Psychodynamic psychotherapy for children and adolescents: an

updated critical review of the evidence-base

Nick Midgleyab, Sally Parkinsonba , Lorna French and Eilis Kennedyc

a Anna Freud National Centre for Children and Families, London, UK

b Research Department of Clinical, Education and Health Psychology,

University College London, London, UK

c Tavistock Clinic, London, UK

Corresponding author: Dr Nick Midgley, 21 Maresfield Gardens, London NW3

5SD

1
Psychodynamic psychotherapy for children and adolescents: an

updated critical review of the evidence-base

Abstract

Introduction

Psychodynamic psychotherapy with children and adolescents is an

approach to working with young people that draws on psychoanalytic ideas,

whilst also integrating ideas from other disciplines, including developmental

psychology, attachment theory and neuroscience (Alvarez, 2012; Kegerreis &

Midgley, 2014; Lanyardo & Horne, 2009). Although the term 'psychodynamic

therapy' covers a range of approaches, most of them share what Kegerreis &

Midgley (2014) refer to as "the central idea ... that behaviour, emotions and

responses have an inherent logic and meaning – a way in which the child’s

problems, despite their apparent unhelpfulness, make some kind of emotional

sense. Their roots lie in the internal world of the child that has been built up

from his earliest experiences and relationships" (p.38)i.

In 2011, we published a critical review of the evidence base for

psychodynamic therapy with children and adolescents (Midgley & Kennedy,

2011). In that paper, we identified 34 studies, published before March 2011,

which formally evaluated therapy outcomes for children aged 3-18. Of these

studies, nine were randomised controlled trials (RCTs), three had a quasi-

experimental design, eight were controlled observation studies and fourteen

were observational studies without a control group. Psychodynamic therapy

2
delivered in a group or dyadic format, or with children below the age of 3, was

not covered by the earlier review, although the evidence-base for

psychodynamic parent-infant psychotherapy has been reviewed in a meta-

analysis conducted by (Barlow, Bennett, Midgley, Larkin, & Wei, 2015).

With regard to the practice and clinical implications of the 2011 review,

the following conclusions were cautiously drawn: [Probably need to revise this

section to avoid self-plagiarism?]

 The existing studies of psychodynamic therapies with children and

adolescents indicate that this treatment can be effective for a range of

childhood disorders, as measured by well-validated, standardised

research instruments.

 Where direct comparisons have been made, psychodynamic treatment

of children and adolescents appears to be equally effective to

comparison treatments, with mixed findings across studies - some

suggesting psychodynamic therapy is more, some less, and some

equally effective as other forms of therapy

 There are some indications that psychodynamic treatment may have a

different pattern of effect to other treatments. For example, when

compared to systemic family therapy, depressed children appeared to

recover more quickly when receiving family therapy, whilst

improvements for those receiving individual psychodynamic therapy

appeared to be slower but more sustained, with some young people

continuing to improve after the end of treatment

 A similar pattern of more gradual improvement, but with improvement

continuing beyond the end of treatment, was found in a study of

3
 children with emotional disorders, giving some evidence of a possible

‘sleeper effect’ in psychodynamic therapy

 Certain children appear to be more responsive to psychodynamic

treatment than others. Where age groups have been directly

compared, younger children appear to benefit more than older ones,

with the likelihood of improvement during treatment declining with age

 However there are also studies that suggest that older children and

adolescents can also benefit from psychodynamic therapy

 Certain disorders appear to be more responsive to psychodynamic

treatment than others. Children with emotional or internalising

disorders seem to respond better than those with

disruptive/externalising disorders

 Children and adolescents with disruptive disorders are more difficult to

engage and more likely to drop out of psychodynamic treatment; but

where they have engaged in treatment there is some evidence that it

can be effectiveThere is an evidence-base emerging for the treatment

of children and young people with depression, which in the UK led to

psychodynamic treatment being identified as an evidence-based

treatment in the NICE guidelines on child and adolescent depression

(NICE, 2005)and in the US it was recommended as a treatment option

in the American Academy of Child and Adolescent Psychiatry Practice

Parameter on Depressive disorders (REF). In samples that can be

assumed to have lesser degrees of difficulty either because of the

setting or selection criteria, short term and even minimal interventions

were shown to be effective

4
  When children present with more marked difficulties e.g. with conduct

disorder or severe emotional disorder, the intensity of the treatment

may be important

 There were some indications of potential adverse affects, for example

that if psychodynamic child psychotherapy was offered without parallel

work with parents, this could have a negative effect on family

functioning and that more intensive work could, in some cases, add to

an adolescent’s sense of ‘stigma’.

A number of conclusions about the nature and quality of research in

this field were draw in the previous review paper (Midgley & Kennedy, 2011).

The key points were that studies tended to have heterogonous clinical

populations studies, variations in the interventions, relatively small samples

and significant methodological limitations, making it difficult to draw any firm

conclusions from the studies. Furthermore, studies often did not reference

each other, build on each other and they tended not to be designed in such a

way that allowed meaningful comparison of findings, restricting the

development of cumulative knowledge about the evidence base for this type

of treatment for children and adolescents.

Since the 2011 review was published, a lively debate has continued in

the field about the science and the politics of evidence based practice, both in

the broader child mental health literature (e.g. Kennedy, 2015), and in regard

to the field of psychodynamic child psychotherapy specifically (e.g. Kegerreis,

2016; Rustin, 2016). A number of treatment manuals for versions of

psychodynamic therapies with children have been made publicly available

5
(e.g. Hoffmann et al., 2015; Cregeen et al., 2016) and in addition, two further

reviews of the evidence-base for psychodynamic child therapy have been

published, each with a somewhat different focus, and drawing somewhat

different conclusions. Given the significance of these reviews, we will discuss

each of them in some depth.

Palmer, Nascimento, & Fonagy (2013) review is of great interest

because it reviews many of the same studies as our 2011 paper, yet reaches

somewhat different conclusions. As such, it illustrates the point that even

when using a systematic approach, rather different conclusions can be drawn

based on the same studies. Palmer et al. aimed to identify, describe, and

review studies published before May 2012, evaluating the efficacy and/ or

effectiveness of psychodynamic treatment for children and adolescents with

mental health problems, by means of a highly systematic review of the

relevant databases. Unlike our 2011 review, Palmer et al. included studies

that evaluated psychodynamic therapy delivered in a group of parent-child

format, and included "studies exploring the effectiveness of therapies that

integrate significant psychodynamic components into a multimodal package"

(p. 154), such as residential treatments informed by a psychodynamic

approach, or Attachment Based Family Therapy (Diamond, 2014). They

justified this broader inclusion on the basis that the distinction between

different modalities of treatment is no longer as clear-cut as it may once have

been, with many treatments, such as Mentalization Based Treatment,

explicitly integrating a psychodynamic approach with elements from other

effective therapies. In order to improve the transparency of the review

process, the Palmer et al. study also made use of RCT-PQRS (Gerber et al.,

6
2011), a scale developed to evaluate the quality of RCTs, made up of 24

items relating to study design, reporting, and execution.

Based on their review, Palmer et al. (2013) drew the following key

conclusions (p.175):

 Currently there is some evidence to support the use of PP for children

whose problems are either internalizing or mixed but with an element of

anxiety and emotional disorder.

 There is also evidence that the support and inclusion of parents is an

important aspect of this treatment.

 There is some evidence that effects tend to increase following the end

of treatment.

 There is evidence that behavioral problems are more resistant at least

to a classical, insight-oriented psychodynamic approach.

 In line with the grouping together of family and individual approaches,

the evidence is stronger for younger children, where parents are almost

always included in treatment, and where a dyadic therapeutic model

exploring the dynamics of the parent–child relationship may be

especially helpful.

The authors concluded by arguing that, in light of the limitations of the

evidence base for CBT in severe childhood disorders, further research into

the effectiveness of alternative treatments, including psychodynamic therapy,

were still called for. But somewhat controversially, they also predicted that

"the techniques that have evolved as part of this approach will not survive

(they are effective, and clinicians, being pragmatic people, will continue to

7
discover and use them), but they will be increasingly absorbed into alternative

models, and the unique approach pioneered by Freud and outlined in this

issue might then not continue" (p.175).

In the same year as the Palmer et al. review was published, Abbass

and colleagues published a meta-analysis which focused specifically on the

evidence base for short-term psychodynamic psychotherapy (STPP) with

children and adolescents (i.e. individual therapy of less than 40 sessions)

(Abbass, Rabung, Leichsenring, Refseth, & Midgley, 2013). In taking a meta-

analytic approach, the authors were able to pool results from a range of

different studies, thereby helping to address the problem of low statistical

power that handicaps much psychotherapy research. Although including a

smaller number of studies (11), these were all randomised controlled trials,

generally considered the 'gold standard' in psychotherapy outcome research.

