by omitting participants with a known donor and tested if this altered results. To identify unhealthy family functioning and levels indicating possible or probable cases of anxiety and depression, family types were compared based on cut-off values on HADS and GAF6+ using chi² tests. Any statistically significant differences were further explored using post hoc tests (Bonferroni).

Furthermore, attrition bias was investigated by comparing responders and non-responders at T5 with regards to variables assessed at the previous wave of data collection (at T4, when the target child was 7–8 years old): relationship with co-parent of target child, psychological distress (HADS), and disclosure of donor conception to target child. Attrition analyses were performed separately for mothers and fathers in the respective groups (OD/SD/IVF). All analyses were performed using IBM SPSS version 28. A p-value < 0.05 was considered statistically significant.

Results

Participant characteristics

There was an equal gender distribution among the parents following OD (n = 73), SD (n = 67) and IVF (n = 65), with women constituting about 60% of the participants. Participants were around 50 years old (women M = 49.7, min = 40, max = 57; men M = 52.3, min = 40, max = 67) with no significant differences between the groups, while parents in the IVF-group had older children (M = 15.8) compared to those in the OD and SD-group (M = 15.2; M = 15.4) (p = 0.003). As presented in Table 1, most parents were still living with the co-parent of the target child, and almost all participants were employed. Most gamete donation parents had conceived using an identity-release donor, except nine parents (4 couples and 1 woman) who conceived with oocytes from a known/directed donor. A large majority of OD parents (85%) and SD parents (94%) reported having disclosed the donor conception to the child.

Table 1. Characteristics of participating parents with adolescent children following oocyte donation, sperm donation and standard IVF.

	Oocyte donation N = 73		Sperm donation N = 67		IVF N = 65		Group difference
	N	%	N	%	N	%	χ2	p
Parent gender							0.627	0.731
Male	32	43.8	25	37.3	27	41.5
Female	41	56.2	42	62.7	38	58.5
Relationship status^a							0.096	0.953
Married/co-habiting	58	79.5	54	80.6	53	81.5
Divorced/separated	15	20.5	13	19.4	12	18.5	0.015	0.993
Occupation
Working	71	97.3	65	97	63	96.9
Unemployed	0	0	0	0	0	0
Other (sick leave, retired)	2	2.7	2	3	2	3.1
Donor type					N/A	N/A	8.828	0.003
Identity-release	64	87.7	67	100
Known/directed	9	12.3	0	0
Disclosure of donor conception to child					N/A	N/A	3.023	0.082
Disclosed	62	84.9	63	94
Not disclosed	11	15.1	4	6

^a Relationship status with the co-parent of the donor conceived child.

* Percentages may not add up to 100% as decimals have been rounded.

Comparisons of responders and non-responders at T5 regarding their responses at T4 revealed significant group differences among SD and OD mothers. In comparison to those who remained in the SSGD, SD and OD mothers who did not respond at T5 were more likely to not have disclosed to the target child at age 7–8 (SD: 27% vs. 56%, [x²(1) = 4.148, p = 0.042]; OD: 26% vs. 62%, [x²(1) = 3.937, p = 0.047]), and non-responding SD-mothers were also more likely to be divorced/separated from the child’s father (8% vs 31%) [x²(1) = 4.668, p = 0.031]. No corresponding group differences were found for IVF-mothers or for any group of fathers. Furthermore, no significant differences were found between responders and non-responders at T5 regarding levels of psychological distress (HADS-A and HADS-D) at T4 (data not shown).

Psychological distress, family functioning and parent-child relationship in OD, SD and IVF families

The majority of mothers and fathers reported below cut-off scores on symptoms of anxiety and depression (HADS), as well as levels indicating healthy family functioning (G6+) (Table 2). No significant differences between family types (OD/SD/IVF) were found regarding depression or poor family functioning, for neither mothers nor fathers. However, a significant difference between family types was found for mothers scoring above/below cut-off levels of anxiety (p = 0.018). Post-hoc comparisons showed that SD mothers reported above cut-off levels of anxiety to a higher extent than OD mothers (31% vs. 7.3%, p = 0.018), and this difference remained statistically significant after excluding participants with a known donor (5 OD mothers) (p = 0.033). Inspection of the individual HADS-A scores of those SD mothers scoring above cut-off showed that most of them (10/13 mothers) scored close to the cut-off (8 or 9 on HADS-A).

Table 2. Mothers' and fathers' responses on HADS and GF6+. Above cut-off levels indicate possible or probable caseness of anxiety, depression, and unhealthy family functioning.

