Prevention of Divorce: Results of a 25-Year Follow-Up Study

There is a universal hope for security, appreciation, and tenderness in a solid partnership¹ [Buss, 2004]. Almost everyone experiences an intimate partnership during their lives, and in western industrialized nations, about 80–90% of those over 18 years marry at least once. In all surveys of life satisfaction, partnership and family are ranked as the most important factors contributing to well-being, followed by health, occupation, or income. Representative studies with samples of newly married couples show that 80–85% of respondents are very happy with their current relationship. Satisfaction with the relationship, however, declines continuously in the first 10 years of marriage [Halford et al., 2010].

In the Federal Republic of Germany, the divorce ratehas risen significantly since the 1960s. In 2015, out of 400,000 new marriages, 163,000 ended in divorce (41%), with most of the divorces taking place in the sixth year of marriage [Statistisches Bundesamt (Federal Statistical Office), 2016]. The case of Switzerland is similar, with a divorce rate of 41% reported for 2015 [Bundesamt für Statistik, BFS, 2016], while the rate in Austria in 2015 was 43% [Statistik Austria, 2016]. In the Federal Republic of Germany, half of the divorces affected minor children (2015: 132,000) [Statistisches Bundesamt, 2016].

Irrespective of the official divorce rate, the separation ratefor unmarried couples living together is estimated to be even higher [Kiernan, 2003]. According to a study by Ermisch and Francesconi [2000], in England only about 20% of couples were still together after 5 years of partnership, and after 10 years only about 10%. Based on the family survey of the Deutsches Jugendinstitut (DJI; German Youth Institute), Eckhard [2015] determined the separation and divorce rate for relationships of 6 years’ duration, based on the years in which the partners were born. This was 23% for those born in the 1950s, 31% for those born in the 1960s, and 46% for those born in the 1970s. The consequences for children of unmarried couples are troubling: Data on the breakup of families from the UK Millennium Cohort Study showed that 35% of unmarried parents had separated before their child’s fifth birthday, compared to 9% of married couples [Callan et al., 2006].

However, not only the high separation and divorce rates are alarming: Many couples (about 10–25% in Germany = 1.6–4 million partners [Job et al., 2014a]) live in stable partnerships with which they are dissatisfied. The reasons for staying together are many and varied, including personal or cultural attitudes to divorce; children; financial problems; and the lack of a new partner [Hahlweg et al., 2010].

Divorce is one of the most stressful events in a person’s life. It is especially important that the conflicts preceding a divorce or separation, and dissatisfaction with the relationship, have many negative consequences for those affected and for society at large [Hahlweg et al., 2010].

The empirical evidence on the short- and long-term consequences of partnership conflicts, separation, and divorce is extensive and well replicated internationally, so that associations with many mental and physical disorders can be repeatedly verified for the affected partners. Some examples are an increased likelihood of substance dependence; affective and anxiety disorders; less favorable outcome of somatic illnesses; poor treatment outcome; and elevated blood pressure and coronary heart disease in women [e.g., Kiecolt-Glaser and Newton, 2001; Orth-Gomer et al., 2000; Robles et al., 2014; Sbarra et al., 2011; Whismann, 1999].

Compared to children from intact families, children from conflict-ridden (divorced) families display a variety of behavioral problems, lower self-esteem, problems in peer relationships, and worse mental as well as physical health [Amato, 2010; Cummings and Davies, 2010; Lansford, 2009]. However, the effects are small, with a mean effect size of d= 0.14 [Amato and Keith, 1991]. These disorders can persist into adolescence and even into adulthood, however, leading to further problems, such as lower academic and vocational qualifications, more frequent delinquency, increased alcohol consumption, and an increased risk of divorcing later in life [e.g., Brown, 2010; Diekmann and Engelhardt, 1995; Hetherington and Elmore, 2004].

Aside from the consequences for those individuals affected, partner dissatisfaction, separation, and divorce are also associated with high economic costs. These include the costs of treatment, absence due to illness, loss of productivity, and the use of social or legal services, all of which must be borne by the partners and the state [Andreß, 2004; Caldwell et al., 2007].

