Worldwide, parental divorce and separation rates are high. These include impactful experiences to all involved, posing risks on well-being and functioning of parents and their children, and leading to increased professional care seeking. The Dutch ‘Divorce ATLAS’ training aims at informing and empowering parents after divorce or separation. The intervention offers preventive support and psychoeducation in a 2-sessions group program. We aimed to assess the effectiveness of Divorce ATLAS for parents in the Dutch setting. We performed a quasi-experimental study in which we compared parents participating in Divorce ATLAS groups (n = 187) and parents on a waiting list for online support (n = 111). Outcomes included well-being of parents and their children, parental sense of competence and social support, and children’s prosocial behavior. We performed regression analyses to assess differences in outcomes between intervention and control group. Additionally, we measured and reported parental knowledge and trigger to act after the intervention. As compared with routine family support of parents in the control group, the Divorce ATLAS group intervention led to increases in parents’ sense of parenting incompetence (Hedges’ g = 0.19), and in perceived social support (Hedges’ g = 0.23). Moreover, most parents indicated that Divorce ATLAS had increased their knowledge and had triggered them to act accordingly. Hence, the increased sense of parenting incompetence in the intervention group may be due to increased participants’ awareness of their parenting behaviors. In sum, Divorce ATLAS is promising for educating and supporting parents after divorce.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Worldwide, the prevalence of parental divorce or separation is high. Regional variations in this prevalence exist, reflecting local differences (e.g., political, economic, cultural, religious). Divorce rates appear to be particularly high in Northern/Western Europe and European settler colonies in North America (United States, Canada) and Oceania (Australia and New Zealand) (Lir Wang and Schofer, 2018). In these countries, the average gross divorce rates, being the number of legally registered divorces per 1000 married people, are about 4 to 6. In Europe, the annual number of divorces per 1000 population, (i.e.,) the crude divorce rate, equals 1.8. This means that in Europe, about 0.8 million marriages end in divorce each year (EU-27, Eurostat, 2021). By comparison, in the United States the crude divorce rate in 2019 was 2.7 per 1000 population (CDC, 2020). These numbers do not include the separation of registered and cohabiting partnerships, which are also numerous, but for which exact numbers are not available (Boertien, 2020).
Divorce (from here on ‘divorce’ refers to both separation and divorce) is considered a negative life event (as is bereavement or unemployment; e.g., Luhmann et al. 2012), posing risks on health, well-being and functioning of those involved (e.g., Pellón-Elexpuru et al. 2024). It marks the loss (Harvey and Miller, 1998) of an important other, in this case the intimate partner. As a result, grief is a common experience in divorce, as it is after the death of a loved one (Papa et al., 2014). It takes time and effort of the divorced to accommodate their loss, and to adapt to new family structures and dynamics. Divorce is found to lead to increased levels of stress and depressive feelings (e.g., Booth and Amato 1991) and of seeking of professional care (e.g., Bracke et al. 2010, Reneflot et al. 2020).
However, outcomes of divorce need not to be just negative. In fact, some studies show that after divorce, subjective wellbeing may decrease initially, but may increase later, and as such be beneficial due to being a relief from a bad marriage (e.g., Luhmann et al. 2012). Moreover, reported consequences of divorce are averages. Some divorces proceed relatively smoothly, while others are more complex. In a large portion of divorce cases, children are involved, complicating the divorced ex-partners’ adaptation after union dissolution and affecting the children involved. In these situations, partners have ended their romantic relationship, but typically remain involved as parents as they continue to have joint legal custody after divorce and jointly raise their children. Following divorce, adults are found to function less optimally as a parent than adults in stable relationships, and parent-child relationship quality in these families is lower (Elam et al., 2019; Margolin et al., 2001; Sandler et al., 2008; Sigal et al., 2011; Van Dijk et al., 2020). Research consistently shows that on average children with divorced parents score significantly poorer on measures of well-being, conduct, psychological adjustment, self-concept, and social relations (Amato and James, 2010: Amato and Keith, 1991; Amato and Sobolewski, 2001; Harland et al., 2002; Theunissen et al., 2017). Similarly, adult children, (i.e.,) aged over 18 at the time of their parents’ divorce, have a hard time to adjust emotionally to the new family structure and experience a negative impact on their relationship with their parents (Shanholtz et al., 2021). Once again, these are averages; children differ in the extent to which they experience (long-term) problems as a result of the divorce (e.g., Amato and Anthony 2014). Research shows that these differences are related to and partly mediated by parent- and family-related factors (e.g., Van Dijk et al. 2020). This concerns, for example, stress and support, bereavement, parenting after divorce (effective parenting, positive parent-child relationships) and cooperation between parents (positive communication and the degree of destructive conflict between parents) (Amato, 2010). The likelihood of problems is higher in case of difficult or lacking contact of the child with one or both parents, and in case of high levels of contention and (destructive) conflict between the parents (Davies et al., 2012; Kopystynska et al., 2017; Lee et al., 2019; Van Dijk et al., 2020). Knowledge and awareness of the mechanisms that mediate post-divorce adjustment, such as social support, grief resolution or accommodation to loss, conflict management and positive communication between parents, and authoritative parenting may serve as protective factors. This therewith offers leads for effective preventive support.