The authors performed a sensitivity analysis and evaluated the risk of bias in

each of the studies included in the review. The key findings of the Abbass et

al. study were as follows:

 The 11 studies included a total of 655 patients covering a broad range

of conditions including depression, anxiety disorders, anorexia

nervosa, and borderline personality disorder.

 Robust (g= 1.07, 95% CI: 0.80–1.34) within group effect sizes were

observed suggesting the treatment may be effective.

 These effects increased in follow up compared to post treatment

(overall, g = 0.24, 95% CI: 0.00–0.48), suggesting a tendency toward

increased gains.

8
  STPP did not separate from what were mostly robust treatment

comparators, but there were some subgroup differences.

 As with the other reviews discussed here, Abbass et al. noted that

heterogeneity was high across most analyses suggesting these data

need be interpreted with caution.

Perhaps the most intriguing finding, which was consistent across the

three review papers, was the support found for a “sleeper effect”, whereby the

gains from therapy continued to increase after the end of therapy. The same

conclusion was drawn by a much broader review of the research on

counselling and psychotherapy with children and young people (McLaughlin,

Carol Holliday, Clarke, & Ilie, 2013), although based primarily on a review of

the same studies.

While these findings suggest that this treatment modality is potentially

effective in treating a range of psychological problems, with sustained effects,

it is difficult to draw firm conclusion because the literature is sparse and there

is a lack of high quality studies. The majority of studies have had small

sample sizes, many of which were conducted in naturalistic settings, lacking a

suitable control group and limiting their generalizability. There are an

insufficient number of high quality studies using comparable treatments to

aggregate findings, to draw any firm conclusions about the efficacy and

effectiveness of psychodynamic psychotherapy for any disorder (Abbass et

al., 2013; Palmer et al., 2013). However, our 2011 paper concluded by noting

that the number of studies evaluating the efficacy and/or effectiveness of

psychodynamic therapy with children and adolescents had increased decade

9
by decade since the 1970s, and that at the time of writing the 2011 review

there were a number of on-going studies that were evaluating psychodynamic

therapy for children with specific diagnostic groups, and that these studies

appeared to be using more sophisticated research designs.

Given the rapid developments in this field, the aim of this review paper

is to provide an update on the evidence base for psychodynamic therapy with

children and adolescents published since March 2011, making use of the

same methodology used in our earlier paper. In particular, the aims of this

review are:

(1) To identify and describe studies of treatment effectiveness/efficacy for

psychodynamic psychotherapy with children and adolescents published

since March 2011.

(2) To categorise those studies according to a hierarchy of evidence of

therapeutic effectiveness.

(3) To examine outcome in different clinical groups of children and

adolescents.

(4) To assess the degree to which these more recent studies support,

challenge or add to the conclusions drawn in our earlier (2011) review.

(5) An additional aim of this study, which was not done in the earlier review,

was to assess the quality of each study included in the review, by rating

each study on the Quality of Evidence Score checklist (Becker & Curry,

2008).

Review Methods

10
 As this paper is an update of the 2011 review paper, the review

methods in this study follow those of Midgley & Kennedy (2011). The search

strategy followed that outlined in an earlier publication (Kennedy, 2004),

covering the period from March 2011 to November 2016. Key psychology and

psychiatry databases were searched, using the same search terms as the

2004 publication. Publication types included were clinical trials, experimental,

follow-up, longitudinal, prospective and treatment outcome studies. The only

change to the review methods of the earlier review paper is the inclusion of a

systematic assessment of the quality of each study, as set out in more detail,

below.

Inclusion and exclusion criteria

The inclusion and exclusion criteria for this review followed that of the earlier

2011 review, which were:

i. Age. Studies of children and adolescents were included, where the

majority of participants were between the age of three and 18 years

old, and no participant was over the age of 25. Studies that focused on

parent-infant work were excluded.

ii. Interventions. Studies were included where they involved individual

psychodynamic or psychoanalytic psychotherapy. Short and long-term

therapies were included. Studies were included where the researchers

specified the treatment as psychodynamic or psychoanalytic, and were

excluded if they did not specify that the treatment was psychodynamic

or psychoanalytic.

11
 iii. Study focus. Studies were included if they were primarily concerned

with evaluating treatment outcomes. Studies focusing on the process of

therapy and clinical case reports were excluded.

iv. Study quality. Studies were not excluded based on quality, but each

study that met the inclusion criteria for this review was assessed for

quality.

v. Other criteria. English-language publications were systematically

included, although when identified during the search non-English

studies were also included. Additional studies were included if

identified by key informants. Unpublished studies were included, but

were identified as such.

Data extraction

Studies that met inclusion criteria for this review were summarised and are

presented in a data extraction table (Table 1). A critical appraisal of each

study was undertaken. Studies were assessed for quality using a checklist

designed for this purpose, which assesses studies across 14 methodological

attributes which are judged as being met (1) or not met or unclear (0). One

item has a possible rating of 0-2, which is based on intent-to-treat (ITT)

analysis (2= ITT analysis; 1= available case analysis; 0= treated case

analysis). These ratings provide a Quality of Evidence Score (QES), with

higher scores reflecting higher study quality (Becker & Curry, 2008). The

second author rated the studies, and the studies were also double rated by a

colleague. Inter-rater reliability was excellent between the raters (report

reliability once ratings completed). The study ratings are presented in Table 2.

12
Findings of this review: outcomes for children and young people

23 studies met the inclusion criteria for this review paper. The studies

will be critically discussed in relation to the specific diagnostic groups they

report on.

Mixed diagnoses

The majority of the studies included in this review paper were

naturalistic, and therefore reported on the outcomes of children presenting

with a range of difficulties receiving psychodynamic psychotherapy.

The only study of mixed diagnoses to use an RCT design was carried

out in Germany. The study examined the effects of psychodynamic treatment

in an inpatient setting, in adolescents who met criteria of a mixed conduct and

emotional disorder at baseline. 68 adolescents (14-19 years old) were

recruited into the trial, and were randomised to receive psychodynamic

treatment in an inpatient setting or to the waitlist group, after which they

received inpatient treatment (Salzer, Cropp, Jaeger, Masuhr, & Streeck-

Fischer, 2013). The authors describe the design as a “hybrid efficacy-

effectiveness RCT”, with the aim of drawing on the strengths of both RCT’s

(e.g. random allocation, manualized treatment and standardised outcome

measures) and effectiveness studies (e.g. treatment conducted in a

naturalistic setting and with few exclusion criteria). The treatment group had a

significantly higher rate of remission (OR = 26.41), and significantly better

outcomes on the SDQ (d = 0.90). These effects were maintained at six-month

follow up. The treatment group did not have significantly better outcomes than

13
the control group on the GSI. This is the first controlled study of a sample with

significant comorbidity, providing preliminary evidence for the use of

psychodynamic treatment in young people with complex difficulties.

Other studies of mixed diagnoses have utilised naturalistic and

observational designs. Stefini and colleagues carried out a study on a

heterogeneous sample of 71 children and adolescents (6-18 years old), who

met criteria for a mental disorder as determined by ICD-10. Participants had a

range of presenting problems (51% internalizing problems; 11% externalizing

problems; 38% mixed diagnoses). They received long-term psychodynamic

psychotherapy, with an average 82 sessions (S.D. = 52.6) (Stefini et al.,

2013). Three quarters of participants achieved reliable and clinically

significant change (ES = 1.95), as measured by the Severity of Impairment

Score for Children and Adolescents (SIS-CA). Further gains were made by

the one-year follow up, with 87% having achieved good outcomes on the SIS-

CA. At baseline, 22.5% were rated as having secure attachments. By the end

of treatment, those with secure attachments had increased to 63.4%, and this

figure increased to 76.6% by one-year follow up. The authors concluded that

there is support for the hypothesis that long-term psychoanalytic treatment

can shift clients’ attachment towards a secure style. Participants with both

secure and insecure attachments were successfully treated with

psychoanalytic treatment in this study, but those with insecure attachments

required more sessions than those who were securely attached. The study

was limited by the lack of control group, and notably, the study also used a

measure of attachment developed specifically for this study: the Heidelberg

Attachment Style Rating for Children and Adolescence (HASR-CA), rather

14
than an existing and validated measure of attachment, which makes it difficult

to draw firm conclusions from this study. Further research is required utilising

validated measures of attachment.

Another observational study had a sample of 218 participants, aged

14-24 years (Edlund & Carlberg, 2016), who received psychodynamic

psychotherapy in a naturalistic setting. The authors report that participants

showed a significant improvement in general functioning with large effect

sizes, as measured by the CGAS (d = 1.54) and the GAF (d = 2.02), as well

as decreased symptom severity with a medium-large effect size, as measured

by the SCL-90 (d=0.76), at the end of treatment. Those receiving longer term

treatment improved more than those whose treatment was shorter in duration.

