The influence of assisted reproduction on family functioning and children’s socio-emotional development: results from a European study

J. Barnes1,9, A.G. Sutcliffe2, I. Kristoffersen3, A. Loft4, U. Wennerholm5, B.C. Tarlatzis6, X. Kantaris2, J. Nekkebroeck3, B.S. Hagberg5, S.V. Madsen7 and M. Bonduelle8

1 Institute for the Study of Children, Families and Social Issues, Birkbeck College, University of London, 7 Bedford Square, London WC1B 3RA, 2 Royal Free and University College Medical School, University College, London NW3 2PF, UK, 3 Vrije Universiteit and 8 Centrum Medische Genetica, Brussels, Belgium, 4 The Fertility Clinic, Rigshospitalet section 4071 and 7 Department of Psychology, Play Therapy and Social Work, Rigshospitalet section 4074, Copenhagen University Hospital, Copenhagen, Denmark, 5 Sahlgrenska University Hospital, East, Goteborg, Sweden and 6 Unit for Human Reproduction, First Department of Obstetrics and Gynaecology, Aristotle University of Thessaloniki, Thessaloniki, Greece

9 To whom correspondence should be addressed. e-mail: jacqueline.barnes{at}bbk.ac.uk


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
BACKGROUND: ICSI is used with increasing frequency, but there is less information about the children born following this method of assisted reproduction than other forms of IVF. Some authors have suggested that it may contribute to more family stress than IVF. METHODS: ICSI conceived children were compared with IVF conceived children and naturally conceived (NC) controls. They were selected in five European countries: Belgium, Denmark, Greece, Sweden and the UK, and seen for psychological testing and a paediatric examination when they were 5 years old. In all countries, except Greece, mothers and fathers were asked to complete questionnaires about parental well-being, family relationships, parenting and child behaviour. RESULTS: Very few differences were found between the ICSI and NC group or the ICSI and IVF group. The only significant differences were that mothers in the ICSI conceived group reported fewer hostile or aggressive feelings towards the child and higher levels of commitment to parenting than the mothers of NC children. CONCLUSIONS: The study confirms the results of previous work with IVF families. This should be encouraging for families using these techniques in the future.

Key words: ICSI/IVF/child behaviour/parent–child relationship/parenting


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Compared with natural contraception, any type of assisted reproductive therapy (ART) may be associated with more parental stress, which could influence adaptation to the parental role, parental relationships, parent–child relationships and children’s socio-emotional development.

Studies performed so far into outcomes of ART have focused primarily on physical problems such as birth defects and neurodevelopmental outcomes (Bonduelle, 2003Go). Some studies have been performed looking specifically at children’s socio-emotional development, but have tended to be small samples (Mushin et al., 1995Go; Gibson et al., 2000aGo) and have not studied the newer types of IVF such as ICSI which, it could be argued, raise additional risks (Sutcliffe, 2000Go, 2002) because the more invasive nature of the procedure may lead to extra stresses on the couples undergoing treatment. Questions have been raised about the physical and socio-emotional development of children born using this technique (Sutcliffe, 2000Go) linked with possible strains on parents and their relationship with their children. Using this technique, particularly useful in the treatment of male infertility, a single sperm is selected and treated, then injected into an oocyte (Palermo et al., 1992Go). There is less information about the children born following this method of assisted reproduction than other forms of IVF.

Potential sources of strain following a period of childlessness and then undergoing ART of any type include the failure to work through feelings towards infertility (Mushin et al., 1995Go). The child may be a constant reminder to one parent of their infertility, possibly acting as a ‘narcissistic injury’ or creating asymmetry in the marital relationship (Colpin et al., 1995Go). In most countries, ICSI has been used primarily for male factor infertility (Sutcliffe, 2002Go), which could have specific implications for fathers.

A review of issues related to the development of IVF children (Van Balen, 1998Go) suggested that the main concerns are: the child might be seen as very ‘precious’ and then be subject to overprotection; parents may have exaggerated expectations of their child; after a long period of infertility, parents may have difficulty adapting to the reality of child rearing; and the child conceived using IVF may be perceived as different by other people in the family’s social network. Nevertheless, on the basis of the studies included in the review, Van Balen (1998Go) concludes that no negative differences can be found in parent–child relationships or children’s psychological development. He remarks on the small size of most of the studies available, although this does not preclude the detection of clinically relevant differences.

