A Literature Review of the Evidence for the Effectiveness of Experiential Psychotherapies

Print Friendly, PDF & Email

Return to Journal Articles

Ben Mullings, PhD, Private Practice, Perth, Western Australia



Experiential psychotherapies originate from the humanistic tradition. This class of approaches include emotion-focused, existential, person-centred, gestalt, psychodrama, focusing-oriented, and expressive therapies. Experiential approaches share a set of core qualities, such as eliciting and exploring reflective experiences in-session, using empathy within the therapeutic relationship as a curative factor, and helping people to explore and make meaning of their own inner lives.

It has frequently been the case that researchers from other therapy traditions have described relationship control groups as ‘supportive’ or ‘non-directive’. An unfortunate consequence has been confusion with inactive controls which are not intended to be genuine therapy approaches. A growing number of contemporary approaches from outside of the humanistic tradition have also now begun to integrate elements of experiential therapy (Angus, Watson, Elliott, Schneider, & Timulak, 2015).

Historically speaking, the experiential tradition has opposed psychiatric classification on the basis that diagnostic categories are conceptually flawed, needlessly based on dehumanising criteria (such as biomedical centrism), and have at times been used to restrict freedom and choice for people who seek psychological care (Cain, Keenan, & Rubin, 2016). Therapists have increasingly relied upon research evidence to support their practices (i.e., evidence-based practice) and to demonstrate that therapeutic interventions are effective, in the context of growing pressures from third party providers who fund mental health care. Therefore, it is now vital to be able to demonstrate that experiential therapy is effective for specific problem areas.


This review draws together findings from meta-analytic reviews, well-controlled randomised clinical trials (RCTs), and key texts or journal articles relevant to the current practice of experiential psychotherapy in Australia.

Search Strategy

Target keywords were entered in the PsycInfo database to generate a primary search list, which was expanded through use of Google Scholar and Web of Science. The list of keywords corresponded with names of each main therapy sub-type from the experiential psychotherapy tradition (e.g., ‘existential psychotherapy’ OR ‘emotion focused therapy’ OR ‘gestalt approach’). The full list of keywords and variations are shown in Appendix 1. Search results were limited to studies and articles published in peer-reviewed journals, written in English, and directly relevant to therapy. This search focused on publications from the last five years internationally, and the last 10 years in Australia. Recently published research was then integrated with findings of prior meta-analytic reviews to build on the current state of knowledge relevant to experiential psychotherapy.

Inclusion and Exclusion Criteria

All identified studies with well-established measures of symptom severity or behaviour change were included. In addition, this review includes qualitative research directly relevant to therapy outcomes. Purely theoretical articles, and process research without a clear link to therapy outcomes, were excluded. Likewise, treatment comparisons without rigorous methodology were excluded.

Due to the deeply relational nature of experiential psychotherapy, studies relevant to couple therapy, group therapy, and family therapy were included alongside individual psychotherapy. For ease of comparison of findings between studies, studies were rejected if the type of intervention was self-help, online, or provided by telephone.


A total of 14,767 search results were identified using the keywords shown in Appendix 1. This pool of initial results was narrowed down to 303 articles by limiting to peer reviewed texts, specifically about therapy and published within the last 10 years. Applying the inclusion and exclusion criteria shown above, a total of 14 new studies was identified for inclusion. The flowchart in Figure 1 provides a more detailed summary of articles included and excluded from the current review.


Five meta-analyses were identified for inclusion in the review.

Elliott, Greenberg, Watson, Timulak and Freire (2013) provide a substantial update on prior reviews which have appeared in Bergin and Garflield’s Handbook of Psychotherapy and Behavior Change. This meta-analytic review adds 77 additional studies since the last revision of the chapter in 2004, bringing the total number of included studies to 186. Data from the meta-analysis shows a revival of outcome studies focused on experiential psychotherapy over the last two decades.

Adding further specificity to the wider meta-analysis mentioned above, Elliott (2013) completed a more targeted review focused on anxiety disorders. This review provides more in-depth analysis and discussion relevant to the question of how recent outcome studies can inform current therapeutic practice in Australia. Elliott’s discussion adds some valuable reflections about the theoretical underpinnings of person-centred therapy, focusing-oriented therapy, and emotion-focused therapy, in populations living with anxiety.

Hölldampf, Behr, and Crawford (2010) review the treatment outcomes for humanistic counselling and play therapy for young people in a chapter of the text Person-centred and Experiential Therapies Work. The review builds on eight major meta-analyses, including data from 94 research projects, with outcomes linked both to specific mental health issues and context-specific problem areas experienced by children and adolescents.

A meta-analysis of outcome studies in existential psychotherapy conducted by Vos, Craig, and Cooper (2014) was also included. The review had limited inclusion criteria, requiring each study to explicitly identify the intervention as being existential in focus. Within those limitations, the reviewers identified 21 RCTs drawn from 15 samples comprised of 1,793 participants. Most studies described meaning-based therapies (e.g., logotherapy) or supportive-expressive therapies provided to medical patients with cancer and other medical illnesses.

This review also includes a qualitative meta-analysis by Timulak and Creaner (2010) to add further descriptive detail about therapy outcomes. This review consists of nine studies exploring the question, “what outcomes/effects are reported in qualitative studies investigating outcomes of humanistic therapies?” (p. 72). The findings of qualitative and process studies have been brought together with the outcome data shown above to summarise the state of current evidence.

Figure 1: Search process for the inclusion of recently published research


Initial search results via PsycInfo,
Google Scholar, and Web of Science 
(N = 14,767)


Did not meet initial criteria (N = 14,464)

Articles with peer review, about therapy, published in last 10 years
(N = 303) 


  • Interpretative phenomenological analysis (IPA) without reference to experiential therapy (N = 157)
  • Used a “constructivist approach” without reference to experiential therapy (N = 14)
  • Studies about management, education, coaching, or law enforcement, referring to “experiential learning” (N = 7)
  • Studies not in English (N = 12)
  • Already included in most recent review by Elliott et al., 2013 (N = 7)
  • Otherwise not relevant: theoretical papers, commentary, book reviews, not bona-fide therapy in study, and other unrelated research (N = 92)
  • Poor methodological rigour (N = 1)

Added to the review
(N = 13 additional studies)

  • Randomised (N = 8)
  • Pre-post Outcomes (N = 9)
  • With follow-up (N=4)
  • Process research (N=4)
  • Therapy comparison (N = 5)
  • With waitlist control (N = 3)

Summary information about the 13 additional studies is provided in Appendix 2.

General Findings

The meta-analysis by Elliott et al. (2013) is the largest and most comprehensive review of humanistic-experiential psychotherapies (HEP) to date. The review included a total of 14,206 individuals, with 62 studies compared to wait-list/control (31 of which were RCTs), and 135 studies directly comparing HEP to other therapies (82 of which were RCTs). The average length of treatment was 20 sessions.

Elliott et al. (2013) combined random effects model significance testing (Wilson & Lipsey, 2001) with equivalence analysis, allowing them to demonstrate the relative levels of equivalence between HEP and non-HEP approaches. They adopted the following conventions for interpreting the practical or clinical implication of differences in effect size: “Equivalent”: within .1 standard deviation of zero (greater than –.1 and less than .1); “Trivially Different”: between .1 and .2 standard deviation from zero; “Equivocal”: between .2 and .4 standard deviation from zero; “Clinically Better/Worse”: at least .4 standard deviation from zero. These criteria for comparing treatment effects are used across the meta-analysis.

