Assignment: Case Study Analyses–Biological and Evolutionary and Integrative Theories

Article

Humanistic and Integrative Therapies for Anxiety and Depression: Practice-Based Evaluation of Transactional Analysis, Gestalt, and Integrative Psychotherapies and Person-Centered Counseling

Biljana van Rijn and Ciara Wild

Abstract

The research described in this article involved a naturalistic, nonrandomized evaluation of transactional

analysis and gestalt psychotherapies, integrative counseling psychology, and person-centered counsel-

ing within a medium-term, community-based service. Routine outcome evaluation used standardized

measures to assess treatment outcomes and the working alliance. Adherence to the model was evaluated

in clinical supervision. The outcomes showed that clients who engaged in treatment made statistically

significant improvements and that transactional analysis and gestalt psychotherapies, integrative coun-

seling psychology, and person-centered counseling can be used effectively in treatment of anxiety and

depression within a community setting. Clients had a choice about the duration of therapy and used dif-

ferent numbers of sessions within the framework of the service. They were also able to change therapists.

Both choices had clinical implications in terms of attrition and outcomes and require further research.

Keywords

transactional analysis psychotherapy, gestalt psychotherapy, person-centered counseling, routine

outcome evaluation, integrative counseling psychology, research clinic

Depression and anxiety are some of the most common difficulties that bring people into therapy.

Type, duration, and cost of therapy depend largely on the setting. In the United Kingdom (UK),

therapy takes place in a variety of settings: statutory, including educational and National Health

Services (NHS); private sector; and voluntary agencies. The NHS is by far the largest provider of

psychological therapies. This service is funded by health departments and is highly impacted by gov-

ernment policies. Therapy is provided free of charge to patients.

Corresponding Author:

Biljana van Rijn, Metanoia Institute, 13 North Common Road, London W5 2QB, UK.

Email: [email protected]

Transactional Analysis Journal 43(2) 150-163 ª International Transactional Analysis Association, 2013 Reprints and permission: sagepub.com/journalsPermissions.nav DOI: 10.1177/0362153713499545 ta.sagepub.com

Voluntary agencies are not-for-profit or charitable organizations. They are frequently grant

funded by statutory agencies to provide a range of counseling and psychotherapy services. Thera-

pists in these organizations operate from a variety of theoretical orientations. Many are students who

work free of charge to gain experience required by their training. These organizations tend to

respond to the needs of their local communities and fill the gaps statutory services are unable

to meet. Clients usually refer themselves through recommendations by their doctors, friends, or

family. Therapy is offered at low cost or free of charge and the length of therapy varies. Even though

they are largely independent, because of their reliance on grants from the statutory sector, they have

also been affected by government health policies in the last 10 years.

The private sector is broadly regulated by professional umbrella bodies. Therapists practice

independently and charge for the service. Even within this sector, a number of therapists are con-

tracted by organizations such as employee assistance schemes and private health insurers, which

are guided by national health policies. This structure of services means that government policies

impact all sectors of counseling and psychotherapy, an influence that has grown substantially over

the last decade.

Department of Health (2002) recognition of the prevalence of depression and anxiety in the

population emphasized that treatments had to be based on research evidence. As a result, within

the UK, treatment is primarily cognitive-behavioral therapy. Research evidence in this climate has

become essential to the recognition of therapeutic approaches and treatments and more difficult to

develop within the nonstatutory sector, which has historically provided a wider range of approaches

and choices for the client. The lack of research has impacted the voluntary agencies and devalued the

work that takes place within them (Moore, 2006).

Transactional analysis and other approaches that are practiced more frequently in private practice

and outside of the health system in the UK have been impacted by this reliance on research evidence.

To address this, a research clinic, Metanoia Institute Counselling and Psychotherapy Service (MCPS),

was developed at Metanoia Institute. It has many of the features of a voluntary agency. It is a low-cost

counseling and psychotherapy service serving a multicultural, multiethnic, inner-city community. The

service became a research clinic in 2010 following an evaluation project in primary care (van Rijn,

Wild, & Moran, 2011). This article focuses on the research clinic outcomes between 2010 and 2011.

Theoretical approaches taught within this academic setting are also practiced within the clinic

and evaluated in this project. Transactional analysis psychotherapy training within Metanoia

Institute leads to a master’s degree-level of academic training and recognized qualification by

the European Association for Transactional Analysis and the International Transactional Analysis

Association. Therapists use the relational model within transactional analysis (Fowlie & Sills,

2011b; Hargaden & Sills, 2002).

The Integrative Psychotherapy and Counseling Psychology (DCP) Training Departments at

Metanoia use the same (integrative) theoretical orientation in their training, which leads to a

master’s or a doctoral degree. The integrative theoretical framework is based on the work of Gilbert

and Orlans (2010) and uses psychodynamic and humanistic theories as well as research to develop

an individualized approach to integration. In this article, this approach is referred to as integrative

counseling psychology and integrative psychotherapy.

Gestalt psychotherapy and person-centered counseling training courses also lead to academic and

national qualifications. All the approaches used at Metanoia Institute are based on relational princi-

ples (Fowlie & Sills, 2011a) and share an emphasis on the centrality of the therapeutic relationship

and cocreation in the therapeutic process (Summers & Tudor, 2000).