We previously studied the effects of a preventive group training for Dutch 6- to 8-years old children after their parents’ divorce (Klein Velderman et al., 2022). The 12-sessions Children of Divorce Intervention Program in the Netherlands (CODIP-NL) led to increases in children’s positive functioning and well-being and decreased emotional and behavioral problems. The findings support wider implementation of CODIP-NL to strengthen children’s adaptation to parental divorce. Less is known however about effective ways to support knowledge, awareness, and post-divorce adaptation of parents. (Court-connected) educational programs for divorcing parents, providing knowledge, and strengthening parents’ awareness, have become available in the majority of U.S. counties since the 90s. Fackrell and colleagues (2011) assessed the effectiveness of such court-affiliated divorcing parents education programs. They found a significant moderate-positive effect (d = 0.39; range 0.19–0.61) based on nineteen studies; participating parents had 50% better outcomes regarding co-parenting conflict, parent-child relationships, parent and child well-being, and relitigation compared with non-participants. The past decades, initiatives to support couple relationships and educate parents after divorce have further grown in number, increasingly using web-based formats such as online divorce platforms or education courses (e.g., Becher et al. 2015, Doss et al. 2016, Hald et al. 2020). Nevertheless, the availability and evidence regarding its effects of educational (group) programs for divorcing parents are still rather limited outside the U.S. Moreover, evidence shows that feelings of isolation may decrease, and social support may increase, as a result from participation in peer-support group programs (e.g., Hoey et al. 2008).
The ‘Divorce ATLAS’ entails an initiative to offer such educational training in the Netherlands, aiming at informing and empowering parents after break-up, offering psychoeducation on effective parenting, coping and communication skills. Figure 1 illustrates the Divorce ATLAS’ model of intervention. The ultimate goal of Divorce ATLAS is to strengthen parenting and the parenting context after divorce, in order to improve the opportunities for and well-being of children. The intervention is available in two varieties: (a) a 2-sessions group program derived from the ACT for Children: Assisting Children through Transition intervention (Pedro-Carroll et al., 2001); and (b) an e-health version inspired by ACT, and the Children in Between program, primarily its e-health format (e.g., Gordon and Arbuthnot 2011). Both versions include the same content. All information is also available in a published workbook for parents (Klein Velderman et al., 2020). The current paper focuses on the 2-sessions Divorce ATLAS group program, offering parents both peer support and educational training.
Fig. 1
Model of divorce ATLAS intervention
×
We aimed to assess the effectiveness of the post-divorce preventive Divorce ATLAS group training for parents in the Dutch setting, regarding well-being of parents and their children, parental knowledge, behavioral intentions, feelings of competence and social support, and children’s prosocial behavior. We tested our hypotheses that Divorce ATLAS would: 1. Increase well-being of participating parents and their children, 2. Strengthen parental sense of competence and social support, 3. Increase children’s prosocial behavior, and 4. Contribute to parental knowledge about the impact of divorce and positive (co-)parenting strategies, and trigger to act accordingly.
Method
We performed a quasi-experimental study, as part of trial NTR6756 (NL-OMON23882) in the Dutch trial register, and we report this following the CONSORT 2010 guidelines (see CONSORT Checklist in Appendix 1). The Leiden Medical Centre Institutional Review Board concluded that the Dutch Medical Research Involving Human Subjects Act (WMO) did not apply to the trial, and accordingly review by the board was not mandatory (September 22nd, 2015). The trial was reviewed and approved by a scientific commission of ZonMw (Dutch institute for health-care research: letter 2015/24796/ZONMW, December 11th, 2015). The trial was carried out according to the approved project plan and was conducted in accordance with ethical principles as published by the World Medical Association in the Declaration of Helsinki.
Participants
We compared two groups: a group of parents receiving Divorce ATLAS group training (n = 187; intervention group); and a group of parents who were on the waiting list for online support (i.e., the e-health module of Divorce ATLAS; n = 111; control group). We used the following inclusion criteria: (a) adults with children (fathers and mothers of children in all ages), (b) divorced or separated or currently in a divorce or separation procedure, and (c) informed consent to participate in the study and training. Consequently, parents were welcome to participate in the study and group training as individual parents (i.e., no need for involvement of the other parent). Separated or divorced adults without own biological children (including stepparents) were excluded in the study. Inclusion in the training was not limited to participants in a specific range of time since divorce. Divorce ATLAS aims at parental post-divorce adaptation. And although many adjustments are made in the first one or two years after divorce, new experiences and challenges (such as formation of new relations, moving house, conflict with ex-partner) can impact post-divorce adaptation and raise parents’ questions and support needs, and therewith lead to participation in the training.
Parents meeting the above-mentioned criteria were included in the intervention group from January 2018 to October 2019, via communication between social work, preventive healthcare, or youth social care professionals and their clients or partners in the field (by means of website, social media, brochures, personal contact). This communication included concrete terms about target group and goals of the intervention. No selection interviews took place prior to intervention start.