Similarly, research into parent–child relationships in families who used IVF has not, by and large, confirmed the predictions of difficulties, although to date most studies have not included large samples. Mothers who became pregnant using IVF are generally reporting more pleasure in their children, more warmth towards them and more parental competence than naturally fertile mothers (Golombok et al., 1995Go; Van Balen et al, 1996Go). An international study including children from Italy, The Netherlands, Spain and the UK (Golombok et al., 1996Go) compared children conceived using either donor insemination (DI) or IVF with naturally conceived (NC) controls and adopted children, aged between 4 and 8 years. During in-depth interviews, mothers in both assisted reproduction groups and the adoptive mothers expressed more warmth about their child than mothers of NC children. The fathers in the IVF and DI groups were more involved and mothers reported less stress overall than mothers in the NC control group.

It has been suggested, however, that these results may be an indication of the tendency to report socially desirable responses (Colpin et al., 1995Go). Some detailed observational work in a small-scale study from Belgium found that, at 2–3 years, comparing only those mothers currently employed and who had used IVF were less able to allow their young children autonomy in a problem-solving task than mothers who had conceived naturally (Colpin et al., 1995Go). However, a study conducted in Australia comparing 65 children conceived using IVF with 61 NC controls did not replicate these findings (Gibson et al., 2000aGo). No differences were identified between IVF and comparison families in observed parent–child attachment, nor were any differences identified in maternal sensitivity, structuring or hostility during videotaped free play. However, this same group found that the IVF fathers expressed lower self-esteem and less marital satisfaction than those in the NC control group (Gibson et al., 2000bGo), although measure of general adjustment and parenting showed no differences.

All the studies reported above have included IVF births, but in the last 10 years there has been increasing use of ICSI (Sutcliffe, 2000Go; Nygren and Nyboe Andersen, 2002Go). Developed initially to address male infertility factors, it has been increasingly used for ‘non-male factors’ where fertilization has failed with normal IVF. However, the implications of this newer technique for family relationships has not been fully investigated. It may have no impact, but there is a possibility that male factor ‘problems’ may lead to more secrecy about disclosure and more marital discord (McWhinnie, 1995Go). The potential for chemical and mechanical damage is greater using this technique, as is the likelihood of introducing foreign material into the oocyte. Some uncontrolled outcome studies have found a higher rate of sex chromosome abnormalities (Bonduelle et al., 1996Go) or delayed development (Bowen et al., 1998Go). Thus parents may be more anxious over time about children conceived using this method. Finally, it may have been used after traditional IVF, thus there could be a longer period of childlessness prior to conception. It is important to investigate both physical and emotional development of children born following ICSI, and the relationship between child behaviour and family characteristics, comparing them both with children conceived using traditional IVF and with NC children.

Here we present the results from a large-scale, multisite European study which was designed to compare ICSI conceived children with NC controls, and to examine the relative levels of stress in the family, as well as parental well-being and adaptation to their parental role, and the children’s socio-emotional development. A group of children conceived using traditional IVF was also included to determine any differences between the two ART groups. Including information from a number of countries, apart from increasing the sample size for the relatively new population of children conceived using ICSI, has enabled there to be an examination of effects on children and parents both within and across different cultures.

The aims of the study are to address the following hypotheses: (i) ICSI is associated with increased family stress and marital discord compared with families who have conceived naturally; (ii) parents who have conceived using ICSI have more mental health problems and difficulties in adapting to the role of parent than those who have conceived naturally; (iii) parents who have conceived using ICSI will report that their children have more psychosocial adjustment problems at the age of school entry than parents who have conceived naturally; and (iv) parents who have conceived using ICSI will be comparable with those who have conceived using IVF.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Participants
The children and their mothers and fathers were recruited in five European countries: Belgium, Denmark, Greece, Sweden and the UK, and were seen when they were 5 years old. However, the Greek team did not distribute parental questionnaires.