The aims of this article are to present the outcomes of transactional analysis and gestalt psy-

chotherapies, integrative counseling psychology, and person-centered counseling in the treatment

of anxiety and depression in routine practice along with differences in outcomes between clients

who engaged in therapy and those who did not.

van Rijn and Wild 151

Literature

There is a wealth of research evidence for the efficacy and effectiveness of psychotherapy in general,

although some approaches are better represented than others. Efficacy research, based on

randomized control trials (RCT), focuses on the effect of treatments on specific diagnostic cate-

gories. A research body of evidence for efficacy of cognitive behavioral therapies for depression and

anxiety has led to it being recognized by the clinical guidelines in the United Kingdom National

Institute for Clinical Excellent (NICE). The policy of Increasing Access to Psychological Therapies

(IAPT) within the UK evaluated cognitive-behavioral therapy for depression and anxiety within the

NHS (Clark et al., 2009) using large-scale, routine outcome evaluation, which showed positive out-

comes for just over 50% of patients. Generic counseling has also been evaluated in primary care in individual studies (Mellor-Clark,

Connell, Barkham, & Cummins, 2001; Stiles, Barkham, Mellor-Clark, & Connell, 2008; Stiles,

Barkham, Twigg, Mellor-Clark, & Cooper, 2006) and systematic reviews (Bower, Rowland, &

Hardy, 2003; Hill, Brettle, Jenkins, & Hulme, 2008). All demonstrated its effectiveness in primary

care. However, approaches such as transactional analysis, gestalt, and integrative psychotherapies

have had limited evaluation, even though they are practiced in a variety of settings and taught in

higher education. In recent years, there has been more research into the general effectiveness of

transactional analysis through quantitative evaluation (van Rijn et al., 2011) and case-study research

into its effectiveness for depression (Widdowson, 2012) and long-term health problems (McLeod,

2012). Despite these new research developments, therapeutic practice outside of the health sector

has had a limited level of evaluation, even though the existing research shows it to be as effective

as the NHS with similar clinical presentations (Moore, 2006).

Research Methodology and Aims

Broadly speaking, quantitative research into psychotherapy outcomes relies on two main types of

methodology, both of which use standardized questionnaires to evaluate outcomes: (1) efficacy

research, which is based on experimental design and used in randomized control trials, and (2) effec-

tiveness evaluation, which focuses on evaluation in naturalistic practice.

RCT designs aim to achieve a high degree of variable control. In psychotherapy research, this

means that therapists practice within a defined treatment manual; there are clearly defined inclusion

and exclusion criteria for clients, usually based on single diagnostic categories; and there is a rando-

mized allocation of clients to experimental and control groups. These trials have a high level of internal

validity and are used as examples of rigorous research by health authorities. Their critics suggest that

they have a limited level of generalizability to practice and therefore a low level of external validity.

Effectiveness research is conducted in naturalistic, practice settings. Therapists practice in their

usual way, and clients are accepted within the normal parameters of the service. Evaluation is

conducted routinely, during practice, and referred to as routine outcomes evaluation. The critics

of this methodology question the fact that the way the therapists practice is not clearly defined or

monitored and that validity is often limited because there is frequently a high percentage of missing

questionnaires due to unplanned endings of therapy (Clark, Fairburn, & Wessely, 2008). These

questions about the methodological or internal validity of naturalistic evaluation have to some extent

been counterbalanced by the external validity of these studies (Stirman, DeRubeis, Crits-Christoph, &

Brody, 2003), which gives a true picture of how therapies perform in real clinical practice. The find-

ings of these types of research thus have the potential to develop clinical practice (Rao, Hendry, &

Watson, 2010).

Because of its applicability to clinical practice, researchers decided to use naturalistic evaluation

while taking measures to address the methodological limitations by using the design developed by

152 Transactional Analysis Journal 43(2)

Nathan, Stuart, and Dolan (2000) in integrating the features of effectiveness and efficacy designs.

This approach was first used by the research team in the evaluation of brief transactional analysis

and integrative counseling psychology in primary care (van Rijn et al., 2011).

The following summarizes the methodology: The project was a naturalistic, nonrandomized

evaluation of routine outcomes of transactional analysis and gestalt psychotherapies, integrative

counseling psychology, and person-centered counseling. Routine outcomes evaluation combined

pre-, mid-, and posttherapy questionnaires with sessional evaluation in order to increase validity.

The research aimed to investigate the effectiveness of these different theoretical approaches and

variables that accounted for change. The design involved monitoring and evaluating the approaches

therapists practiced by developing and using questionnaires about adherence to the theoretical

model.

The hypothesis of the research team, based on common factors research outcomes, was that

transactional analysis and gestalt psychotherapies, integrative counseling psychology, and person-

centered counseling would demonstrate effectiveness and lead to a statistically significant reduction

in the symptoms of depression and anxiety.

The Research Setting. The counseling and psychotherapy service where the research clinic was

established has been operating since 1995. Metanoia Institute Counselling and Psychotherapy Ser-

vice provides low-cost counseling and psychotherapy to the general public. Therapy may be

extended to up to a year, depending on the client’s need and availability.

