Relationship education and counselling

Recent research findings

Content type
Policy and practice paper

July 2015


Cathryn Hunter, Joanne Commerford

Download Policy and practice paper


This paper reviews the research on relationship education and relationship counselling. The paper investigates the effectiveness of these strategies in working with couples who are at varying stages of their relationship - from highly satisfied at the beginning of their relationship, to highly distressed and considering separation. The paper addresses some of the complexities and issues surrounding how and why these strategies work in order to assist practitioners in engaging more effectively with couples and families.

Key messages

  • Relationship education programs have been found to produce significant moderate improvements in couple communication and relationship satisfaction in the short term. There is a lack of research demonstrating the effectiveness of relationship education in the longer term.

  • There is an inconsistent evidence base, with a lack of long-term follow up, supporting the use of relationship education as a universal prevention program. Relationship education may work best as a selective intervention aimed at couples at high-risk of relationship deterioration, although this is still a growing area of research with some inconsistent findings.

  • Many groups, such as cohabiting couples, are under-served by relationship education programs. Incorporating relationship education in to other programs, such as those run by nurses and midwives at the transition to parenthood, offers a potential opportunity to reach such under-serviced couples.

  • The effectiveness of relationship education programs may be influenced by the individual characteristics of the couples undertaking the program. Some couples are at higher or lower risk for relationship distress, and the couples’ experiences of these risk factors may affect the likelihood of a successful outcome.

  • Several approaches to relationship counselling have been found to be moderately effective in reducing relationship distress or increasing relationship satisfaction compared to no treatment. However, there has been limited new research in the past decade. There are a number of widely practiced approaches to relationship counselling that still remain largely untested.



The ways in which intimate couple relationships1 are entered in to and sustained have altered significantly over the last few decades (Moloney & Weston, 2012), with many unprecedented changes to how couples form and dissolve relationships and make decisions to have children (Weston & Qu, 2013). Couples choosing to live together without being married, getting married at increasingly later ages and having greater access to divorce, are some of the trends in relationships that are important to consider when designing programs and delivering services to couples and families (Weston & Qu, 2013).

A significant amount of research has reported considerable personal and social repercussions for couples and their children stemming from relationship distress and dissolution (e.g. Halford, Markman, & Stanley, 2008; Markman & Rhoades, 2012). Marital distress, for example, has been associated with an increased risk of psychological disorders (Whisman, 2007). Further, children whose parents separate, or have high levels of relationship conflict, perform worse on outcomes ranging from infant development to adolescent social adjustment (Markman & Rhoades, 2012). Research has also identified that being in a happy, satisfying marriage is one of the strongest factors across a range of cultures that determines life satisfaction for adults (Halford, 2011). It is these findings, along with the demographic shifts noted, that have led to research and policy interest in prevention and intervention strategies aimed at reducing relationship distress and breakdown.

The serious and wide-ranging negative effects that relationship distress can have on individuals and their children highlight the need for a greater understanding of the effectiveness of prevention and intervention strategies. There has been a recent increase in the amount of research available investigating relationship education as a prevention strategy. In contrast, although widely used as an intervention for couples experiencing distress, relationship counselling research has been somewhat neglected, with limited new research taking place over the past decade. This paper reviews the literature available on the effectiveness of relationship education and counselling. A search of the recent literature identified meta-analyses and a number of reviews focusing on the effectiveness of either relationship education or counselling strategies. Where possible, large-scale primary research and/or Australian findings are reported. Due to differences in the focus of the two strategies (i.e., prevention versus intervention), the paper has been divided into two sections; addressing, firstly, relationship education and, secondly, relationship counselling.

Implications for practitioners working with families and children are discussed in the concluding section.

1 In the context of the current paper a relationship is considered any “long-term committed union of romantic partners” (Lebow, Chambers, Christensen, & Johnson, 2012, p.2) and includes married and cohabiting couples.

Relationship education

Relationship education

What is relationship education?

Relationship education2 generally works with couples who are currently satisfied with their relationship and hoping to prevent relationship distress and breakdown through a focus on “building the foundations for a positive life together” (Halford, 2011, p. 3). Broadly defined, relationship education is:

Efforts or programs that provide education, skills and principles that help individuals (a person not in a relationship or a person without his or her partner) and couples (both partners participating) increase their chances of having healthy and stable relationships. (Markman & Rhoades, 2012, p. 169)

In Australia more than one third of couples go to premarital relationship education before they get married (Halford, 2011). Relationship education incorporates more programs, and is available in more settings, than these traditional premarital classes. Relationship education is not restricted to engaged couples preparing for marriage, and is not only church-based (Simons & Parker, 2002), although in Australia programs are most commonly offered through religious organisations (Halford, 2011). Relationship education programs can also be run as part of other programs including antenatal classes, parenting programs, the school curriculum, and rebuilding after separation or divorce programs, in which relationship education material can be embedded (Simons & Parker, 2002; Halford, 2011). Markman and Rhoades (2012) described the types of programs included under the umbrella term “relationship education” as:

  • couple enrichment and enhancement programs: typically marital health promotion interventions aimed at helping couples increase their levels of marital satisfaction;
  • communication programs: designed to teach couples effective communication skills; and
  • prevention programs: which are divided into three groups:
    • universal prevention programs: for all couples, usually starting when couples are young and happy, aiming to keep couples happy;
    • selective intervention programs: for couples at risk for distress or divorce; and
    • indicated programs: for couples in the early stages of distress.

Marriage and relationship education programs are typically offered to couples in a group setting (Ooms, 2010); however, the couple, or individual, can also direct their own relationship education in what is known as self-directed relationship education. Self-help books and DVDs are common forms of self-directed relationship education, and recent developments have seen an increase in the number of self-directed relationship programs available on the Internet (McAllister, Duncan, & Hawkins, 2009).

There are two evidence-based strategies commonly used in relationship education programs that are fairly widely adopted: assessment with feedback and curriculum-based knowledge and skills training (Halford, 2011).

Assessment with feedback

Assessment with feedback approaches usually comprise inventory-based couple assessments followed by a feedback session with a relationship educator who informs the couple of their current relationship strengths and weaknesses (Halford & Snyder, 2012).

The three most common inventories used in practice are PREmarital Preparation And Relationship Enhancement (PREPARE), the Facilitating Open Couple Communication Understanding and Study (FOCCUS) and RELATionship Evaluation (RELATE) (Halford, 2011).

These inventories share a number of commonalities. Partners complete the inventories independently, and the content of each inventory is similar (Halford, 2011). Questionnaires are used to score couples across dimensions such as shared realistic relationship expectations, effective communication and conflict resolution skills, and emotional health and effective individual stress management. Computers are used to score the results for all inventories, with each providing the couple with a summary report used to predict couples’ future relationship satisfaction and stability (Halford & Snyder, 2012).

There are some differences, however. The RELATE questionnaire is more comprehensive and is available online, not requiring any involvement from a relationship educator (Halford, 2011).

Although these inventory assessments are relatively quick and inexpensive to administer, Halford (2011) identified several weaknesses with the research behind the assessment with feedback approach, including the lack of published systematic evaluations of the long-term effects on relationship outcomes and the reliance on couples to self-report their behaviours rather than researchers observing the couples interacting.3

Curriculum-based knowledge and skills training

Curriculum-based approaches are relationship education programs concentrating on the active training of key relationship skills and place a significant focus on building knowledge (Halford & Snyder, 2012).

Some commonly used curriculum-based programs include the Relationship Enhancement (RE) program, the Prevention and Relationship Enhancement Program (PREP), the Couple Commitment and Relationship Enhancement (Couple CARE) program, Couples Communication Program (CCP) and Couples Coping Enhancement Training (CCET) (Halford, 2011).

These programs share many similarities, including a focus on skills training in positive communication, conflict management and positive expression of affection (Halford & Snyder, 2012); however, they also vary in significant ways (Halford, 2011), particularly in their emphasis on certain content such as the development of partner empathy in Relationship Enhancement and the development of relationship self-regulation in Couple CARE (Halford & Snyder, 2012). The topics chosen for inclusion in curriculum-based approaches are those that are thought to predict relationship outcomes, such as couple communication (Halford, 2011). This focus on communication could be considered a potential weakness of this approach, however, due to inconsistent evidence supporting the positive effect of communication on couple relationship satisfaction, as discussed later.

What does the research tell us about the effectiveness of relationship education?

Overall effectiveness of relationship education

Relationship education programs have been found to be generally successful in the short-term in improving the two most commonly tested outcomes: couples’ communication and couples’ relationship satisfaction (Markman & Rhoades, 2012; Wadsworth & Markman, 2012). Meta-analyses4 have shown that, overall, relationship education generates significant, moderate effect sizes on these outcomes (Halford & Bodenmann, 2013; Hawkins, Blanchard, Baldwin, & Fawcett, 2008), similar to those found for other prevention programs such as parent effectiveness training and maternal sensitivity to newborns programs (Hawkins et al., 2008). Blanchard and colleagues’ (2009) meta-analysis investigating the effects of marriage and relationship education on communication skills, for example, found modest evidence for the universal application of relationship education, where well-functioning, happy couples improved or maintained communication skills learned through relationship education. The researchers noted that most of the programs that have been evaluated included married (rather than premarital) couples who were happy in their relationship (Blanchard et al., 2009).