This was a naturalistic study drawing on cases from a clinic in Sweden

between 2002 and 2009. However, it is important to note the limitations of this

study, as there was no control group, participants were not followed up

beyond the end of treatment, and they excluded participants from the analysis

if they attended fewer than six sessions.

Another naturalistic study, with a sample of 207 participants aged 4-12

years (Edlund, Thorén, & Carlberg, 2014) found psychodynamic

psychotherapy was associated with a significant improvement in functioning,

as measured by the CGAS, with a large effect size (d = 1.35). The authors

reported that 40% of participants achieved clinically significant change on the

CGAS. Effect sizes were large for all diagnostic groups (anxiety; attention-

deficit and disruptive behaviour; pervasive developmental; adjustment

disorders). Improvement measured on the SDQ subscales were found with

small-medium effect sizes (d = 0.21-0.50). Younger children (4-6 year olds)

15
showed larger improvements in general functioning at the end of treatment

than older children (10-12 years old). The study was naturalistic, conducted

retrospectively using a clinic research database. This was an uncontrolled

study, with no control group or follow up, limiting the extent to which any firm

conclusions can be drawn from this study.

One study of 28 young people receiving psychodynamic psychotherapy

found that adolescents, their parents and therapists reported a significant

reduction in symptomology by the end of treatment, across measures of

somatic, mental and social impairment. The strengths of this study are that

change was reported from multiple perspectives (Seiffge-Krenke & Nitzko,

2011). The authors report a waitlist condition, but do not report the outcomes

of the control group, and therefore while the study does suggest the therapy

was effective, it cannot be reported whether this improvement was beyond

what would be been observed by spontaneous remission.

Two publications have resulted from a naturalistic study of adolescents

receiving psychodynamic psychotherapy in outpatient clinics in Israel. The

sample comprised of 72 adolescents (aged 15-18), and the comparison group

comprised of a non-clinical community control group. The authors report that

those in the treatment group became less rigid in their interpersonal patterns

and improved significantly in their symptoms. No such changes were

observed in the community sample (Atzil-Slonim, Shefler, Gvirsman, &

Tishby, 2011). The second published study from this research focused on

adolescents’ changes in internal representations of relationships with their

parents. The authors reported that adolescents’ internal representations of

their relationships with their parents changed significantly throughout

16
treatment, as the treatment group increased in “close and supportive

interactions” and abandonment representations reduced (Atzil-Slonim,

Shefler, Slonim, & Tishby, 2013). The limitations of these studies were the

lack of clinical control group, which makes it impossible to account for

potential spontaneous changes in a clinical population. The authors also

excluded those who dropped out of therapy from the analyses.

A natural follow-up study drawing on hospital records in a child

psychiatry setting examined the outcomes of all children below the age of 16

between 1996-2005, which received individual psychotherapy. Their sample

comprised of 118 children (aged 4-15), and a comparison group of 118 age

and sex matched children who received other psychosocial treatments

(Ryynänen, Alen, Koivumaa-Honkanen, Joskitt, & Ebeling, 2015). While this

study was not specifically about psychodynamic treatment, 93% of the

therapists were psychodynamic. The authors found that children with

internalizing problems benefitted the most from psychotherapy, while family

violence and child protection intervention were associated with poorer

prognosis among psychotherapy patients. As this study was based on

hospital records, it reflects routine clinical practice, yet the conclusions that

can be drawn are limited as although 93% of the therapists were classed as

psychodynamic therapists, the extent to which they drew on the

psychodynamic model with these cases is unknown as treatment integrity was

not measured. A further limitation of this study is that long-term follow up was

based on hospital records of future psychiatric care. It is therefore unknown

whether clients may have sought treatment elsewhere or if they had future

problems but did not seek help.

17
 A small scale study sought to examine the feasibility and clinical use of

a Goal Based Outcome Measure (GBOM; Law, 2009), a collaborative

measure for patient and clinicians to use together to establish appropriate and

achievable goals for treatment. Goals are rated on a 0-10 scale, with zero

reflecting the patient not having dealt with the goal at all, and ten reflecting the

goal having been fully met. 34 participants receiving psychoanalytic

psychotherapy (either individual or group) in a CAMHS setting in the UK used

the GBOM (Emanuel, Catty, Anscombe, Cantle, & Muller, 2014). The mean

improvement was 3.2 points on the scale, which reached statistical

significance. This suggests psychoanalytic treatment was beneficial for these

young people in terms of their own goals for treatment. However, it is

important to note that this study was not specifically aiming to report on the

outcomes of psychoanalytic treatment, and as the focus was on reporting the

feasibility of using the GBOM, so at this stage it is unclear what should be

considered as clinically meaningful change on this measure. Furthermore, the

lack of control group makes it impossible to know what spontaneous

improvements may have been observed in those not receiving treatment.

Another observational study was carried out, to investigate

psychodynamic psychotherapy in a sample of 30 children and adolescents

(Krischer et al., 2013). The authors report significant improvements on the

CBCL, with medium effect sizes, although parent-rated quality of life was note

found to improve over the course of the therapy. However, the study had a

small sample and lacked a control group.

A recent observational study was published which reported on the

outcomes of 23 children and adolescents, who received weekly

18
psychodynamic psychotherapy, and their parents received fortnightly support

sessions alongside their child’s therapy (Gatta, Sisti, Sudati, Miscioscia, &

Simonelli, 2016). After one year of treatment, a statistically significant

reduction in internalizing symptoms was found. The authors also assessed

family interactions during the therapy, from which they found that family

interactions remained similar by the one year follow up. The study was a small

scale pilot study, lacking a control group, limiting the conclusions that can be

drawn from this research.

One study carried out a long-term follow-up with eight males who had

received psychoanalytic psychotherapy in adolescence for a range of mental

health problems (Sugar & Berkovitz, 2011a). They completed a questionnaire

between 20 and 30 years after their psychoanalytic psychotherapy in

adolescence began. The authors reported that some had good outcomes and

had fulfilling adulthoods, while those who had poorer outcomes were those

who had had a poorer relationship with their therapist, higher symptom

severity and physical illness during adolescence. While this study benefits

from a long-term follow up, this was an uncontrolled study, without a control

group and the authors did not use a standardized outcome measure, making

it difficult to draw any strong conclusions from this research.

Depression

Three studies in this review focused specifically on psychodynamic

psychotherapy in the treatment of depression.

The largest and best-designed RCT study of psychoanalytic

psychotherapy to date is the IMPACT study; a pragmatic trial comparing two

19
specialist therapies, Short-Term Psychoanalytic Psychotherapy (STPP) and

Cognitive-Behavioural Therapy (CBT), with a brief psychosocial intervention

(BPI), in the treatment of depression in adolescents (aged 11-17) (Goodyer et

al., 2016). 465 participants who met criteria for moderate to severe

depression were recruited into the trial. Participants were clinically referred

and therefore reflect clients routinely referred into NHS services in the UK,

with 47% of the young people receiving STPP having one or more co-morbid

psychiatric diagnosis (most frequently generalised anxiety disorder, social

phobia, post-traumatic stress disorder and oppositional defiant disorder), 35%

having a recorded lifetime suicide attempt and 54% reporting non-suicidal

self-injury episodes.

Young people in all three arms of the study were found to have

sustained reduced depressive symptoms. STPP was found to be equally as

effective as CBT and BPI in maintaining reduced depressive symptoms a year

after the end of treatment, with an average of 49-52% reduction in depressive

symptoms one year after the end of treatment. There were no significant

differences in total costs between the three treatment groups by the end of

study. Although no superiority effects for STPP at long-term follow up were

found, 85% of adolescents receiving STPP no longer met diagnostic criteria

for depression one year after the end of treatment, compared with 75% and

73% in the CBT and BPI arms respectively. This difference was not found to

be statistically significant, but does provide an indication of the effectiveness

of STPP in terms of long-term depression remission.

Further improvements were observed in the young people receiving

STPP, including a 59% reduction in anxiety symptoms, 43% reduction in

20
obsessive-compulsive symptoms and a 45% reduction in functioning

impairment, which were similar to the improvements observed in the other two

treatment arms. Only 4% of young people in the STPP treatment arm had

relapsed by the time of the one-year follow-up, compared to 11.6% in BPI and

16.5% in CBT. However the study was not powered for treatment group

comparisons, and these differences were not statistically significant.

Nevertheless, these findings are the strongest support to date for the long-

term effectiveness of psychoanalytic psychotherapy in the treatment of

adolescent depression, which can be observed across a range of symptoms.