Therapists. The therapists in the study were second-year students at Metanoia Institute who were just

starting to practice within their approach. They had regular clinical supervision at a ratio of 1 hour of

supervision for every 4 hours of clinical practice. There were 67 practitioners during the year. Table

1 shows the theoretical approaches within the group.

Clients. Clients self referred to the service. There were 321 clients during the year. The profile of the

clients reflected the expected gender ratio and age span within a community service. The ethnic mix

reflected the local area. Seventy-two percent of the clients were female, 67% were white British, and 16.25% were Asian and black. The average age was 38. The majority of clients were between the ages of 20 and 49 (82.17%), with 15.29% over 50 and 1.27% under 20.

To assess levels of anxiety and depression, within the clinic clients were given two standardized

questionnaires at the assessment session: the Patient Health Questionnaire 9 (PHQ-9) and the

General Anxiety Measure 7 (GAD-7). The research clinic used a clinical cutoff point of 9 for the

PHQ-9 and 7 for the GAD 7, just as was used within the health service as a measure of severity

of depression and anxiety. Prior to the assessment:

� 57.9% of clients were above the clinical cutoff for depression (PHQ-9) � 74.9% of clients were above the clinical cutoff for anxiety (GAD-7) � 89.5% of clients were above the clinical cutoff for anxiety and depression

Table 1. Therapists.

Orientation Practitioners Whole group %

TA 7 10.9

Gestalt 9 12.8

Person-centered 23 34.6

Integrative counseling psychology 28 41.7

Total 67

van Rijn and Wild 153

The clinic did not take clients who had psychotic disorders, fully developed personality disorders,

or active addiction.

Treatment. After the initial contact, clients had an assessment session with a trained clinical assessor.

The assessment format was previously developed for the service by the head of clinical and research

services at Metanoia Institute (Bager-Charleson & van Rijn, 2011) and highlighted presenting issues

(such as current symptoms and functioning), developmental history, and risk.

Following the assessment session, clients were referred to practitioners for the initial four

exploratory sessions. Assessors usually talked to clients about their preferences for a way of working

or the person of the therapist (mostly gender and ethnicity). If clients decided to change therapist

at this stage, they would be referred to another practitioner. Practitioners could also decide if they

were unable to meet the needs of a particular client. A client would then be referred on. The reason

for the exploratory sessions was to offer additional safety to clients and therapists, taking into

account the relative inexperience of the therapists. The exploratory period was an opportunity for

the therapist to reflect on his or her level of competency. For the client, it was an opportunity to have

a trial period of therapy and decide whether the therapist and his or her way of working were suit-

able. When clients asked to change therapists, they sometimes expressed a preference for a different

way of working or just asked for a change of day and time.

Based on the research into importance of feedback and therapist responsiveness (Horvath, Del

Re, Fluckiger, & Symonds, 2011; Lambert & Shimokawa, 2011; Miller, Duncan, Brown, Sorrel, &

Chalk, 2006), therapists were instructed to use the outcome measures as a part of therapy as well as

for research. Questionnaires were treated as an important part of the therapeutic dialogue. When clients

handed measures in, therapists inquired into them and used them to inform a contract for the session.

They always addressed deterioration, risk, and any ruptures and misattunements evident in the

Working Alliance Inventory (WAI).

Questionnaires

Adherence to the Theoretical Approaches. All sessions were audio recorded. Clinical supervisors

listened to segments of the recordings for each client once every six sessions and assessed whether

the approach they heard being used matched the theoretical approach. The role of the supervisor was

both to assess and to support the students in developing their adherence to the model. This role was

formalized by the use of adherence questionnaires. Adherence questionnaires have been designed by

the tutor teams at Metanoia for each theoretical approach. Adherence to the model was evaluated

using a five-point scale ranging from ‘‘no adherence’’ (1) to ‘‘full adherence’’ (5).

Clinical Questionnaires. At the assessment, the sixth session, and the end of therapy, each client was

given Beck’s Depression Inventory (BDI-II) (Beck, 1996), a 21-item questionnaire measuring

depression.

After each session, clients were given a Patient Health Questionnaire (PHQ-9) (Kroenke, Spitzer, &

Williams, 2001), a 9-item questionnaire that distinguished between clinical and nonclinical

populations, and a General Anxiety Measure (GAD -7) (Spitzer, Kroenke, Williams, & Lowe,

2006), a 7-item questionnaire that was initially developed to evaluate generalized anxiety disorder

and found to have sensitivity for other anxiety disorders (Kroenke, Spitzer, Williams, Monahan, &

Lowe, 2007)

Therapeutic Alliance Questionnaire. After each session, clients were given the short form of the

Working Alliance Inventory (WAI) (Horvath, 1986), a 12-item questionnaire developed to measure

the working alliance as defined by Bordin (1979).

154 Transactional Analysis Journal 43(2)

Ethical Considerations. Assessors gave written information about the research to clients, answered

further questions about the research, and sought consent. Clients who decided not to take part, or

who withdrew from research during treatment, continued to receive the service. Outcomes of

questionnaires were discussed transparently between therapists and clients and became a part of the

therapeutic dialogue. All the data was confidential and anonymized before analysis. The Metanoia

Institute Ethics Committee (an independent body approved by Middlesex University) gave ethical

consent to the project.