Although the goal of relationship education programs is to improve and enhance couple relationships over the long term, there are a limited number of studies that address long-term effects (Hawkins et al., 2008). When considering the effect of programs on relationship satisfaction, Halford (2011) writes there is “no convincing evidence that CRE [Couple Relationship Education] provided universally to all couples prevents deteriorating relationship satisfaction” (Halford, 2011, p. 55), noting that studies which did find evidence of a universal effect had very short follow-up assessments (Halford, 2011). A strong test of whether relationship education programs are successful as universal prevention strategies would require well-functioning couples to demonstrate the maintenance of skills for a period of 2 to 3 years (Blanchard et al., 2009).

As noted, relationship education research commonly assesses for the effects of the education on couples’ communication skills and relationship satisfaction or quality separately. The next section discusses these results.

Does relationship education improve couples’ communication skills?

Underlying relationship education programs is the assumption that learning skills and principles usually associated with healthy stable relationships will lead to improved couple relationships and the prevention of relationship problems (Wadsworth & Markman, 2012). Due to a supportive evidence base that suggests that premarital communication skills are positively associated with good marital outcomes (Fawcett et al., 2010), communication skills are thus an important focus of many relationship education programs. Communication is considered in relationship education to be a feature of relationships that can be changed to lead to immediate and future improvements in relationship quality. In other words, communication is one of the “potentially modifiable variables that predict relationship outcomes” (Halford & Bodenmann, 2013, p. 513).

Hawkins et al.’s (2008) and Blanchard et al.’s (2009) meta-analyses found relationship education produced significant, moderate, changes in couples’ communication patterns within the short term. The number of studies included in these analyses that included a follow-up period for participants of 6 months or longer was small, so there is not enough information available to know if improvements are maintained over a longer-term period. Fawcett et al. (2010) conducted a meta-analysis of premarital relationship education studies that similarly found the programs were moderately effective in improving couples’ communication. Overall, this significant, moderate effect equates to an improvement of between 50% and 60% in couples’ communication skills (Hawkins & Ooms, 2012).

There is, however, inconsistent evidence in the relationship education field that improving communication skills leads to positive changes in relationship satisfaction. Halford and Bodenmann (2013) conducted a meta-analysis studying the effects of relationship education on couple relationship satisfaction, including only studies that had longer-term follow-ups of participants of 12 months or more. They found inconsistent results regarding the effect that changes to communication had on the outcome of relationship education. Wadsworth and Markman (2012) state “we know very little about why and how CRE [Couple Relationship Education] works in general, and if the increased communication skills in particular are linked to successful outcomes”.

Does relationship education improve couples’ relationship quality and satisfaction?

Findings on the effect of relationship education on relationship quality are mixed. Hawkins and colleagues’ (2008) meta-analysis found slightly weaker results for relationship education’s effect on relationship quality than it did for communication; however, the results still showed a moderate significant effect on couples’ relationship quality in the short term. In terms of relationship quality, this amounted to couples being 40-50% better off after relationship education (Hawkins & Ooms, 2012).

A meta-analysis specifically investigating the effectiveness of relationship education for only premarital couples by Fawcett and colleagues (2010) found “these prevention programs do not improve relationship quality/satisfaction” (p. 235) when the entire body of published and unpublished research on premarital education is included in the analysis, and did not find good evidence “for a positive effect of premarital education on relationship quality/satisfaction, at least over the short time frame of the typical study” (p. 235).

Participants in studies on premarital relationship education, compared to those included in relationship education programs more broadly, are engaged to be married and typically have high levels of relationship satisfaction before starting the program (Fawcett et al., 2010), therefore results on relationship quality/satisfaction could be influenced by the “ceiling effect” whereby couples have little room for improvement (Fawcett et al., 2010). It is possible that relationship education’s effects on relationship quality/satisfaction may only be evident in the long term (and this is particularly the case for premarital programs) as the decline in satisfaction within initially satisfied couple relationships is usually a gradual one. This makes it difficult for researchers to detect any effects of relationship education on this satisfaction, particularly when the follow-up with participants is any less than 3 or 4 years after program delivery and when sample sizes are small (Halford, 2011; Halford & Bodenmann, 2013).

As the results of research into the effectiveness of relationship education as a universal prevention program have been equivocal, there has been a shift in the focus of research to examine if it may be better suited as a selective or targeted intervention. This is discussed below.

Key messages

  • Relationship education programs have been found to produce significant moderate improvements in couple communication and relationship satisfaction in the short term. There is a lack of research demonstrating the effectiveness of relationship education in the longer term.
  • There is an inconsistent evidence base supporting the use of relationship education as a universal prevention program. There is a lack of evidence with long-term follow-up supporting the universal application of relationship education. It can be difficult for studies to capture improvements in couples who enter the relationship education program already highly satisfied with their relationship.

Targeting couples at higher risk of relationship distress

Wadsworth and Markman (2012) suggested that researchers working to understand how effective relationship education programs are for higher-risk couples have used varying, and often limited, definitions of “high risk”. Factors placing couples at higher risk of relationship distress have been identified across three categories including personal characteristics (such as race/ethnicity), couple risk dynamics (such as high vs low conflict couples) and external contexts (such as situational stressors).

Many factors that fall into these three categories potentially influence not only the effectiveness of relationship education but also the likelihood of ongoing relationship quality. These factors include but are not limited to:

  • personal characteristics:
    • level of formal education;
    • socio-economic background;
    • personality traits;
    • attachment styles;
    • psychological disorders;
    • family of origin experiences;
    • relationship history of the couple and the individual; and
    • personal problems (e.g., problem drinking or substance abuse, depression and major psychiatric disorders).
  • couple risk dynamics:
    • holding shared and realistic relationship expectations;
    • working to sustain and strengthen the relationship (i.e., relationship self-regulation); and
    • how partners think about their relationship (or their “couple bond”).
  • external contexts (or life events):
    • transition to parenthood; and
    • loss of work or increased stress at work.

(Source: Halford et al., 2008; Halford, 2011)

Box 1 explores recent research aimed at understanding how individual and couple characteristics may affect outcomes in relationship education.

Although there have been a limited number of meta-analyses that include a focus on implementing relationship education as a selected (targeting couples at risk of distress) or indicated (targeting couples in the early stages of distress) intervention, there is growing research. For example, Halford and Bodenmann (2013) included almost all selective intervention programs in their meta-analysis of relationship education programs with a follow-up of more than 12 months, and demonstrated that in 14 of the 17 studies included, relationship education helped couples to maintain relationship satisfaction “for some years in at least some high-risk couples” (Halford & Bodenmann, 2013, p. 523). Future research needs to clarify which higher-risk groups would benefit most from relationship education.

Australian research by Halford, Sanders and Behrens (2001) and Halford and Wilson (2009) suggested that relationship education may work to prevent the more pronounced decline in satisfaction for high-risk couples and assist high-risk couples to have a similar trajectory of relationship satisfaction to low-risk couples. Although the results indicated that relationship education may work selectively to enhance relationship satisfaction for high-risk couples and not for low-risk couples, the Halford and Wilson (2009) study was limited by the lack of a control group. Further to this, and as previously noted, in the relationship education research, risk is often measured in varied and limited ways, with some studies focusing on single risk factors and others focusing on varied combinations of factors. For example, Halford and Wilson (2009) defined high-risk as couples in which the woman’s parents were divorced or the male partner had been a witness to inter-parental violence in his family of origin. These definitions of risk make generalisations or comparisons between findings difficult.

Box 1: Moderators of the effects of relationship education

Relationship education as an intervention may differ in its effectiveness based on couple and individual characteristics. If couples are at high or low risk for relationship distress then this may moderate the effect of the intervention. Researchers suggest that studying risk as a moderator of relationship education effectiveness should be a major focus of future studies, as understanding moderators can help to determine who relationship education is most effective for (Wadsworth & Markman, 2012).

If relationship education can modify the factors putting couples at high risk for relationship distress then it may be more effective; if relationship education cannot modify the factors putting couples at high risk for relationship distress it may be less effective (Halford & Wilson, 2009; Halford & Bodenmann, 2013). The research suggests risk factors that may be unmodifiable are low relationship commitment and severe male-perpetrated inter-partner violence (Halford & Bodenmann, 2013). Comparatively, some low-risk couples might sustain high levels of relationship quality without relationship education (Halford & Wilson, 2009) and little benefit would be evident for these couples (Halford & Bodenmann, 2013; Halford & Snyder, 2012), particularly if there were no factors placing them at risk.

The successful outcomes of relationship education (for instance improving couple communication and relationship satisfaction) can also be influenced, or modified, by couples’ experiences or risk factors prior to undertaking the intervention. Halford and Bodenmann (2013) explain:

One plausible explanation of the inconsistent findings about the association of changes in couple communication after RE [relationship education] and future relationship satisfaction is that mediation effects [changes in communication] might be moderated by pre-intervention levels of the moderator [previous communication style]. (Halford & Bodenmann, 2013, p. 521)

If, for instance, couples enter into a relationship education program that focuses on improving communication (as a mediating factor) with initially high levels of negative communication (a relationship moderator that makes them more likely to experience declining satisfaction), then undertaking the program would increase the likelihood of improved relationship satisfaction, compared to couples undertaking the program with already low levels of negative communication who have little to improve upon in terms of communication.