Parents were included in the control group between October 2017 and December 2019, via several channels: We placed open invitations to participate in the online training on social media, local and national newspapers and newsletters, and provided for an informative website. We also informed professionals in the field (e.g., mediators, family lawyers, youth social care professionals) about the availability of the online training and requested them to find parents willing to participate. Moreover, we distributed flyers and held presentations about the training at meetings of professionals. Half of the parents who consented to participate in the online training, were randomly assigned to a waiting list for online support. This group of parents formed our control group.
Intervention
Parents in the intervention group received the Divorce ATLAS (in Dutch denoted as “ScheidingsATLAS”) group educational program, consisting of two 3-hours sessions, 3 to 4 weeks apart. Each group consisted of up to 10 participants. The intervention was offered to them, free of charge, at a divorce center or at the local health center in two rural regions in the Netherlands (The Hague and Almere area). Part of the groups was scheduled at daytime (during working hours), the other groups at evenings. No childcare or meals were provided, but drinks (e.g., coffee, tea, water) were.
The Divorce ATLAS group intervention has resulted from a co-creative and participatory research approach with participation of scientific experts on parenting after divorce, experienced professionals, and separated parents. Initial inspiration for the content of the Divorce ATLAS group intervention came from content of the ACT for the Children intervention (Pedro-Carroll et al., 2001). ACT for the Children consists of two components: one on the legal process and one on parent and child well-being. Divorce ATLAS’ main focus is on the latter, which means that Divorce ATLAS, as does ACT for the Children, addresses the impact of the divorce on parents and children, ways to have a good relationship with and provide support for children, the importance of keeping children out of conflicts and the importance of emotional security, ways to interact with the other parent in a conflict reducing way, and the importance for parents of taking care of themselves and ask for support where necessary to enable quality parenting. A detailed program outline for group leaders, specific program content such as video fragments and the “circle of loss”, and a Dutch-published workbook were co-created for this aim. In accordance with the structured program outline, two certified group leaders offer support and psychoeducation (see Fig. 1) focused on six themes: 1. The impact of divorce, 2. (Co-)parenting after divorce, 3. Communication with the other parent, 4. Dealing with conflict, 5. Parenting strategies and parent-child relations, 6. Feeling empowered after divorce. Regarding all six intervention themes, group leaders provide participants with information and positive strategies, as well as facilitate positive group processes, such as self-disclosure, giving and receiving feedback, by welcoming group members to reflect upon their own situation and take part in group dialogues.
The first Divorce ATLAS group session includes the first three themes. To start with, group leaders educate parents about loss and stages of grief, using the “circle of loss” (Klein Velderman et al., 2020). This circle is a simplified model representing the grief task model as proposed by Worden (1991) and Weijers and Penning (2001). The circle of loss begins with the divorce itself and the recognition that the loss is definite. Then it shows the (four) grief tasks that are needed to process the grief. Parents are welcomed to reflect on their own situation and are taught how the circle can help them better understand their child, the other parent, and others involved in the divorce. Next, session 1 educates parents about the impact of divorce on children, as well as common responses of children in different age groups and strategies to respond to these as a parent. Additionally, parents learn about protective and risk factors found in literature, impacting on child outcomes after divorce and playing an important role in post-divorce adjustment. Regarding theme 2, group leaders introduce the concepts of cooperative and parallel parenting after divorce (Pedro-Carroll et al., 2011), using an animation video. It is discussed how to share parenting responsibilities with the other parent, according to their personal situations. And finally, positive communication skills (Rosenberg, 2015) are introduced (including use of the I message; Gordon and Edwards 1997) to help discuss difficult topics constructively, (i.e.,) theme 3. Group leaders use video fragments with and without the use of the I message, or alternatively use role plays. The group reflects on situations and possible benefits of positive communication on children.
The second group session regards the other three themes. The first is dealing with conflict, (i.e.,) theme 4. Group leaders explain the negative impact of high levels of interparental contention or (destructive) conflict between parents on children (e.g., Kopystynska et al. 2017, Van Dijk et al. 2020). To explain the concept of conflict levels or stages (e.g., Glasl 2020), a simplified thermometer is used, ranging from green colored conflict (“ratio”; shared goals, constructive) to red (“fight”; enemies in conflict, hopeless, destructive). Parents reflect on their own situation. Besides, video fragments with different ways of conflict resolution are watched and discussed, and group leaders provide practical tips. The second regards parenting strategies and parent-child relations, theme 5. Group leaders provide information about the two parenting dimensions; (a) parental warmth and responsiveness, and (b) monitoring and control, resulting in the four parenting styles permissive, uninvolved or neglectful, authoritarian, and authoritative (Baumrind, 1966; Maccoby and Marin 1983). It is explained why children profit from an authoritative parenting style. Again, group leaders work with video, reflection, and group discussion. The final theme of Divorce ATLAS is about the balance between stressors and challenges on the one hand, and supportive factors and resilience on the other hand. In this balance or resilience model, the divorce in itself is not perceived as the end of the world for parents and children. Post-divorce coping depends to a large extent on how these stressors and supportive factors are balanced. Group leaders demonstrate theory of resilience and coping, using the two-armed scale with weights to “illustrate (dis)balance between stressors and challenges on the one side (e.g., the divorce, finance, stress, worries, arguments with the ex-partner), and resilience and positive factors on the other side (e.g., a good friend, positive character, hobbies, coping strategies)”. Group members reflect on what is on either arm of their personal scale. In Divorce ATLAS reaching out for support is explicitly regarded as a sign of strength and not failure. In the group session as well as parents’ workbook, options for further support are provided.