Research Sample. Table 2 shows that there were altogether 346 cases during the year. The number of

cases included clients who had been reallocated within the assessment period, which is why the

number of cases is higher than the overall number of clients (321). A proportion of clients were not

accepted into the service. Some clients opted out of the research but continued in therapy. Outcomes

were divided into three groups:

� Group 1 represented clients who engaged in therapy after the assessment period (assessment session and the four exploratory sessions).

� Group 2 represented clients who did not engage in therapy past the assessment period. � Group 3 was not analyzed because of the lack of the adherence questionnaires for the

therapeutic approach used.

There is a full analysis of outcomes and adherence to the model for Group 1. Outcomes for Group

2 were analyzed as far as possible to determine the outcomes for clients who did not engage past the

assessment period.

Number of Sessions. The average number of sessions for Group 1 was 17.48 sessions. There was a

difference in the length of therapy, as shown in Table 3. The average number of sessions for Group 2

was 2.5 sessions. However, 62.8% of clients in Group 2 asked to be reallocated to another practitioner. After reallocation, their average number of sessions rose to 12.

Data Completeness. Data completeness outcomes have implications for the validity of evaluation.

Table 4 shows a percentage of completed questionnaires for Groups 1 and 2. There was a higher

percentage of data completeness for questionnaires given each session (PHQ-9, GAD-7, and WAI)

Table 2. Research Sample.

Research

cases SPSS

Cases not

accepted into

the service

Cases

opted out

Group 1: Five

sessions or more

Group 2: Four or

fewer sessions

Group 3: Seven or

more sessions

without adherence

No 346 16 45 138 110 37

% 100 4.6 13 40 32 10.7

Table 3. Group 1 Number of Sessions.

Frequency Percent Cumulative percent

Valid 1-12 weeks 51 37.0 37.0

13-24 weeks 57 41.3 78.3

25-plus weeks 30 21.7 100.0

Total 138 100.0

van Rijn and Wild 155

than for questionnaires given at the beginning, middle, and end of therapy (BDI-II). In addition, data

completeness for Group 2 was lower than for Group 1.

Outcomes

Tables 5 and 6 show the descriptive statistics for Groups 1 and 2. The measures of central tendency

were close in value, which suggests that few or no outliers affected the sample. However, the stan-

dard deviations for all the measures were large in comparison to the mean. This demonstrates a wide

spread of scores from the mean in the sample on all of the measures and across all of the groups. It

infers that clients entered therapy with a wide range of levels of distress, ended at different levels,

and achieved change differently from each other. The wide spread of scores was examined statisti-

cally using a Kolmogrov-Smirnov test to establish whether the sample had a normal distribution of

scores. The test showed a mixture of distributions for all the measures. This would be expected in

this type of sample in which clients usually had higher scores at the beginning than at the end.

These figures illustrate a negative skew in scores for all measures except the WAI, which had a

positive skew in scores, or floor and ceiling effects, respectively. This shows that by the end of

Table 4. Data Completeness.

Data completeness % PHQ-9 GAD-7 BDI-II WAI

Group 1 99.3 99.3 74.5 95.8

Group 2 57.2 59.2 12.6 49.5

Table 5. Descriptive Statistics for Group 1.

Group 1

BDI-II PHQ-9 GAD-7 WAI

Pre Post Diff S1 Post Diff S1 Post Diff S1 Post Diff

Mean 24.44 11.56 12.43 11.08 5.89 4.91 10.57 5.75 4.86 63.18 70.42 8.55

Median 22.0 8.50 12.0 10.0 4.5 4.0 10.0 5.0 5.0 64.50 72.0 7.00

Mode 20 4 12 a

8 0 a

5 7 0 5 67 a

84 0

Sd 10.83 10.32 10.58 5.91 5.01 6.33 4.76 4.76 5.63 12.712 13.11 12.02

a. Multiple modes exist. The smallest value is shown.

Table 6. Descriptive Statistics for Group 2.

Group 2

BDI-II PHQ-9 GAD-7 WAI

Pre Post Diff S1 Post Diff S1 Post Diff S1 Post Diff

Mean 23.27 22.55 7.33 9.73 7.79 1.42 9.05 8.90 .52 55.44 63.31 3.22

Median 25.0 24.0 5.0 9.0 6.0 1.0 8.0 9.0 .00 58.0 64.5 2.00

Mode 26 a

4 0 a

9 4 a

0 a

7 a

6 0 57 60 a

0

Sd 10.84 14.4 7.74 5.88 5.7 4.42 5.5 5.7 5.13 17.40 14.62 11.50

a. Multiple modes exist. The smallest value is shown.

156 Transactional Analysis Journal 43(2)

therapy the majority of clients reported less distress on the measures. This was different for the

working alliance, which showed high scores at the end of therapy.

Improvement Rates. Criteria for improvement were calculated by taking the difference between

scores at the start of therapy and at the end using the percentage of improvement, no change, and

deterioration. The descriptive statistics show that posttherapy scores were mainly low, with the

exception of the WAI, which was high. Tables 7 and 8 show the percentage improvement scores for

Groups 1 and 2.