Lower-income couples

Lower-income couples are at greater risk of having relationship difficulties and encounter many life challenges and hardships that may make improving couple relationships a more difficult task (Hawkins & Fackrell, 2010). A meta-analysis of 15 couple education programs by Hawkins & Fackrell (2010) researched programs servicing predominantly lower-income participants. Their findings suggested relationship education programs could produce small to moderate reliable improvements in couple relationship quality and communication. The authors state these findings are noteworthy considering the stressful lives of the couples being studied and the modest amount of education offered by the programs (Hawkins & Fackrell, 2010). However, limitations to the studies included self-reporting measures, no control groups for most studies, and limited long-term follow-up of participants.

Two large-scale studies that have included lower-income couples (both married and unmarried) in the United States are explored in greater detail in Box 2.

Box 2: Recent findings from large-scale relationship education studies with lower-income couples

Building Strong Families

The Building Strong Families (BSF) project was a large-scale randomised controlled trial of relationship education involving more than 5,000 couples (Wood, Moore, Clarkwest, Killewald, & Monahan, 2012). Eight organisations across the United States ran programs included in the trial with each following a set of research-based program guidelines (Wood et al., 2012). Each organisation could choose from one of three different curricula, which were specially adapted from programs used for married couples to meet the needs of unmarried couples (Wood et al., 2012). Couples were included in the trial if they were: romantically involved, expecting a baby together (or had a baby that was less than 3 months old), unmarried, over the age of 18, and both wanted to participate in the program (Wood et al., 2012). Although no eligibility criteria was applied in relation to income, the programs “targeted and typically served low-income parents” (Wood et al., 2012, p. 4), with the average combined income of the couple in the year prior to applying for the program averaging US $20,475 (Wood et al., 2012).

Despite the trial providing significant incentives for couples to attend, including cash incentives, gift cards and baby products, on top of the provision of child care, food, transportation and flexible session times, 45% of couples randomly assigned to the treatment group did not attend even one session (Wood et al., 2012). The remaining couples averaged 21 hours of group session attendance (Wood et al., 2012), much less than was intended. Results from the trial at the three-year follow-up found the program had had no effect on relationship quality and did not make couples more likely to stay together or get married (Wood et al., 2012). There was no effect on the couples’ co-parenting relationship; in fact the program had small negative effects on some aspects of father involvement (Wood et al., 2012).

Supporting Healthy Marriages

The Supporting Healthy Marriages (SHM) program, similar to the BSF program, ran across eight different locations in the United States, offering a total of between 24-30 hours of curriculum (Lundquist et al., 2014). The eight services included in the trial could choose between four different curricula that were tailored to meet the needs of lower-income couples (Lundquist et al., 2014). Over 6,200 couples were enrolled in the randomised controlled study and met the following criteria: low-income couple, married (or considered themselves to be married), over the age of 18, and either expecting a child or parents of a child,under the age of 18 years (Lundquist et al., 2014). Couples were economically disadvantaged with most couples reporting low to modest incomes, and came from diverse racial and ethnic backgrounds (Lundquist et al., 2014).

Results from the 30-month follow-up are somewhat mixed. Although attendance was stronger than the BSF program (83% of couples attended at least one group session; Halford & Bodenmann, 2013), couples receiving the relationship education were no more likely to stay together than couples receiving no relationship education. However, the program had a small but statistically significant positive effect on couples marital quality, which was maintained at the 30-month follow-up (Lundquist et al., 2014). The authors state, “while SHM did improve marital quality for program group couples, these effects were likely too small to appreciably affect marital stability, parenting and children’s adjustment and well-being” (Lundquist et al., 2014, p. ES-11).

As can be seen, findings of the efficacy of relationship education for lower-income couples are inconsistent. In contrast to Hawkins and Fackrell’s (2010) finding that relationship education has small positive effects on lower-income couples, the two large-scale studies found limited evidence for its effectiveness. Comparisons between studies are difficult, particularly due to the differences in risk factors measured and couples’ levels of economic disadvantage. For example, while Petch, Halford, Creedy, and Gamble (2012a) demonstrated that relationship education may be better implemented as a targeted strategy focusing on couples at higher risk for relationship distress, their study identified a household income of AUD $50,000 or less as a risk factor compared to the average combined income in the Building Strong Families (BSF) program (US $20,475).

These higher rates of economic disadvantage may play a role in lessening the effectiveness of relationship education (Wood et al., 2012). The authors of the BSF report suggested that if unmarried low-income parents have higher levels of economic disadvantage, they may be less likely to fully engage with the program and put their newly learnt skills in to practice (Wood et al., 2012).

In addition to low income, there may be other potentially important moderators of relationship education effectiveness that can affect outcomes. Researchers in the BSF project reported that low levels of relationship commitment could be an important factor. Low relationship commitment not only places couples at higher risk for relationship problems but is also potentially associated with low levels of engagement with relationship education (Halford & Bodenmann, 2013). Levels of relationship commitment may have been lower in BSF than in other studies, with a proportion of BSF couples “not even cohabiting despite having a child together, and almost a quarter of all couples in the study separated in the first 15 months after recruitment” (Halford & Bodenmann, 2013, p. 518).

Key messages

  • The effectiveness of relationship education may differ based on couple and individual characteristics. If couples are at a high or low risk for relationship distress then this may moderate the effect achieved, and may also be influenced by the type of risk factor each couple experiences.
  • Despite offering many incentives for couples to attend, the Building Strong Families program had a low attendance rate, highlighting the difficulties faced by practitioners in reaching unmarried lower-income couples.
  • Low levels of relationship commitment and high rates of economic disadvantage may affect couples’ ability to fully engage with the relationship education program.
High- and low-risk couples at the transition to parenthood

The decline in relationship satisfaction may be particularly notable for couples after the birth of a child (Parker & Hunter, 2011; Halford & Petch, 2010). Considering the difficulties faced in encouraging couples to attend relationship education, finding other opportunities to reach couples, such as when they are intending to have children, is important (Petch, Halford, Creedy, & Gamble, 2012b). The transition to parenthood may offer a potential opportunity to reach some couples who would not usually attend relationship education but would attend prenatal classes, and occurs at a time when the couple is at higher risk for relationship distress.

In an Australian study, Petch and colleagues (2012a) conducted a randomised controlled trial of Couple CARE for Parents, investigating the differences in effects for high- and low-risk couples. Couples were assessed for risk of future relationship problems by the number of risk factors that were present:

  • parental divorce in the woman or man’s family of origin;
  • lack of university education in either partner;
  • annual household income of AUD $50,000 or less;
  • unplanned pregnancy; and
  • presence of inter-partner violence.

Couple CARE for Parents, compared to the mother-focused parenting program Becoming a Parent, which the control group attended, helped to prevent deterioration in relationship satisfaction for high-risk women (Petch et al., 2012a). Although not significant, high-risk men showed a trend for higher relationship satisfaction. There was no effect found for low-risk men and women (Petch et al., 2012a). The findings suggest that selectively offering relationship education to high-risk couples at the transition to parenthood stage may improve high-risk women’s relationship outcomes (Petch et al., 2012a).

Another study by the same authors (2012b) investigated whether the transition to parenthood stage could be an opportunity to reach high-risk couples who would not normally access relationship education. Couples were defined as high risk if they had three or more of the following risk factors:

  • low income;
  • low education (defined as either partner completing less than 12 years of schooling);
  • cohabitation;
  • psychological distress;
  • low relationship satisfaction;
  • unplanned pregnancy; and
  • low-level inter-partner violence.

The authors concluded “making CRE [Couple Relationship Education] accessible through hospitals, antenatal clinics, maternity and child health care is highly desirable” (Petch et al., 2012b, p. 507). Providing relationship education outside of the traditional format, delivery and service, in this case incorporating elements of it in to transition to parenthood programs, appears to have been successful in attracting high-risk couples to receive a form of relationship education. Box 3 further elaborates on the importance of making programs relevant to other potentially under-served populations, cohabiting couples and step-families.

Box 3: Relationship education and family structure

Cohabiting couples

Cohabiting couples, compared to married couples, are at greater risk of relationship distress, experiencing higher rates of negative communication and relationship aggression (Petch et al., 2012b), yet relationship education generally does not directly target this population (Markman & Rhoades, 2012). Fawcett et al. (2010), in their meta-analysis of premarital education, argue that relationship educators and researchers need:

more critical and creative thinking about how to do premarital education for the 21st century ... the timing of premarital education may be a particularly salient aspect to reconsider given family formation patterns in recent decades. (Fawcett et al, 2010, p. 236)

Not all couples choose to marry; however, of those who do, 78% have lived together before their wedding day (Weston & Qu, 2013). As such, an engagement period (formal or informal) has shifted to be more of a middle stage of the couple relationship rather than the beginning (Fawcett et al., 2010), with couples undertaking premarital education after they have lived together for a number of years and may already have had children together.

Finding new ways to target couples who are not planning to marry, but are wanting to make a stronger commitment to one another, is an important consideration for practitioners (Green & Miller, 2013); in fact, limiting such relationship interventions to only couples planning to marry “could be detrimental and isolating to a number of couples” (Green & Miller, 2013, p. 267). Research suggests that relationship education counsellors working with cohabiting couples should not only work on the couples’ communication skills but also discuss and consider the couples’ expectations for the future and include information on commitment (Markman & Rhoades, 2012).

Managing to attract greater numbers of high-risk couples to undertake the relationship education offered through the transition to parenthood program than could be achieved through premarital programs, “partially reflects that cohabiting couples are not offered premarital CRE [Couple Relationship Education], and that the time of becoming parents might be the first time they have been offered CRE” (Petch et al., 2012b, p. 506). Implementing relationship education in this context may present an opportunity for practitioners to better reach cohabiting couples.