The strengths of this study were the large sample, long-term follow up, the

use of standardised outcome measures, the fact that outcome assessors

were blind to treatment allocation, and that treatment fidelity was assessed by

independent raters.

The 2011 review paper reported Trowell et al's (2007) RCT comparing

psychodynamic psychotherapy and family therapy in the treatment of

depression in children and adolescents. Since then, a study carrying out

secondary analyses on data from Trowell et al’s (2007) study was published.

This study focused on the effect of psychodynamic psychotherapy and family

therapy on self-esteem and social adjustment (Kolaitis et al., 2014). The

authors found significant improvements in depression severity, self-esteem

and social adjustment at the end of treatment and six-month follow-up in both

treatment arms, suggesting that both treatments were equally effective across

these domains. The authors also found a significant interaction between type

of treatment and social adjustment with friends, revealing that social

21
adjustment with friends improved more for those who received

psychodynamic psychotherapy than those in the family therapy arm.

A smaller study was published in 2013 reported on the outcomes of 53

children and young people (3-21 years old) who took part in a quasi-

randomised study (Weitkamp et al., 2014). Participants were allocated to

either a psychoanalytic psychotherapy or a waitlist condition. At the end of

therapy, there was a reduction in depression pathology in the treatment

group, with a large effect size based on child (d = 0.81) and parent-report (d =

1.09). A significant reduction in depressive pathology was also found in the

waitlist group based on parent report (d = 0.64), but not based on child report.

In the treatment group, an improvement in quality of life was also found with

moderate to large effect sizes (child report d = 0.56; parent report d = 0.83).

At one-year follow-up, 53% of the treatment group did not have any

psychiatric disorder, suggesting potential sustained long-term effects of

psychoanalytic psychotherapy. However, data was not available on quality of

life or long-term psychiatric diagnoses in the waitlist group. While this study

offers some further support for the use of psychoanalytic psychotherapy for

young people with depression, no firm conclusion can be drawn from this

study as it was not sufficiently powered, limited data were available with the

waitlist control group and participants were not randomised to treatment

groups, as allocation was based on the availability of clinicians.

Taken together with the studies reported in our 2011 review, these

additional studies support the view that young people with moderate to severe

depression have at least equally good outcomes in psychodynamic therapy as

in other well-supported approaches, such as CBT and family therapy; and

22
supports the guidance of the National Institute of Clinical Health and

Excellence (NICE) that STPP should be made available as part of clinical care

for this population. However the fact that psychodynamic therapy has not

been compared to a waiting-list or no-treatment group in a randomised

controlled trial limits the confidence with which we can conclude that STPP is

responsible for the clinical changes observed, which explains why

psychodynamic therapy did not feature as an empirically-supported treatment

in the network meta-analysis carried out by Zhou et al. (2015).

Self-harm

The only study in this review to investigate treatment for reducing self-

harm compared Mentalization Based Treatment for Adolescents (MBT-A) with

Treatment As Usual (TAU). MBT-A was a year long, manualized,

psychodynamic treatment, comprising of weekly individual sessions and

monthly family sessions. 80 participants were recruited into this pragmatic

RCT (Rossouw & Fonagy, 2012). The authors found significant reductions in

self-harm and risk taking behaviours in both groups (as measured by the Risk

Taking and Self-Harm Inventory for Adolescents; RTSHIA; Vrouva, Fonagy,

Fearon, & Roussow, 2010). These reductions were significantly greater for the

MBT-A group, with a 44% recovery rate in the MBT-A group compared to 17%

in the TAU group. The MBT-A group also made moderately greater

improvements in depression scores compared to the TAU group (d = 0.49).

No difference in risk taking were found between the groups at 12 months,

although it is important to note that the MBT-A group had significantly more

23
risk-taking at baseline than the TAU group. Overall the study found modest

effect sizes, but does suggest potential in this treatment for reducing self-

harm in young people. This was a well-designed study, yet it did have a

relatively small sample size and is the only known study to investigate a

psychodynamic intervention in the treatment of self-harm. Larger scale

studies are required to strengthen the evidence base for the treatment of self-

harm.

Anxiety disorders

Two studies focused on samples of youth with anxiety disorders, both

of which were pilot studies with small sample sizes. The first of these studies

investigated manualized psychodynamic psychotherapy which was delivered

twice weekly over a 12-week period (Milrod et al., 2013). Ten participants

were recruited, aged 8-16, and had a diagnosis of an anxiety disorder. Except

for the one participant who dropped out, all participants no longer met criteria

for their primary diagnosis by the end of treatment, and clinically significant

improvements were found across outcome measures, including anxiety

symptoms and general functioning. These gains were maintained at six-month

follow up. However, the study was limited by a very small sample size and

lack of control group.

The limitations of the previous study were somewhat addressed in a

later study, which used a quasi-experimental design. 30 children aged 4-10

years old who met criteria for an anxiety disorder were recruited. 18 were

allocated to receive 20-25 sessions of Psychoanalytic Child Therapy (PaCT),

and 12 were allocated to a waitlist control group, after which they also

24
received PaCT (Göttken, White, Klein, & von Klitzing, 2014). PaCT is a play

focused and emotion oriented manualized psychoanalytic treatment. Based

on intent-to-treat analyses, 60% of the treatment group no longer met

diagnostic criteria for any anxiety disorder, whereas in the waitlist group, no

participants had remitted by the end of the waitlist. Parent and teacher

reported improvements were statistically significant on the Strengths and

Difficulties Questionnaire (SDQ; Goodman, 1997) subscales, and child

internalizing problems as measured by the Berkeley Puppet Interview (BPI;

Measelle, Ablow, Cowan, & Cowan, 1998). At six-month follow-up, the effects

of treatment were maintained on parent and teacher-report, although the child

report did not show significant effects of treatment. There was no evidence of

a sleeper effect in this study. The limitations of this study were a small sample

and participants were not randomised; allocation was determined by the

availability of therapists and outcome assessors were not blind to treatment

allocation, which poses the risk of bias in the study. While these studies offer

some preliminary support for the use of psychoanalytic psychotherapy in the

treatment of anxiety disorder in children and adolescents, these studies were

uncontrolled with small samples so it is impossible to draw any firm

conclusions about the effectiveness of psychoanalytic treatment for anxiety

disorders.

Disruptive behavior disorders

One study investigated psychoanalytic psychotherapy in the treatment

of disruptive ? behavior disorders. 73 participants, aged 6-11 years old, with

oppositional defiant disorder and/or attention deficit hyperactivity disorder

25
were recruited into this controlled observational study (Laezer, 2015).

Participants were allocated to receive psychoanalytic psychotherapy, or

behavioural therapy and/or medication. Both treatment groups demonstrated

significant symptom reduction, with no significant differences between the two

groups. An interesting finding in this study was that the majority of children in

the behavioural/medication group were still receiving medication at follow-up,

with no better results than the psychoanalytic psychotherapy group. The

authors note that it is generally assumed that psychoanalytic psychotherapy

takes longer than any other form of treatment, whereas in fact medication was

found to be the longest form of treatment, suggesting that psychoanalytic

psychotherapy may offer a viable alternative to medication. The study had a

relatively small sample size and allocated to the treatment arms was

naturalistic, so these findings should be viewed as preliminary, and future

studies should utilise an RCT design.

Personality disorders

Two studies investigated psychodynamic psychotherapy in the

treatment of borderline personality disorder. One of these was an

observational study of 28 participants receiving psychodynamic

psychotherapy (Salzer, Cropp, & Streeck-Fischer, 2014). At the end of

treatment, 39% of the participants had remitted and statistically significant

improvements were observed on a range of other measures. This study may

be viewed as preliminary support for the use of psychodynamic

psychotherapy in the treatment of personality disorders. However, given the

small sample, lack of control group and long-term follow-up, further research

26
is needed to draw any conclusions and the effectiveness of this treatment in

this clinical population.

The second study reported on three females who were treated for

borderline personality disorder during their adolescence, and were followed

up 15-20 years later. The authors concluded that all three were considered to

be in remission and “fulfilling, successful adult lives, despite not being entirely

free of psychopathology” (Sugar & Berkovitz, 2011b). It is not possible to draw

any firm conclusions from this uncontrolled study because of the small

sample, lack of control group or standardized outcome measures.

Children in foster care

One study reported on the outcomes of children who received long-

term, psychoanalytic, relational play therapy. The intervention was long-term,

lasting “for as long as it takes”. The study reported on the outcomes of 20

children, and found they had statistically significant reductions in mental

health symptoms, improved peer relationships and reduced school problems

(Clausen, Ruff, Wiederhold, & Heineman, 2012). These findings present some

support for the use of long-term psychoanalytically orientated treatment for

children in foster care, yet the study was limited by a small sample size, lack

of control group and change was measured based on therapist report, which

poses the risk of bias as therapists may overestimate change.