The percentage improvement clearly supported the descriptive statistics. Large percentages of

improvement showed low scores at the end of therapy in comparison to the start of therapy. To

examine this further, the data were tested to establish if these improvements rates were significant.

As there was a mixture of normal and non-normal distributions in the sample, a Wilcoxon Signed

Ranks test was used to examine the difference between pre- and posttherapy scores. They showed

that the difference between pre- and post-scores for all measures was significant at P < 0.01 for

Group 1, and the direction of the difference was represented by the negative Z score and effect size

in Table 9, in which scores decreased from pre- to posttherapy. All the changes show large effect

sizes as they have a value greater than 0.5. Group 2 did not achieve a statistically significant change.

Tables 9 and Table 10 show the Z scores for groups 1 and 2.

Table 7. Improvement Rates for Group 1.

% BDI-II PHQ-9 GAD-7 WAI

Improve 59.9 77.5 77.5 71.7

No change 2.9 6.9 5.1 7.2

Deteriorate 3.6 15.9 17.4 19.6

No data 33.6 1.4

Table 8. Improvement Rates for Group 2.

% BDI-II PHQ-9 GAD-7 WAI

Improve 4.6 34.6 28.4 14.8

No change 0.9 9.3 10.1 4.6

Deteriorate 0.9 16.8 22.9 9.3

No data 93.6 39.3 38.5 71.3

Table 9. Z Scores for Group 1.

Test Statistics c

Post-BDI1 –

Pre-BDI

Post-Score 1 –

PHQ-9S1

Post-Score 2 –

GAD-7S1

Post-ScoreTotal -

WAIS1Total

Z �7.433a �7.136a �7.345a �5.088b Asymp. Sig.

(2-tailed)

.000 .000 .000 .000

Effect Size �0.83 �0.64 �0.65 �0.58

a. Based on positive ranks.

b. Based on negative ranks.

c. Wilcoxon Signed Ranks test.

van Rijn and Wild 157

Which Variables Accounted for Change? A regression was carried out to investigate which of the

variables accounted for the greatest change in clients’ scores from pre- to posttherapy and which

variable had the greatest impact on posttherapy scores. Total attendance, prescore (severity), adher-

ence score, and WAI total score were entered stepwise into the regression. A regression could only

be performed on Group 1 data because Group 2 had too many missing cases. The regression showed

that severity (prescores) and WAI accounted for a significant proportion of the variation in the

regression model for postscores and change scores on all measures as follows:

� BDI-II Post - b ¼ .47, t ¼ (142) ¼ 4.71, severity and b ¼�.29, t ¼ (142) ¼�2.96; WAI explained a significant proportion of the variance in scores, where adjusted R

2 ¼ .252, F (2, 75) ¼13.96, which is a large effect. � BDI-II Change - b ¼ .42, t ¼ (142) ¼ 4.16, severity and b ¼ .216, t ¼ (142) ¼ 2.16;

WAI explained a significant proportion of the variance in scores, adjusted R 2 ¼ .227,

F (4, 77) ¼ 12.63, which is a large effect. � PHQ-9 Post - b ¼ .29, t ¼ (142) ¼ 3.4, severity and b ¼ .29, t ¼ (142) ¼ �3.43; WAI

explained a significant proportion of the variance in scores, adjusted R 2 ¼ .149, F (2, 115)

¼ 11.24, which is a large effect. � PHQ-9 Change - b ¼ .682, t ¼ (142) ¼ 10.68, severity and b ¼ .231, t ¼ (142) ¼ 3.62;

WAI explained a significant proportion of the variance in scores, adjusted R 2 ¼ .523,

F (2, 115) ¼ 65.1, which is a large effect. � GAD Post - b ¼ .232, t ¼ (142) ¼ 2.69, severity and b ¼�.307, t ¼ (142) ¼�3.56; WAI

explained a significant proportion of the variance in scores, adjusted R 2 ¼ .125, F (2, 116) ¼

9.47, which is a large effect.

� GAD Change - b ¼ .6, t ¼ (142) ¼ 8.5, severity and b ¼ 0.25, t ¼ (142) ¼ 3.5; WAI explained a significant proportion of the variance in scores, adjusted R

2 ¼ .423, F (2, 116) ¼ 44.18, which is a large effect.

These results suggest that severity of depression and anxiety at the outset accounted for the

greatest variation in scores. Those with the highest scores pretherapy showed the greatest difference

on the outcome measures between pre- and posttherapy or the greatest amount of change during

therapy. The regression model indicated that severity and working alliance were good predictors

of clients’ therapy outcomes.

Theoretical Orientation. The absence of normal distribution warranted checking the pretherapy

scores to see if there were differences between the theoretical orientations that might have impacted

the analysis. A Kruskal Wallis test was carried out. That is a nonparametric test that does not assume

Table 10. Z Scores for Group 2.

Test Statistics c

Post-BDI1 –

Pre-BDI

Post-Score1 –

PHQ-9S1

Post-Score2 –

GAD-7S1

Post-Score Total –

WAIS1Total

Z �1.428a �2.702a �3.364a �3.232b Asymp. Sig.