The intervention fidelity was evaluated using structured logbooks, filled out per session by the group leaders, as we did in the evaluation of a preventive group training for children who experienced parental divorce (Klein Velderman et al., 2022). These logbooks showed that group leaders deviated from the method in only a very limited number of instances. In those specific instances, the group leaders still aimed at the same goal, but used a different method, with a different activity (e.g., a different group familiarization exercise). As group dialogue and the first two themes took more time, group leaders occasionally finished the third program theme in the second instead of in the first session. In none of the groups, group leaders omitted parts of the program.
Control Group
Parents in the control group received no support or psychoeducation as part of the study. However, they had access to routine family support if needed. In the Netherlands, such services are provided by local services, paid by municipalities, and accessible without charge. Parents gained access to the e-health version of the Divorce ATLAS education program after finalizing their posttest questionnaire. The Divorce ATLAS e-health program consists of 6 chapters of one hour of self-study each, about the same themes as the Divorce ATLAS group training.
Procedure
We evaluated the intervention based on self-report from parents. Parents received a questionnaire before the start (pre), and next after (post) the second session (intervention group) or after 4 weeks (control group). Parents completed questionnaires to evaluate well-being of themselves and their children, sense of competence and social support, and children’s prosocial behavior (i.e., at pre- and post-test). Additionally, parental knowledge about the impact of divorce and positive (co-)parenting strategies as well as the extent to which parents felt triggered to act, were measured after the intervention (post-test). In order to prevent the questionnaires from becoming to lengthy and to support uniformity, parents reported child outcomes about one child only; their eldest child, meeting the following criteria: aged up to eighteen years, on whom the parent has (joint) custody, and who lives with this parent at least part of the time. Prior to the start of the intervention, parents’ written informed consent was obtained, regarding participation in the study and the intervention. Control group parents received written procedural and privacy information, and consented to participate as part of filling out the pre-test questionnaire.
Measures
Well-being
As a measure of parent and child well-being, parents indicated on a 1–10 scale how well they and their child were doing at that time (adapted continuous “Cantril ladder”; Cantril 1965; as previously used in Klein Velderman et al. 2022). Higher scores indicate more optimal well-being.
Parental sense of competence
We used the Dutch parental perceptions of parenting questionnaire: the “Ouderlijke Opvattingen over Opvoeding” (OOO; Van der Zanden et al. 2010), to measure parents’ Feelings of Incompetence (6 items) and Feelings of Competence (5 items). Items are scored on a 6-point scale (1 = completely disagree, 6 = completely agree). An example of the Incompetence scale: “Parenting my child is more difficult than I thought it would be.” An example of the Competence scale: “I feel confident in my role as parent.” Higher scores indicate more (in)competence as experienced by parents. Cronbach’s alphas in the current study were 0.81 (pre-test) and 0.84 (post-test) for the Incompetence scale, and 0.78 (pre-test) and 0.84 (post-test) for the Competence scale.
Perceived social support
As a measure of perceived social support, we used two items of the Dutch Standardized youth monitoring questionnaire (RIVM, 2005). The first item measures Daily support: “Do you receive support from your family, friends, acquaintances, or neighbors regarding everyday things in taking care of your child (e.g., babysitting your child, advise on daily care or parenting)?”. The second item measures Support with problems: “Can you talk to your family, friends, acquaintances or neighbors if you have a problem with the care of your child?”. Parents answer on 4-point scales (1 = often, 2 = regularly, 3 = occasionally, 4 = rarely or never). Items were recoded reversely (1 = 4, 2 = 3, 3 = 2, 4 = 1). As a result, higher scores indicate more perceived support.
Children’s prosocial behavior
Children’s prosocial behavior was measured using the 5-items prosocial behavior subscale of the Strengths and Difficulties Questionnaire (Goodman, 1997) for ages 4–16 (Parent Form, SDQ-PF). Items of the SDQ are scored on a 3-point scale (0 = not true, 1 = somewhat true, and 2 = certainly true), the scale being composed of the mean of (at least 3) items times 5 (i.e., to compose a sum score of the 5 items). Higher scores indicate more prosocial behavior of the child. Cronbach’s alphas in the current study were 0.76 (pre-test) and 0.79 (post-test).
Parental knowledge and trigger to act
Two items in the post-test questionnaire asked intervention participants to rate the impact of Divorce ATLAS on their knowledge, and the extent to which Divorce ATLAS triggered them to act. Items were rated on a 5-point scale (1 = not at all, 5 = absolutely): “Divorce ATLAS has increased my knowledge”. “Divorce ATLAS gave me an impulse/push to get started with this.”
Demographic and divorce-related characteristics
The following demographics were measured: child’s sex and age, participant’s sex, age, country of birth, and highest education. We also measured a selection of divorce-related variables, namely: who took the divorce initiative (joint, participant, ex-partner), child custody arrangements (no, single, joint), availability of a temporary or definite parenting plan, child support and parent alimony agreements, whether or not the participant and ex-partner had a new partner, and presence of stepchildren.