Couples who re-partner with a child or children from one or more previous relationships, forming step-families (or step-couples), are considered a high risk or high need group of couples with needs and stressors unique to this experience (Lucier-Greer & Adler-Baeder, 2012). A meta-analysis by Lucier-Greer and Adler-Baeder explored the effectiveness of relationship education programs designed specifically for these couples. Their meta-analysis assessed 14 studies conducted from as early as 1982 and up to 2011, highlighting the lack of research in this area, and found relationship education programs targeting step-families (including both married and non-married couples) had small yet significant effects overall. The authors concluded the programs were modestly effective in influencing the overall functioning of participants, as well as family functioning and parenting; however, they noted that the effects were considerably smaller and less effective than those found in studies for generalised relationship education (e.g., in Hawkins et al. (2008) and Blanchard et al. (2009) meta-analyses). These findings need to be interpreted cautiously as there were several limitations to this meta-analysis; particularly that studies included did not measure for long-term effects, and that due to the small number of programs included, may not have been representative of all programs offered to step-families (Lucier-Greer & Adler-Baeder, 2012).

Key messages

  • Relationship education may work best as a selective intervention aimed at couples at high-risk of relationship deterioration, although this is still a developing area of research with some inconsistent results.
  • Incorporating relationship education into other programs such as those run by nurses and midwives at the transition to parenthood offers a potential opportunity to reach some usually under-serviced couples such as those in cohabiting relationships.

Limitations and gaps in the relationship education research

Given the mixed findings for the effectiveness of relationship education, it is important to highlight a number of limitations in the research.

  • The range of relationship outcomes studied is limited. When researching marriage and relationship education programs, studies most commonly assess for the effect the program has on couples’ relationship quality and communication skills. Important features of healthy relationships (i.e., relationship virtues) such as commitment, sacrifice and forgiveness are included in the curriculum of many relationship education programs; however, the effect the intervention has on these virtues has rarely been assessed (Hawkins et al., 2008). Divorce, or separation, is a crucial outcome that is not often examined (Hawkins et al., 2008) and rarely studied by relationship education researchers due to the need for large sample sizes and long-term follow-up to accurately establish any effect from the programs (Markman & Rhoades, 2012). Relationship aggression, while important to the quality of the couple relationship and wellbeing of any children in the relationship, is also rarely studied as an outcome (Hawkins et al., 2008).
  • Observational methods used to assess communication may be overstating the effects of relationship education (Hawkins et al., 2008). Researchers often use observational methods to record participants demonstrating improvements in communication skills (Hawkins et al., 2008). Some researchers have found stronger effects for observational methods compared to self-report (Fawcett et al., 2010). This may indicate that although participants demonstrate these skills in an observational setting, the success participants have when implementing them into their day-to-day lives may be more modest (Fawcett et al., 2010; Blanchard et al., 2009; Hawkins et al., 2008).
  • It is still unclear why or how relationship education works. A main focus of relationship education programs is on improving communication skills; however, the evidence behind the effects these skills have on relationship quality and satisfaction is not clear. Addressing which risk factors are modifiable, or which are not, by relationship education is an area that requires further research.
  • There is a lack of studies investigating the long-term effects of relationship education. Although the short-term effects on communication and relationship satisfaction or quality have been widely established and replicated, the capacity of relationship education to assist couples to maintain this in the longer term is less clear (Halford & Bodenmann, 2013). Whether relationship education, as an intervention, is responsible for producing longer lasting changes in areas of relationships that are important to couples, hasn’t been shown in long-term studies (Bradbury & Lavner, 2012).
  • Research suggesting relationship education may be more effective for some couples than others, and recommending a selective targeted approach, is still a growing area of research. Results from studies investigating the effectiveness of relationship education for couples at higher risk of relationship distress are inconsistent.

2 “Relationship education” is used in this report to include “couple relationship education”, “marital education”, “premarital education” and “marital and relationship education”. Where differences are apparent the appropriate term has been specified.

3 Strengths and weaknesses have been identified with both self-report and observational methods. For more information on the strengths and weaknesses of self-report and observational measures see the Limitations section of this report and Halford (2011), Fawcett, Hawkins, Blanchard, & Carrol (2010) and Blanchard, Hawkins, Baldwin, & Fawcett (2009).

4 A meta-analysis is a way of reviewing groups of studies. A statistic that quantifies the amount of change in particular variables is calculated and used to indicate whether the effect of a type of program (e.g., parenting competence, adolescent resilience) is small, medium or large. The larger the combined effect, the more effective the program.

Relationship counselling

Relationship counselling

The remainder of the paper focuses on the more interventionist end of the relationship-related services spectrum, exploring the evidence for the effectiveness of relationship counselling in treating relationship distress. Compared with the relationship education research, there has been relatively little research focusing on relationship counselling in the past decade. Reviews and meta-analyses of the research, along with relevant primary studies, have been examined and the findings that are pertinent to practitioners and policy-makers are presented.

What is relationship counselling?

Relationship counselling5 usually refers to interventions that involve a couple attending multiple sessions with one counsellor, generally together but individual sessions for one or both partners may also be included (Markman & Rhoades, 2012). Most couples who attend relationship counselling differ from those who attend relationship education (see Box 4) and tend to be experiencing high levels of distress within their relationship. Couples come to counselling with the intention of decreasing distress and increasing the health of their relationship (Markman & Rhoades, 2012). Schofield and colleagues suggest that relationship counselling approaches “typically aim to improve communication skills and commitment, and ameliorate conflict, relationship distress and dissatisfaction in couples who recognise that they are experiencing relationship difficulties” (In press, p. 4).

Box 4: Characteristics of Australian relationship education and counselling participants

There are significant differences between couples presenting for relationship education and relationship counselling. A recent Australian study by Schofield and colleagues (in press) compared the demographic, relationship, health and wellbeing characteristics of 368 participants attending counselling, compared to 92 participants attending relationship education.

Some differences between the two participant groups in Schofield and colleagues study included couples counselling participants being more likely to be married (65%) than relationship education participants (46%), and much less likely to be university educated. Relationship counselling participants were significantly younger when they started their relationship, had been in the relationship for a significantly longer period and had more children than relationship education participants (Schofield et al., in press). Relationship education participants in the study had a higher socio-economic background than the general Australian public, and found it easier to manage their finances than relationship counselling clients (Schofield et al., in press). They were also more likely to be in better health, with lower levels of depression, and were significantly less likely to nominate “serious reasons” as their motivation for attending (Schofield et al., in press). Relationship education participants found communication significantly more important as a reason for attending than couples counselling participants (Schofield et al., in press).

Practitioners working with couples and families may find that couples presenting for couples counselling have higher levels of relationship discord and reduced health (Schofield et al., in press). In fact, relationship counselling clients in Australia reported four times the rate of very high psychological distress, compared to the rate found in the general Australian population (Petch, Murray, Bickerdike, & Lewis, 2014).

The reasons given for attending relationship counselling can be wide ranging with relationship counsellors tending to focus on the specific issues that the couple present with. As noted by Snyder and Balderrama-Durbin, “couple therapists confront a tremendous diversity of presenting issues, marital and family structures, individual dynamics and psychopathology, and psychosocial stressors characterising couples in distress” (2012, p. 14). Carson and Casado-Kehoe (2012) further suggested that presenting issues in counselling are highly varied and can include problems such as infidelity, domestic violence and addiction as well as common concerns such as communication difficulties, conflicts over beliefs, values and roles, financial disagreements, problems with child rearing or as a step towards separating. Due to the complex nature of the issues and high levels of distress couples may present with, relationship counselling often involves multiple sessions over an extended period of time (Halford & Snyder, 2012).

Negative effects associated with relationship distress and conflict

Relationship distress and conflict have been found to be associated with a range of negative factors affecting the individual, couple and family. For example, in a US population-based survey of married individuals, Whisman (2007) found that marital distress was associated with an increased risk of a range of psychological disorders. The strongest associations were between marital distress and bipolar disorder, alcohol-use disorders and generalised anxiety disorder. Further to this, in a review of the research on the treatment of couple distress, Lebow and colleagues (2012) reported that the relationship between couple distress and psychopathology (such as depression) is cyclical with both negatively affecting each other. Relationship distress is also related to “social role impairment with family and friends, impaired work functioning, general distress, poorer health, and increased likelihood of suicidal ideation” (Lebow, Chambers, Christensen, & Johnson et al., 2012, p. 3). Relationship distress has also been associated with poor physical health, including depressed immune functioning (Bambling, 2007).

The negative effects of relationship distress are not limited to the couple. Researchers have found links extending the effects to children with it being associated with mental health issues such as childhood depression, substance abuse and behavioural problems (Bambling, 2007). Further to this, interparental conflict has been found to be negatively associated with parenting behaviours; meta-analytic findings suggest that conflict between parents is related to impaired parenting practices with the strongest effects found between interparental conflict and higher levels of harsh parental discipline and lower levels of parental acceptance (e.g., expressed love, support and sensitivity; Krishnakumar & Buehler, 2000).

It is the serious and wide-ranging nature of the negative effects of relationship distress on the individuals experiencing it and their children that highlight the importance and need for effective relationship counselling.