(2-tailed)

.153 .007 .001 .001

Effect Size �0.45 �0.62 �0.6 �0.67

a. Based on positive ranks.

b. Based on negative ranks.

c. Wilcoxon Signed Ranks test.

158 Transactional Analysis Journal 43(2)

a normal distribution and can be used with large variations in scores. The test indicated that there

were significant differences at P < 0.05 between orientations in Group1 only on the Core 34 X2

(4, N ¼ 107) ¼ 11.98. A chi-squared through Crosstabs analysis was run to search for a difference between theoretical orientations and association with outcomes. There was no difference found

between orientation and improvement in any of the groups.

Adherence to the Therapeutic Model. Each practitioner and supervisor completed an adherence

form specific to his or her modality. On average, adherence to models was high, which is demon-

strated below along with a demonstration of how scores were grouped into low, medium, and high

adherence.

Gestalt: Low 1-24, Medium 48-85, and High 86-125; average score 86.5

Integrative/DCP: Low 1-40, Medium 41-81, and High 82-120; average score 92.1

Transactional Analysis: Low 1-26, Medium 27-53, and High 54-80; average score 57.9

Person-Centered Counseling: Low 1-18, Medium, 19-37, and High 38-55; average score 54.9

Discussion

Limitations of the Study. The limitations of this research are contained within the naturalistic meth-

odology. Therefore, the research suggests effectiveness in clinical practice but not causality. There

was no follow up, and we have no indication whether changes were maintained over time. Clients

presented with a range of issues and were not chosen especially for the research or randomly allo-

cated to treatment. There was no control group. This was the same for the therapists and their super-

visors. Because of this, the outcomes showed how these therapies performed in practice but cannot

claim that they were the single cause of change.

Effectiveness. The outcome measures showed that clients who engaged in therapy achieved a high

rate of improvement (77.5%) on sessional measures for depression and anxiety. The outcomes were statistically significant, with a large effect size, which confirms the hypothesis of the

research team.

Adherence to the model was high, on average, as would be expected with trainee therapists who

were being trained in their model and did their best to practice it. High completion rates for the

GAD-7, PHQ-9, and WAI, completed by over 90% of the clients, suggested that these outcomes were reliable. Completion rates for the BDI-II were lower and reflected unplanned endings. How-

ever, the sessional evaluations showed that the clients had improved by the time they ended therapy,

if the ending was not planned.

Higher severity at the outset resulted in more change. However, some of this might have been due

to the sensitivity of the measures because they could not capture change from clients with moderate

or low pretherapy scores. The length of therapy varied following the assessment period but did not

have a significant impact on the outcomes. This suggests that the optimal number of sessions for

clients was individual and that therapists’ responsiveness (Stiles et al., 2008) and contracting about

the length of therapy were more important than the actual number of sessions.

Outcomes showed no difference in effectiveness between theoretical approaches. These

outcomes are supported by the evidence of common factors research and meta-analytic research

(Lambert & Bergin, 1994; Lambert & Ogles, 2004; Wampold, 2001).

Overall, these outcomes strongly demonstrate that transactional analysis and gestalt psychothera-

pies, integrative counseling psychology, and person-centered counseling can be used as effective

treatments for anxiety and depression.

Use of Questionnaires. The design of this research created a particular therapeutic process. It

involved therapists and clients in a structured dialogue about therapy that included clinical

van Rijn and Wild 159

questionnaires as well as an overt discussion about the working alliance during therapy. Clients took

away the questionnaires at the end of their sessions and completed them during the week, engaging

in a period of structured reflection about which they then reported to the therapists. Therapists and

clients integrated questionnaires into therapy in different ways, usually contracted for at the begin-

ning of each session. Sometimes clients wanted to spend time talking about the particular issues

brought up by the questionnaire, and sometimes they wanted to focus on other issues. In each case,

therapists looked through the questionnaires briefly at the beginning of each session, which gave

them an opportunity to comment on or inquire into what the client had said. Working Alliance

Inventory questionnaires gave clients the overt message that the therapeutic relationship was a bona

fide area for discussion and that their therapist was interested in knowing how they experienced the

working relationship.

It is likely that this had an impact on outcomes and could be an area that has potential to enhance

effective therapeutic practice when used respectfully and dialogically, even outside of the research

inquiry. This suggestion is supported by research showing that client feedback increases engagement

in therapy and improves outcomes (Lambert & Shimokawa, 2011; Lambert et al., 2002; Miller et al.,

2006). Similarly, previous research has suggested that attention to the working alliance and ruptures

to it has a positive impact on therapeutic outcomes (Horvath & Bedi, 2002; Horvath et al., 2011;

Safran, Muran, & Eubanks-Carter, 2011).

Transactional analysis has always been based on the principles of client empowerment and trans-

parency, and this approach could be particularly well suited to it.

Changing Therapists. A high level of attrition is one of the clinical realities of low-cost clinics and

health settings (Ogrodniczuc, Joyce, & Piper, 2005; Reis & Brown, 1999). This was evident in this

research by those in Group 2, or clients who did not proceed after the assessment period. In the few

sessions these clients had (an average of 2.5), they did not achieve much change. However, a particular

feature of the Metanoia clinic is that it allowed clients to change therapists. Eighty-eight percent (12)

of clients decided to change therapists within the assessment period; only 28% (7) of clients then had four sessions or less. The remaining 71.2% of clients engaged in therapy past the assessment period. They had indistinguishable outcomes to other clients in Group 1. This again highlighted the importance

of therapist responsiveness (Stiles, Barkham, Connell et al., 2008) and the early working alliance to suc-

cessful outcomes.