About 1.6 single parents were living in Germany in 2014. Life after separation is considerably more expensive, since maintaining two households costs much more than one. The economic consequences of separation and divorce are different for men and women. If there are children to be provided for, the income requirements of the custodial parent – in 89% of cases, the mother – are higher. At the same time, childcare requirements often limit employment opportunities. Furthermore, there is a great risk that the custodial parent will not be paid half of the childcare costs by the other parent [Andreß, 2004]. The women’s income gap has to be filled by public transfers. In Germany, children whose parents are separated face a greater risk of living in poverty than their peers who live with both parents. About 2.3 million children are growing up in a so-called single-parent family and about 1 million are living on Hartz IV welfare benefits [Andresen and Galich, 2015]. Fully 21% of children are affected by poverty, which has long-term negative effects on academic success and health.

This situation is ample justification for comprehensive preventive measures to reduce marital, partnership, and pediatric mental disorders. Early prevention (when the couple is still happy or at an early stage of conflict) can effectively prevent these adverse developments. The research into risk and protection factors described above [Bodenmann, 2016; Hahlweg et al., 2010; Halford and Bodenmann, 2013; Heinrichs et al., 2008] showed that the objective of many preventive programs is to reduce the frequency of negative communication during partners’ quarrels and to elicit a more positive approach to facilitate problem solving [for a review of various programs, see Job et al., 2014a]. Since partnership conflicts are an important component of life with another person, the point is not to prevent them, but to improve the couple’s skills in dealing with future conflicts.

In the Federal Republic of Germany, EPL – Ein Partnerschaftliches Lernprogramm (A Partnership Learning Program) [Job et al., 2014b] is the program most commonly used; it was developed at the end of the 1980s with state and church funding as an adaptation of the American Premarital Relationship Enhancement Program, PREP [Markman et al., 1993]. During EPL training, couples learn to address their feelings openly, express their expectations in a comprehensible way, and listen to and understand each other using specified speaker and listener rules. Another program is Paarlife[formerly Freiburger Stresspräventions­training für Paare (Freiburg Stress Prevention Training for Couples);Bodenmann, 2016], which, in addition to promoting partnership communication and problem solving, aims to improve the partners’ dyadic coping.

These programs can be implemented both in a group and with a couple on their own, in six 2- to 2.5-h weekly sessions or on a weekend (Saturday to Sunday afternoon). Group sizes vary from 3 to 8 couples, with a coach for every 2 couples. The couples work on the theoretical course material in the large group and then practice the skills individually, in separate rooms with a coach.

Couples generally report high satisfaction with the program after their participation. Hawkins et al. [2008] examined in a meta-analysis 117 studies on the effectiveness of prevention programs on partnership quality and the couple’s communication. For randomized controlled trials, effect sizes were between d =0.30 and 0.44. Follow-up studies confirmed the stability of the findings, obtaining pre- to 1-year follow-up effect sizes between d =0.40 and 0.45 [Blanchard et al., 2009]. Fawcett et al. [2010] summarized N= 47 studies that only looked at marriage preparation programs. Effect sizes for partnership satisfaction were d= 0.22 and for communication behavior, obtained through observational studies, d= 0.45.

Long-term effectiveness is of especially great interest in prevention programs. Unfortunately, here too there are only a few studies internationally, as frequently only 1-year catamneses have been performed [Blanchard et al., 2009]; in the United States, there was a 5-year follow-up by the Markman group [1993]. So far, only in Germany have longer catamneses of up to 11 years been performed. Thurmaier et al. [1999], in a 5-year follow-up study of couples who had participated in the EPL I marriage preparation course, showed that the EPL couples exhibited more positive and less negative communication in all subsequent surveys than did the comparison couples, who were given either another intervention that took a similar amount of time or no intervention. The EPL I couples also had a lower divorce rate (4 vs. 24%). The number of couples who had children after 5 years was significantly different: 81% of the EPL couples compared to 62% of the control group (CG) couples (p= 0.03).