According to centre size, the aim was to recruit groups of ICSI, IVF and NC children with 175 children per group in Britain/Belgium and 66 children per group in Denmark/Sweden. The final totals were close to this (see Table I). Children were recruited so that at the time of assessment they would be between 4.5 and 5.5 years. Children were eligible if they were singleton, Caucasian, born after at least 32 weeks gestation, first or second born and whose mother tongue was respectively English, Dutch, Danish, Swedish or Greek.


View this table:
[in this window]
[in a new window]
 
Table I. Numbers of participants in the study, participation rates (where known) and response rates by mothers and fathers to the CBLC in each country
 
In the UK and Belgium, ICSI conceived children were recruited mainly from established cohorts already assessed in their second year (Bonduelle et al., 1998Go; Sutcliffe et al., 2001Go, 2004). Additional children were recruited from major fertility clinics. In Sweden, ICSI children were conceived after treatment at two fertility clinics and recruited from their consecutive birth records. In Denmark, most ICSI children were recruited from one clinic, but additional children were recruited from three other clinics. In Greece, the children were recruited from several clinics. In Belgium and the UK, the ICSI families had been contacted when their children were younger. However, the impact of this in terms of bias is unclear since participation was higher in the UK than in the other countries, but lowest in Belgium (see Table I).

A comparison group of children conceived with standard IVF (i.e. without ICSI) was also recruited using the same criteria. The IVF comparison groups were recruited from participating fertility clinics in all countries in a similar manner to the ICSI conceived children.

NC controls were selected according to the above criteria and the groups were matched as closely as possible in each country to the ICSI and IVF groups for maternal age, maternal education and parental socio-economic status, gender and birth order. Due to differing national laws and quality of child health records, strategies for the recruitment of NC control and comparison groups differed from country to country. In the UK, Belgium and Greece, local schools and nurseries were used to recruit matched NC controls. In Sweden, NC controls were recruited via the Swedish medical birth registry. In Denmark, recruitment was from the participating hospital’s birth registry. The participation rate (where known) is indicated in Table I. The different recruitment strategies used for each groups, and across countries, may have introduced some selection bias. The ideal situation would have been to have a national birth register in each country. In the absence of that resource, the control groups recruited in Belgium and the UK may have been biased towards children and families with few problems, or alternatively to those with many problems who wished to share these with a psychologist.

Procedures
Close to their fifth birthday, children received a paediatric examination and an assessment of their cognitive development using the WPPSI-R and subtests from the McCarthy Scales of Child Development (the results of medical and the cognitive testing are reported elsewhere; Kristoffersen et al., 2003Go; Sutcliffe et al., 2003Go).

Children were also presented with a procedure designed to investigate their perception of parents, the Bene–Anthony Family Relations Test (Bene, 1985Go). This test allows them to attribute 16 positive feelings [e.g. Who does (child) like to play with? Who smiles at (child)?] and 16 negative feelings [e.g. Who is naughty? Who makes (child) cry?] to their mother, father, siblings, themselves or to nobody in the family. An additional eight items reflect dependence [e.g. Who should help (child) get dressed in the morning?] giving a total score for each family member that can range from 0 to 40. The family members are represented by cardboard shapes attached to a ‘posting box’ base into which each statement can be posted.

The countries used different strategies for this, some asking parents to complete the questionnaires at the same time as the child assessment, other countries using postal distribution and return. In all tables, the sample size is indicated together with the countries included in the analysis. The response rates of mothers and fathers in each country to the most frequently completed questionnaire [the Child Behaviour Checklist (CBCL)] are shown in Table I.

Parental well-being and family functioning
Measures were only included that assessed current well-being and family functioning. It would have been useful to include other instruments examining a history of mental health difficulties and family stresses (such as life events) during the 5 years since birth, but there was pressure to keep the number of questionnaires to a minimum. As an indication of current emotional well-being, parents completed the 28 item version of the General Health Questionnaire (GHQ; Goldberg and Hillier, 1979Go) describing four dimensions: somatic symptoms, anxiety, social dysfunction and severe depression, each with seven items (scored 0, 1, 2 and 3) with scores ranging from 0 to 21 and a total symptom score (range 0–84). Current stress regarding parenting was examined using the Parenting Stress Index (PSI), Short Form (Abidin, 1990Go) which has a subscale indicating Parental Distress (PD) consisting of 12 items with a 5-point response scale (range 12–60; e.g. there are quite a few things that bother me about my life).