Few distressed couples seek professional help

Despite the negative effects of relationship distress, the limited research available suggests that only a small number of those experiencing it actually seek professional assistance. For example, a Relationships Australia (2011) study reported that of the 1,204 adults they interviewed, only 22% sought professional help when experiencing relationship issues. The majority had discussed their concerns with friends (35%), partners (20%) or family (16%). Further to this, a US study (n = 2,323) found that among respondents who had divorced, 37% had received counselling prior (with 34% seeking assistance from a therapist, 42% from a religious leader, and the remaining 24% from both; Johnson et al., 2002). Of those interviewed who were currently married, 19% had sought counselling for their marriage (Johnson et al., 2002). It appears that a minority of couples seek assistance when experiencing relationship difficulties.

Key messages

  • Relationship distress and conflict are associated with a range of negative factors affecting the individual, couple and family.
  • The connection between relationship distress and mental health problems (such as depression) is cyclical with both negatively affecting each other. As the two are interlinked, practitioners working in mental health should assess for relationship difficulties and those working with couples should assess for individual mental health issues.
  • The majority of those experiencing relationship difficulties do not seek professional assistance.

What does the research tell us about the effectiveness of relationship counselling?

This section provides an overview of the research available on the effectiveness of relationship counselling, focusing largely on meta-analyses and including some of the issues that have been raised with the research.

Although relationship counselling is widely practiced (e.g., one international study found that 70% of psychotherapists reported working with couples; Orlinksy & Ronnestad, 2005 reported in Lebow et al., 2012) and has been for many years, there has been relatively little growth in research in to its efficacy. Reports on the research in the 2000s had very similar findings to a review of the research in the 1990s, suggesting that little has changed. For example, in 2000, Johnson and Lebow reported:

The main models of couple therapy at the end of the 1990s are very similar to those identified in Sprenkle’s (1990) review of the field a decade ago, namely cognitive-behavioral, narrative, solution-focused and emotionally-focused. Other models such as feminist, Bowenian, psychodynamic, and integrative, have also flourished. However, it is still the case that only a few methods of intervention have been subject to research validation. (p. 25)

The most recent reviews of research on relationship counselling continue to report that there are a small number of models of therapy that have been evaluated but a large number that continue to be practiced with little or no research validation of their effectiveness:

The challenge continues for the numerous forms of couple therapy other than [the limited number that have had some form of evaluation] to demonstrate their efficacy. The last decade has seen no additional broad approaches to couple therapy moving toward becoming empirically tested. (Lebow et al., 2012, p. 14)

Overall effectiveness of relationship counselling

There are several meta-analyses and studies that provide evidence for the efficacy of a small number of models of relationship counselling. Several models have been evaluated using multiple randomised control trials (RCTs) while others have had at least one randomised control trial.6 For example, Shadish and Baldwin (2003) reviewed 20 published and unpublished meta-analyses of marriage and family interventions (MFT; of which 11 included marital therapy). The findings from the meta-analysis suggested that both marriage and family interventions were effective compared to no treatment. The authors reported that MFT interventions were at least as effective as other types of treatment such as individual therapy and showed clinically significant results in 40-50% of those treated. In summarising their findings for clinicians, the authors noted that there did not appear to be significant differences in findings for different types of MFT interventions, that there had been little research done on the clinical representativeness of the interventions (i.e., how closely the conditions under which the studies were done compared to how treatment would generally be carried out in “real world” clinical settings), and there is little known regarding which factors may moderate the effects of MFT interventions, although how the research was done appeared to have as strong an effect on outcomes as the model of intervention used (Shadish & Baldwin, 2003). For example, the setting in which the research takes place - university or non-university, measurement reactivity, measurement specificity, measurement manipulability and number of subjects have all been found to moderate the effects of types of therapy.

The same authors performed a meta-analysis of 30 randomised experiments of distressed couples that compared Behavioural Marital Therapy (BMT) consisting of communication training, behaviour exchange, cognitive restructuring and emotional expressiveness training with no-treatment controls (Shadish & Baldwin, 2005). The studies, dated from 1973 to the early 1990s, included unpublished dissertations and published studies, and were limited to studies aimed at reducing marital or psychological distress. The authors reported that, overall, BMT was significantly more effective in reducing marital or psychological distress than no treatment, although less effective than reported in previous meta-analyses. The authors suggested that the inclusion of unpublished dissertations with small sample sizes and small or negative effect sizes led to the lower effect size that they reported and that previously reported higher effect sizes might have been inflated. They also noted that communication training/problem-solving training was the only component of BMT that predicted greater effectiveness, and cognitive restructuring actually predicted lower effect sizes. Further to this, Shadish and Baldwin found no significant association between effect size and clinical representativeness or effect size and number and length of sessions.

Finally, in a review of research on treating couple distress, Lebow and colleagues (2012) reported on a range of studies that supported the efficacy of relationship counselling with the authors concluding “studies continue to show that most couple therapy has an impact, with about 70% of cases showing positive change. These build on the already two decades of existent findings suggesting couple therapy is an effective mode of treatment” (p. 148).

Differences in effectiveness between theoretical approaches to relationship counselling

While there is evidence to support the efficacy of relationship counselling compared to no treatment, as noted by Shadish and Baldwin (2003), there is limited available research identifying why relationship counselling is effective or why there are few differences in efficacy found between theoretical models of relationship interventions. One longitudinal study of 134 highly distressed couples compared traditional behavioural couple therapy (TBCT) and integrative behavioural couple therapy (IBCT; Christensen, Baucom, Atkins, & Yi, 2010) and found few differences in effectiveness at follow-up. The researchers found that although couples who received IBCT reported higher marital satisfaction than those who received TBCT in the earlier stages of follow-up, by the 5-year follow-up the effects of each treatment were no longer significantly different. Both treatments reported similar significant improvements in marital satisfaction. Approximately a quarter of couples in each treatment group were separated or divorced at the 5-year follow-up, with the authors noting that the couples who divorced or separated were more likely to have started treatment somewhat more distressed, to have improved little and to have shown significant deterioration following the completion of therapy.

Further to this, a meta-analysis of 23 studies (20 published and 3 unpublished dissertations) by Wood, Crane, Schaalje, and Law (2005) investigated whether particular relationship counselling approaches were more effective in treating different levels of relationship distress. The study identified 41 treatment groups of which seven were rated mildly distressed, 33 moderately distressed and one severely distressed. Treatments were coded into previously established treatment models— behavioural marital therapy (BMT), emotionally focused therapy (EFT), mixed (if the treatment involved components of different treatment models), BMT components (if the treatment was limited to only one component of BMT) and others. The majority of studies involved BMT. Although the authors noted that the small number of studies found limited the meta-analysis and that there was a need for more replication studies focusing on different models of treatment, they reported that differences in treatment approaches were found once differences in levels of distress were identified. The authors concluded that: couples experiencing different levels of distress should not be viewed as one homogenous group; that any intervention for mildly distressed couples would be better than no treatment as there were no significant differences found between treatment models; and that moderately distressed couples should receive a full treatment model rather than isolated components. The authors suggested that further research is required to tease out potential differences in the effectiveness of various approaches to relationship counselling in treating couples experiencing disparate levels of relationship distress.

See Table 1 for a brief summary of selected approaches to relationship counselling that have reported positive outcomes in treating relationship distress.

Table 1: Brief descriptions of selected relationship counselling approaches that have undergone outcome evaluationsa
Notes: a The table is not intended as an exhaustive list of all approaches to relationship counselling but rather an overview of models with some empirical support for their effectiveness. b These three therapies have demonstrated positive outcomes in treating couple distress in one trial only (IOCT: Snyder & Wills (1989); ISCT: Goldman & Greenberg (1992); IBCT: Jacobson & Christensen (1996) and Christensen et al. (2004) cited in Snyder et al., 2006). c Snyder & Mitchell, 2008.
Behavioural couple therapy (BCT)/ Behavioural marital therapy (BMT; including couple cognitive behavioural therapy [CBT])The most well-researched approaches used to treat distressed couple relationships (Bambling, 2007), BCT and BMT work on increasing positive behaviours and decreasing negative behaviours to improve relationship satisfaction (Halford & Snyder, 2012; Baucom, Sevier, Eldridge, Doss, & Christensen, 2011). These approaches include several treatment components: communication training, problem-solving training, contingency contracting, behaviour exchange, desensitisation, cognitive restructuring and emotional expressiveness training (Snyder, Castellani, & Whisman, 2006; Shadish & Baldwin, 2005). Cognitive behavioural therapy (CBT) techniques, often added to BCT, assume that how couples interpret their partner’s behaviour is as important as the actual behaviour in influencing distress or satisfaction (Bambling, 2007). CBT helps couples identify problematic thoughts, feelings and reactions, and to identify unrealistic expectations (Bambling, 2007).
Emotion focused couple therapy (EFCT)Influenced by attachment theory, EFCT works on the assumption that relationship distress disrupts attachment to create strong problematic reactions in the couple due to their fears of abandonment (Bambling, 2007). Relationship behaviours that are problematic are considered to be secondary responses to this abandonment fear (Bambling, 2007). This therapy works to reduce hostile emotions and angry behaviours, and increase the expression of emotional vulnerability and attachment needs (Halford & Snyder, 2012; Bambling, 2007).
Integrative behavioural couple therapy b (IBCT)Recent research has begun to explore IBCT, which combines traditional BCT techniques with strategies focusing on fostering emotional acceptance amongst couples (Snyder et al., 2006). Therapists using this therapy explore partners’ emotional reactions to one another’s messages, with the assumption that this will help develop more effective and sensitive communication for couples (Baucom et al., 2011).
Insight oriented couple therapy b (IOCT; also known as Affective-Reconstructive Couple Therapy c)IOCT interprets maladaptive relationship patterns from a developmental perspective, working to develop better ways for partners to interact through helping them to understand and change any enduring dysfunctional patterns (Snyder, 2009). This therapy considers unmet needs and unresolved anxieties, stemming from prior relationships, as linked to patterns of emotional and behavioural responses in their current relationship (Snyder, 2009).
Integrated systemic couple therapy b (ISCT)Attempting to disrupt any negative repetitive, self-perpetuating, cycles of interaction in the couple’s relationship, ISCT works by changing the meaning the couple attributes to these interactions (Snyder et al., 2006). The process followed by therapists to achieve this includes: restructuring interactions, reframing problems through the use of positive connotations, prescribing the symptom, urging the couple to go slow, and then prescribing a reenactment of the negative interactions (Snyder, 2009).