Overall, pre-post effect sizes were large (d = .96). Weighted effect size for studies with a control group remained large (dw = .76), a finding which remained consistent in randomised samples. These findings indicate that approximately 80% of pre-post gains can be attributed to experiential therapy. Effect size differences between therapies were non-significant (-.02) in treatment comparison studies. Equivalence analyses revealed that HEP therapies were ‘equivalent’ to other approaches in general, however when compared to CBT interventions HEP obtained ‘trivially worse’ therapy outcomes (-.13). By contrast, HEP studies showed ‘trivially better’ (.17) effect sizes when compared to approaches other than CBT.

Researcher allegiance favoured HEPs in 65% of studies, however in studies with treatment comparisons, the allegiance was significantly lower at just 31% pro-HEP. This finding must be seen in the context of researchers from other theoretical orientations using relationship control conditions which tend to be labelled as ‘supportive’ or ‘non-directive’ therapy. The contribution of researcher allegiance has been found to represent a moderate and consistent effect on psychotherapy outcome studies across meta-analyses (Munder, Brütsch, Leonhart, Gerger, & Barth, 2013). Compounding this issue, allegiance often goes unreported in meta-analyses and only a tiny fraction of RCTs (3.2%) report it (Dragioti, Dimoliatis, & Evangelou, 2015). When equivalence analyses were repeated controlling for researcher allegiance, Elliott et al. (2013) found that all differences between therapies were reduced to ‘equivalent’ effect sizes (dw = -.03 to .06).

Experiential therapies vary in the degree to which therapists guide the process. Traditional person-centred therapy and supportive therapy generally have a lower degree of process-guiding, whereas other experiential approaches such as emotion-focused therapy have a much higher degree of process-guiding elements. The review by Elliott et al. (2013) classified HEP approaches as either low or high on process-guiding, then compared the effect sizes to see whether there were any measurable differences. Experiential approaches which were low on process-guiding were found to be trivially worse than CBT (-.16), whereas those approaches which were high on process-guiding were equivalent. This finding will be discussed further in the sections which follow, related to each problem area.


Overall, the review by Elliott et al. (2013) found equivalent effect sizes between experiential therapies and other approaches in treatment comparison trials. Therapies which were more process-guiding (EFT and Gestalt) had a clinically significant advantage over less process-guiding techniques (dw = .44). Elliott et al. identified three well-defined RCTs showing equivalent or superior efficacy of EFT for depression. Applying the revised criteria of Chambless and Hollon (1998) for empirically supported therapies (shown in Appendix 3), these results indicate that EFT is ‘efficacious and specific’ for the treatment of depression. Likewise, four well-designed RCTs for person-centred therapy confirm that PCT is ‘efficacious and specific’ for the treatment of perinatal depression. This review identified three new studies focused on depression which were not included in other meta-analyses.

A Korean study by Seo, Kang, Lee and Chae (2015) compared treatment outcomes of an 8-session model of narrative therapy with an emotional approach (NTEA) versus treatment as usual for a sample of chronically depressed individuals at a community mental health centre. Significant improvements were found on measures of hope, positive affect, and depressed mood in the NTEA group (p<.01). No change in self-awareness was detected, leading the researchers to conclude patients may need more time in therapy to make personally meaningful change.

Cornish and Wade (2015) studied treatment outcomes for 26 individuals who were experiencing excessive guilt over past wrongdoings, randomised either to individual EFT or waitlist. The researchers developed a brief manualised intervention which incorporates the four R’s of genuine self-forgiveness in EFT (i.e., responsibility, remorse, repair, and renewal). Those in the treatment condition reported statistically significant improvement, with reduced self-condemnation and psychological distress, in addition to increased self-forgiveness and self-compassion (Hedge’s g>0.7).

Ellison, Greenberg, Goldman and Angus (2009) provide the first study exploring maintained gains for emotion focused experiential therapy. Participants with moderate to severe depression were randomised to receive either client-centred therapy or EFT for 16 to 20 sessions. Those who responded positively to therapy (72% of those who completed treatment) were monitored across 18 months. At the 6-month post-therapy interval there were no statistically significant differences between groups. A difference emerged after 18 months, where survival analyses showed that first signs of depressive relapse presented at an average of 68 weeks in the EFT group versus 53 weeks in the client-centred therapy group.

Relationship and Interpersonal Issues

Elliott et al. (2013) highlight that the most consistently positive outcomes for experiential therapies are associated with clients who present with relational distress. The weighted mean effect size across 23 samples was large but variable (dw = 1.27) with larger effects found for couple or family therapy (dw = 1.5) by comparison to individual therapy (dw = .97). Of the 11 controlled comparisons (with seven RCTs) the review identified very large treatment effects (dw = 1.39). Superior treatment effects for relationship difficulties were found for experiential therapies compared to other therapies, including CBT (dw = .34). Emotion focused therapy (for individuals or couples) produced significantly larger effect sizes (dw = .69) than person-centred therapy (dw=-.08). Applying the criteria of empirically supported therapies (Chambless & Hollon, 1998), the data shows that EFT for both individuals and couples is ‘efficacious and specific’ in the care of people presenting with unresolved relationship issues. It is noteworthy that this population includes those who have survived abuse or have found themselves in relationships fractured by infidelity and/or trauma.

This review identified four additional studies relevant to relationship problems which were not included in the recent meta-analysis by Elliott et al. (2013).

A study by MacIntosh and Johnson (2008) tracked therapeutic outcomes for a small group of couples (N = 10) seeking EFT where one partner reported child sexual abuse and met the criteria for PTSD. The couples completed an average of 19 sessions. Half of those with a traumatic background of child sexual abuse reported clinically significant improvement to trauma symptoms at post-treatment. All survivors of child abuse showed clinically significant improvement on the Clinician Administered PTSD Scale, with 80% no longer meeting the criteria for PTSD at the end of the study. Qualitative analyses of client experiences in session identified emotional flooding, emotional numbing, dissociation, restricted range of affect, hypervigilance to attachment figures, and issues with sexuality in the relationship. The researchers recommend that therapists use additional structure in such work, gradually increasing the level of emotional challenge as therapy progresses. One particularly helpful suggestion is for therapists to carefully intervene during moments where trauma-based shame is being expressed. The intention here is to help survivors notice the love and care their partners feel for them in the present.

Dalgleish et al. (2015) conducted a pre-post intervention study investigating the potential predictor variables of positive outcomes with EFT for couples. A total of 31 couples showing attachment insecurity were assessed for attachment anxiety, attachment avoidance, emotional control, and relationship trust. Hierarchical linear modelling showed that those experiencing higher levels of emotional control and attachment anxiety reported greater changes in marital satisfaction after receiving therapy. Baseline measures of attachment avoidance and relationship trust were not significant predictors of outcome. The authors conclude that EFT for couples represents an opportunity for emotional connection for those with fears about disconnection, recommending that practitioners could identify such people prior to commencing therapy. Overall, the intervention showed a strong effect size (Cohen’s d = .81) on the Dyadic Adjustment Scale with 64% of couples meeting criteria for reliable change at post-test.