Marital relationships were key to this study since some previous studies had found more discord in families who have conceived using IVF. The Dyadic Adjustment Scale (DAS; Spanier, 1976Go) measures the quality of the parental relationship. It has four subscales: consensus (range 0–65; e.g. agreement making major decisions); satisfaction (range 0–50; e.g. ever discussed divorce?); affectional expression (range 0 – 12; e.g. not showing love); and cohesion (range 0 – 24; e.g. share outside interests) and a total adjustment score (range 0–151).

Undergoing ART involves a substantial commitment to becoming a parent. Measures were included to examine current commitment to parenting (range 17–102; e.g. I cannot imagine a satisfying life without children) and (for most adults their other major role) to their identity as a worker outside the home (range 17–102; e.g. I find that I put work responsibilities ahead of family responsibilities) (Greenberger and Goldberg, 1989Go). Each scale consists of 17 items with a 6-point Likert-type agree to disagree response scale. A higher score indicates more commitment.

Parent–child relationship
There have been suggestions of overprotection or higher levels of warmth in relation to IVF children. However, stress within the parent–child relationship is also relevant. The PSI has a ‘Parent–Child Dysfunctional Interaction’ subscale (P-CDI) derived from 12 items indicating difficulties in the parent–child relationship (range 12–60) (e.g. my child smiles at me much less than I expected). The Parental Acceptance–Rejection Questionnaire (PARQ; Rohner, 1999Go) asks parents how they usually treat their child and measures both warmth and negative aspects of the parent–child relationship. It consists of 60 items each with a 4-point scale (strongly agree to strongly disagree). Scores are derived for warmth/affection (range 20–80; e.g. I say nice things about my child), aggression/hostility (range 15–60; e.g. I nag or scold my child when he/she is bad), neglect/indifference (range 15–60; e.g. I ignore my child) and rejection (range10–40; e.g. I wonder if I really love my child). By taking the warmth score from 100, the four scales make up a total acceptance/rejection score, with higher scores indicating more rejection in the parent–child relationship.

Childrens’ socio-emotional development
To understand parenting, it is important to have information about the children themselves. While it would have been preferable to obtain this independently (for instance from teachers), this was not possible. Consequently, parents were asked to complete a personality or ‘temperament’ scale, the McDevitt and Carey (1978Go) temperament questionnaire. The 72 items each have a 7-point Likert-type scale ranging from ‘hardly ever’ to ‘almost always’ which reflect how much the parent considers each of the behaviours is typical of their child (e.g. when my child objects to wearing certain clothing he/she argues loudly, yells and cries). Nine temperamental dimensions (activity, rhythmicity, approach, intensity, mood, persistence, distractibility, threshold and adaptability) are derived by taking the mean of the eight item scores for that dimension so that each dimension’s score ranges from 1 to 7 and the total temperament score ranges from 9 to 63 (a higher score indicating a more ‘difficult’ temperament).

The CBCL (Achenbach, 1991Go) gives information about children’s emotional and behavioural problems. Descriptions of 113 problems are presented and the parent indicates whether they are ‘not true’, ‘somewhat or sometimes true’ or ‘very true or often true’ of their child, with item scores of 0, 1 or 2. The total score (adding all items) is converted to a T score to normalize the distribution and to take into account child age (range 23–100), with different conversion tables for boys and girls. The symptoms are subdivided further into externalizing problems (e.g. fights, temper tantrums) and internalizing problems (e.g. worries) which are also converted to normalized T scores, with a score of 50 representing the 50th percentile. A T score of 60 is at the bottom of the clinical range, with T scores of 64 or more representing marked problems.

The PSI Short Form (Abidin, 1990Go) Difficult Child (DC) subscale consists of 12 items (e.g. I feel that my child is very moody and easily upset) to which the parent has to agree or disagree (5-point Likert-type scale), resulting in a total ‘difficult child’ score ranging from 12 to 60.