Data Handling and Analyses
Regarding data handling, we performed multiple imputations on the final data, of which 18.9% of values were missing. Missing data were imputed using the R-package mice (Van Buuren and Groothuis-Oudshoorn, 2011). A total of 100 imputations were used. Convergence of the imputation procedure was evaluated using a visual inspection of trace plots; this convergence was achieved for all variables within 10 iterations. The imputation model contained all outcome parameters, as well as demographic variables and divorce-related background variables measured at baseline. We used the poly regression method to impute categorical variables and predictive mean matching for continuous variables. Given the high amount of relevant background, divorce related, and outcome variables in our imputation model, we believe missing at random (MAR; White et al., 2010) is likely to be the missingness mechanism in our model.
In the analyses, we first described the flow of participants. Next, we assessed the demographic and divorce-related characteristics of the two groups. Third we assessed the effectiveness of the ATLAS group training regarding well-being of parents and their children, parental knowledge, behavioral intentions, feelings of competence and social support, and children’s prosocial behavior, using linear models. In these models, with and without adjustment for (I) demographic characteristics or (II) demographic characteristics and time since divorce, we predicted outcomes post-test out of intervention arm and outcomes pre-test. We then calculated Hedges’ g as measure of effect sizes from the performed regression analyses in our study. Hedges’ g represents a standardized mean difference between two groups with a correction for sample size, to come to a comparable metric independent of design. We calculated the Hedges g-values from the Beta-coefficients of our regression analyses. Each estimate was divided by the pooled standard deviation and corrected for sample size. To obtain the pooled standard deviations, we subtracted the variability attributable to treatment assignment from the variability in the outcome variables, as described by Wilson (2016). The outcomes parental knowledge and trigger to act were only measured at the post-test in the intervention group. A second series of data imputation was performed on the subset that followed the group intervention, with the only addition being the inclusion of these two variables. An intercept only model was used to establish if answers differed significantly from three on a Likert scale from one to five.
Results
Participants
A total of 298 parents participated in this study. Figure 2 illustrates the progress of participants through the trial (CONSORT diagram) by group (n): intervention (187) and control group (111). All 187 parents in the intervention condition took part in Divorce ATLAS. They participated in one of 22 intervention groups of up to 10 participants. In the intervention group, 34 parents (18.2%) did not fill out the post-test questionnaire, and so did 46 parents (41.4%) in the control group (no reason given, see Fig. 2). We conducted a drop-out analysis, comparing people who did (n = 80) and did not (n = 218) drop-out between pre- and post-test (see Fig. 2) on all demographic background variables as well as on the time since divorce variable. Results showed that drop-out was mostly not selective, except that participants with a lower level of education were relatively more likely to end their participation in the study, F(1, 232) = 0.98, p = .03. Missingness analysis revealed no selective missingness in terms of higher nonresponse on specific outcome variables. All parents with baseline data were included in the study.
Fig. 2
CONSORT flow diagram of trial study subjects. Because of imputation of the data, participants with baseline data could be included in the analysis
×
In Table 1 we summarize baseline demographic characteristics of parents and their children per group, as measured at the pre-test. The sample consisted of 99 fathers (33.2%) and 199 mothers (66.9%). They were on average 41 years old (M = 40.5, SD = 6.1). Parents in the control group were on average three years older than parents in the intervention group (t(85.21) = 3.04, p = 0.003). The eldest child of parents was on average nine years of age (M = 8.9, SD = 3.8), 53.4% of them being daughters, and 46.6% being sons. The majority (85.6%) of participating parents were born in the Netherlands: this tended to be more so in the control group (91.0% of parents born in the Netherlands) than in the intervention group (82.4% of parents born in the Netherlands; χ2 = 3.54, n = 298, p = 0.06). The proportion of high-educated parents was larger in the control group than in the intervention group (χ2 = 16.49, n = 297, p = 0.0003).