	Oocyte donation		Sperm donation		IVF
	Below cut-off	Above cut-off	Below cut-off	Above cut-off	Below cut-off	Above cut-off
	n (%)	n (%)	n (%)	n (%)	n (%)	n (%)	χ2^c	P^d
Mothers
Anxiety (HADS-A)^a	38 (92.7)	3 (7.3)	29 (69.0)	13 (31.0)	32 (84.2)	6 (15.8)	8.004	0.018
Depression (HADS-D)^a	37 (90.2)	4 (9.8)	36 (85.7)	6 (14.3)	36 (94.7)	2 (5.3)	1.820	0.403
Family functioning (GF6+)^b^,^e	38 (92.7)	3 (7.3)	38 (90.5)	2 (4.8)	36 (94.7)	2 (5.3)	0.235	0.889
Fathers
Anxiety (HADS-A)^a	29 (90.6)	3 (9.4)	22 (87.5)	3 (12.5)	21 (77.8)	6 (22.2)	2.126	0.345
Depression (HADS-D)^a	30 (93.8)	2 (6.3)	23 (92.0)	2 (8.0)	26 (96.3)	1 (3.7)	0.436	0.804
Family functioning (GF6+)^b	29 (90.6)	3 (9.4)	24 (96.0)	1 (4.0)	25 (92.6)	2 (7.4)	0.616	0.735

^a Scale from 0 to 21, with cut-off scores ≥ 8 indicating possible or probable caseness of anxiety or depression.

^b Scale from 1 to 4, with cut-off at scores >2 indicating unhealthy family functioning.

^c Pearson Chi-Square test (χ²) for treatment type differences (oocyte donation, sperm donation, and IVF with own gametes).

^d P: p-value (level of significance).

^e Missing n = 2.

Table 3 presents participants’ levels of symptoms of anxiety and depression, family functioning and parent-child relationship and shows no significant differences between family types (OD/SD/IVF), separately for mothers and fathers. Overall, participants’ median scores indicate low levels of psychological distress and good family functioning, as well as high levels of closeness and low levels of conflict with their adolescent child.

Table 3. Comparison of parents’ psychological distress, family functioning, closeness and conflict with adolescent child between family types, by parent gender.

	Oocyte donation			Sperm donation			IVF
	n	Median	Range	n	Median	Range	n	Median	Range	H^d	P^e
Mothers
Anxiety (HADS-A)^a	41	3	15	42	6	14	38	5	13	5.443	0.066
Depression (HADS-D)^a	41	2	15	42	2	13	38	2	11	0.169	0.919
Family functioning (GF6+)^b^,^f	41	1.17	1.33	40	1.17	1.50	38	1.25	2.67	1.623	0.444
Closeness^c	41	4	3	42	4	1	38	4	3	0.104	0.949
Conflict^c	41	3	2	42	3	3	38	3	3	1.049	0.592
Fathers
Anxiety (HADS-A)^a	32	3.5	11	25	3	11	27	5	11	2.485	0.289
Depression (HADS-D)^a	32	2.5	8	25	2	16	27	2	8	0.870	0.647
Family functioning (GF6+)^b	32	1.08	2	25	1	2	27	1.33	1.5	1.919	0.383
Closeness^c	32	4	2	25	4	2	27	4	1	0.542	0.763
Conflict^c	32	3	2	25	3	3	27	3	2	0.585	0.746

^a Scale from 0 to 21, with scores ≥ 8 indicating possible or probable caseness of anxiety or depression.

^b Scale from 1 to 4, with scores >2 indicating unhealthy family functioning.

^c Single item ranging from 1 to 4, where a higher score indicates more closeness and less conflict, respectively.

^d H-value (Kruskal-Wallis test) for family type differences (OD, SD, IVF).

^e p-value (level of significance).

^f Missing n = 2.

Discussion

The present study investigated differences between heterosexual-couple parents with adolescent children following oocyte donation, sperm donation or standard IVF in terms of psychological distress, family functioning, and parent-child relationship. A majority of parents, regardless of group, reported within normal range levels of anxiety, depression, and family functioning, as well as experiencing closeness and infrequent conflicts with their child. No significant differences were found between parents who had used gamete donation and parents who had used IVF with their own gametes.

Parents who conceived using OD and SD reported low levels of psychological distress in terms of anxiety and depression symptoms, which is in line with previous research on SD parents with pre-adolescent children (Owen & Golombok, 2009). OD and SD parents’ levels of anxiety and depression symptoms did not differ significantly from those reported by parents who had adolescent children following IVF with their own gametes. The IVF parents’ levels of psychological distress were in line with those of a general population sample of men and women of similar age (Hinz & Brähler, 2011) indicating that the lack of significant differences between groups was not due to the IVF group radically differing from the general population. Our results showed high levels of family functioning and positive parent-child relations with no significant group differences between OD, SD and IVF parents with adolescent children. These findings are in line with a meta-analysis showing no difference in positive and negative parenting between donation families and IVF-families with young children (≤12 years of age) (Zanchettin et al., 2022). Furthermore, our findings confirm previous results of similar parent-child relationships when comparing OD and SD mothers with NC mothers of adolescent children (Ilioi & Golombok, 2015) and 20-year-old children (Golombok et al., 2023).