In another study of indicated prevention, N = 67 couples with a partnership of at least 3 years, of whom 70% were dissatisfied with their relationship, were randomly assigned to participate in the EPL or a CG without intervention [EPL II; Kaiser et al., 1998]. The average age of the men was 40 years and that of the women 37 years. The couples had lived together for an average of 10 years. An 11-year follow-up showed that the EPL II couples had a divorce rate of 26% as compared to 56% in the comparison group who had declined to participate in the EPL for various reasons. Of all the EPL and control couples who were still living together after 11 years, 75% were happy with their partnership [Hahlweg and Richter, 2010]. In another, uncontrolled study, N= 62 couples were evaluated to see whether the efficacy of the EPL II could be enhanced by completing 2 individual refresher sessions, after 1 and after 3 months [EPL II-B; Braukhaus et al., 2001]. The results of the EPL II study were replicated; after 11 years, a divorce rate of 20% was found [Hahlweg and Richter, 2010]. In all of the studies, the couples reported a high level of satisfaction with the EPL program. The high rate of acceptance is also shown by the fact that the dropout rates in the courses were on average well below 5%. This is also the case when the training extends over several weekly sessions, up to 8 weeks.

The overall objective of the present study was to examine the divorce rate of couples who were in the EPL I study 25 later. The hypotheses were: (a) EPL I couples have a significantly lower divorce rate than CG couples who received either no marriage preparation or a different Catholic marriage preparation program; (b) after 25 years, partnership satisfaction of the EPL couples is significantly higher than that of the CG couples. We also ascertained how often the partners still used the EPL speaker and listener rules that they had been taught and whether they would recommend the EPL to other couples.

From 1988 to 1994, the 5-year long-term study “Marriage Preparation – A Partnership Learning Program” (EPL I) was conducted at the Institut für Forschung und Ausbildung in Kommunikationstherapie e.V. (Institute for Research and Training in Communication Therapy e.V.) in Munich [Thurmaier et al., 1999]. The EPL courses were offered in the context of the Roman Catholic Church’s marriage preparation in the Archdiocese of Munich-­Freising and were conducted in groups of 3–4 couples by 2 specially trained instructors. The intervention was either on a weekend (Friday evening to Sunday afternoon) or was spread over six 2.5-h weekly sessions. The comparison group comprised couples who either participated in another marriage preparation class that took a similar amount of time (“Bridal Couple Weekend”) or who received no preparatory instruction and were recruited via a newspaper advertisement. Originally, a randomized assignment to the experimental group and the CG was planned, but this did not work out, because the courses were advertised in advance over a long period of time and the couples could sign up for the individual classes. The study thus has a quasi-experimental character. There were no significant differences between the “Bridal Couples” and the couples recruited from advertisements with respect to sociodemographic and partnership variables and questionnaire and communication variables, so the 2 groups were combined and designated as the CG for the purpose of statistical analysis. Measurements were made both in the short term (before and after the training) and in the long term (1.5, 3, and 5 years after the postmeasurement) [for a more detailed description of the training program, see Job et al., 2014b].

A total of 109 heterosexual couples participated in the study, of whom 77 were assigned to the EPL I group and 32 to the CG. In international prevention studies, notably the PREP study of Markman et al. [1993], only couples with satisfactory partnership quality were accepted. This was defined by a mean adjustment score >100 in the Marital Adjustment Test [MAT; Locke and Wallace, 1959; see below], so that the EPL I sample size for premeasurement was reduced by 13 to 64 couples (see flowchart; Fig. 1). The women were on average 25.8 years old (SD = 3.8), the men about 3 years older (mean = 28.9 years, SD = 4.7). One quarter of the partners (25%) had a university degree, 12% a Fachhochschulabschluss (technical college diploma), 28% an Abitur (high school diploma with qualification for university entrance), 8% a Fachhochschulreife (advanced technical certificate), 18% a mittlere Reife (secondary school-leaving certificate), and 9% a Hauptschulabschluss (certificate of secondary education). The couples had been together for an average of about 3.5 years (in months: mean = 43.1, SD = 26.6, range 3 months to 11 years); 83% of the participants were Catholic. The EPL I group and CG did not differ with respect to socioeconomic variables (age, religion, and religiosity) or partnership variables (partnership duration, partnership satisfaction MAT, and number of problem areas at premeasurement). The findings for recruitment and communication behavior and overall satisfaction at the 5-year follow-up can be found in Hahlweg et al. [1998] and Thurmaier et al. [1999].

A complex search process was used to obtain the highest possible rate of participation. The participants were contacted by mail and asked to participate. 83% of the data collection was by means of questionnaires. After 5 years, 57 couples (EPL I, n= 44; CG, n= 13) participated in the follow-up. After 25 years, N= 41 couples could still be motivated to participate (EPL I, n= 31, of whom 2 partners had died; CG, n =10) (see Fig. 1). This corresponds to a retention rate of 62%, which is considered good.