Table 1
Baseline demographic characteristics and divorce related background characteristics: number and (percentages) of the total sample, and of intervention and control groups separately
Characteristics
Total
I vs. C
Intervention (I)
Control (C)
Demographics characteristics
Na
%
p
Na
%
Na
%
Sex child
0.95
Boy
138
(46.6)
87
(47.0)
51
(45.9)
Girl
158
(53.4)
98
(53.0)
60
(54.1)
Age child M (SD)
8.9
(3.8)
0.05
8.6
(3.6)
9.5
(4.1)
Sex participant
0.11
Male
99
(33.2)
69
(36.9)
30
(27.0)
Female
199
(66.8)
118
(63.1)
81
(73.0)
Age participant M (SD)
40.5
(6.1)
0.003**
39.7
(5.7)
42.7
(6.7)
Country of birth
0.06
Netherlands
255
(85.6)
154
(82.4)
101
(91.0)
Other
43
(14.4)
33
(17.6)
10
(9.0)
Education
<0.001**
Low
29
(9.8)
22
(11.8)
7
(6.3)
Medium
96
(32.3)
73
(39.2)
23
(20.7)
High
172
(57.9)
91
(48.9)
81
(73.0)
Divorce related variables
Median
IQR
p
Median
IQR
Median
IQR
Time since divorce
14
(29.0)
0.68
16.5
(27.3)
10
(30.8)
in months
Range
1–132
1–132
1–123
Na
%
p
Na
%
Na
%
Time since divorce (2 years)
0.48
≤24 months
178
(66.9)
110
(65.1)
68
(70.1)
>24 months
88
(33.1)
59
(34.9)
29
(29.9)
Divorce initiative
0.18
Joint
53
(17.8)
39
(21.0)
14
(12.6)
Participant
141
(47.5)
86
(46.2)
55
(49.5)
Ex-partner
103
(34.7)
61
(32.8)
42
(37.8)
Child custody
0.84
Joint
245
(82.8)
152
(82.2)
93
(83.8)
Other
51
(17.2)
33
(17.8)
18
(16.2)
Official parenting plan
0.14
Yes, definite or temporary
140
(47.2)
94
(50.8)
46
(41.4)
Yes, co-parenting
95
(32.1)
61
(33.0)
34
(30.6)
Not (yet) applicable
61
(20.6)
30
(16.2)
31
(27.9)
Child support agreement
0.41
Yes, definite or temporary
175
(59.7)
113
(62.1)
62
(55.9)
No
53
(18.1)
33
(18.1)
20
(18.0)
Not (yet) applicable
65
(22.2)
36
(19.8)
29
(26.1)
Parent alimony agreement
0.95
Yes, definite or temporary
71
(24.1)
46
(25.1)
25
(22.5)
No
91
(31.0)
55
(30.1)
36
(32.4)
Not (yet) applicable
132
(44.9)
82
(44.8)
50
(45.0)
Has new partner
0.55
Yes
88
(29.5)
58
(31.0)
30
(27.0)
No
210
(70.5)
129
(69.0)
81
(73.0)
Ex-partner has new partner
0.03*
Yes
111
(37.5)
63
(34.1)
48
(43.2)
No
102
(34.5)
60
(32.4)
42
(37.8)
Don’t know
83
(28.0)
62
(33.5)
21
(18.9)
Stepchildren
0.83
Yes
28
(9.9)
18
(10.5)
10
(9.0)
No
254
(90.1)
153
(89.5)
101
(91.0)
Student t-tests (age, time since divorce) and Chi-square tests (other variables) were used to evaluate between group differences
IQR interquartile range
*p < 0.05
**p < 0.01, two-tailed significance
aTotals per variable may slightly deviate from total N due to missings
Table 1 also summarizes divorce related variables as measured at the pre-test. This shows that time since divorce for parents who sought for information and support and consequently joined Divorce ATLAS in this study (I or C) ranged to over 10 years post-divorce. The majority of parents (66.9%) participated within the first two years after divorce. There were no differences in time since divorce between the I and C group. Only in a minority of families (17.8%), the divorce was a joint initiative. Almost half of participating parents initiated the divorce. In most families, parents shared custody with the other parent (82.8%) and had agreements in the form of definite or temporary (co-)parenting plan (79.4%). Out of the participants’ families 59.7% had child support, and 24.1% had parent alimony agreements. Less than a third (29.5%) of the participating parents had a new partner, compared with 37.6% of their ex-partners. Regarding the latter, shares were higher in the control group (43.2%) than in the intervention group (34.1%; χ2 = 7.42, n = 296, p = 0.02). One in ten participants had one or more stepchildren of a new partner.
Effects of Divorce ATLAS
Table 2 presents our comparisons of the differences in outcomes between the intervention and control group. In terms of wellbeing, no intervention effects were found on well-being of parents (Hedges’ g = 0.11), or children (Hedges’ g = 0.18). Nor did the intervention impact on child prosocial behavior (Hedges’ g = 0.05). In regard to feelings of competence, no between-group differences were found in increases of competence (Hedges’ g = 0.03). However, parents’ sense of parenting incompetence had increased after participation (Hedges’ g = 0.19), see Fig. 3. This difference between intervention and control group did not remain significant after adjustment for parental age and education (Hedges’ g in the adjusted model I = 0.16) and for time since divorce (Hedges’g in the adjusted model II = 0.16). As can also be seen in Table 2, our analyses pointed to a small increase in perceived social support (daily support, Hedges’ g = 0.23; social support with problems, Hedges’ g = 0.18; see also Fig. 3), with similar (although non-significant) results in the adjusted models (daily support, Hedges’ gI = 0.19 and gII = 0.18; social support with problems, Hedges’ gI = 0.19 and gII = 0.19). Supplementary subset analysis on participants who were within the first two years after divorce, revealed small, non-significant differences in the same directions; with somewhat larger differences for feelings of incompetence (Hedges’g = 0.24) and perceived social support with problems (Hedges’g = 0.26; see Appendix 2). Finally, concerning parental evaluation of knowledge and trigger to act, according to the majority of parents (60.7 and 65.3%, respectively), Divorce ATLAS had increased their knowledge and triggered them to act accordingly (see Table 2).