Statistical analysis
For each questionnaire, data were only included in the analysis if the total response rate for the country was at least 60%. In all countries, there were lower response rates for fathers’ questionnaires. In some cases, this led to severely reduced sample size for fathers since the countries with the largest samples (Belgium and the UK) had the lowest response rates. In these cases, to increase sample size, results are presented including countries with at least a 50% response rate.

Univariate analyses were conducted, using ANOVA in SPSS 11.5 for Windows, to determine differences in mean scores between the children conceived using ICSI, IVF and the NC controls. Where a group effect was identified post hoc at a significance level of P < 0.001, Tukey’s test was conducted. The level of significance was chosen to take into account alpha inflation.

In those instances where there was a significant group effect on mean values at P < 0.001, linear regression analyses were conducted in SPSS 11.5, using the ‘Enter’ method, to control for demographic differences between the conception groups. Dummy variables were created for each conception group, and the reference group was the NC controls. All the demographic descriptors for which there was a significant univariate conception group effect were included: maternal education, maternal age, paternal education and paternal age. Country was also entered along with child’s gender if there was a significant difference between boys and girls on the outcome variable. The reference country was chosen for each regression as the one with the lowest score for the outcome variable under consideration.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Differences between the groups
Table I shows the final sample and the response rate for the most frequently completed questionnaire, the CBCL. The mean child age at the time of assessment was similar for all three groups (ICSI 5.02 ± 0.33; IVF 5.04 ± 0.31; NC control 5.05 ± 0.33; means ± SD). There were, however, significant differences between the conception groups in terms of maternal age and paternal age; parents in both the ICSI and IVF groups were significantly older than the NC control group, but similar to each other [maternal age (years) ICSI 33.4, IVF 33.9, NC control 30.6; F = 64.21, P < 0.001; paternal age (years) ICSI 34.6, IVF 35.2, NC control 31.9; F = 52.47, P < 0.001]. Both maternal and paternal education were significantly related to conception group; more parents in the NC control group tended to have higher levels of qualifications and fewer had none (maternal {chi}2 = 44.44, P < 0.000; paternal {chi}2 = 27.43, P < 0.002). Maternal occupation, paternal occupation and family social class were not associated with conception group.

Parental well-being and family functioning
There were no conception group differences in maternal well-being according to the GHQ total score or any of the subscales (see Table II). Furthermore, no group differences were identified in the total GHQ for fathers. Rates of problems on the subscales were generally low, the only difference being a conception group effect on social dysfunction (see Table II). However, this difference was not evident for either group in the regression analyses, after taking demographic factors and country into account (ICSI standardized {beta} = –0.09, t = –1.76, P < 0.08; IVF standardized {beta} = –0.12, t = –2.38, P < 0.02).


View this table:
[in this window]
[in a new window]
 
Table II. Mean parental mental health (SD) from the GHQ by conception type
 
There were no significant conception group differences in levels of parental stress, according to the PSI, PD subscale (see Table III). Similarly, there were no differences between the three groups in marital discord according to the total DAS or any of the subscales, as reported by either mothers or fathers. In fact, the three groups were remarkable for their similarity across each of the subscales of the questionnaire (see Table IV).


View this table:
[in this window]
[in a new window]
 
Table III. Mean parental stress and parent–child dysfunctional interactions (SD) according to the PSI by conception type
 

View this table:
[in this window]
[in a new window]
 
Table IV. Mean marital satisfaction (SD) from the DAS by conception type
 
Parent–child relationship
There were no significant group effects on the P-CDI scale of the PSI (see Table III).

Mothers who had conceived using ICSI reported fewer negative feelings towards their children than mothers in the naturally conceived group (PARQ total) and were specifically less likely to report aggressive or hostile feelings (PARQ aggression; see Table V). There were no significant differences on the father’s responses to the PARQ (see Table V).


View this table:
[in this window]
[in a new window]
 
Table V. Mean parent–child relationships (SD) according to the PARQ by conception type
 
The effect of conception method on maternal hostility/aggression scores remained in the regression analyses after taking into account demographic variables that differentiated the conception groups, and country (PARQ hostility: ICSI versus NC {beta} = –0.17, t = –3.47, P < 0.001).