A study by McRae, Dalgleish, Johnson, Burgess-Moser and Killian (2014) looked specifically at predictor variables for blamer softening events in EFT for couples. In emotionally-focused couple therapy, the blamer softening event is a key indicator of treatment success. Blamer softening events occur when “a previously hostile/critical partner asks, from a position of vulnerability and within a high level of emotional experiencing, for reassurance, comfort, or for an attachment need to be met” (Bradley & Furrow, 2004, p. 234). Hence, blamer softening is theoretically regarded to correspond with repair of the relationship. A sample of 32 heterosexual couples with no diagnosed mental health conditions participated in an average of 21 sessions of EFT. Blamer softening events were not predicted by any of the identified intake variables, such as level of emotional control or emotional self-awareness. However, the researchers did observe that levels of emotional experiencing increased over the course of therapy. Although men reported significantly lower levels of emotional experiencing in best-rated sessions, gender was not predictive of blamer-softening events.

Another important process study by McKinnon and Greenberg (2013) examined the role of trust within the relationship and the expression of vulnerability in therapy. In this study, 25 couples received up to 12 sessions of EFT, where at least one person in the relationship reported feeling angry or hurt associated with a recent emotional injury. Couples rated sessions significantly more positively (p=.008) when vulnerable emotions were expressed, with no significant differences in partner perspective for that rating. Those couples who had at least one session where vulnerable emotions were expressed showed significantly greater trust at post-test (p=.016), but no significant differences on measures of dyadic adjustment or unfinished business.


The meta-analysis by Elliott et al. (2013) identified a significant pre-post effect size for the treatment of anxiety with experiential therapies overall (dw = .94), however the confidence interval for this calculation revealed significant variability in outcomes. Supportive therapy had considerably smaller treatment effects (dw = .66) than other experiential approaches. Experiential therapy was least effective for the treatment of anxiety, with a consistent and moderately significant effect favouring CBT (dw =-.39). Further analyses of the subtype of anxiety disorder show that the evidence clearly favours CBT for generalised anxiety disorder (dw=-.44), somewhat favours CBT for panic and agoraphobia (dw=-.39), and is equivocal for phobia and social anxiety (dw=-.15). There is a noticeable gap in the research on the effectiveness of experiential therapy for PTSD. After researcher allegiance effects were controlled, the overall difference in effect size was reduced to -.21 which is in the ‘equivocal’ range. Given these results, experiential therapies meet classification as possibly efficacious for the treatment of anxiety (Chambless & Hollon, 1998). Elliott et al. suggest that anxiety disorders may respond better to more structured therapy approaches.

In broader terms, the meta-analysis by Elliott et al. (2013) found that experiential approaches with more process-guiding elements were equivalent to less process-guiding therapies in RCTs (.08). However, the RCTs included in that meta-analysis show that experiential therapies with more process-guiding elements are trivially better than CBT (.12). In a separate meta-analytic study focused on anxiety, Elliott (2013) makes the point that experiential avoidance and a dysregulated sense of self are commonly reported by those with anxiety. He argues that a lack of structure in the early stages of therapy can be an unbearable experience for those struggling with anxiety. A more active and process-guiding stance may enhance experiential therapy when working with anxious client populations.

In his review of anxiety research, Elliott (2013) concludes that experiential therapies show reasonable pre-post effects (.88), with weaker effect sizes in controlled studies (.5), and negative effect sizes in comparison trials (-.39). It is worth noting that all but one of the comparisons were with some form of CBT, however, even when researcher allegiance effects were controlled statistically, the negative result for comparative effect size was still statistically significant (-.21). Elliott draws attention to the small number of studies investigating the use of experiential therapy with anxiety, but shows that more recent studies (such as the Strathclyde Social Anxiety Project) have found superior comparative effect sizes for EFT (.62) over non-directive person-centred approaches of up to 20 sessions.

A process study by Holowaty and Paivio (2012) relevant to anxiety disorders was identified in this review of recent studies. In this study, nine therapists provided individualised EFT for 29 people with memories of child abuse. Half of the sample met diagnostic criteria for PTSD with a mean level of symptom severity in the moderate range. Client-rated helpful events in therapy were associated with significantly greater levels of emotional arousal and were significantly more focused on memories of child abuse. The emotions identified by clients were primarily anger, followed by sadness and fear. Although depth of experiencing was found to be relatively consistent across therapy events, those who showed the deepest level of experiencing identified helpful events which evoked significantly greater depth for them. The researchers suggest that this finding may reflect that individuals differ in terms of their capacity to engage with emotional depth over time. This study also highlights the positive function of providing space for clients to express anger safely in therapy when working on childhood trauma.

Coping with Chronic Medical Conditions

Support groups using an experiential approach have commonly been applied to help people with serious medical illnesses. The meta-analysis by Elliott et al. (2013) found a moderate but highly inconsistent pre-post effect size (dw = .57) for experiential support groups overall. Weighted effect sizes varied by type of health condition, ranging from .68 for autoimmune conditions, .55 for early stage cancer, .62 for early/late cancer combined, to as low as .42 for other medical problems. Direct comparisons of experiential therapy to non-experiential therapy show a consistent finding of equivalence, with a weighted effect size of exactly zero. This general finding of equivalence to other established treatments for the same medical conditions led Elliott et al. to cautiously suggest that experiential therapies meet the criteria to be identified as efficacious for psychological coping (Chambless & Hollon, 1998), setting aside the problem of clustering various medical conditions together in the same category. It should be noted that a Cochrane review has concluded that there is insufficient evidence to support psychological group therapies of any kind as a routine treatment for women diagnosed with metastatic breast cancer (Edwards, Hulbert-Williams, & Neal, 2008). Given the moderate and positive post-therapy effect on psychological outcomes, obvious questions are raised about how to identify those cases where treatment is indicated.

This review identified two additional studies which could shed some light on how that question might be addressed. McLean, Walton, Rodin, Esplen and Jones (2013) carried out a randomised trial comparing EFT for couples with treatment as usual for 42 patients with end-stage terminal cancer. Up to eight sessions of support were provided to patients and their carers (as couples) over a 3-month period. A large post-treatment effect size was found for EFT (Cohen’s d = 1.00) on measures of marital functioning (p < .0001), which was maintained at 3-month follow-up. Improvement to marital distress was in the clinically significant range for 91% of patients in the EFT group, compared to 28% in the control group (χ2 = 16.8, p<0.0001). In addition, patients who received EFT reported a statistically significant improvement to their perceptions of carer empathy. Although there was no significant improvement in depression over that brief period of therapy, the reported gains to relationship functioning and perception of empathy are clearly important to consider in the management of terminal cancer.

Another study by Herschbach et al. (2009) explored the role of group therapy in treating fears of illness progression among an in-patient sample of 348 patients with a diagnosis of either chronic arthritis or cancer. Patients were randomised to receive either supportive expressive group therapy, CBT group therapy, or treatment as usual across four sessions. Outcomes were assessed at post-test and follow up at three months and 12 months. Cancer patients showed improvement to dysfunctional fears of illness progression, with equivalent treatment effect sizes for both CBT (.54) and supportive-expressive (.5) group therapy. It is worth noting that measures of depression, health-related quality of life, and life satisfaction decreased across time for cancer patients staying in hospital, however, it was only those receiving group therapy who continued to improve after discharge. Neither of the active treatments were successful in significantly improving outcomes for those diagnosed with chronic arthritis. Given that people with chronic arthritis reported higher baseline fears of illness progression and lower physical quality of life (both at p<0.001), the researchers conclude that specific illness characteristics must be factored into consideration when selecting appropriate forms of psychological care for individual patients.