Table 2
Comparison of differences in outcomes between intervention (I, n = 187) and control group (C, n = 111)
Pre
Post
ΔPre-post
Difference I vs C
Adjusted difference-Ia I vs C
Adjusted difference-IIb I vs C
Outcome
Group
M
(SD)
M
(SD)
M
(SD)
B (SE)
g
B (SE)
g
B (SE)
g
Well-being of parent
I
7.15
(1.44)
7.16
(1.63)
0.01
(1.41)
0.19 (0.19)
0.11
0.12 (0.19)
0.07
0.12 (0.19)
0.07
C
6.27
(1.74)
6.47
(1.59)
0.19
(1.64)
Well-being of child
I
6.94
(1.42)
7.23
(1.45)
0.29
(1.39)
0.26 (0.17)
0.18
0.33 (0.18)
0.22
0.33 (0.18)
0.22
C
6.66
(1.49)
6.81
(1.51)
0.15
(1.36)
Feelings of competence
Competence
I
22.86
(3.90)
23.90
(3.29)
1.03
(4.16)
0.09 (0.45)
0.03
0.12 (0.47)
0.04
0.13 (0.47)
0.04
C
22.83
(4.24)
23.80
(3.80)
0.97
(4.60)
Incompetence
I
14.22
(5.65)
14.95
(6.01)
0.72
(3.87)
1.14 (0.57)*
0.19
1.00 (0.59)
0.16
0.99 (0.59)
0.16
C
16.50
(5.63)
15.65
(6.09)
−0.85
(4.45)
Prosocial behavior of child
I
8.15
(1.96)
8.36
(1.91)
0.21
(1.43)
0.09 (0.19)
0.05
0.13 (0.20)
0.07
0.14 (0.20)
0.07
C
7.89
(2.18)
8.09
(2.07)
0.20
(1.62)
Perceived social support
Daily support
I
2.61
(0.99)
2.77
(1.00)
0.16
(0.93)
0.23 (0.11)*
0.23
0.19 (0.12)
0.19
0.18 (0.12)
0.18
C
2.55
(1.03)
2.51
(0.95)
−0.04
(0.95)
Support with problems
I
3.00
(0.91)
3.09
(0.88)
0.09
(0.89)
0.16 (0.10)
0.18
0.17 (0.10)
0.19
0.17 (0.10)
0.19
C
3.04
(0.91)
2.95
(0.86)
−0.09
(0.76)
M
(SD)
% >3
Knowledge
I
3.68
(1.05)
60.7%
Trigger to act
I
3.62
(1.13)
65.3%
Statistical significance of regression analyses: post-test outcomes were predicted out of intervention arm and outcomes at pre-test, with the exception of knowledge and trigger to act (post-test data in intervention group only)
*p < 0.05, two-tailed significance.
aAdjusted model I was adjusted for centered age, and centered squared age of participants, and educational level.
bAdjusted model II was adjusted for centered age, centered squared age, educational level of participants, and time since divorce, dichotomized >24 months
Fig. 3
Mean pre-test and post-test scores (standard deviations) in the intervention (I) and control (C) groups on well-being of parent and child, parent’s feelings of competence and incompetence, perceived daily support and support with problems, and prosocial behavior of the child (*p < 0.05, Two-tailed significance of non-adjusted between group differences)
×
Discussion
We assessed the effectiveness of the post-divorce preventive Divorce ATLAS group training for parents in the Dutch setting in a quasi-experimental study. We found small effects regarding perceived social support by parents, and no effects regarding parent and child wellbeing, parents’ feelings of competence, and child prosocial behavior. Divorce ATLAS contributed to participants’ knowledge and trigger to act accordingly. In contrast to our hypotheses, parents’ feelings of incompetence increased after participation.
The effects of the intervention on social support, knowledge and trigger to act may be explained using the model of intervention (Fig. 1), and accordingly Divorce ATLAS’ intervention aims and key elements. More specifically, we suggest that primarily the following two possible mechanisms add: (1) group support, and (2) psychoeducation. Regarding group support, Divorce ATLAS aims at strengthening parents’ social support, by offering a small peer-support group in which participants can benefit from group processes, like modeling, comparison, feedback, identification, and exchange with other divorced parents (Borek and Abraham, 2018). In addition, the balance between stressors and challenges on the one hand, and supportive factors and resilience on the other hand is one of the key themes of Divorce ATLAS. Feelings of stress resulting from divorce are normalized, and at the same time participants are stimulated to reach out for support where needed, to rebalance and cope with negative feelings. The positive change in perceived social support after participation in Divorce ATLAS may be due to these intervention elements. Based on the results, we cannot determine whether parents have actually sought and/or received more emotional support (i.e., quantity) or mainly perceived the support that they already received as more sufficient after clarification of their stressors and challenges (i.e., quality). Either way, Divorce ATLAS may thus contribute positively to participants’ stress-support balance, and thus to post-divorce adjustment.