N= 68 partners were still married to their first partner (87%), 3 couples (8%, 2 EPL, 1 CG) were still married but had separated, and 2 partners had died. Of the 34 couples still living together, 5 had 1 child (13%), 9 had 2 children (23%), 16 had 3 children (41%), 3 had 4 children (8%), and 1 couple had 5 children (4%). This corresponds to an average of 2.6 children (SD = 0.95), which is significantly higher than the German birth rate, which was 1.4 in 2014. There were no significant differences between the EPL I and CG couples (t(66) = 0.041, p= 0.686). During the 25-year period, 14 partners (7%) had visited a marriage counseling center (2 of them were from the CG).

Satisfaction with Marriage/Partnership (MAT).The German version of the Marital Adjustment Test (MAT) [Locke and Wallace, 1959] was prepared by Scholz [1978]. Up to the end of the 1970s, the MAT was most frequently used to determine marriage satisfaction and for evaluation of therapy, especially in the English-speaking world. It consists of 15 items, such as “Do you ever wish you had not married?,” “How much agreement or disagreement is there between you regarding: finance, freedom, demonstrations of affection, friends, sexuality, manners, philosophy of life, and relationship with relatives?” (with a range of 6 possible responses from “always agree” to “never agree”). Cronbach’s α is r= 0.90. The total score can theoretically vary between 2 and 158, with a score ≥100 considered the limit for good marital quality.

The program is divided into 6 units, each building on the one before [see Job et al., 2014b] and is usually conducted in a group of 4 couples with 2 coaches. The couples work on the theoretical course material, such as communication rules and theoretical understanding of problem-solving processes, in the large group and then practice the learned skills individually, in separate rooms with a coach. The partners are viewed as experts in their own relationship, while the coach provides them with a scientifically substantiated tool for conflict resolution. In order to encourage the couple to take responsibility for themselves, the coach takes a back seat during the course. The systematic structure and intensity of the training are intended to enable the partners to achieve the greatest and longest-lasting learning outcome in a relatively short time. The 6 EPL components are described in greater detail below.

Component 1: Using a short film clip of a partnership conflict, all the participating couples work on the 5 communication rules for the speaker (e.g., “Using the word I,” “Staying on topic”) and for the listener (e.g., “Receptive listening,” “Summarizing”). This is followed by the first exercise for couples where they practice the expression of positive feelings and wishes, following the rules that were previously learned. Component 2:Then, exercises on how to express negative feelings that can occur in conflicts follow. These are first done in role playing and afterwards with a topic chosen by the partners. Component 3:A 5-step scheme is introduced to promote the partners’ problem-solving skills; they first discuss a problem of their own, then compile possible solutions and select the one that is best for them. Component 4:The partners are supervised when using a questionnaire to discuss their own values and expectations of the partnership. The objective here is to describe complex terms (e.g., trust, friendship) using concrete examples relevant to the relationship (e.g., “I felt trust when you hugged me yesterday, after ...”). Component 5:The partners are encouraged to talk about their shared sexuality. They are instructed to arrange and exchange little cards on which various terms about sexuality are printed (e.g., erotic, fantasy, orgasm). The goal is to connect the concepts with concrete situations and experiences. Component 6:In the last session, depending on the needs of the couple or the setting in which the EPL is offered, different topics can be discussed (for example, in a church setting, the meaning of Christian marriage for one’s own relationship, or parenthood with expectant parents).

The training of the EPL I coach included didactic training (e.g., reading scientific background literature) as well as intensive training in conversational role playing interventions. Two 2.5-day training workshops were held, each with 1 instructor for 4 trainees. On the first weekend, trainees paired up to form different couples for role playing; each time one trainee, continuously guided by an instructor, practiced conversations that were on a steadily increasing level of difficulty, and the other trainee alternately paused and gave feedback as an observer. They also studied the detailed EPL I manual together. During the second training weekend, the participants continued to practice the EPL coaching skills in supervised role playing, while also learning to present the program appropriately to a group of make-believe couples. Supervision was provided throughout the course of the study.