Children and Young People

A review by Hölldampf et al. (2010) provides a recent update of outcome studies about person-centred and experiential psychotherapies with children and young people. The review identified 83 unique studies of which 34 were RCTs. Broadly speaking, the provision of experiential therapy led to positive outcomes in the treatment of mood disorders, anxiety, adjustment disorder, PTSD, mental retardation and developmental disorders, and ADHD. The largest treatment effects were found in the treatment of children experiencing symptoms of distress related to adjustment reactions to stressful life events, trauma, and anxiety disorders. In addition to symptom measures, the authors note that most studies included other relevant variables to the target problem, including social skills, emotional adjustment and assessment of the parental relationship. The chapter concluded with a brief mention of a meta-analysis of outcome studies aiming for publication in the coming years.

McArthur et al. (2013) explored the therapeutic outcomes of school-based humanistic counselling in the UK. Young people aged 13 to 16 were randomly assigned to either therapy or waitlist with pre and post-test of psychological distress and adjustment after 12 weeks. Those who were assigned to humanistic counselling showed significantly greater reduction of distress than controls on the CORE (Hedges g=1.14) with strong treatment effects also found on measures of adjustment (g=.73).

In Australia, Havighurst, Wilson, Harley and Prior (2009) evaluated a parenting program informed by the emotion-focused approach called ‘Tuning in to kids’. A sample of 218 parents of children aged four to five years were randomised to either waitlist or treatment in a 6-week group program on parenting. Parent self-evaluations before and after treatment showed large (h2) treatment effect sizes for increased emotion coaching and decreased emotion dismissing. Child problem behaviour was significantly improved, with only 13% of parents who participated in the group indicating problems in the clinically significant range at post-test. By comparison, parents in the waitlist indicated that clinically significant child problem behaviour was relatively unchanged (26% at pre-test and 28% at post-test).

A more recent trial by Havighurst et al. (2015) evaluated a similar intervention for a larger sample of parents of children in lower school aged between five and eight years. Child behaviour problems were identified by high scores on combined parent and teacher ratings. In this study both parents and children were provided with an 8-session program, with teachers of those children also being provided with six hours of training on building more emotionally responsive teacher-student relationships. The researchers report significant decreases in emotion dismissing and increased empathy, with moderate effect sizes for both. Although some positive improvement occurred in the waitlist across time, children who participated in the intervention showed significantly greater improvement in their emotional understanding.


Elliott et al. (2013) offer cautious conclusions regarding the state of current evidence relating to work with psychosis. The reviewers point out that guidelines in the UK (NICE, 2014) state: “Do not routinely offer counselling and supportive psychotherapy (as specific interventions) to people with schizophrenia”. The impact of such a statement in terms of public doubt has been profoundly negative. Support groups for people living with schizophrenia across the last few decades in the UK are now no longer offered. Elliott et al. reanalysed the same data from the nine studies referred to in the NICE guidelines, revealing a trivially small advantage of CBT with psychosis (d = -.19) with significant variability in outcomes. Elliott points out that many of the therapy approaches described as ‘person-centred’ by the NICE guideline were in fact a mixture of supportive interventions, only some of which were bona fide experiential approaches.

The wider meta-analysis by Elliott et al. (2013) includes six outcome studies on the treatment of psychosis with a large weighted effect size overall (dw = 1.08). There were no studies comparing humanistic experiential therapies to waitlist controls. Five comparative treatment trials in the review showed a consistent and moderately positive effect size (dw = .39). Using the Chambless and Hollon (1998) criteria, this finding classifies experiential therapy as ‘possibly efficacious’.

Other Conditions

Studies about various other problem areas were also included in the review by Elliott et al. (2013), including substance abuse and eating disorders. Experiential approaches for substance misuse have been found to be superior to non-treatment controls and either equivalent or superior to an already established treatment (i.e., CBT) in at least two independent trials. This meets criteria as an ‘efficacious’ treatment approach, with a strong effect size for working with substance misuse (dw=.68). As the review only identified two small studies about overeating, with a finding of no difference between treatments, the results remain equivocal for eating disorders.

Qualitative Outcomes

Timulak and Creaner (2010) carried out a qualitative meta-analytic synthesis which drew together descriptive report of clients about positive and negative outcomes of humanistic therapies. Data were drawn from nine studies involving 108 clients using a variety of qualitative methods; in most cases emotion-focused therapy or person-centred therapy. Positive outcomes were nested in three main meta-categories (1) appreciating experiences of self, (2) appreciating experiences of self in relation to others, and (3) a changed view of self/others. Negative outcomes included a sense that problems were not fully resolved, being overwhelmed, feeling harmed by the therapist, fears of changing, and disappointment at not being understood. The reviewers note the overlap of their findings with prior meta-analyses on the same topic (Elliott, 2002), identifying that their analysis revealed several qualitatively distinct additional categories: appreciating vulnerability, enjoying change, and feeling supported. It is interesting to note that these subjective outcomes align with the goals and underlying philosophy of experiential approaches.

Facilitative Conditions

The core therapist-offered conditions proposed by Rogers (1957) as necessary and sufficient for change have been explored by researchers both individually and in the context of efficacy research for person-centred therapy (PCT). Early meta-analyses identify medium effect sizes for PCT (Smith, Glass, & Miller, 1980) with larger effects in practice settings (d = 1.32), equivalent to CBT (d = 1.27) and psychodynamic therapy (d = 1.23) (Stiles, Barkham, Twigg, Mellor-Clark, & Cooper, 2006). The review by Elliott et al. (2013) found PCT to be only equivocally less effective than CBT, however it was noted that PCT was often used as a minimum-training relationship control condition, meaning that in most cases what was described as PCT was not actually intended to be therapeutic.

Taken separately, a significant proportion of outcome variance can be apportioned to each of the facilitative conditions. Wampold and Imel (2015) report that up to 9% of outcome variance can be apportioned to therapist empathy. However, Elliott, Bohart, Watson and Greenberg (2011) point out that empathy is a complex construct. Empathy appears to be less predictive of outcomes with more experienced therapists. Moreover, not all clients respond well to open expressions of empathy. Empathy also seems to be more closely linked to outcome in CBT (r=.31) but less associated with outcome in psychodynamic therapy (r=.19) compared to experiential therapy (r=.26). A review by Farber and Doolin (2011) shows that positive regard is more closely associated with positive outcomes in psychodynamic therapy (r=.52) than other approaches (r=.12 to .33). Kolden, Klein, Wang and Austin (2011) review the status of research about the contribution of congruence, reporting that outcomes are mixed but mainly positive. Effect sizes range from negative (r = -.26) to strongly positive (r = .69), with the overall weighted mean effect size for congruence showing a moderately positive effect (r = .24).