Key messages

  • Several approaches to relationship counselling have been found to be moderately effective in reducing relationship distress or increasing relationship satisfaction compared to no treatment, although there has been limited new research in the past decade.
  • Few differences in effectiveness have been found between different theoretical approaches to relationship counselling.
  • Initial levels of distress may be an important factor in the success or failure of counselling and it may be of value for practitioners to assess for this. Initial research suggests that moderately distressed couples should receive a full treatment model and that highly distressed couples may be less likely to experience positive outcomes from counselling.

Mechanisms of change in relationship counselling

Although there is evidence for its effectiveness, there is limited understanding of the mechanisms of change (i.e., how or why it works) in relationship counselling. There is debate regarding whether treatment effectiveness is related to the specific theoretical model used or variables common to all relationship counselling (Davis, Lebow & Sprenkle, 2012; Sprenkle, 2012). There is also discussion on whether integrated approaches that encompass elements from multiple models of relationship counselling could lead to increases in the effectiveness of treatment (Christensen et al., 2010; Snyder & Balderrama-Durbin, 2012). The debate is particularly relevant to those teaching relationship counselling. For example, should students be taught one particular theoretical model of therapy, several models of therapy or should there be more focus on common factors, such as therapeutic alliance, if these are actually responsible for a large proportion of the change seen in therapy? There is currently limited research available to conclusively support one or other of these theories and it may be that it is a combination of both; for example, as posited by Davis and colleagues (2012), well-evidenced theoretical approaches may provide the structure and organisation through which common factors operate. What is known is that:

  • several theoretical approaches to relationship counselling have been found to be moderately effective in improving a range of couple functioning outcomes;
  • at least one integrated approach has evidence of its efficacy although little evidence of greater effectiveness than other single theory approaches over the longterm (integrative behavioural couple therapy; Christensen et al., 2010); and
  • there is evidence to support the positive impact of certain common factors in therapy (e.g., Bambling, 2007; Halford et al., 2012; Lebow et al., 2012; Shimokawa, Lambert, & Smart, 2010; See Box 5 for further details of common factors).

Box 5: Common factors in relationship counselling

The research on common factors that may be effective across theoretical approaches, particularly in the field of relationship counselling (compared to research on individual therapy), is limited and needs expansion but there are several common factors that have evidence for their importance in affecting therapy outcomes.

  • Working or therapeutic alliance: The quality and strength of the relationship between each client and the therapist has been found to be strongly related to, and necessary for, positive counselling outcomes (Davis et al., 2012; Sprenkle, Davis, & Lebow, 2009). Research suggests that the alliance needs to be formed early in therapy to ensure that couples do not drop out of therapy. Clients’ gender and views of the alliance have also been found to affect treatment outcomes (Bambling, 2007; Lebow et al., 2012). Couples who share a similar view of the strength of the alliance with each other and the counsellor are more likely to experience positive therapy outcomes (Bambling, 2007). Finally, it has been reported that the most highly distressed couples are more likely to form poorer alliances (Lebow et al., 2012).
  • Client expectancy and hope: Couples’ beliefs that the counselling is credible and likely to help them as well as having hope of a positive outcome are linked to more positive outcomes in therapy (Bambling, 2007; Davis et al., 2012). The therapist’s ability to present treatment that is consistent with clients’ expectations has also been linked to favourable outcomes in relationship counselling (Bambling, 2007).
  • Feedback: Research (largely exploring individual psychotherapy) has suggested that providing progressive formal feedback (either from the clients, themselves, or a clinical supervisor) on the progress of therapy throughout counselling may help cut drop-out rates and improve outcomes, particularly for those who are considered unlikely to benefit from therapy (Anker, Duncan, & Sparks, 2009; Shimokawa et al., 2010; Snyder & Halford, 2012).
  • Client and therapist characteristics: Similarly to the findings on the effectiveness of relationship education, although the evidence is limited, other variables such as client characteristics (e.g., relationship commitment, individual mental health issues, initial levels of relationship distress, motivation and engagement in therapy; Davis et al., 2012) and therapist characteristics (e.g., sensitive to unique client needs including culturally sensitive, friendly and positive towards client) have also been found to affect relationship counselling outcomes.

In a review of relationship counselling, Snyder and colleagues (2006) suggested that the evidence available supports the benefits of training relationship counsellors in both common factors and the ability to practice integratively across theoretical approaches while also suggesting that further research is needed to identify and overcome the barriers to transferring relationship counselling research in to practice.

Key messages

  • There is continued debate, although currently limited evidence, regarding exactly why relationship counselling is effective, with research focusing on explanations of change based on specific theoretical models or factors common to all approaches.
  • Several common factors, although requiring further investigation, have been found to affect outcomes in relationship counselling: the therapeutic alliance; expectancy and hope; the provision of feedback during counselling, and client and therapist characteristics.

Limitations of relationship counselling

Although randomised control trials and meta-analyses provide support for the efficacy of relationship counselling there have been a number of limitations raised regarding its effectiveness, some of which have been noted in previous sections of the paper and have been raised in multiple reviews of the research.7

  • Many couples experiencing relationship distress do not seek professional assistance. As noted above, only a small proportion of those couples experiencing distress seek relationship counselling.
  • The effects of relationship counselling tend to deteriorate over time. For example, it has been reported that 30-60% of couples show a significant reduction in the effects of relationship counselling at 2-year or longer follow-ups (Snyder et al., 2006).
  • Relationship counselling is not effective for all couples. Regardless of approach, there appears to be approximately 25-30% of couples for whom relationship counselling does not work, that is, they show little improvement in relationship distress or relationship satisfaction after treatment (Christensen et al., 2010; Halford & Snyder, 2012). Also, within couples there are often differences in the effects of treatment and when these are taken into consideration the effectiveness of relationship counselling is further reduced (Bambling, 2007).
  • Couples having the most relationship difficulties tend to benefit less from relationship counselling. Research suggests initial levels of relationship distress are highly related to treatment outcomes, with those experiencing greatest distress showing least improvement in distress levels (Snyder et al., 2006). As noted by Christensen and colleagues (2010) in their longitudinal study of IBCT and TBCT, those couples that went on to separate or divorce were more likely to have reported high levels of relationship distress and lower marital satisfaction prior to treatment than those who did not go on to separate or divorce, although all couples in the study had been chosen because they were experiencing chronically high levels of distress.

Concerns regarding limitations of the research itself have also been raised and these are outlined briefly in Box 6.

Box 6: Limitations of relationship counselling research

There are a number of limitations of the research on relationship counselling that may limit the findings and their generalisability. These are the most commonly reported limitations.

  • There are a large number of approaches to relationship counselling that have not been rigorously evaluated. As noted previously, although several approaches to relationship counselling have been investigated for their effectiveness, there are a large number still widely practised that have not been the subject of rigorous testing (Johnson & Lebow, 2000; Sprenkle, 2012). This is not to say that these therapies are necessarily ineffective but that they have not undergone rigorous evaluation of their effectiveness. Considering that some of these therapies have been around for decades this should not be the case (Lebow et al., 2012).
  • There is not a lot of “transportation” of research studies to the “real world” of therapy. As noted by Sprenkle (2012), most of the research on relationship counselling relates to “efficacy” (i.e., “the effects of interventions that are studied in controlled clinical trials under specified conditions, usually in a university or hospital or university or researcher-controlled clinic setting” [p. 5]) rather than “effectiveness” (i.e., the “normal” therapeutic circumstances under which most therapy takes place). It is, therefore, worth interpreting the research findings and their applicability to the “real world” of therapy with caution. For example, in an examination of 50 published clinical trials of couple therapy outcome studies, Wright, Sabourin, Mondor, McDuff, and Mamodhoussen (2006) reported the clinical representativeness 8of relationship counselling as only “fair” based on scores overall being lower than the midpoint of their clinical representativeness scale. They did find that although clinical representativeness of patient problems was high with patients generally presenting with clinical levels of couple distress and a broad range of problems, it was lower on other dimensions (e.g., advertisements for free relationship counselling were often used to recruit couples; graduate students who were given intensive training and close supervision were often used to conduct the therapy; and the number of sessions was often set and the treatment highly structured) suggesting that while overall the trials were rated as fair, there were many areas that were not clinically representative.
  • Research continues to be largely on married, middle-class, heterosexual white couples (Bambling, 2007; Sprenkle, 2012). There is little research on the effectiveness of relationship counselling for other groups such as low-income couples, lesbian and gay couples or those from minority ethnic backgrounds, thus limiting the generalisability of the findings.
  • The majority of research has been done by a small number of researchers who are highly invested in the models of relationship counselling they are testing. As noted by Sprenkle (2012), the developers of the intervention models also generally run the randomised control trials, which may lead to an allegiance effect (i.e., if the researcher or therapist has a bias towards a particular approach this may positively influence the results of that approach). This, along with the limited testing of therapy in the “real world”, Sprenkle suggests, raises questions regarding how effective the treatments may be when performed by “typical therapists in real-world settings” (p. 5).
  • There is limited understanding of why or how different approaches to relationship counselling are effective. As discussed in the previous section, research has found little difference in the effectiveness of different approaches to relationship counselling and although it is a growing area of research, there is still little understanding of the mechanisms of change in therapy (Bambling, 2007; Lebow et al., 2012). Some researchers have suggested that factors common to all approaches to relationship counselling may be responsible (the “common factors” approach) while others have suggested it is the unique elements and mechanisms of particular approaches that are responsible for their effectiveness (the “model-driven change paradigm”, p. 37; Davis et al., 2012).