For n= 46 of the EPL I individual partners who remained in the 25-year survey, we were able to evaluate the extent to which they still applied the EPL rules for speakers and listeners that had been taught them during their training. In the questionnaire, the rules were summarized and the following possible responses were offered: “never/very rarely – rarely – sometimes – often – very often.” For the speaker rules (e.g., talking about one’s own feelings and desires), only 11% stated that they never or rarely did this, 26% answered “sometimes,” 39% “often,” and 24% “very often.” Similar percentages were found in the use of the listener rules (e.g., open-ended inquiries): “never/rarely” 9%, “sometimes” 33%, “often” 39%, and “very often” 20%. All respondents said they would recommend the EPL (98%) to others, except for 1 person who preferred another program. Overall, couples seem to incorporate the rules into their everyday communications.

Couples who no longer attended the follow-up appointments and about whom no information was available regarding separation or divorce were defined as dropouts. The rate varied considerably over time (see Fig. 1). At the 25-year follow-up, there was no significant difference between the dropout rates of EPL I couples (n= 24; 38%) and CG couples (n= 13; 41%). A comparison of prevalues between those still married, separated, and divorced at the 25-year follow-up and the dropouts revealed nonsignificant results with respect to socioeconomic variables (age, education, religiosity) and relationship variables (MAT partner satisfaction, problem frequency).

Of the original 96 couples who had been included in the sample 25 years before, 57 (44 EPL I and 13 CG couples) came back for the 5-year survey (see Fig. 1). By this time, 25% of the original group could no longer be reached (dropouts), 11% had separated before marriage (EPL: 10%, CG: 14%), and 10% had already been divorced (EPL: 4%, CG: 24%; adjusted for dropouts: EPL 4%, CG 24%; χ² (1, N= 64) = 7.29, p < 0.01, Fisher’s test p= 0.016; Fig. 2).

After 25 years, 44 of the 57 couples (33 EPL and 11 CG couples) who had remained for the 5-year survey could still be reached. Of these, 39 were still living together (29 EPL and 10 CG couples). 5% of the EPL couples had separated shortly before the 25-year survey but had not (yet) initiated divorce proceedings. For the CG, this was also true of 5%. Unfortunately, in 2 (3%) of the EPL couples, 1 partner had died. No further divorces were reported since the 5-year survey, so that the significant difference in divorce rates remained between the EPL and CG couples (adjusted for dropouts: EPL: 5%, CG: 26%, χ² (1, N= 46) = 5.66, p < 0.05, Fisher’s test p= 0.029; Fig. 2). Over the 25 years, there were 3 divorces among those who had participated in the “Bridal Couple Weekend” (nat pre = 18) and 2 divorces among the couples who had no intervention (nat pre = 14). This means that CG couples who had an intervention and those who did not also did not differ significantly in their divorce rate.

Only partners who had also participated in the 25-year survey were evaluated (N= 64; EPL I: n= 46; CG: n= 18). Marital quality throughout the study for both the EPL I and CG partners was, on average, at least 1 standard deviation above the MAT value of 100, which was considered the cutoff score for satisfactory marital quality (Fig. 3). There were no significant gender and group effects (p >0.393). After 25 years, 83% of the EPL I partners were still happy; for the CG partners, the rate was 72%. These differences were not significant (p= 0.353).

In the Federal Republic of Germany as well as in Austria and Switzerland, divorce rateshave continuously risen since the 1960s and in 2015 were at approximately 40%; over half of the divorces affected minor children. The separation rateofunmarried couples is much higher. In light of these high separation and divorce rates and as partnership duration increases, an overall declining satisfaction with the relationship, and the many associated negative consequences for those affected [Robles et al., 2014] and for their children [Cummings and Davies, 2010], the prevention of dysfunctional partnerships should become increasingly important for public health.

Current research indicates that active, action-oriented training of partnership skills (communication, problem solving, dyadic coping) is extremely important to prevent partnership problems [Hahlweg et al., 2010]. Although in recent decades, numerous prevention programs have been developed worldwide and their evidence has been validated, unfortunately, there is a lack of long-term studies that are so crucial for preventive purposes and which go beyond a 5-year follow-up period [Blanchard et al., 2009].