Physical health

The only study on physical health was a pilot randomized controlled

trial, investigating brief psychodynamic psychotherapy in the treatment of

27
idiopathic headache (Balottin et al., 2014). Participants were randomly

allocated to receive brief psychodynamic psychotherapy or care as usual. The

authors reported statistically significant greater gains for the treatment group

on the frequency, intensity and duration of headache attacks and

improvement in the CGI scores, as rated by a physician. The study provides

preliminary support for brief psychodynamic psychotherapy in the treatment of

headaches compared with care as usual. As this was a pilot study, the sample

size was small (N = 36) highlighting the need for a sufficiently powered study

to build on these preliminary findings. The study had a short follow up period

of six months which further limits the findings, so the long-term benefits of the

therapy are unknown. Nevertheless, these findings suggest that

psychodynamic psychotherapy may be effective in treating physical health

problems, yet this area remains understudied, with this being the only study of

its kind during the past five year period.

Assessment of study quality

The majority of the studies (78%) met the attribute for having explicit

hypotheses and/or objectives amenable to statistical analysis. A high

proportion of the studies (61%) specified the primary and secondary

outcomes. Despite this, only four (24%) reported the process for determining

sample size and four studies were sufficiently powered to detect differences

between the groups.

Six of the studies (26%) were randomized, thus fulfilling the criteria that

each participant had an unpredictable, independent chance of receiving each

intervention. However, only three of these described the study in adequate

28
detail to determine that treatment allocation of participants was not identifiable

to the research team.

Seven of the studies (30%) had an active comparison group, three of

which were treatment as usual (Balottin et al., 2014; Rossouw & Fonagy,

2012; Salzer et al., 2013), while the other four compared psychoanalytic

treatment to therapeutic approaches including cognitive-behavioural therapy,

family therapy and psychosocial interventions (Goodyer et al., 2016; Kolaitis

et al., 2014; Laezer, 2015; Ryynänen et al., 2015). Of the remaining studies,

four (24%) had a passive comparison group as they had a waitlist control

group (Edlund & Carlberg, 2016; Edlund et al., 2014; Göttken et al., 2014;

Weitkamp et al., 2014), while two (12%) used a community comparison group

(Atzil-Slonim et al., 2011, 2013) and five (29%) had no control group or did not

report outcomes of the control group (Clausen et al., 2012; Emanuel et al.,

2014; Gatta et al., 2016; Krischer et al., 2013; Milrod et al., 2013; Salzer et al.,

2014; Seiffge-Krenke & Nitzko, 2011; Stefini et al., 2013; Sugar & Berkovitz,

2011a, 2011b). The finding that only a third of the studies had a suitable

active comparison group is perhaps unsurprising given that many of the

studies were observational and carried out in naturalistic settings.

For the next attribute relating to whether the authors presented

baseline demographic and clinical data by treatment condition, studies were

rated zero if they did not include a comparison group. The majority of studies

(65%) did report baseline data by treatment condition. Almost half of the

studies used a manualised treatment (44%), while fewer assessed treatment

adherence (30%)

29
 While the RCT’s were the studies rated with the highest study quality,

interestingly, only one of them met the attribute for collateral report. However,

seven studies did meet this attribute, reflecting that the use of collateral report

has been used in more naturalistic and observational studies, but is one area

that is lacking from many RCT’s. Eight (35%) studies fulfilled the attribute for

intent-to-treat analysis. One of the least reported attributes of these studies

was the use of blind assessors, which were only reported in three (13%) of

the studies.

Discussion

This updated review identifies five Randomised Controlled Trials (one

of which was a secondary analysis of an RCT reported in the previous review

paper) which have been published since our previous review. One of these

RCTs, the IMPACT study (Goodyer, 2016) is the largest study to date to

include a psychodynamic treatment arm either in children or adults (n=465)

The findings of this study found that 85% of adolescents receiving STPP no

longer met criteria for depression one year after the end of treatment,

compared with 75% and 73% in the CBT and BPI arms respectively. This

would suggest that extending the evaluation of psychodynamic psychotherapy

as a treatment option for children and young people with a variety of clinical

diagnoses where it shows promise as an intervention is warranted. The three

other Randomised Controlled Trials published since the last review (Balottin

et al., 2014; Rossouw & Fonagy, 2012; Salzer et al., 2013) had relatively

small sample sizes (n=33, 80 and 66 respectively), yet all studies showed

potential benefits of a psychodynamic treatment for patients with complex and

30
severe difficulties (self-harm and depression; and adolescent with co-morbid

diagnoses), indicating that further randomized evaluation involving a larger

sample of adolescents could more definitively evaluate whether this is a

treatment that might benefit young people with such complex conditions. Of

the remaining studies, three were quasi-experimental designs, 12 were

observational studies without a control group and three were observational

studies with a comparison group. Such study designs limit the conclusions

that can be made regarding the effectiveness of the intervention. Yet there

were indications that psychodynamic psychotherapy might be a beneficial

intervention for children and young people with a range of clinical diagnoses.

The majority of studies focused on children and young people with mixed

diagnoses (9 out of 17) and most of the remaining studies included

participants with anxiety or depression, and one study focused on self-harm

with co-morbid depression.

The studies included in this review were assessed for study quality,

and the highest quality studies were the RCT’s, which had investigated

psychoanalytic treatment for depression, self-harm, headaches and

adolescents with significant co-morbidity. No sufficiently high quality studies

were found in samples of children and adolescents with anxiety, disruptive

behavior problems or personality disorders. The findings from the quality

assessment ratings reveal a lack of studies that were randomized, sufficiently

powered to detect meaningful differences between treatment conditions, and

with independent, blind outcome assessors.

It is important to note that an increasing number of studies are

integrative, drawing on psychodynamic techniques alongside other

31
approaches (Müller et al., 2015; Rothschild-Yakar, Lacoua, & Stein, 2013).

Such studies make it less straightforward to review the evidence fort the

effectiveness of specific treatment modalities; yet also reflect the potential of

more integrative approaches which may draw on the strengths of different

approaches.

One of the barriers to further research evaluating of psychodynamic

psychotherapy as a treatment for children and young people is the relative

underfunding of research both in psychological therapies and focused on

children and young people (MQ, 2015). Within psychological therapies

research, psychodynamic psychotherapy is one the least well funded

therapies (1.96% of total funding compared to 27.55% for CBT) and this

inevitably limits the scope for undertaking rigorous evaluation of it’s

effectiveness (MQ, 2015). In the rare situation where psychodynamic

psychotherapy has been evaluated within a high quality, adequately powered

RCT it has been found to be at least as clinically and cost effective as other

treatments (Goodyer et al., 2016), strengthening the case for further

evaluations of psychodynamic psychotherapy as a treatment modality for

different clinical conditions and different age ranges.

Conclusion

References

Abbass, A. a., Rabung, S., Leichsenring, F., Refseth, J. S., & Midgley, N.

(2013). Psychodynamic psychotherapy for children and adolescents: A

meta-analysis of short-term psychodynamic models. Journal of the

32
 American Academy of Child and Adolescent Psychiatry, 52(8), 863–875.

Alvarez, A. (2012). The Thinking Heart: Three Levels of Psychoanalytic

Therapy with Disturbed Children (1st ed.). Hove, UK: Routledge.

Atzil-Slonim, D., Shefler, G., Gvirsman, S. D., & Tishby, O. (2011). Changes

in rigidity and symptoms among adolescents in psychodynamic

psychotherapy. Psychotherapy Research, 21(6), 685–697.

Atzil-Slonim, D., Shefler, G., Slonim, N., & Tishby, O. (2013). Adolescents in

psychodynamic psychotherapy: changes in internal representations of

relationships with parents. Psychotherapy Research, 23(2), 201–17.

Balottin, U., Ferri, M., Racca, M., Rossi, M., Rossi, G., Beghi, E., … Termine,

C. (2014). Psychotherapy versus usual care in pediatric migraine and

tension-type headache : a single- blind controlled pilot study. Italian

Journal of Pediatrics, 40(6), 1–7.

Barlow, J., Bennett, C., Midgley, N., Larkin, S., & Wei, Y. (2015). Parent-infant

Psychotherapy for Improving Parental and Infant Mental Health: A

Systematic Review. Campbell Systematic Reviews, 6.

Baruch, G. (1995). Evaluating the outcome of a community-based

psychoanalytic psychotherapy service for young people between 12–25

years old: work in progress. Psychoanalytic Psychotherapy, 9(3), 243–67.