Client ratings of therapist empathy predict improvement in client attachment style post-therapy: leading to reports of less insecurity, more self-acceptance, and greater levels of self-protection (Watson, Steckley, & McMullen, 2014). Recent studies highlight the mediating role of affect regulation on in-session emotional processing, working alliance, and outcome (Watson, McMullen, Prosser, & Bedard, 2011). This evidence is consistent with the view that empathy offered by a therapist cultivates self-empathy in a client, which may in turn assist people to make positive life changes (Barrett-Lennard, 2015). In summary, whether investigation has targeted the facilitative conditions as separate elements or as a unified treatment approach in the form of PCT, these core conditions for therapy remain central to psychotherapy practice.

Therapeutic Alliance

Elliott et al. (2013) report that the accumulated evidence indicates that there is a moderately strong relationship between the therapeutic alliance and outcome, setting aside some of the complex factors involved and methodological problems in the research to date. Watson and Geller (2006) have noted the conceptual overlap of facilitative conditions with the therapeutic alliance, drawing attention to the strong and positive correlation of ratings on the Barrett-Lennard Relationship Inventory (BLRI) and alliance measures (.72). The review by Elliott et al. (2013) draws together a range of studies which suggest that relationship variables in therapy may play an important role in subsequent changes to client affect regulation, attachment style, and other positive therapy outcomes. Despite the attention that the alliance has received in research in recent decades, significant questions remain about the definition of the construct and the direction of the relationship between alliance and outcome (Doran, 2016). Nevertheless, the therapeutic alliance remains the most robust predictor of outcome across the common factors (Tracey, Lichtenberg, Goodyear, Claiborn, & Wampold, 2003).

Depth of Experiencing

Watson, Greenberg and Lietaer (2010) review the research relevant to both the therapeutic alliance and the client’s experiential self-reflection (aka ‘depth of experiencing’). These two psychotherapy process factors are central to the theory of psychological change proposed by Carl Rogers (1957). Although Watson et al. (2010) note that interpretation of the current state of research is limited by common methodological flaws, the authors do reveal some valuable findings. With widely used measures such as the Experiencing Scale, researchers have explored the degree to which people are engaged in their experiences, from detached narration through to a more focused, open, and free-flowing synthesis of feelings and meanings. Watson et al. (2010) report a consistent relationship between depth of experiencing and positive outcomes, but that depth of experiencing does not appear to increase over time, as initially proposed by Rogers (1961). Conversely, some research does suggest that depth of experiencing may increase from early to late stages of therapy when ratings are obtained from segments of therapy linked more directly to the core problems identified by the client (e.g., Goldman, Greenberg, & Pos, 2005).

Watson et al. (2010) review studies showing that exploration of emotions in session has been found to be beneficial across a range of psychotherapy approaches (e.g., PCT, CBT, psychodynamic). People who achieve more favourable results from psychotherapy begin, continue, and end with greater depth of exploration, refer to their emotions more often, are more inwardly focused, and draw from their experiences to find new meaning and purpose.

A frequency of 25% of time spent in therapy focused on moderate to high levels of emotional arousal has been associated with optimal therapy outcomes. Consistent with emotion-focused theories of change, the expression of greater levels of emotion at mid-treatment has been linked with better outcomes (Pos, Greenberg, & Warwar, 2009), however there is some indication that the strength of this effect may be constrained by earlier processes in therapy.

Structured approaches

Vos et al. (2014) conducted a meta-analysis with entry criteria strictly limited to studies which explicitly stated that the approach being investigated was existential in focus. A total of 21 RCTs were identified from 15 samples with 1,792 participants, with most studies describing meaning-based therapies or supportive-expressive approaches for clients with cancer or other chronic physical illness. For these structured approaches, large post-treatment effect-sizes were identified post-intervention (d = 0.65) and follow-up (d = 0.57), with moderate effects on psychopathology (d = 0.47) and self-efficacy (d = 0.48). No significant effects were found on measures of self-reported physical health. By comparison, supportive-expressive therapy had small effects at post-treatment and follow-up on psychopathology (d = 0.20 and 0.18, respectively). No significant effects on self-efficacy and self-reported physical well-being were identified. Vos et al. (2014) note the similarity of outcomes for structured existential approaches by comparison to other psychological interventions widely used for this client group (e.g., mindfulness and acceptance-based therapies and support groups).


This review has found clear and consistent evidence for the effectiveness of experiential approaches for depression and unresolved relational issues. It is noteworthy that this therapeutic modality meets the highest standard of evidence (i.e., ‘specific and efficacious’). Yet there is no indication that experiential therapies will be added to the list of approved therapies for use in government-funded programs in Australia, for example the Better Access initiative. Research on the application of EFT to cases of relationship dysfunction involving PTSD and attachment insecurity requires more replication to confirm the early promising findings identified in this review.

The evidence supporting experiential therapy for anxiety disorders is inconsistent and comparatively weaker. The ‘possibly efficacious’ status of experiential therapy warrants further attention by researchers to clarify whether bringing further structure into these approaches could improve outcomes. Given that all but one of the RCT comparisons were made to CBT with ‘supportive therapy’ being generally intended as a control group, it is unclear whether bona fide therapies were being applied. The present state of uncertainty about therapies used as control conditions makes it difficult to interpret findings on a wider scale. Researchers should ensure that training and allegiance of therapists in both types of therapy being compared should be reasonably consistent in future RCTs. Alternatively, researchers should consider more clearly labelling specific types of therapy (rather than simply calling them ‘supportive’) so that reviewers do not erroneously include relationship control groups which were not intended to be therapeutic.

This review highlights the need for more comparison trials to be carried out on the use of experiential therapy to manage psychosis. While the current data indicates that this class of therapy is ‘possibly efficacious’, there are similar problems in the psychosis research to those described for anxiety disorders. At present, the NICE guidelines in the UK specify that people with psychosis should be routinely offered CBT rather than supportive therapy (NICE, 2014), however there is little, if any data, to suggest the superiority of CBT. For example, a Cochrane review found that CBT is no better or worse than other talking therapies for schizophrenia (Jones, Hacker, Cormac, Meaden, & Irving, 2012). Other recent reviews show modest effects of CBT for psychosis, which has raised concerns that public treatment guidelines over-promote CBT and understate other viable alternatives (Jauhar, McKenna, Radua, Fung, Salvador, & Laws, 2014).

The research on experiential therapies for medical conditions is of considerable importance. Rates of co-morbid mental health disorders remain high for those diagnosed with cancer, with estimates ranging from 20% to 50% of such patients also meeting diagnostic criteria for depression or anxiety (Pasquini & Biondi, 2007). Conversely, longitudinal research spanning 18 years has now identified depression as an independent risk factor for coronary heart disease in women, with a greater magnitude of association between these conditions than other typical and atypical risk factors (O’Neill et al., 2016). The close association of high prevalence mental disorders and serious physical illness highlights the need for flexible and personally relevant psychotherapy approaches. Rather than attempting to directly target physical symptoms, the aim of such therapies is to improve quality of life, increase levels of interpersonal support, and optimise meaningful recovery wherever possible.

This review identifies that experiential therapies are equivalent to other well-established therapies aimed at improving psychological coping. There is also evidence that structured meaning-based existential therapies can effectively reduce psychopathology and increase self-esteem for people with cancer and other chronic health problems. The benefits of therapy were not merely temporary but sustained at follow-up. On this basis, it is suggested that experiential approaches should be made accessible to people with co-occurring mental disorders and serious health conditions such as cancer.