The aim of the present study was, first of all, to investigate the 25-year effectiveness of EPL I (A Learning Program for Premarital Couples) in the prevention of divorce. The second goal was to determine the effects of the training on long-term partnership satisfaction and the lasting use of EPL communication rules by the partners. The basis for the results reported here is a quasi-experimental study, begun in 1988, for which 96 couples were recruited. After 25 years, N= 41 couples could still be motivated to participate. The dropout rate was 38% (corresponding to an annual dropout rate of 1.5%), which is considered good. According to Wölke and Söhne [1997], long-term studies result in up to 10% annual dropouts. These couples seem to have no effect on the results, as there was no significant difference in the dropout rate between EPL I and CG couples. The results confirmed the hypothesis: After 25 years, the EPL I couples had a significantly lower divorce rate (5%) than the CG couples (26%). Otherwise, the divorce rate of the CG corresponds approximately to the 30% rate reported by the Bavarian State Statistics Office in 2015 for couples who had gotten married in Bavaria in 1989.

These results replicate the findings of the randomized controlled trial [Kaiser et al., 1998], which also illustrated the effectiveness of the EPL as an indicated prevention program (EPL II). In another study, 62 couples were evaluated to see whether the effectiveness of the EPL II could be enhanced by completion of 2 individual refresher sessions after 1 and after 3 months [EPL II-B; Braukhaus et al., 2001]. A follow-up 11 years later [Hahlweg and Richter, 2010] showed that only 28% of the EPL II couples were divorced compared to 53% of the control couples who had not participated in the EPL. The EPL II-B study replicated these findings: The divorce rate was 20%.

Overall, the results of these 3 studies are impressive in demonstrating the effectiveness of the EPL training program in reducing rates of separation and divorce. The studies were conducted in different locations (Munich [M], Braunschweig [BS], and Kiel [K]) at different times (M: 1980s; BS, K: 1990s) by different research groups and different coaches with couples who were quite different in the variable “religion” (M: 83% Catholic; BS, K: 11% Catholic, 57% Protestant, 32% no religious affiliation). Both for marriage preparation (EPL I) with young, happy couples (universal prevention) and also with older couples who had been married for an average of 10 years and who were mainly dissatisfied with their relationships (indicated prevention), clear differences were shown: a reduction by about 50% in separation and divorce rates compared to the control couples. That is, the marriage survival of the EPL couples is significantly better than that of the CG couples. Satisfaction with the EPL is very high and, depending on the study, has a recommendation rate between 89 and 98%. There are similarly positive findings for the EPL-based programs KEK (Konstruktive Ehe und Kommunikation [Constructive Marriage and Communication], for couples with a longer relationship) and KOMKOM (Kommunikationskompetenz [Communication Competence] – training in couples counseling for distressed couples) [Job et al., 2016].

For the nondivorced couples, partner satisfaction remains high in both groups: Rates of 72–83% were found across the different studies. However, the hypothesis that, after 25 years, the EPL I couples’ partnership satisfaction would be significantly higher than that of the CG couples could not be confirmed, since no significant differences were found. It should be noted that the divorce rates were considerably higher in the CG couples than in the EPL couples, so that after 11 or 25 years, many dissatisfied CG couples had already been eliminated.

One objective of the EPL preparatory courses also seems to have been fulfilled: participating in the EPL to see whether the relationship is really “the right one” and whether it would be better not to marry. In the EPL I, a number of couples explicitly had this goal and split up soon after the course ended.

It must be made clear that the EPL is not designed to prevent separation and divorce “at any price.” Separations and divorces can liberate people from an unhappy, conflict-ridden, and potentially even humiliating or violent partnership and give each partner the opportunity to further develop themselves and to enter into new, more harmonious and fulfilling relationships. Above all, it is important that this separation will not be at the expense of the children and not lead to ugly divorce wars. Today, the ideal of “living separately and raising the children together” plays a major role for parents and children, in social legislation and – very specifically – in family courts.

At a conference organized by the Bundesministerium für Familie, Senioren, Frauen und Jugend (BMFSFJ; Federal Ministry for Family Affairs, Senior Citizens, Women and Youth), the presiding judge at the Berlin Appeals Court, Prof. Dr. Rüdiger Ernst, declared: “Joint upbringing of the child does not fail, or not primarily so, because of differing opinions about factual matters (Which school should the child go to?) ... It is rather the parents’ inexperience and ineptitude that are an obstacle, after the partnership breaks up, to talking to each other ... in a quasi “businesslike” way(Which words do I choose when I phone the other parent = ex-partner? When I send a text?) ... Based on my experience and judgment, raising children ‘together separately’ in many cases (also) depends, quite banally, on these language skills” [Ernst, July 11, 2017].