Baruch, G., Fearon, P., & Gerber, A. (1998). Evaluating the outcome of a

community– based psychoanalytic psychotherapy service for young

people: one year repeated follow up. In M. DAVENHILL, R. and

PATRICK (Ed.), Rethinking Clinical Audit. London: Routledge.

Becker, S. J., & Curry, J. F. (2008). Outpatient interventions for adolescent

substance abuse: a quality of evidence review. Journal of Consulting and

33
 Clinical Psychology, 76(4), 531–543.

Boston, M., & Lush, D. (1994). Further considerations of methodology for

evaluating psychoanalytic psychotherapy with children: reflections in light

of research experience. Journal of Child Psychotherapy, 20(5), 205–9.

Chanen, A. M., Jackson, H. J., McCutcheon, L. K., Jovev, M., Dudgeon, P.,

Yuen, H. P., … McGorry, P. D. (2008). Early intervention for adolescents

with borderline personality disorder using cognitive analytic therapy:

randomised controlled trial. The British Journal of Psychiatry : The

Journal of Mental Science, 193(6), 477–484. Comparative Study, Journal

Article, Randomized Controlled Trial, Research Support, Non-U.S. Gov’t.

Clausen, J. M., Ruff, S. C., Wiederhold, W. Von, & Heineman, T. V. (2012).

For As Long As It Takes: Relationship-Based Play Therapy for Children

in Foster Care. Psychoanalytic Social Work, 19(1–2), 43–53.

Deakin, E., & Nunes, M. L. T. (2009). Effectiveness of child psychoanalytic

psychotherapy in a clinical outpatient setting. Journal of Child

Psychotherapy, 35(3), 290–301.

Diamond, G. M. (2014). Attachment-based family therapy interventions.

Psychotherapy, 51(1), 15–19.

Edlund, J. N., & Carlberg, G. (2016). Psychodynamic psychotherapy with

adolescents and young adults: Outcome in routine practice. Clinical Child

Psychology and Psychiatry, 21(1), 66–80.

Edlund, J. N., Thorén, A., & Carlberg, G. (2014). Outcome of psychodynamic

child psychotherapy in routine practice. European Journal of

Psychotherapy & Counselling, 16(3), 228–244.

Emanuel, R., Catty, J., Anscombe, E., Cantle, A., & Muller, H. (2014).

34
 Implementing an aim-based outcome measure in a psychoanalytic child

psychotherapy service: insights, experiences and evidence. Clinical Child

Psychology and Psychiatry, 19, 169–83.

Eresund, P. (2007). Psychodynamic psychotherapy for children with disruptive

disorders. Journal of Child Psychotherapy, 33(2), 161–180.

Fonagy, P., & Target, M. (1994). The efficacy of psychoanalysis for children

with disruptive disorders. J Am Acad Child Adolesc Psychiatry, 33(1), 45–

55.

Fonagy, P., & Target, M. (1996). Predictors of outcome in child

psychoanalysis: a retrospective study of 793 cases at the Anna Freud

Centre. Journal of the American Psychoanalytic Association, 44(1), 27–

77.

Gatta, M., Sisti, M., Sudati, L., Miscioscia, M., & Simonelli, A. (2016). The

Lausanne Trilogue Play within the outcome evaluation in infant mental

health: A preliminary report. Research in Psychotherapy:

Psychopathology, Process and Outcome, 19(1), 19–30.

Gerber, A. J., Kocsis, J. H., Milrod, B. L., Roose, S. P., Barber, J. P., Thase,

M. E., … Leon, A. C. (2011). A quality-based review of randomized

controlled trials of psychodynamic psychotherapy. The American Journal

of Psychiatry, 168(1), 19–28.

Gilboa-Schechtman, E., & Foa, E. (2010). prolonged exposure vs dynamic

therapy for adolescent PTSD: a pilot randomized controlled trial. … of

Child & Adolescent …, 49(10), 1034–1042.

Goodman, R. (1997). The Strengths and Difficulties Questionnaire: A

Research Note. Journal of Child Psychology and Psychiatry, 38(5), 581–

35
 586.

Goodyer, I. M., Reynolds, S., Barrett, B., Byford, S., Dubicka, B., Hill, J., …

Fonagy, P. (2016). Effectiveness And Cost-Effectiveness Of Cognitive

Behaviour Therapy And Short-Term Psychoanalytic Psychotherapy

Compared With Brief Psychosocial Intervention In Maintaining Reduced

Depressive Symptoms 12 months after end of treatment in Adolescents

with U. Lancet Psychiatry.

Göttken, T., White, L. O., Klein, A. M., & von Klitzing, K. (2014). Short-term

psychoanalytic child therapy for anxious children: a pilot study.

Psychotherapy, 51(1), 148–58.

Heede, T., Runge, H., Storebø, O. J., Rowley, E., & Hansen, K. G. (2009).

Psychodynamic milieu-therapy and changes in personality—What is the

connection? Journal of Child Psychotherapy, 35(3), 276–289.

Heinicke, C. M., & Ramsey-Klee, D. M. (1986). Outcome of Child

Psychotherapy as a Function of Frequency of Session. Journal of the

American Academy of Child Psychiatry, 25(2), 247–253. article.

Horn, H., Geiser-Elze, A., Reck, C., Hartmann, M., Stefini, A., Victor, D., …

Kronmuller, K.-T. (2005). [Efficacy of psychodynamic short-term

psychotherapy for children and adolescents with depression]. Praxis der

Kinderpsychologie und Kinderpsychiatrie, 54(7), 578–597. English

Abstract, Journal Article, Multicenter Study, Randomized Controlled Trial.

Kegerreis, S. (2016). Further thoughts on research: a response to Michael

Rustin. Journal of Child Psychotherapy, 42(2), 198–207.

Kegerreis, S., & Midgley, N. (2014). Psychodynamic approaches. In A. B. Sue

Pattison, Maggie Robson (Ed.), The Handbook of Counselling Children &

36
 Young People.

Kendall, P. C., & Southam-Gerow, M. A. (1996). Long-term follow-up of a

cognitive–behavioral therapy for anxiety-disordered youth. Journal of

Consulting and Clinical Psychology. JOUR PMID - 8803362, US:

American Psychological Association.

Kennedy, E. (2004). Child and adolescent psychotherapy: A systematic

review of psychoanalytic approaches. London: North Central Health

Strategic Health Authority.

Kennedy, E. (2015). Editorial: Developing Interventions In Child and

Adolescent Mental Health Services: Do We Really Know What Works for

Whom? Clinical Child Psychology and Psychiatry, 20, 529–531.

Kolaitis, G., Giannakopoulos, G., Tomaras, V., Christogiorgos, S., Pomini, V.,

Layiou-Lignos, E., … Tsiantis, J. (2014). Self-esteem and social

adjustment in depressed youths: A randomized trial comparing

psychodynamic psychotherapy and family therapy. Psychotherapy and

Psychosomatics, 83(4), 249–251.

Krischer, M. K., Trautmann-Voigt, S., Kaspers, S., Voigt, B., Flechtner, H.-H.,

& Lehmkuhl, G. (2013). [Effectiveness of psychodynamic psychotherapy

in children and juveniles - results of a pilot study]. Zeitschrift fur Kinder-

und Jugendpsychiatrie und Psychotherapie, 41(2), 87–97. English

Abstract, Journal Article. http://doi.org/10.1024/1422-4917/a000216

Kronmüller, K., Postelnicu, I., Hartmann, M., Stefini, A., Geiser-Elze, A.,

Gerhold, M., … Winkelmann, K. (2005). [Efficacy of psychodynamic

short-term psychotherapy for children and adolescents with anxiety

disorders]. Praxis Der Kinderpsychologie Und Kinderpsychiatrie, 54(7),

37
 559—577. article.

Kronmüller, K., Stefani, A., Gesier-Elze, A., Horn, H., Hartmann, M., &

Winklemann, K. (2010). The Heidelberg study of psychodynamic

psychotherapy for children & adolescents. In J. Tsiantis & J. Trowell

(Eds.), Assessing Change in Psychoanalytic Psychotherapy of Children

and Adolescents. London: Karnac.

Laezer, K. L. (2015). Effectiveness of Psychoanalytic Psychotherapy and

Behavioral Therapy Treatment in Children with Attention Deficit

Hyperactivity Disorder and Oppositional Defiant Disorder. Journal of

Infant, Child, and Adolescent Psychotherapy, 14(2), 111–128.

Lanyardo, M., & Horne, A. (2009). The Handbook of Child and Adolescent

Psychotherapy (2nd ed.). London: Routledge.

Law, D. (2009). Goal Based Outcomes (GBOs): Some useful information.

London, UK: CAMHS Outcome Research Consortium.

Lock, J., Le Grange, D., Agras, W. S., Moye, A., Bryson, S. W., & Jo, B.