This review highlighted a collection of studies that confirm the ongoing relevance of the therapist-offered facilitative conditions proposed by Rogers (1957), most notably therapist empathy. Emerging process research on therapist empathy suggests a role in helping clients develop self-compassion and self-care, which again is consistent with humanistic theories of change. It is unclear from the current state of research whether depth of experiencing increases as therapy progresses, however the expression of deeper levels of emotion in therapy in the early and middle stages of therapy has been associated with better outcomes. Further research is needed to establish a clearer link between process factors in experiential therapy and later outcomes.


Reviewers across this body of literature highlight methodological challenges which limit the conclusions which can be drawn from the existing research. Elliott et al. (2013) draw attention to variation in the level of rigour across qualitative studies, and the need for more consistently high standards. Process and outcome studies of experiential therapy face challenges around the use of inconsistent measures, relying on small samples without follow-up, and variability in the level of therapist experience (Krycka & Ikemi, 2016; Watson et al., 2010). The review noted significant differences between studies for which a segment of therapy was sampled, in which detection of the key therapeutic process is paramount. Murphy and Joseph (2016) raise concerns that the common single group pre-test post-test design of studies of humanistic therapies does not allow us to draw conclusions about comparative efficacy. In addition, it is noteworthy that comparisons of high versus low process-guiding in most cases are carried out by researchers who have an allegiance with high process-guiding therapies (generally EFT). Allegiance effects remain an ongoing issue in psychotherapy research, and as this review has shown, they can have subtle but far ranging consequences.


There is a need for experiential therapists to co-develop and take part in large-scale research projects. The limitations and gaps in the research identified in this review can be addressed, but will require support from universities and organisations dedicated to promoting psychotherapy such as PACFA. A major obstacle for Australia is the recent closure of multiple postgraduate training programs where experiential approaches were a key component of the course (Di Mattia & Grant, 2016). It is unlikely that research of the required type will be carried out in Australia if there are no postgraduate training institutions bringing new therapists and researchers into these approaches to therapy.

Despite these challenges, counselling and psychotherapy practitioners remain dedicated and engaged with experiential therapy, a modality which clearly deserves greater recognition and support in Australian public policy. The findings of this review make it clear that this therapy modality should be approved for use with depressive disorders and in cases of significant relationship dysfunction. It is recommended that structured experiential approaches should also be made accessible to people with co-morbid medical issues, to align with client treatment preferences. Experiential work with anxiety disorders should be managed more carefully, supporting clients who prefer these approaches with a greater level of structure and increased therapist guidance around therapeutic processes.


Angus, L., Watson, J. C., Elliott, R., Schneider, K., & Timulak, L. (2015). Humanistic psychotherapy research 1990–2015: From methodological innovation to evidence-supported treatment outcomes and beyond. Psychotherapy Research, 25(3), 330–347.

Barrett-Lennard, G. T. (2015). The relationship inventory: A complete resource and guide. Chichester, England: John Wiley & Sons.

Bradley, B., & Furrow, J. L. (2004). Toward a mini-theory of the blamer softening the moment by moment process. Journal of Marital and Family Therapy, 30, 233–246.

Cain, D. J., Keenan, K., & Rubin, S. (2016). Humanistic psychotherapies: Handbook of research and practice (2nd ed.). Washington, DC: American Psychological Association.

Chambless, D. L., & Hollon, S. D. (1998). Defining empirically supported therapies. Journal of Consulting and Clinical Psychology, 66(1), 7–18.

Cornish, M. A., & Wade, N. G. (2015). Working through past wrongdoing: Examination of a self-forgiveness counseling intervention. Journal of Counseling Psychology, 62(3), 521–528.

Dalgleish, T. L., Johnson, S. M., Moser, M. B., Lafontaine, M., Wiebe, S. A., & Tasca, G. A. (2015). Predicting change in marital satisfaction throughout emotionally focused couple therapy. Journal of Marital and Family Therapy, 41(3), 276–291.

Di Mattia, M., & Grant, J. (2016). Counselling psychology in Australia: History, status and challenges. Counselling Psychology Quarterly, 29(2), 139–149.

Doran, J. M. (2016). The working alliance: Where have we been, where are we going? Psychotherapy Research, 26(2), 146–163.

Dragioti, E., Dimoliatis, I., & Evangelou, E. (2015). Disclosure of researcher allegiance in meta-analyses and randomised controlled trials of psychotherapy: A systematic appraisal. British Medical Journal Open, 5(6), 1–10.

Edwards, A. G., Hulbert-Williams, N., & Neal, R. D. (2008). Psychological interventions for women with metastatic breast cancer. Cochrane Database Systematic Reviews, (3). doi: 10.1002/14651858.CD004253.pub3.

Elliott, R. (2002). Render unto Caesar: Quantitative and qualitative knowing in research on humanistic therapies. Person-centred and Experiential Psychotherapies, 1, 102–117.

Elliott, R. (2013). Person-centred/experiential psychotherapy for anxiety difficulties: Theory, research and practice. Person-centred and Experiential Psychotherapies, 12(1), 16–32.

Elliott, R., Bohart, A. C., Watson, J. C., & Greenberg, L. S. (2011). Empathy. In J. C. Norcross (Ed.), Psychotherapy relationships that work (2nd ed., pp. 132–152). New York, NY: Oxford University Press.

Elliott, R. Greenberg, L. S., Watson, J., Timulak, L., & Freire, E. (2013). Research on humanistic-experiential psychotherapies. In M. J. Lambert (Ed.), Bergin & Garfield’s Handbook of psychotherapy and behavior change (6th ed., pp. 495–538). New York, NY: Wiley.

Ellison, J. A., Greenberg, L. S., Goldman, R. N., & Angus, L. (2009). Maintenance of gains following experiential therapies for depression. Journal of Consulting and Clinical Psychology, 77(1), 103–112.

Farber, B. A., & Doolin, E. M. (2011). Positive regard and affirmation. In J. C. Norcross (Ed.), Psychotherapy relationships that work (2nd ed., pp. 168–186). New York, NY: Oxford University Press.

Goldman, R. N., Greenberg, L. S., & Pos, A. E. (2005). Depth of emotional experience and outcome. Psychotherapy Research, 15, 248–260.

Havighurst, S. S., Duncombe, M., Frankling, E., Holland, K., Kehoe, C., & Stargatt, R. (2015). An emotion-focused early intervention for children with emerging conduct problems. Journal of Abnormal Child Psychology, 43(4), 749–760.

Havighurst, S. S., Wilson, K. R., Harley, A. E., & Prior, M. R. (2009). Tuning into kids: An emotion-focused parenting program – Initial findings from a community trial. Journal of Community Psychology, 37(8), 1008–1023.

Herschbach, P., Berg, P., Waadt, S., Duran, G., Engst-Hastreiter, U., Henrich, G., … Dinkel, A. (2009). Group psychotherapy of dysfunctional fear of progression in patients with chronic arthritis or cancer. Psychotherapy and Psychosomatics, 79(1), 31–38.

Hölldampf, D., Behr, M., & Crawford, I. (2010). Effectiveness of person-centred and experiential psychotherapies with children and young people: A review of outcome studies. In M. Cooper, J. C. Watson, & D. Hölldampf (Eds.), Person-centred and experiential therapies work (pp. 16–44). Ross-on-Wye, England: PCCS Books.