Another explicit area for application of the EPL (or other evidence-based training) could therefore be teaching the parents “language skills,” e.g., in combination with the group program “Kinder im Blick” (“Focus on Children”) [Walper and Krey, 2013]. It would be essential, of course, not to wait until the child “falls into a well,” but rather to attend such training sessions as early as possible in the relationship, either before the marriage/cohabitation, during pregnancy, after the birth of the child, or when the child moves out of the family home. If it turns out after the training that a separation is desired, participation in the course should at least convey the basics of common “language skills,” so that crises can be dealt with in a more suitable and solution-oriented way. Courses for those already divorced/separated could be useful in a modified form, but these still need to be developed and evaluated.

The strengths of the study (e.g., longest follow-up in international research, good retention rate, first-ever assessment of the long-term use of therapeutic rules, high long-term participant satisfaction with very high rate of recommendations) are constrained by some factors that limit the validity of the results. (1) This is not a randomized CG study but rather a general framework for a quasi-experimental investigation, as described in the Methods section. (2) Two-thirds of the couples had a high level of education, so the results cannot be generalized to the population at large. Studies of couples with a lower level of education are urgently needed. (3) The sample size, especially the comparison group, did not permit any more differentiated statistical analyses. (4) The data about quality of partnership is based on self-assessments, as it was not possible for financial reasons to make behavioral observations after 25 years.

This study shows for the first time internationally that short cognitive-behavioral prevention programs based on active training can achieve very long-term sustainable effects on partner stability. Such interventions should therefore play a major role in public health strategies. This could also reduce the high child poverty rate of 21% – at least for children who live with single mothers whose separation or divorce might have been prevented. There is an initial study by Birch et al. [2004] to that effect. In a quasi-experimental study, the authors compared 122 counties in the United States in which Community Marital Initiatives (CMIs) had been introduced, primarily through religious institutions, with control counties matched according to their divorce rates before introduction of the CMIs. They were able to show a significant decline in divorce rates in the CMI counties: Over a 7-year period, around 700,000 divorces would have been expected in the control counties, but there were 30,000 fewer divorces in the CMI counties. This corresponds to an annual reduction rate of 2%.

It seems illusory, however, to rely only on personnel-intensive training to reduce the rate of separation since, unfortunately so, few couples participate. More than 1,800 EPL coaches have been trained since 1988 by the Institut für Kommunikationstherapie (Institute for Communication Therapy) in Munich (www.institutkom.de). Of these, about 200 EPL coaches work outside Germany in 13 other countries (mainly Austria and Switzerland). The EPL is thus the most widely used and scientifically best-studied prevention training for couples in German-speaking countries. Over 25 years, a total of about 30,000 couples have participated in the EPL, but with approximately 400,000 marriages a year, this is an absolute “drop in the bucket.” This very low rate of participation will not be able to significantly change the rate of separation and divorce and the prevalence of chronic conflicts between partners.

The couples themselves are one of the reasons for the low utilization: In the first years of the relationship, satisfied couples tend to have an unrealistically positive view of the future of their relationship. For example, the majority of happily married couples, despite all the information available about high divorce rates, considered the likelihood of their own marriage ending in divorce as zero [Fowers et al., 1996].

In order, however, to successfully counteract the full impact of partnership dissatisfaction and conflict-laden separations and divorces in the long term, further options and interventions are needed. For this reason, Job et al. [2014a] propose a multi-level model for couple interventions, providing a wide range of interventions so that each couple can receive as much help as needed. In addition to competence-oriented couple training with close supervision, this model provides the following additional interventions:

We make available here scientific information and incentives for couples who want to improve their relationship. Various cost-effective media technologies are used to reach couples who do not want direct contact with professionals. This includes self-help materials such as books [e.g., Bodenmann, 2007; Engl and Thurmaier, 2012a, 2012b; Schindler et al., 2017] and DVDs. Examples of DVDs for people with existing partnership in German-speaking countries include Glücklich zu zweit trotz Alltagsstress (Happy Despite Everyday Stress)[Bodenmann et al., 2008] and the series Gelungene Kommunikation … damit die Liebe bleibt (Successful Communication ... to Keep Love Alive)for young couples, parents, and retired (and not so retiring) couples (Engl and Thurmaier, 2007, 2010, 2012). This series was funded by the Bayerisches Sozialministerium (Bavarian Ministry of Social Affairs) and costs EUR 10 per DVD, with an extensive accompanying brochure (available from www.institutkom.de). In addition to the use of the communication rules in the EPL program, the DVD for young couples will be given away to all Bavarian newlyweds during the civil marriage ceremony.