Key messages

  • Research to date suggests that relationship counselling does not work for approximately 25-30% of those who attend, regardless of approach. Couples facing the most relationship difficulties and highest levels of distress tend to have worse outcomes in counselling.
  • There are a number of widely practised approaches to relationship counselling that still remain largely untested.

5 The term “relationship counselling” is used throughout and is used interchangeably with other common terms in the literature such as “couples counselling”, “couple therapy”, “relationship therapy” and “marital therapy”. Where differences are apparent the appropriate term has been specified.

6 Although considered the most rigorous method of evaluating effectiveness, research has highlighted potential concerns with the practicality and ethicality of running RCTs in social services (e.g., Child Family Community Australia, 2013; Dixon et al., 2014; Fairhurst & Dowrick, 1996; Sibbald & Roland, 1998).

7 For reviews of the research including its limitations see Halford & Snyder, 2012; Lebow et al., 2012: Snyder et al., 2006; Snyder & Halford, 2012; Sprenkle, 2012.

8 Clinical representativeness refers to how closely the conditions under which studies are performed compares to how treatment is generally carried out in “real world” clinical settings.

Implications and conclusions

Implications and conclusions

The research findings from both areas have implications for policy, practice and future research. For example, while relationship education research has expanded since the mid 2000s to study a wider range of couples, such as those with lower incomes, unmarried couples who have children together, couples where one partner has an illness or couples who have fostered or adopted a child, this still may not be reflective of the diversity of couples in the wider community (Markman & Rhoades, 2012; Halford & Bodenmann, 2013). There is potential to expand this wider application of relationship education to the many groups who still remain under-served, such as older couples, gay, lesbian or transgendered couples, separated and divorced people, cohabiting couples, individuals looking for a relationship (Markman & Rhoades, 2012), couples with ageing parents, couples forming stepfamilies and couples transitioning to retirement (Halford & Bodenmann, 2013). Couples in these diverse groups are still “not well understood or represented in relationship education programs” (Bradbury & Lavner, 2012, p. 115). These findings can be extended to relationship counselling research where there has also been limited research focusing on these groups (Lebow et al., 2012). It is important, however, that any programs or interventions catering for diverse couples are evidence-based and that this evidence is generated from studies that sample a diverse range of couples (Johnson, 2012) so that it is clear from the evidence that these couples will benefit from the relationship education or counselling offered to them.

An implication for relationship educators to consider in their practice could be that offering a fixed curriculum for all couples may be less effective than offering a tailored relationship education program (Halford & Bodenmann, 2013). If the program can offer content that addresses the risk factors specific to individual couples, then that material may be more relevant and meaningful. There is limited evidence for the universal application of relationship education programs, particularly over a longer-term period, and there is developing evidence suggesting some couples may be more likely to benefit from programs than others based on shared and individual risk factors. Bradbury and Lavner (2012) write that offering interventions to all couples regardless of their levels of risk for relationship deterioration “ignores important information, is likely to compromise the effects of interventions, and devotes resources to couples least likely to need them” (Bradbury & Lavner, 2012, p. 117).

The findings from the two large-scale trials of relationship education in the United States, Building Strong Families and Supporting Healthy Marriages, highlight implications for policy makers. Relationships for lower-income couples can come under added stress from a range of other factors such as “lack of economic resources, poor educational opportunities, [un]stable jobs, unsafe neighbourhoods, drug addictions, traumatised childhoods, and so on” (Hawkins & Ooms, 2012, p. 543). Hawkins and Ooms (2012) highlighted suggestions that a more effective policy tool would be to direct funding towards some of these underlying causes of relationship instability, rather than toward the relationship itself. Hawkins and Ooms (2012), however, still see the value in supporting marriage and relationship programs for disadvantaged groups, where the intervention “is intended to supplement other antipoverty efforts, not replace them” (Hawkins & Ooms, 2012, p. 543). Furthermore, there is potential for relationship education programs to act as a gateway for program participants who may need help from other services (Hawkins & Ooms, 2012).

Further to this, with the high co-occurrence and cyclical nature of relationship distress and psychopathology (such as depression, anxiety and substance misuse), there is merit in mental health workers assessing for relationship difficulties and practitioners working with couples assessing for individual psychopathology. There is opportunity for workers in these areas to provide referrals to relationship counselling or mental health services as required, thus potentially increasing the likelihood of positive outcomes for clients.

Finally, while relationship education and counselling may differ in their target groups (those in relatively satisfying relationships versus those experiencing distress) and aims (e.g., prevention of relationship issues versus intervening in relationship issues), there are commonalities across the two. As highlighted by Halford and Snyder (2012), although research in relationship education and counselling has found several evidence-based approaches to be effective, there is still little understanding in both fields of the precise mechanisms of change and also that “all approaches to couple therapy and relationship education have significant limitations in their efficacy, and to date it has not been possible to find a general approach that is reliably more effective than pre-existing approaches” (p. 8). Further research is required to understand why effective interventions work and, importantly, why there are people for whom these interventions do not work in order to potentially increase the overall effectiveness of relationship education and counselling interventions in the future.