Qualitative research could give further, more in-depth insight into this process and assist thera-

pists and organizations in developing services.

Implications for Practice and Research

The research described in this article highlighted several implications for practice and further

research:

� Transactional analysis and gestalt psychotherapies, integrative counseling psychology, and person-centered counseling can be used effectively in the treatment of anxiety and depres-

sion within a community setting. A randomized control trial would be required to establish

the efficacy of these approaches.

� Features of overt collaboration and therapist responsiveness suggest the importance and centrality of the therapeutic relationship, which is one of the principles of relational trans-

actional analysis (Fowlie & Sills, 2011a). This research illustrated that standardized quan-

titative questionnaires could be used in relational therapeutic practice and have good impact

on outcomes. In-depth qualitative research could investigate this impact further.

160 Transactional Analysis Journal 43(2)

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publi-

cation of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Bager-Charleson, S., & van Rijn, B. (2011). Understanding assessment in counselling and psychotherapy.

Exeter, England: Learning Matters.

Beck, A. (1996). Beck depression inventory. United States of America Patent No. 0154018392.

Bordin, E. S. (1979). The generalizability of the psychoanalytic concept of the working alliance. Psychother-

apy: Theory, Research, Practice, 16(3), 252–260.

Bower, P., Rowland, N., & Hardy, G. (2003). The clinical effectiveness of counselling in primary care: A

systematic review and meta-analysis. Psychological Medicine, 33(2), 203–215.

Clark, D. M., Fairburn, C. G., & Wessely, S. (2008). Psychological treatment outcomes in routine NHS

services: A commentary on Stiles et al. (2007). Psychological Medicine, 38(5), 629–634.

Clark, D. M., Layard, R., Smithies, R., Richards, D. A., Sucksling, R., & Wright, B. (2009). Improving access to

psychological therapy: Initial evaluation of two demonstration sites. Behaviour Research and Therapy, 47,

910–920.

Department of Health. (2002). A national service framework for mental health. London, England: Author.

Fowlie, H., & Sills, C. (2011a). Introduction. In H. Fowlie & C. Sills (Eds.), Relational transactional analysis:

Principles in practice (pp. xxv–xxxii). London, England: Karnac Books.

Fowlie, H., & Sills, C. (Eds.). (2011b). Relational transactional analysis: Principles in practice. London,

England: Karnac Books.

Gilbert, M., & Orlans, V. (2010). Integrative therapy: 100 key points and techniques. London, England:

Routledge.

Hargaden, H., & Sills, C. (2002). Transactional analysis: A relational perspective. Hove, England: Brunner-

Routledge.

Hill, A., Brettle, A., Jenkins, P., & Hulme, C. (Eds.). (2008). Counselling in primary care: A systematic review

of the evidence. Lutterwoth, England: British Association for Counselling and Psychotherapy.

Horvath, A. O. (1986, April). Another approach to assess the counseling relationship: The working alliance.

Paper presented at the American Educational Research Association, San Francisco, CA, USA.

Horvath, A. O., & Bedi, R. P. (2002). The alliance. In J. C. Norcross (Ed.), Psychotherapy relationships that

work: Therapist contributions and responsiveness to patients (pp. 37–69). New York, NY: Oxford Univer-

sity Press.

Horvath, A. O., Del Re, A. C., Fluckiger, C., & Symonds, D. (2011). Alliance in individual psychotherapy. In

J. C. Norcross (Ed.), Psychotherapy relationships that work. Evidence based responsiveness (pp. 25–69).

New York, NY: Oxford University Press.

Kroenke, K., Spitzer, R. L., & Williams, J. B. (2001). The PHQ-9: Validity of a brief depression severity mea-

sure. Journal of General and Internal Medicine, 16, 606–613.

Kroenke, K., Spitzer, R. L., Williams, J., Monahan, P. O., & Lowe, B. (2007). Anxiety disorders in primary

care: Prevalence, impairment, comorbidity, and detection. Annals of Internal Medicine, 146, 317–325.

Lambert, M. J., & Bergin, A. E. (1994). The effectiveness of psychotherapy. In A. E. Bergin & S. L. Garfield

(Eds.), Handbook of psychotherapy and behavior change (4 th

ed.) (pp. 143–189). New York, NY: Wiley.

van Rijn and Wild 161

Lambert, M. J., & Ogles, B. M. (2004). The efficacy and effectiveness of psychotherapy. In M. J. Lambert

(Ed.), Bergin and Garfield’s handbook of psychotherapy and behavior change (5 th

ed.) (pp. 139–193).

New York, NY: Wiley.

Lambert, M. J., & Shimokawa, K. (2011). Collecting client feedback. In J. C. Norcross (Ed.), Psychotherapy

relationships that work: Evidence-based responsiveness (2 nd

ed.) (pp. 203–223). New York, NY: Oxford

University Press.