While comparison of OD, SD and IVF mothers’ levels of symptoms of anxiety did not reach statistical significance, SD mothers to a statistically significantly larger extent reported anxiety levels above cut-off compared to OD mothers. Although this finding should be regarded with caution, we suggest that it may be related to the specific challenges of parenting following identity-release donation. In our sample, most OD and SD parents had talked with their adolescent child about his/her donor conception and were aware that their child within a few years could request the donor’s identity from the fertility clinic. Recent findings from our group indicate that children’s interest in their identity-release sperm donor can be challenging for the family, particularly for the father and for the father-child relationships (Widbom et al., 2021, 2024). Thus, although SD fathers did not differ from OD and IVF fathers regarding psychological distress, it is possible that SD mothers experienced a kind of ‘bystander stress’ related to perceived or anticipated challenges for the family due to their child’s conception with identity-release sperm donation. In contrast, OD mothers live and get to process the experience of non-genetic parenthood and many feel that the gestational bond reinforces the feeling that the child is their own (Imrie et al., 2020). More research exploring the experiences of families with adolescent and adult children are needed to increase our understanding of long-term psychological consequences of identity-release gamete donation.

Methodological considerations

The SSGD includes a large population-based sample of parents following gamete donation and standard IVF with high initial response rates. However, as in all longitudinal studies, there is a risk of attrition bias. In the present long-term follow-up (T5) we reached relatively high response rates for parent groups but also found that dropout was significantly more likely among OD and SD mothers who had not started the disclosure process at the previous follow-up (T4, at child age 7–8). However, in a recent study focusing on predicting parents’ disclosure behaviour using T4 and T5 data, we found equally high future disclosure probability among responders and non-responders at T5 (Paulin et al., 2024), which suggests that attrition may be related to other factors than disclosure-related issues. Also, the present study concerns heterosexual-couple parents and was conducted in the context of the Swedish legislation that gives donor-conceived offspring the right to obtain the donor’s identity, which limits the generalizability to other groups and contexts. Strengths of the present study include the use of validated instruments for assessment of psychological distress and family functioning. Parent-child relationship was assessed with two items assessing closeness and conflict, which have been described as important predictors of child development (Driscoll & Pianta, 2011), but it should be noted that parent-child relationship was assessed from the perspective of the parent only and that ratings by the child may very well differ. Finally, additional sociodemographic data such as ethnicity and history of mental ill-health among the participants was not collected, which constitutes a limitation.

Clinical implications

Based on the present findings and previous results on good psychological wellbeing in families following donor conception (Blake et al., 2014; Brewaeys, 2001; Golombok et al., 2023), we support that counsellors should discuss the implications of using donor gametes and the distinction between nature and nurture with patients contemplating donor conception (ESHRE Working Group on Reproductive Donation, Kirkman-Brown et al., 2022). While patients should be informed that research overall shows that a genetic link is not necessary for positive family functioning and parent-child relationships, it is important to address that family building with donor conception may include specific challenges, e.g. concerns about non-genetic parenthood and the role of the donor (Widbom et al., 2021, 2024). As donor conception has life-long consequences, adequate informational resources to support families, as well as counselling should be available before, during and after donor conception (ESHRE Working Group on Reproductive Donation, Kirkman-Brown et al., 2022).

Conclusions

The present results add to the growing body of research on the long-term implications of gamete donation by focusing on heterosexual-couple families with adolescent children conceived within an identity-release donation program. Our findings support that genetic parent-child links are not necessary for a healthy family functioning, and indicate that children´s right to obtain the donor´s identity does not have a negative impact on parents’ psychological well-being and family relationships.

Authors’ roles

A.W., C.L., A.S.S., and G.S. conceived of the design of the study. J.P. and C.L. performed the data collection. A.W. and J.P. performed the statistical analyses. A.W. wrote the initial draft of the manuscript. All authors contributed to the interpretation of the results, revised the manuscript and approved the final manuscript.

Acknowledgements

We want to thank all the parents who participated in the project.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data that support the findings of this study are available on request from the corresponding author, JP. The data are not publicly available due to their containing information that could compromise the privacy of research participants.

Additional information

Funding

This work was supported by the Swedish Research Council under Grants 2013-2712 and 2021-03174. The funder had no involvement in any stage of the study process.