(2010). Randomized clinical trial comparing family-based treatment with

adolescent-focused individual therapy for adolescents with anorexia

nervosa. Archives of General Psychiatry, 67(10), 1025–1032.

Lush, D., Boston, M., Morgan, J., & Kolvin, I. (1998). Psychoanalytic

Psychotherapy with Disturbed Adopted and Foster Children: A Single

Case Follow-Up Study. Clinical Child Psychology and Psychiatry, 3(1),

51–69.

McLaughlin, C., Carol Holliday, Clarke, B., & Ilie, S. (2013). Research on

counselling and psychotherapy with children and young people: a

systematic scoping review of the evidence for its effectiveness from

38
 2003-2011. Leics: BACP.

Measelle, J. R., Ablow, J. C., Cowan, P. A., & Cowan, C. P. (1998). Assessing

young children’s views of their academic, social, and emotional lives: an

evaluation of the self-perception scales of the Berkeley Puppet Interview.

Child Development, 69(6), 1556–1576.

Midgley, N. (2006). The “inseparable bond between cure and research”:

clinical case study as a method of psychoanalytic inquiry. Journal of Child

Psychotherapy, 32(2), 122–147.

Midgley, N., & Kennedy, E. (2011). Psychodynamic psychotherapy for

children and adolescents: a critical review of the evidence base. Journal

of Child Psychotherapy, 37(3), 232–260.

Milrod, B., Shapiro, T., Gross, C., Silver, G., Preter, S., Libow, A., & Leon, A.

C. (2013). Does manualized psychodynamic psychotherapy have an

impact on youth anxiety disorders? American Journal of Psychotherapy,

67(4), 359–366.

MQ. (2015). MQ Landscape Analysis, April 2015: UK Mental Health Research

Funding. Mq.

Müller, J. M., Averbeck-Holocher, M., Romer, G., Fürniss, T., Achtergarde, S.,

& Postert, C. (2015). Psychiatric Treatment Outcomes of Preschool

Children in a Family Day Hospital. Child Psychiatry & Human

Development, 46(2), 257–269. http://doi.org/10.1007/s10578-014-0465-3

Muratori, F., Picchi, L., Apicella, F., Salvadori, F., Espasa, F. P., Ferretti, D., &

Bruni, G. (2005). Psychodynamic psychotherapy for separation anxiety

disorders in children. Depression and Anxiety. United States.

Muratori, F., Picchi, L., Bruni, G., Patarnello, M., & Romagnoli, G. (2003). A

39
 two- year follow up of psychodynamic psychotherapy for internalizing

disorders in children. Journal of the American Academy of Child and

Adolescent Psychiatry, 42(3), 331–9.

Muratori, F., Picchi, L., Casella, C., Tancredi, R., Milone, A., & Patarnello, M.

G. (2002). Efficacy of brief dynamic psychotherapy for children with

emotional disorders. Psychother Psychosom, 71(1), 28–38.

Odhammar, F., Sundin, E. C., Jonson, M., & Carlberg, G. (2011). Children in

psychodynamic psychotherapy: Changes in global functioning. Journal of

Child Psychotherapy, 37(3), 261–279.

Palmer, R., Nascimento, L. N., & Fonagy, P. (2013). The State of the

Evidence Base for Psychodynamic Psychotherapy for Children and

Adolescents. Child and Adolescent Psychiatric Clinics of North America,

22(2), 149–214.

Rossouw, T. I., & Fonagy, P. (2012). Mentalization-based treatment for self-

harm in adolescents: A randomized controlled trial. Journal of the

American Academy of Child and Adolescent Psychiatry, 51(12).

Rothschild-Yakar, L., Lacoua, L., & Stein, D. (2013). Changes in Patient

Measures as Predictors of Therapists’ Ratings of Treatment Collaboration

and Change in Eating Disorder Subgroups. Assessment, 20(6), 752–763.

Rustin, M. (2016). Grounded Theory methods in child psychotherapy

research. Journal of Child Psychotherapy, 42(2), 179–197.

Ryynänen, T., Alen, M., Koivumaa-Honkanen, H., Joskitt, L., & Ebeling, H.

(2015). Implementation and outcome of child psychotherapy compared

with other psychiatric treatments in a naturalistic clinical setting. Nordic

Journal of Psychiatry, 69(3), 179–187.

40
Salzer, S., Cropp, C., Jaeger, U., Masuhr, O., & Streeck-Fischer, A. (2013).

Psychodynamic therapy for adolescents suffering from co-morbid

disorders of conduct and emotions in an in-patient setting: a randomized

controlled trial. Psychological Medicine, 44, 2213–2222.

Salzer, S., Cropp, C., & Streeck-Fischer, A. (2014). Early Intervention for

Borderline Personality Disorder: Psychodynamic Therapy in Adolescents.

Zeitschrift Fur Psychosomatische Medizin Und Psychotherapie, 60(4),

368–382.

Schachter, A., & Target, M. (2009). The adult outcome of child

psychoanalysis: The Anna Freud Centre Long-term Follow-up Study. In

N. V. T. and U. C. Midgley N, Anderson J, Grainger E (Ed.), Child

psychotherapy and research: New approaches, emerging findings. (pp.

146–156). London: Routledge.

Seiffge-Krenke, I., & Nitzko, S. (2011). [How effective are long-term

psychoanalytic treatments in adolescents? A comparison between the

evaluations of the patients, their parents and their therapists]. Zeitschrift

Für Kinder- Und Jugendpsychiatrie Und Psychotherapie, 39(4), 253–64.

Sinha, U. K., & Kapur, M. (1999). Psychotherapy with emotionally disturbed

adolescent boys: outcome and process study. National Institute of Mental

Health and Neurosciences Journal, 17(2), 113–30.

Smyrnios, K. X., & Kirkby, R. J. (1993). Long-term comparison of brief versus

unlimited psychodynamic treatments with children and their parents.

Journal of Consulting and Clinical Psychology, 61(6), 1020–1027.

Stefini, A., Hildegard, H., Winkelmann, K., Geiser-Elze, A., Hartmann, M., &

Kronmuller, K. T. (2013). Attachment styles and outcome of

41
 psychoanalytic psychotherapy for children and adolescents.

Psychopathology, 46(3), 192–200.

Sugar, M., & Berkovitz, I. H. (2011a). Male Adolescent Treatment Outcome: A

Case Series of Eight Men Treated with Psychoanalytic Psychotherapy.

Adolescent Psychiatry, 1(2), 169–178.

Sugar, M., & Berkovitz, I. H. (2011b). Treatment outcome of three female

adolescents with borderline personality disorder. Adolescent Psychiatry,

1(1), 6–19.

Szapocznik, J., Rio, A., Murray, E., Cohen, R., Scopetta, M., Rivas-Vazquez,

A., … Kurtines, W. (1989). Structural family versus psychodynamic child

therapy for problematic Hispanic boys. Journal of Consulting and Clinical

Psychology, 57(5), 571–578.

Target, M., & Fonagy, P. (1994). The efficacy of psychoanalysis for children:

prediction of outcome in a developmental context. Journal of the

American Academy of Child and Adolescent Psychiatry, 33(8), 1134–44.

Tishby, O., Raitchick, I., & Shefler, G. (2007). Changes in interpersonal

conflicts among adolescents during psychodynamic therapy.

Psychotherapy Research, 17(3), 297–304.

Tonge, B. J., Pullen, J. M., & Hughes, G. C. (2009). Effectiveness of

psychoanalytic psychotherapy for adolescents with serious mental illness:

12 month naturalistic study. Australian and New Zealand Journal of

Psychiatry, 43(5), 467–75.

Trowell, J., Joffe, I., Campbell, J., Clemente, C., Almqvist, F., Soininen, M., …

Tsiantis, J. (2007). Childhood depression: a place for psychotherapy.

European Child & Adolescent Psychiatry, 16(3), 157–167.

42
Trowell, J., Kolvin, I., Weeramanthri, T., Sadowski, H., Berelowitz, M., Glaser,

D., & Leitch, I. (2002). Psychotherapy for sexually abused girls:

psychopathological outcome findings and patterns of change. The British

Journal of Psychiatry : The Journal of Mental Science, 180(April 2002),

234–247.

Trowell, J., Rhode, M., Miles, G., & Sherwood, I. (2003). Childhood

depression: Work in progress. Individual child therapy and parent work.

Journal of Child Psychotherapy, 29(2), 147–169.

Vrouva, I., Fonagy, P., Fearon, P. R. M., & Roussow, T. (2010). The risk-

taking and self-harm inventory for adolescents: development and

psychometric evaluation. Psychological Assessment, 22(4), 852–65.