Holowaty, K. A. M., & Paivio, S. C. (2012). Characteristics of client-identified helpful events in emotion-focused therapy for child abuse trauma. Psychotherapy Research, 22(1), 56–66.

Jauhar, S., McKenna, P. J., Radua, J., Fung, E., Salvador, R., Laws, K. R. (2014). Cognitive-behavioural therapy for the symptoms of schizophrenia: Systematic review and meta-analysis with examination of potential bias. The British Journal of Psychiatry, 204(1), 20–29

Jones, C., Hacker, D., Cormac, I., Meaden, A., & Irving, C. B. (2012). Cognitive behavioural therapy versus other psychosocial treatments for schizophrenia. Retrieved from http://onlinelibrary.wiley.com/doi/10.1002/14651858.CD008712.pub2

Kolden, G. G., Klein, M. H., Wang, C., & Austin, S. B. (2011). Congruence/genuineness. In J. C. Norcross (Ed.), Psychotherapy relationships that work (2nd ed.) (pp. 187–202). New York, NY: Oxford University Press.

Krycka, K. C., & Ikemi, A. (2016). Focusing-oriented experiential psychotherapy: From research to practice. In D. J. Cain, K. Keenan, & S. Rubin (Eds.), Humanistic psychotherapies: Handbook of research and practice (2nd ed., pp. 251–282). Washington, DC: American Psychological Association.

MacIntosh, H. B., & Johnson, S. (2008). Emotionally focused therapy for couples and childhood sexual abuse survivors. Journal of Marital and Family Therapy, 34(3), 298–315.

McKinnon, J. M., & Greenberg, L. S. (2013). Revealing underlying vulnerable emotion in couple therapy: Impact on session and final outcome. Journal of Family Therapy, 35(3), 303–319.

McLean, L. M., Walton, T., Rodin, G., Esplen, M. J., & Jones, J. M. (2013). A couple-based intervention for patients and caregivers facing end-stage cancer: Outcomes of a randomized controlled trial. Psycho-Oncology, 22(1), 28–38.

McRae, T. R., Dalgleish, T. L., Johnson, S. M., Burgess-Moser, M., & Killian, K. D. (2014). Emotion regulation and key change events in emotionally focused couple therapy. Journal of Couple and Relationship Therapy, 13(1), 1–24.

Munder, T., Brütsch, O., Leonhart, R., Gerger, H., & Barth, J. (2013). Researcher allegiance in psychotherapy outcome research: An overview of reviews. Clinical Psychology Review, 33(4). 501–511.

Murphy, D., & Joseph, S. (2016). Person-centred therapy: Past, present, and future orientations. In D. J. Cain, K. Keenan, & S. Rubin (Eds.), Humanistic psychotherapies: Handbook of research and practice (2nd ed., pp. 185–218). Washington, DC: American Psychological Association.

National Institute for Health and Care Excellence (NICE). (2014). Psychosis and schizophrenia in adults: Prevention and management. NICE clinical guideline (CG178). Retrieved from https://www.nice.org.uk/guidance/cg178

O’Neill, A., Fisher, A. J., Kibbey, K. J., Jacka, F. N., Kotowicz, M. A., Williams, L. J., … Pasco, J. A. (2016). Depression is a risk factor for incident coronary heart disease in women: An 18-year longitudinal study. Journal of Affective Disorders, 196, 117–124.

Pasquini, M., & Biondi, M. (2007). Depression in cancer patients: A critical review. Clinical Practice and Epidemiology in Mental Health, 3(1), 2. doi: http://doi.org/10.1186/1745-0179-3-2

Pos, A. E., Greenberg, L. S., & Warwar, S. (2009). Testing a model of change for experiential treatment of depression. Journal of Consulting and Clinical Psychology, 77(6), 1055–1066.

Rogers, C. R. (1957). The necessary and sufficient conditions of therapeutic personality change. Journal of Consulting Psychology, 21, 95–103.

Rogers, C. R. (1961). On becoming a person. Boston, MA: Houghton Mifflin.

Seo, M., Kang, H. S., Lee, Y. J., & Chae, S. M. (2015). Narrative therapy with an emotional approach for people with depression: Improved symptom and cognitive-emotional outcomes. Journal of Psychiatric and Mental Health Nursing, 22(6), 379–389.

Smith, M. L., Glass, G. V., & Miller, T. I. (1980). The benefits of psychotherapy. Baltimore, DC: John Hopkins University Press.

Stiles, W. B., Barkham, M., Twigg, E., Mellor-Clark, J., & Cooper, M. (2006). Effectiveness of cognitive-behavioural, person-centred and psychodynamic therapies as practiced in UK National Health Service settings. Psychological Medicine, 36, 555–566.

Timulak, L., & Creaner, M. (2010). Qualitative meta-analysis of outcomes of person-centred and experiential psychotherapies. In M. Cooper, J. C. Watson, & D. Hölldampf (Eds.), Person-centred and experiential therapies work (pp. 65–90). Ross-on-Wye, England: PCCS Books.

Tracey, T. J. G., Lichtenberg, J. W., Goodyear, R. K., Claiborn, C. D., & Wampold, B. E. (2003). Concept mapping of therapeutic common factors. Psychotherapy Research, 13(4), 401–413.

Vos, J., Craig, M., & Cooper, M. (2015). Existential therapies: A meta-analysis of their effects on psychological outcomes. Journal of Consulting and Clinical Psychology, 83(1), 115–128.

Wampold, B. E., & Imel, Z. E. (2015). The great psychotherapy debate (2nd ed.). New York, NY: Routledge.

Watson, J. C., Greenberg, L. S., & Lietaer, G. (2010). Relating process to outcome in person-centred and experiential psychotherapies. In M. Cooper, J. C. Watson, & D. Hölldampf (Eds.), Person-centred and experiential therapies work (pp. 132–263). Ross-on-Wye, England: PCCS Books.

Watson, J. C., McMullen, E. J., Prosser, M. C., & Bedard, D. L. (2011). An examination of the relationships among clients’ affect regulation, in-session emotional processing, the working alliance, and outcome. Psychotherapy Research, 21(1), 86–96.

Watson, J. C., Steckley, P. L., & McMullen, E. J. (2014). The role of empathy in promoting change. Psychotherapy Research, 24(3), 286–298.

Wilson, D. B., & Lipsey, M. W. (2001). Practical meta-analysis. Thousand Oaks, CA: Sage.


Appendix 1

The list of keywords for the name of each approach was as follows:

Experiential, focusing oriented, process experiential, emotion focused (or emotionally focused), existential, phenomenological, phenomenology, intersubjective, constructivist, logotherapy, psychodrama, gestalt, Hakomi,

The above list was paired with the following keywords:

Approach, therapy, psychotherapy, counselling, counseling, enquiry, psychology, analysis

Appendix 2

Table 1: Studies meeting inclusion criteria




Targeted problem Area(s)

Participant Characteristics


Cornish & Wade (2015)

Randomised trial comparing treatment to waitlist with 2-month follow-up.