On the basis of children’s responses to the Bene–Anthony, there were no conception group effects for children’s positive or negative feelings towards or total involvement with either mother or father (see Table VI)


View this table:
[in this window]
[in a new window]
 
Table VI. Mean child perceptions of parent–child relationships from the Bene–Anthony Family Relations Test (SD) by conception group
 
Parental attitudes to family life
Univariate comparisons identified significant effects of conception type on mothers’ commitment to work and to parenting, and fathers’ commitment to parenting. Mothers with children conceived through IVF tended to be less committed to their work role than the NC controls, and mothers who had conceived using ICSI were more committed to their role as a parent than either NC controls or the IVF group. There were no significant differences between the groups in father’s commitment to work (see Table VII).


View this table:
[in this window]
[in a new window]
 
Table VII. Mean parental commitment to work and commitment to parenting (from the Greenberger scales) (SD) by conception type
 
However, the conception group effect (IVF greater) on maternal commitment to work did not remain once demographic variables (maternal and paternal age and education) and country had been taken into account in the regression. The only significant effects were country variability (Denmark, Sweden and Belgium all higher than the UK, {beta} = 0.23, 0.16 and 0.54; t = 5.99, 3.85 and 12.05, respectively; all significant at P < 0.0001), and the highest maternal education group (degree or professional qualification) was higher than the lowest (no qualifications; {beta} = 0.16, t = 2.63, P < 0.001)

The conception group effect remained for maternal commitment to parenting. Those mothers who had used ICSI had more commitment to their parental role than the NC group (adjusted R2 = 0.11, F = 6.76, P < 0.0001; ICSI versus NC standardized {beta} = 0.20, t = 4.94, P < 0.0001). In addition, there was a significant country effect, with mothers in Belgium and the UK reporting more commitment than those in Sweden ({beta} = 0.27 and 0.13, t = 5.38 and 2.50, respectively). Parental age and education were unrelated.

Child personality and socio-emotional development
There were no significant conception group differences in total temperament of the children according to mothers or fathers (see Table VIII), nor were there any differences on the subscales. On the basis of the DC scale or the PSI, there were no significant differences between conception groups (see Table VIII ).


View this table:
[in this window]
[in a new window]
 
Table VIII. Mean child socio-emotional problems (SD) and percentage above the clinical cut-off by conception type according to the CBCL, mean scores on the PSI DC scale and mean total temperament scores from the Carey–McDevitt scale
 
Children in the assisted reproduction groups also did not differ overall in terms of behaviour problems as rated on the CBCL, nor were there differences in the proportion of children above the clinical cut-off points (see Table VIII ).


    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
This study includes a substantial number of children and parents from different European countries. Overall it confirms and substantiates the results of previous smaller studies in that mothers and fathers who have experienced conception using IVF or the newer technique of ICSI are no more likely to have marital difficulties, mental health problems or family stress than families who have conceived naturally. Few differences were identified between parents who had conceived using IVF and those who had used the newer technique of ICSI. Therefore, concerns for parental well-being related to differences in the technique would seem unnecessary.

Parents of children born using both methods of assisted reproduction reported their children as having similar temperaments and similar levels of behaviour problems to the NC group.

Mothers (but not fathers) in the ICSI group were more committed to being a parent than the NC control mothers. Duncan and Edwards (1999Go) have identified three ‘gendered moral rationalities’ used by women to explain their roles, based on identities and responsibilities to children: those who see themselves as primarily a mother; those who see themselves primarily as a worker; and those with an integrated mother–worker role. The mothers who have experienced ICSI are closest to the primarily mother identity. This may indeed have some implications in years to come. Women who have negated their role in the ‘outside world’ in order to commit to parenting may, as they see their chances of career development receding, regret this approach. Thus it is vital to follow these families as their children grow and the families have different priorities.

However, at this point in their children’s lives, there is no evidence of any resentment. The PARQ allows the respondent to report on the parent–child relationship, including a number of questions elicited. Using this, we are able to report that mothers who have used ICSI were less likely to be hostile or aggressive about their child than those mothers in the NC control group, although, in contrast to Golombok et al. (1995Go), we did not find higher levels of expressed warmth. Possibly, as Van Balen (1996Go) suggests, they may be trying to be overprotective by suppressing negative feelings about their child, which might also have implications during adolescence, a time when children become more independent from their parents and may display a range of emotional and behavioural problems not typical of younger children. Following these and similar families over time will be required to address this issue.