  • Anker, M. G., Duncan, B. L., & Sparks, J. A. (2009). Using client feedback to improve couple therapy outcomes: A randomized clinical trial in a naturalistic setting. Journal of Consulting and Clinical Psychology, 77(4), 693–704.
  • Bambling, M. (2007). Does it work? Research on the effectiveness of couple therapy. In E. Shaw & J. Crawley (Eds.), Couple therapy in Australia: issues emerging from practice (pp. 49–76). Kew, Vic.: PsychOz Publications.
  • Baucom, K. J. W, Sevier, M., Eldridge, K. A., Doss, B. D., & Christensen, A. (2011). Observed communication in couples two years after integrative and traditional behavioural couple therapy: Outcome and link with five-year follow-up. Journal of Consulting and Clinical Psychology, 79(5), 565–576.
  • Blanchard, V. L., Hawkins, A. J., Baldwin, S. A., & Fawcett, E. B. (2009). Investigating the effects of marriage and relationship education on couples’ communication skills: A meta-analytic study. Journal of Family Psychology, 23(2), 203–214.
  • Bradbury, T. N., & Lavner, J. A. (2012). How can we improve preventive and educational interventions for intimate relationships? Behaviour Therapy, 43,113–122.
  • Carson, D. K., & Casado-Kehoe, M. (2012). Introduction: Couple therapy in contemporary society and a theory-based case study approach. InD. K. Carson & M. Casado-Kehoe (Eds.), Case studies in couple therapy: Theory-based approaches. Hoboken: Taylor and Francis.
  • Child Family Community Australia. (2013). Planning for evaluation I: Basic principles (Practitioner resource). Melbourne: Child Family Community Australia, Australian Institute of Family Studies. Retrieved from <>
  • Christensen, A., Baucom, B., Atkins, D. C., & Yi, J. (2010). Marital status and satisfaction five years following a randomized clinical trial comparing traditional versus integrative behavioral couple therapy. Journal of Consulting and Clinical Psychology, 78(2), 225–235.
  • Davis, S. D., Lebow, J. L., & Sprenkle, D.H. (2012). Common factors of change in couple therapy, Behaviour Therapy, 43, 36–48.
  • Dixon, J., Biehal, N., Green, J., Sinclair, I., Kay, C., & Parry, E. (2014). Trials and tribulations: Challenges for randomised controlled trials of social work with children. British Journal of Social Work, 44, 1563–1581.
  • Fairhurst, K., & Dowrick, C. (1996). Problems with recruitment in a randomized controlled trial of counselling in general practice: Causes and implications. Journal of Health Services, Research & Policy, 1(2), 77–80.
  • Fawcett, E. B., Hawkins, A. J., Blanchard, V. L., & Carrol, J. S. (2010). Do premarital education programs really work? A meta-analytic study. Family Relations, 59, 232–239.
  • Green, A. R., & Miller, L. D. (2013). A literature review of the strengths and limitations of premarital preparation: Implications for a Canadian context. Canadian Journal of Counselling and Psychotherapy, 47(2), 256–271.
  • Halford, W. K. (2011). Marriage and relationship education: What works and how to provide it. New York: Guilford Press.
  • Halford, W. K., & Bodenmann, G. (2013). Effects of relationship education on maintenance of couple relationship satisfaction. Clinical Psychological Review, 33,512–525.
  • Halford, W. K., Hayes, S., Christensen, A., Lambert, M., Baucom, D. H., & Atkins, D. C. (2012). Toward making progress feedback an effective common factor in couple therapy. Behavior Therapy, 43, 49–60.
  • Halford, W. K., Markman, H. J., & Stanley, S. (2008). Strengthening couples’ relationships with education: Social policy and public health perspectives. Journal of Family Psychology, 22(3), 497–505.
  • Halford, W.K., & Petch, J. (2010) Couple psychoeducation for new parents: Observed and potential effects on parenting, Clinical Child & Family Psychology Review, 13, 1641–80.
  • Halford, W. K., Sanders, M. R., & Behrens, B. C. (2001). Can skills training prevent relationship problems in at-risk couples? Four-year effects of a behavioral relationship education program. Journal of Family Psychology, 15(4), 750–768.
  • Halford, W. K., & Snyder, D. K. (2012). Universal processes and common factors in couple therapy and relationship education. Behavior Therapy, 43, 1–12.
  • Halford, W. K., & Wilson, K. L. (2009). Predictors of relationship satisfaction four years after completing flexible delivery couple relationship education. Journal of Couple & Relationship Therapy, 8,143–161.
  • Hawkins, A. J., Blanchard, V. L., Baldwin, S. A., & Fawcett, E. B. (2008). Does marriage and relationship education work? A meta-analytic study. Journal of Consulting and Clinical Psychology, 76(5), 723–734.
  • Hawkins, A. J., & Fackrell, T. A. (2010). Does relationship and marriage education for lower-income couples work? A meta-analytic study of emerging research. Journal of Couple & Relationship Therapy, 9, 181–191.
  • Hawkins, A. J., & Ooms, T. (2012). Can marriage and relationship education be an effective policy tool to help low-income couples form and sustain healthy marriages and relationships? A review of lessons learned. Marriage & Family Review, 48(6), 534–554.
  • Johnson, M. D. (2012). Healthy marriage initiatives: On the need for empiricism in policy implementation. American Psychologist, 67(4), 296–308
  • Johnson, S., & Lebow, J. (2000). The “coming of age” of couple therapy: A decade review. Journal of Marital and Family Therapy, 26(1), 23–28.
  • Johnson, C. A., Stanley, S. M., Glenn, N. D., Amato, P. R., Nock, S. L., & Markman, H. J. (2002). Marriage in Oklahoma: 2001 baseline statewide survey on marriage and divorce (S02096 OKDHS). Oklahoma City: Oklahoma Department of Human Services. Retrieved from <>.
  • Krishnakumar, A., & Buehler, C. (2000). Interparental conflict and parenting behaviors: A meta-analytic review. Family Relations, 49(1), 25–44.
  • Lebow, J. L., Chambers, A. L., Christensen, A., & Johnson, S. M. (2012). Research on the treatment of couple distress. Journal of Marital and Family Therapy, 38(1), 145–168.
  • Lucier-Greer, M., & Adler-Baeder, F. (2012). Does couple and relationship education work for individuals in stepfamilies? A meta-analytic study. Family Relations, 61, 756–769.
  • Lundquist, E., Hsueh, J., Lowenstein, A. E., Faucetta, K., Gubits, D., Michalopoulos, C., & Knox, V. (2014). A Family-Strengthening Program for Low-Income Families: Final Impacts from the Supporting Healthy Marriage Evaluation. Washington, DC: Office of Planning, Research and Evaluation, Administration for Children and Families, US Department of Health and Human Services. Retrieved from <>.
  • Markman, H. J., & Rhoades, G. K. (2012). Relationship education research: Current status and future directions. Journal of Marital and Family Therapy, 38(1), 169–200.
  • McAllister, S., Duncan, S. F., & Hawkins, A. J. (2009). Examining the early evidence for self-directed marriage and relationship education: A meta-analytic study. Family Relations, 61, 742–755.
  • Moloney, L., & Weston, R. (2012). Changing patterns of marriage and intimate relationships. Brief reflections on research, relationship counselling and relationship education. Threshold, Dec., 14–17.
  • Ooms, T. (2010). Marriage and relationship education and couples and marriage therapy/counseling: What’s the difference? National Health and Medical Research Council Brief. Oklahoma City: National Healthy Marriage Resource Center. Retrieved from <>.
  • Parker, R., & Hunter, C. (2011). Supporting Couples Across the Transition to Parenthood. (Australian Family Relationships Clearinghouse). Melbourne: Australian Institute of Family Studies.
  • Petch, J., Halford, W. K., Creedy, D. K., & Gamble, J. (2012a). A randomized controlled trial of a couple relationship and coparenting program (Couple CARE for Parents) for high- and low-risk new parents. Journal of Consulting and Clinical Psychology, (80)4, 662–673.
  • Petch, J., Halford, W. K., Creedy, D. K., & Gamble, J. (2012b). Couple relationship education at the transition to parenthood: A window of opportunity to reach high risk couples, Family Process, (51)4, 498–511.
  • Petch, J., Murray, J., Bickerdike, A., & Lewis, P. (2014). Psychological distress in Australian clients seeking family and relationship counselling and mediation services. Australian Psychologist, 49,28–36.
  • Relationships Australia. (2011). Issues and concerns for Australian relationships today: Relationship indicators survey 2011. Canberra: Relationships Australia. Retrieved from <>.
  • Schofield, M. J., Mumford, N., Jurkovic, I., Jurkovic, J., Chan, S. P., & Bickderdike, A. (In press). Understanding profiles of couples attending community-based couple counselling and relationship education services. Journal of Couple and Relationship Therapy.
  • Shadish, W. R., & Baldwin, S. A. (2003). Meta-analysis of MFT interventions. Journal of Marital and Family Therapy, 29(4), 547–562.
  • Shadish, W. R., & Baldwin, S. A. (2005). Effects of behavioral marital therapy: A meta-analysis of randomized control trials. Journal of Consulting and Clinical Psychology, 73(1), 6–14.
  • Shimokawa, K., Lambert, M. J., & Smart, D. W. (2010). Enhancing treatment outcome of patients at risk of treatment failure: Meta-analytic and mega-analytic review of a psychotherapy quality assurance system. Journal of Consulting and Clinical Psychology, 78(3), 298–311.
  • Sibbald, B., & Roland, M. (1998). Understanding controlled trials: Why are randomised controlled trials important? The BMJ, 316, 201. Retrieved at <>.
  • Simons, M., & Parker, R. (2002). Relationship education services. Family Matters, 63,77–79.
  • Snyder, D. (2009). Insight-oriented couple therapy. In H. Reis, & S. Sprecher (Eds.), Encyclopedia of human relationships (pp. 860–863). Thousand Oaks, CA: SAGE Publications, Inc.
  • Snyder, D. K., & Balderrama-Durbin, C. (2012). Integrative approaches to couple therapy: Implications for clinical practice and research. Behavior Therapy, 43, 13–24.
  • Snyder, D. K., Castellani, A. M., & Whisman, M. A. (2006). Current status and future directions in couple therapy. Annual Review of Psychology, 57, 317–344.
  • Snyder, D. K., & Halford, W. K. (2012). Evidence-based couple therapy: Current status and future directions. Journal of Family Therapy, 34, 229–249.
  • Snyder, D. K., & Mitchell, A. E. (2008). Affective-reconstructive couple therapy: A pluralistic, developmental approach. In A.S. Gurman (Ed.). Clinical handbook of couple therapy (4th ed.) (pp. 353–382). New York: Guildford Press.
  • Sprenkle, D. H. (2012). Intervention research in couple and family therapy: A methodological and substantive review and an introduction to the special issue. Journal of Marital and Family Therapy, 38(1), 3–29.
  • Sprenkle, D. H., Davis, S. D., & Lebow, J. L. (2009). Common factors in couple and family therapy. New York: The Guildford Publications.
  • Wadsworth, M. E., & Markman, H. J. (2012). Where’s the action? Understanding what works and why in relationship education. Behavior Therapy, 43, 99–112.
  • Weston, R., & Qu, L. (2013). Working out relationships (Australian Family Trends No. 3). Melbourne: Australian Institute of Family Studies.
  • Whisman, M. A. (2007). Marital distress and DSV-IV psychiatric disorders in a population-based national survey. Journal of Abnormal Psychology, 116(3), 638–643.
  • Wood, N. D., Crane, R. D., Schaalje, G. B., & Law, D. D. (2005). What works and for whom: A meta-analytic review of marital and couples therapy in reference to marital distress. The American Journal of Family Therapy, 33, 273–287.
  • Wood, R. G., Moore, Q., Clarkwest, A., Killewald, A., & Monahan, S. (2012). The Long-Term Effects of Building Strong Families: A Relationship Skills Education Program for Unmarried Parents (OPRE Report # 20122–8A). Washington DC: Office of Planning, Research and Evaluation, Administration for Children and Families, U.S. Department of Health and Human Services (OPRE).
  • Wright, J., Sabourin, S., Mondor, J., McDuff, P., & Mamodhoussen, S. (2006). The clinical representativeness of couple therapy outcome research. Family Process, 46(3), 301–316.

Cathryn Hunter is a Senior Research Officer and Jo Commerford was a Research Officer with the Child Family Community Australia information exchange at the Australia Institute of Family Studies at the time of writing this.  

The authors wish to acknowledge the valuable contributions of Robyn Parker, Senior Manager, Research and Evaluation at Interrelate, Rosalie Pattenden, Counselling Psychologist, and Elly Robinson, Manager, Child Family Community Australia (CFCA) information exchange, for their feedback.

Featured image: © Shutterstock/Monkey Business Images