The so-called Internet-based self-help programs, which are used by the client alone on a PC (or notebook/tablet/smartphone) [Schindler et al., 2016] take a considerably more complex approach. Depending on the setting, such programs can be used with very different degrees of contact between the person seeking help and the professional [Berger, 2015]. They range from “unguided” (used on one’s own) to “guided” (sporadic contact with the practitioner) to the so-called “blended format” (a combination of face-to-face and Internet-based approaches).

A similar American program is available internationally for partnership problems and relationship competence (ourrelationship.com [OR]); it was developed by Doss et al. [2016], based on the Integrative Behavioral Couple Therapy (IBCT) approach [Christensen et al., 2010] and funded by the National Institutes of Health. The guided OR program is designed as a 7- to 8-week intervention, requiring a total of 7 to 8 h and the participation of both partners. An initial effectiveness study showed significant improvement in both the relationship quality and personal well-being of the individual partners [Doss et al., 2016].

In Germany, the program “PaarBalance” (“CoupleBalance”), developed by Schindler, Gastner, and Metz [2016], an interactive online program for greater satisfaction in the couple relationship, has been available since 2016 (https://www.paarbalance.de/). During its development, special emphasis was placed on 2 aspects:

(1) It should be possible for the web-based program to be used both by one partner aloneand by both partners at once, because if the simultaneous use by both partners is required, many couples will not even be reached, because only a small proportion of affected partners is open to outside support at the same time. If one partner gets help and begins to constructively change his or her personal role in the relationship, the principle of reciprocity says that there is a high probability that the other partner will follow suit. The Internet-based approach also increases the chance that the second partner will start the program later on and make his or her own contribution to improving the quality of the relationship. (2) The program should be appropriate in practice for both guided use (accompanying treatment) and unguided use (as a self-help program). Implementation without personal contact with a counselor has proven to be effective for prevention and early intervention [Berger, 2015].

The interactive online program includes: (a) a comprehensive initial and follow-up diagnosis with detailed strength/weakness analysis of the relationship, (b) 18 interactive sessions (each with a coaching video, training, and homework), (c) motivational thought-provoking points for preparation and reworking of the individual topics (automated information), (d) the possibility of printing out the relationship profile as well as important results or sharing them as a PDF by email with one’s partner or consultant by e-mail, and (e) recommendations for in-depth reading (book suggestions), etc. An evaluation study is expected to start in 2020. A 3-month subscription costs EUR 129.

The Internet and smartphones are changing society’s communication culture at lightning speed. Younger people especially spend a great deal of time looking at the display and they adjust their means of communication to the technical circumstances. The smartphone has virtually become a life companion. In order to convey the topic of “successful partnership” to a wide audience today, one has to consider changed “user” habits. That is why the Münchner Institut für Kommunikationstherapie e.V. (Munich Institute for Communication Therapy e.V.) is developing, with appropriately specialized companies, a modern app (“Paaradies”) and an accompanying website. The project is funded by the Bayerisches ­Staatsministerium für Arbeit und Soziales, Familie und Integration (Bavarian Ministry of Labor and Social Affairs, Family and Integration). The app is expected to be available in 2020 from the Apple and Google stores for free download. Its focus will be interaction with one another that is appropriate for a partnership. With numerous functions, the app is intended to encourage the partners to regularly invest time in worthwhile conversations, to value each other, and to record special moments of their relationship. Those interested can obtain additional, more in-depth information on the accompanying website. Focal points here are information worth knowing about couple communication and interesting questionnaires.

We conclude with a quote from Loewit from 1991: “Hardly any occupation is still possible today without training; only for the ‘occupations’ of a spouse and a parent is there still not even the awareness that appropriate preparation is necessary” (p. 214).

Translated by Susan Welsh

The study was approved by the Ethics Committee of the Technical University of Braunschweig and is in accordance with the 1964 Declaration of Helsinki. All participants gave their informed consent prior to their inclusion in the study.

On behalf of all authors, the corresponding author states that there is no conflict of interest.