Nevertheless, the limitations of the study need to be kept in mind. The variable, and low in some cases, response rate in each country for the questionnaires may have influenced the results. Families experiencing difficulties may have been less likely to complete the questionnaires, especially in the ART groups who have ‘worked harder’ to become parents. Another possible factor which inevitably distorts the comparison is that the families who have conceived from fertility treatment implicitly represent the successful ones, i.e. the ones who have followed the stressful processes involved to completion. They are a selected group and their intrinsic qualities undoubtedly have contributed to the success of their treatment, whereas when children are the result of natural conception, it may or may not have been planned, which could have implications for parenting. Therefore, this should be considered when interpreting the more positive feeling couples report towards their children after ART. Finally, it would have been useful to know about the experiences of these families in the 5 years since their child was conceived. There may have been stressful events, or periods of marital discord that differentiated the groups. A more detailed retrospective examination of family histories in face to face interviews could have examined that, but the available funding did not allow that amount of face to face contact with parents. This could be explored in future research.

The results from this study should be of immense importance in understanding the implications of using ICSI, both for the child outcomes and for the impact on family life. Temperament most closely reflects any inherited aspects of children’s style of responding to the world, and the children conceived by ICSI cannot be distinguished from those conceived using standard IVF or those conceived naturally. Families are coping well and no strains were identified in marital or parent–child relationships. Therefore, clinicians can confidently assure families that there is no additional risk of negative socio-emotional impact for either parents or children.


    Acknowledgements
 
We would like to thank the families who have participated in our study, the fertility clinics in all five countries assisting with recruitment, and likewise the schools in the countries who assisted with recruitment. This project was funded by EU project grant QLRT-1999-30545.

The protocol was approved by the ethics committee of each institution in accordance with national regulations in each country. The European Union 5th framework quality of life programme contract QLG4-CT-2000-00545 paid for this project entitled—An international Collaborative Study of ICSI: Child and Family Outcomes (ICSI-CFO). The funding source had no responsibility for study design or interpretation of data.


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Abidin R (1990) Parenting Stress Index Test Manual. Pediatric Psychology Press, Charlottesville, VA.

Achenbach TM (1991) Manual for the Child Behaviour Checklist/4-18 and 1991 Profile. University of Vermont, Department of Psychiatry, Burlington, VT.

Bene E (1985) Manual for the Family Relations Test, 2nd edn. NFER, Slough, UK.

Bonduelle M (2003) ICSI-Related Risk for the Children. PhD in Medical Science. Vrije Universiteit Brussels.

Bonduelle M, Joris H, Hofmans K, Liebaers I and Van Steirteghem A (1998) Mental development of 201 ICSI children at 2 years of age. Lancet 351,1553.[CrossRef][Medline]

Bonduelle M, Wilikens A, Buyssc A et al. (1996) Prospective follow-up study of 877 children born after Intracytoplasmic sperm injection (ICSI), with ejaculated epididymal and testicular spermatozoa and after replacement of cryopreserved embryos obtained after ICSI. Hum Reprod 11(suppl 4) 131–155.

Bowen JR, Gibson FL, Leslie GI and Saunders DM (1998) Medical and developmental outcomes at 1 year for children conceived by intracytoplasmic sperm injection. Lancet 351,1529–1534.[CrossRef][Medline]

Colpin H, Demyttenaere K and Vandemeulebroecke L (1995) New reproductive technology and the family: the parent–child relation following in vitro fertilization. J Child Psychol Psychiatry 36,1429–1441.[Medline]

Duncan S and Edwards D (1999) Lone Mothers, Paid Work and Gendered Moral Rationalities. Palgrave Macmillan Publishers Ltd, Basingstoke, UK.