Lambert, M. J., Whipple, J. L., Vermeersch, D. A. D., Smart, W., Hawkins, E. J., Nielsen, S. L., & Goates, M.

(2002). Enhancing psychotherapy outcomes via providing feedback on client progress: A replication.

Clinical Psychology & Psychotherapy, 9(2), 91–103.

McLeod, J. (2012). Process and outcome in pluralistic transactional analysis counselling for long-term health

conditions: A case series. Counselling and Psychotherapy Research, 13(4), 1–12.

Mellor-Clark, J., Connell, J., Barkham, M., & Cummins, P. (2001). Counselling outcomes in primary health

care: A CORE system data profile. European Journal of Psychotherapy & Counselling, 4(1), 65–86.

Miller, S., Duncan, B. L., Brown, J., Sorrel, R., & Chalk, M. B. (2006). Using formal client feedback to improve

retention and outcome: Making ongoing real-time assessment feasible. Journal of Brief Therapy, 5(1), 5–21.

Moore, S. (2006). Voluntary sector counselling: Has inadequate research resulted in a misunderstood and

underutilised resource? Counselling and Psychotherapy Research, 6(4), 221–226.

Nathan, P. E., Stuart, S. P., & Dolan, S. L. (2000). Research on psychotherapy efficacy and effectiveness:

Between Scylla and Charybdis? Psychological Bulletin, 126(6), 964–981.

Ogrodniczuc, J., Joyce, A., & Piper, W. (2005). Strategies for reducing patient-initiated premature termination

of psychotherapy. Harvard Review of Psychiatry, 13(2), 57–70.

Rao, A. S., Hendry, G., & Watson, R. (2010). The implementation of routine outcome measures in a tier 3 psy-

chological therapies service: The process of enhancing data quality and reflections of implementation chal-

lenges. Counselling & Psychotherapy Research, 10(1), 32–38.

Reis, B., & Brown, L. (1999). Reducing psychotherapy dropouts: Maximizing perspective convergence in the

psychotherapy dyad. Psychotherapy, 36(2), 123–136.

Safran, J. D., Muran, J. C., & Eubanks-Carter, C. (2011). Repairing alliance ruptures. In J. C. Norcross (Ed.),

Psychotherapy relationships that work: Evidence-based responsiveness (2 nd

ed.) (pp. 224–238). New York,

NY: Oxford University Press.

Spitzer, R. L., Kroenke, R., Williams, J. B., & Lowe, B. (2006). A brief measure for assessing generalized

anxiety disorder: The GAD-7. Archives of Internal Medicine, 166, 1092–1097.

Stiles, W. B., Barkham, M., Connell, J., & Mellor-Clark, J. (2008). Responsive regulation of treatment duration

in routine practice in United Kingdom primary care settings: Replication in a larger sample. Journal of

Consulting and Clinical Psychology, 76, 298–305.

Stiles, W. B., Barkham, M., Mellor-Clark, J., & Connell, J. (2008). Effectiveness of cognitive-behavioural,

person-centred, and psychodynamic therapies in UK primary-care routine practice: Replication in a larger

sample. Psychological Medicine, 38(5), 677–688.

Stiles, W. B., Barkham, M., Twigg, E., Mellor-Clark, J., & Cooper, M. (2006). Effectiveness of cognitive-

behavioural, person-centred and psychodynamic therapies as practised in the UK national health service

settings. Psychological Medicine, 36(4), 555–566.

Stirman, S. W., DeRubeis, R. J., Crits-Christoph, P., & Brody, P. E. (2003). Are samples in randomized

controlled trials of psychotherapy representative of community outpatients?: A new methodology and initial

findings. Journal of Consulting and Clinical Psychology, 71(6), 963–972.

Summers, G., & Tudor, K. (2000). Cocreative transactional analysis. Transactional Analysis Journal, 30,

24–40.

van Rijn, B., Wild, C., & Moran, P. (2011). Evaluating the outcomes of transactional analysis and integrative

counselling psychology within UK primary care settings. International Journal of Transactional Analysis

Research, 2(2), 34–43.

Wampold, B. (2001). The great psychotherapy debate. Mahway, NJ: Lawrence Erlbaum Associates.

162 Transactional Analysis Journal 43(2)

Widdowson, M. (2012). TA treatment of depression: A hermeneutic single-case efficacy design study—

‘Denise.’ International Journal of Transactional Analysis Research, 3(2), 3–14

Author Biographies

Biljana van Rijn, D.Psych., is head of clinical and research services at Metanoia Institute in

London, where she has developed a research clinic. A Teaching and Supervising Transactional Ana-

lyst (psychotherapy), she tutors at Metanoia Institute and works as a psychotherapist and supervisor

in private practice. She can be reached at Metanoia Institute, 13 North Common Road, London W5

2QB, United Kingdom; email: [email protected].

Ciara Wild, MSc (forensic psychology), is a researcher and clinic administrator at Metanoia

Institute in London. She can be reached at Metanoia Institute, 13 North Common Road, London

W5 2QB, United Kingdom; email: [email protected].

van Rijn and Wild 163

Copyright of Transactional Analysis Journal is the property of Sage Publications Inc. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.