Brief model of
EFT for individuals
(8 sessions)

Psychological distress (CORE)

Self-condemnation, self-condemnation, and self-compassion

26 individuals with guilt over prior wrongs

Significantly reduced self-condemnation and psychological distress, in addition to increased self-forgiveness and self-compassion (Hedge’s g>.07). Gains were maintained at 2-month follow-up.

Dalgleish et al. (2015)

Pre-post design outcome study

EFT for couples
(21 session mean)

Predictor variables

(attachment anxiety, emotional control, and trust)

Relationship adjustment

31 couples

Strong effect size (Cohen’s d = .81) on the Dyadic Adjustment Scale (DAS) with 64% showing reliable change

High emotional control and attachment anxiety associated with greater change. Baseline attachment avoidance and relationship trust not significantly predictive.

Elliott et al. (2013)


Humanistic Experiential Psychotherapies
(20 session mean)


14,206 individuals

Large pre-post effects (d = .96) with treatment comparisons generally non-significant (-.02). CBT trivially better (-.13) however researcher allegiance reduced to equivalent effect size range (dw = -.03 to .06)

Ellison et al. (2009)

Randomised comparison of treatment with 6-month and 18-month follow-up

EFT for individuals versus client-centred therapy
(16 to 20 sessions)

Depression (moderate to severe)

43 individuals

No statistically significant differences between groups at 6-months. Survival analysis shows first depressive relapse occurs at average of 53 weeks in client-centred group versus 68 weeks in EFT group.

Havighurst et al. (2009)

Randomised trial comparing treatment to waitlist

Emotion-focused group intervention for parents

(6 weeks)

Behaviour problems

Parental management of child’s emotions

218 parents of pre-school children with behaviour issues

Child problem behaviour significantly improved in treatment group. Parent self-evaluation improved on management of child’s emotions.

Havighurst et al. (2015)

Randomised trial comparing treatment to waitlist

Emotion-focused group intervention for parents, teachers, and children
(8 sessions)

Behaviour problems


204 primary caregivers and their children, with identified behaviour issues

Significant decreases in emotion dismissing and increased empathy, with moderate effect sizes for both. Although some positive improvement occurred in the waitlist across time, children who participated in the intervention who showed significantly greater improvement to emotional understanding and the targeted behaviour problems

Herschbach et al. (2009)

Randomised comparison of treatments

Supportive-expressive group therapy versus CBT group therapy versus TAU
(4 sessions only)

Dysfunctional fears of illness progression, depression, health-related quality of life, and life satisfaction

348 patients diagnosis with either chronic arthritis or cancer

Cancer patients report significant improvement to dysfunctional fears of illness progression over time. Only those patients who attended group therapy continue to improve after discharge. Neither form of group therapy improved outcomes for patients with chronic arthritis.

Hölldampf et al. (2010)


Person-centred and experiential therapies
(variable length)


Children from 83 individual studies

Reports positive outcomes for mood disorders, anxiety, adjustment disorder, PTSD, mental retardation and developmental disorders, and ADHD. Largest effects were found in the treatment distress related to stressful life events, trauma, and anxiety disorders

Holowaty & Paivio (2012)

Process study

EFT for individuals

Client-rated helpful events, emotional arousal, and depth of experiencing

29 individuals with memory of child abuse

Helpful events identified by clients were associated with significantly greater emotional arousal and linked to memories of child abuse.

MacIntosh & Johnson (2008)

Pre-post design outcome study with process variables


EFT for couples
(19 session mean)

Symptoms of trauma and PTSD

Relationship satisfaction

Subjective report of client experiences in session

10 couples with partner history of child sexual abuse and PTSD

Half of couples show clinically significant improvement to trauma symptoms and satisfaction with relationship at treatment completion. 80% of trauma survivors no longer met PTSD criteria at completion.

McArthur et al. (2013)

Randomised trial with waitlist control group

School-based humanistic counselling versus waitlist control
(up to 9 sessions)

Psychological distress (YP-CORE)

33 young people aged 13 to 16 years

Young people who received humanistic counselling showed significantly greater reduction of distress than controls on the CORE (Hedges g=1.14) with strong treatment effects also found on measures of adjustment (g=.73)

McKinnon & Greenberg (2013)

Process study

EFT for couples
(up to 12 sessions)

Partner trust, dyadic adjustment, and unfinished business


25 couples with a partner angry/hurt by recent injury to emotions

Couples rated sessions significantly more positively (p=.008) when vulnerable emotions were expressed. Greater levels of trust when couples have at least one session where vulnerable emotions are expressed, but no difference to unfinished business or dyadic adjustment.

McLean et al. (2013)

Randomised trial comparing to waitlist with 3-month follow-up

EFT for couples
(up to 8 sessions)

Depression, marital functioning, and perception of carer empathy

42 patients with terminal cancer

Large post-treatment effect (d = 1.00) on measures of marital functioning maintained at 3-month follow-up. Improvement to marital distress was in the clinically significant range for 91% of patients in the EFT group compared to 28% in the control group. Statistically significant improvement to perceptions of carer empathy. No significant improvement to depression.

McRae et al. (2014)

Process study

EFT for couples
(21 session mean)

Predictor variables for blamer softening events in therapy

32 couples

Blamer softening not predicted by intake variables such as emotional control or emotional self-awareness. Levels of emotional experiencing increase over course of therapy.

Seo et al. (2015)

Quasi-experimental trial comparing to TAU

Narrative therapy with emotional approach versus treatment as usual
(8 sessions)


50 individuals with chronic depression medicated for 10+ years

Significant improvements to hope, positive affect, and depression (p<.01). No significant change to self-awareness.

Timulak & Creaner (2010)

Meta-analytic synthesis

Humanistic therapies
(variable length)

Qualitative outcomes of therapy

108 clients

Three main meta-categories identified for outcomes of humanistic therapy: (1) appreciating experiences of self, (2) appreciating experiences of self in relation to others, and (3) a changed view of self/others.

Vos et al. (2014)


Existential therapy


1792 individuals

Large post-treatment effect-sizes for meaning-based existential therapies (d=.65), maintained at follow-up (d=.57), with moderate effects on psychopathology (d=.47) and self-efficacy (d=.48) at post-intervention.


Appendix 3

The revised criteria proposed by Chambless and Hollon (1998) for designating levels of empirical support for psychological interventions include the following.

First, in order to meet basic standards, research about psychological intervention is expected to meet certain quality criteria:

  1. reasonable sample size (n > 25 per group);
  2. use of treatment manual or adherence checks;
  3. a specific client population defined by reliable, valid inclusion criteria;
  4. use of reliable, valid outcome measures, including measurement of targeted client difficulties;
  5. appropriate data analysis (e.g., direct comparisons, evaluation of all outcome measures).

The three levels of efficacy are defined as:

  1. Possibly efficacious: One controlled study in absence of conflicting evidence.
  2. Efficacious: In at least two independent research settings, the treatment is either (a) superior to no treatment or another treatment, or (b) equivalent to an established treatment using studies of reasonable size (n > 25 per group). With conflicting evidence, the preponderance of the well-controlled studies supports the treatment.
  3. Efficacious and specific: In at least two independent research settings, the treatment must have been shown to be statistically significant and superior either (a) to a non–bona fide treatment (e.g., a “placebo”) or (b) to an alternative bona fide treatment. With conflicting evidence, the preponderance of the well-controlled studies supports the treatment.



Return to Journal Articles