Gibson FL, Ungerer JA, McMahon CA, Leslie GI and Saunders DM (2000a) The mother–child relationship following in vitro fertilisation (IVF): infant attachment, responsivity and maternal sensitivity. J Child Psychol Psychiatry 41,1015–1023.[CrossRef][Medline]

Gibson FL, Ungerer JA, Tennant CC and Saunders DM (2000b) Parental adjustment and attitudes to parenting after in vitro fertilisation. Fertil Steril 73,565–574.[CrossRef][Medline]

Goldberg DP and Hillier VF (1979) A scaled version of the General Health Questionnaire. Psychol Med 9,139–145[Medline]

Golombok S, Cook R, Bish A and Murray C (1995) Families created by the new reproductive technologies: quality of parenting and social and emotional development of the children. Child Dev 64,285–288.

Golombok S, Brewaeys A, Cook R, Giavazzi MT, Guerra D, Mantovani A, van Hall E, Crosignani PG and Dexeus S (1996) The European study of assisted reproduction families: family functioning and child development. Hum Reprod 11,2324–2331.[Abstract]

Greenberger E and Goldberg W. (1989) Work, parenting, and the socialization of children. Dev Psychol 25,22–35.[CrossRef]

Kristoffersen I, Bonduelle M, Barnes J, Sutcliffe A, Nekkebroeck J, Loft A, WennerholmU, Tarlatzis BC, Kantaris X and Hagberg B (2003) Cognitive and motor developmental outcomes of 1523 ICSI, IVF naturally conceived 5 years old children. Paper presented at the 19th Annual Meeting of ESHRE, Madrid, July.

McDevitt SC and Carey WB (1978) The measurement of temperament in 3 to 7 year old children. J Child Psychol Psychiatry 19,245–253.[Medline]

McWhinnie A (1995) A study of parenting of IVF and DI children. Med Law 14,501–508.[Medline]

Mushin DN, Spensley JC and Barreda-Hanson MC (1995) Children of IVF. Clin Obstet Gynecol 12, 865–876.

Nygren KG and Nyboe Andersen A (2002) Assisted reproductive technology in Europe, 1999. Results generated from European registers by ESHRE. Hum Reprod 17,3260–3274.[Abstract/Free Full Text]

Palermo G, Joris H, Devroey P and Van Steirteghem AC (1992) Pregnancies after Intracytoplasmic injection of single spermatozoon into an oocyte. Lancet 340,17–18.[Medline]

Rohner RP (1999) Handbook for the Study of Parental Acceptance and Rejection. University of Connecticut, Storrs, CT.

Spanier GB (1976) Measuring dyadic adjustment: new scales for assessing the quality of marriage and similar dyads. J Marriage Fam 38,15–28

Sutcliffe AG (2000) Intracyctoplasmic sperm injection and other aspects of new reproductive technologies. Arch Dis Child 83,98–101.[Free Full Text]

Sutcliffe AG (2002) Health risks in babies born after assisted reproduction. Editorial. Br Med J, 325,117–118.[Free Full Text]

Sutcliffe AG, Taylor B, Saunders K, Thornton S, Lieberman BA and Grudzinskas JG (2001) Outcome in the second year of life after in-vitro fertilisation by intracytoplasmic sperm injection: a UK case–control study. Lancet 357,2080–2084.[CrossRef][Medline]

Sutcliffe AG, Loft A, Wennerholm UB, Tarlatzis B and Bonduelle M (2003) International collaborative study of ICSI-child and family outcomes—physical development at 5 years. Paper presented to the 19th Annual Meeting of the EHSRE, Madrid, July.

Sutcliffe AG, Edwards PR, Beeson C and Barnes J (2004) Comparing parents’ perceptions of IVF conceived children’s behaviour with naturally conceived children. Infant Mental Health Journal 25(2),163–170.[CrossRef]

VanBalen F (1996) Child rearing following invitro fertilization. J Child Psychol Psychiatry, 37,687–693.[Medline]

VanBalen F (1998) Development of IVF children. Dev Rev 18,30–46.[CrossRef][Medline]

VanBalen F, Trimbos-Kemper TCM and Naaktgeboren N (1996) IVF: the experience of treatment, pregnancy and delivery. Hum Reprod 11,95–98.[Abstract]

Submitted on November 13, 2003; accepted on February 27, 2004.