Second regarding psychoeducation, Divorce ATLAS aims at increasing knowledge, awareness, and positive behavioral intentions, by offering parents information about coping, parenting and communication after divorce (Fig. 1). We suggest that particularly these intervention elements may have led to knowledge gains and behavioral intentions (i.e., trigger to act) of parents after participation. That is, providing information about the desired situation (i.e., subjective norms), health and parenting behaviors (e.g., bereavement, authoritative parenting, using I-messages), may have helped to improve the perception of the own behaviors and their importance among the informed persons. Increased knowledge (e.g., how does bereavement work and which behaviors matter) and interest may in turn have resulted from this information transfer and can in turn have contributed positively to awareness (e.g., awareness of what behavior the parent does or does not exhibit) and attitude. In this process, the Divorce ATLAS’ reflection and feedback exercises may have played an important role by increasing parents’ awareness of what behavior they do or do not exhibit, as having appropriate reflective skills may be necessary to gain awareness (Lane and Roberts, 2022). In line with this, the increased feelings of incompetence may indicate increased awareness of own parenting behaviors among parents in the intervention group compared to the control group. The conscious competency learning model (Broadwell, 1969; cited in Lane and Roberts 2022) suggests four stages in learning skills: ultimately conscious and unconscious competence develop from unconscious incompetence (ignorance) and next conscious competence (i.e., awareness). Thus, parents in the intervention group may have gained more consciousness of their incompetence. Parents’ recognition of their incompetence is hypothesized to pave the way for conscious learning of new skills. Increased awareness can sequentially have formed the basis for a positive behavioral intention. The enhancement of knowledge and behavioral intentions after participation as reported by parents in our study, can thus be interpreted as a result of this mechanism.
This study has as main strengths its quasi-experimental design embedded in routine care, employing a comparative pre-post-test design. Moreover, missing data were imputed. However, this study also has some limitations. First, its quasi-experimental design may have affected the comparability of the two groups, potentially affecting outcomes. That is, parents in the intervention group were 3 years younger and less educated than parents in the control group. This may be due to differences in recruitment for these two groups: parents either registered for a peer-support group (intervention group) in response to written or in-person communication by professionals in social work, preventive healthcare or youth social care, or registered for e-health training (control group) in response to public communication. Older, higher-educated parents may have found their ways to the latter e-health training more easily. However, we adjusted for this by adding pretest scores as covariates to the analyses. Also, our model adjusted for between-group differences (i.e., participants’ age and education) showed results with effect sizes being similar as in our non-adjusted model, indicating that these group-differences did not affect our findings in an important way. Second, in terms of study limitations, the use of self-report may have led to less reliable effect estimations, leading to some underestimation of real effects.
Our findings of positive effects of Divorce ATLAS on support, knowledge, and trigger to act accordingly suggest that this intervention deserves further and wider implementation. An implementation plan (in Dutch) is available to support implementation elsewhere. Implementation outside of the Netherlands would foremost require translation of program content and adaptation to the local (cultural) context (e.g., by filming new role play videos). Moreover, research is needed into longer term impact of Divorce ATLAS, (i.e.,) longer than 4 weeks, as parents probably need more time to actually act in accordance with their changed awareness, beliefs, and intentions.
In this study, the 2-sessions peer-support group program Divorce ATLAS was found promising and feasible to educate and inform parents after divorce. The intervention showed a positive impact on participants’ direct intervention targets, namely social support and knowledge. Our findings could guide people interested in providing educational preventive group support to parents post-divorce in various countries. As evidence-based preventive educational programs in the Netherlands are currently rather scarce, these promising results findings support a broader use of Divorce ATLAS in order to strengthen both parents’ and children’s post-divorce adaptation in the Netherlands. A combination of Divorce ATLAS for parents with the previously proven effective CODIP-NL for children (Klein Velderman et al., 2022) may contribute to post-divorce adaptation of full families. Finally, the Divorce ATLAS intervention and its effectiveness deserves a further evaluation in a larger scaled randomized trial with longer follow-up, and in other countries.
This research was supported by a grant from a major Dutch non-profit funding body, ZonMw (the Netherlands Organization for Health Research and Development: Grant No. 729310002). The authors were independent of the funders and the funders had no role in the project. This intervention study into Divorce ATLAS received the national ZonMw “Pearl Award” for strong innovative power (September 2021). We gratefully acknowledge Mrs. Wendy van Vliet, Minddistrict, Dr. J. Pedro-Carroll, and participating parents for their contribution to the development of the current version of Divorce ATLAS, as well as group leaders for leading the Divorce ATLAS groups in this study. We thank I. Wildeman, MSc, group leaders, and parents for their contribution to data collection.
Compliance with ethical standards
Ethical approval
This manuscript has not been and will not be published in this or a similar form (in print or electronically, including on a web site) and is neither accepted for publication elsewhere nor under consideration by another journal. This research was supported by a grant from ZonMw (the Netherlands Organization for Health Research and Development: Grant No. 729310002). The authors were independent of the funders and the funders had no role in the project. All procedures performed in this study involving human participants were in accordance with the ethical principles as published by the World Medical Association in the Declaration of Helsinki. The authors do not have any conflict of interest and all have contributed significantly to the manuscript. All authors have confirmed that they have read and approved the final version of the paper and that they believe that the paper represents honest work. They are able to verify the validity of the results reported.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Met BSL Psychologie Totaal blijf je als professional steeds op de hoogte van de nieuwste ontwikkelingen binnen jouw vak. Met het online abonnement heb je toegang tot een groot aantal boeken, protocollen, vaktijdschriften en e-learnings op het gebied van psychologie en psychiatrie. Zo kun je op je gemak en wanneer het jou het beste uitkomt verdiepen in jouw vakgebied.