Weitkamp, K., Daniels, J. K., Hofmann, H., Timmermann, H., Romer, G., &

Wiegand-Grefe, S. (2014). Psychoanalytic psychotherapy for children

and adolescents with severe depressive psychopathology: Preliminary

results of an effectiveness trial. Psychotherapy, 51(1), 138–147.

What is child and adolescent psychotherapy? (2016). Retrieved November

22, 2016, from http://childpsychotherapy.org.uk/child-psychotherapy/faqs

Winkelmann, K., Stefini, A., Hartmann, M., Geiser-Elze, A., Kronmüller, A.,

Schenkenbach, C., … Kronmüller, K.-T. (2005). [Efficacy of

psychodynamic short-term psychotherapy for children and adolescents

with behavioral disorders]. Praxis Der Kinderpsychologie Und

Kinderpsychiatrie, 54(7), 598—614. article.

43
 Table 1. Table of studies from 2011 to 2016

Type of Control Type of Primary outcome

Authors n Location Age Design
Problem group Therapy measures

Randomized Brief Headache frequency,

Balottin et al. Idiopathic Care as
33 Italy 6-18 years Controlled psychodynamic intensity and duration;
(2014) headache usual
Trial psychotherapy EuroQoL; CBCL
Observation
al study Psychoanalytic,
Clausen et al. No control
20 USA 5-10 years Foster care without relational play
(2012) group
control therapy
group

Children’s Global
Observation Assessment Scale
Edlund & Carlberg 14-24 Mixed al study No control Psychodynamic (CGAS);
218 Sweden
(2014) years diagnoses without group psychotherapy Global Assessment of
control Functioning (GAF);
group Symptoms Checklist-
90 (SCL-90)
Observation
CGAS;
al study
Edlund, Thoren & Mixed No control Psychodynamic Strength and
207 Sweden 4-12 years without
Carlberg (2014) diagnoses group psychotherapy Difficulties
control
Questionnaire (SDQ)
group

44
 Observation Psychoanalytic
Emanuel, Catty,
al study psychotherapy
Anscombe, Cantle Mixed No control Goal Based Outcome
34 London ?? without (study included
& Muller (2014) diagnoses group Measure (GBOM)
control individual and
and evidence.
group group therapy)
Observation
Psychodynamic
al study
Mixed No control psychotherapy Child Behaviour
Gatta et al. (2016) 23 Italy 4-17 years without
diagnoses group and parental Checklist (CBCL)
control
support
group
Cognitive
Behavioural
Randomised
Goodyer et al 11-17 Therapy; Psychoanalytic
465 UK Depression Controlled MFQ, KSADS
(2016) years Brief Psychotherapy
Trial
Psychosocial
Intervention
Preschool Age
Quasi- Psychiatric
Gottken, White,
Anxiety experimenta Psychoanalytic Assessment (PAPA);
Klein & 30 Germany 4-10 years Waitlist
disorders l controlled psychotherapy SDQ;
von Klitzing (2014)
study Child Behaviour
Checklist (CBCL)
Kolaitis, CDI;
Giannakopoulos, Culture-Free Self-
Finland.
Tomaras, Randomised Esteem Inventory
Greece. Family Psychodynamic
Christogiorgos, 72 9-15 years Depression Controlled (CFSEI);
United therapy psychotherapy
Pomini, Layiou- Trial Social Adjustment
Kingdom
Lignos, Tzavara, Scale-Self-Report
Rhode, Miles, (SAS-SR)

45
Joffe, Trowell &
Tsiantis (2014)

Observation
al study
Krischer et al. Mixed Psychodynamic Child Behaviour
30 Germany without N/A
(2013) diagnoses psychotherapy Checklist (CBCL)
control
group
Diagnostic System for
Mental Disorders in
Children and
Adolescents (DISYPS-
Attention
KJ); Conners Parent
deficit
Rating Scale (CPRS);
hyperactivity Quasi- Behavioural
Conners Teacher
disorder expermintal therapy Psychoanalytic
Laezer (2015) 73 Germany 6-11 years Rating Scale (CTRS);
and/or controlled and/or psychotherapy
CBC; Teacher Report
oppositional study medication
Form (TRF);
defiant
Inventory for the
disorder.
Assessment of the
Quality of Life in
Children and
Adolescents (IAQLiCA)

46
 Anxiety
disorders
(Generalize
d anxiety Observation Pediatric Anxiety
Milrod, Shapiro,
disorder, al study Rating Scale (PARS);
Gross, Silver, Psychodynamic
10 US 8-16 years social without N/A CGAS;
Preter, Libow & psychotherapy
phobia, control Clinical Global
Leon (2013)
and/or group Impression scale (CGI)
Separation
Anxiety
Disorder)
Mentalization
Randomised Based Risk-Taking and Self-
Rossouw & Fonagy 12-17 Treatment As
80 London Self-harm Controlled Treatment for Harm Inventory
(2012) years Usual
Trial Adolescents (RTSHIA)
(MBT-A)
93% of
therapists
Age and sex reported as
matched being
Ryynänen, Alen,
children psychodynamic
Koivumaa- Mixed Observation
236 Finland 4-15 years undergoing psychotherapsit CGAS
Honkanen, Joskitt diagnoses al study
other s - but not
& Ebeling (2015)
psychosocial explicitly a
treatments study of
psychodynamic
therapy

47
 Remission rates;
Observation
GAF;
Salzer, Cropp & Borderline al study
Psychodynamic GSI;
Streeck-Fischer 28 Germany ?? personality without N/A
psychotherapy SDQ;
(2014) disorder control
IIP;
group
BPI scores
Salzer, Cropp, Remission rates;
Randomised Psychodynamic
Jaeger, Masuhr & 14-19 Mixed Global Severity Index
66 Germany Controlled Waitlist / TAU psychotherapy
Streeck-Fischer years diagnoses (GSI);
Trial (inpatient)
(2013) SDQ

Observation
al study
Seiffge-Krenke & Mixed No control Psychodynamic YSR;
60 Germany >11 years without
Nitzko (2011) diagnoses group psychotherapy CBCL
control
group
Core Conflictual
Relationship Theme
Community method (CCRT);
Slonim, Shefler,
15-18 Mixed Observation control group Psychodynamic Youth-Outcome Ques-
Gvirsman & Tishby 72 Israel
years diagnoses al study (non-clinical psychotherapy tionnaire Self-Report
(2011)
sample) (Y-OQ-SR);
Target Complaints
Scale (TCS)

48
 Community
Slonim, Shefler, CCRT;
15-18 Mixed Observation control group Psychodynamic
Slonim & Tishby 72 Israel Y-OQ-SR;
years diagnoses al study (non-clinical psychotherapy
(2013) TCS
sample)
Heidelberg Attachment
Style Rating for
Stefini, Horn, Observation Children and
Winkelmann, al study Adolescents (HASR-
Mixed Psychodynamic
Geiser-Elze, 71 Germany 6-18 years without N/A CA);
diagnoses psychotherapy
Hartmann & control Severity of Impairment
Kronmuller (2013) group Score
for Children and
Adolescents (SIS-CA)
Observation
Borderline al study
Sugar & Berkovitz Psychoanalytic Questionnaire
3 USA 14-20 personality without N/A
(2011a) psychotherapy assessing functioning
disorder control
group
Observation
al study
Sugar & Berkovitz 16-19 Mixed Psychoanalytic Questionnaire
8 USA without N/A
(2011b) years diagnoses psychotherapy assessing functioning
control
group

49
Weitkamp, Daniels,
Hofmann, Quasi CDI;
Psychoanalytic
Timmermann, 53 Germany 4-21 years Depression randomised Waitlist KIDSCREEN;
psychotherapy
Romer & Wiegand- trial KSADS
Grefe (2014)

50
Table 2. Quality ratings of studies

Item

Article 1 2 3 4 5 6 7 8 9 10 11 12 13 Total

Balottin et al. (2014) 1 0 0 1 1 1 1 1 1 0 1 1 0 9

Clausen et al. (2012) 0 0 0 0 0 0 0 0 0 0 0 0 0

Edlund & Carlberg (2014) 1 0 0 0 0 0 0 1 0 0 0 0 0 2

Edlund, Thoren & Carlberg 1 0 0 0 0 0 0 1 0 0 1 1 0 4

(2014)

Emanuel et al. (2014) 0 0 0 1 0 0 0 0 0 0 0 0 0 1

Gatta et al. (2016) 1 0 0 1 0 0 0 0 0 0 0 0 0 2

Goodyer et al. (2016) 1 1 1 1 1 1 1 1 1 1 0 2 1 13

Gottken et al. (2014) 1 0 0 1 0 0 0 1 1 1 1 2 0 8

Kolaitis et al. (2014) 1 1 0 1 1 0 1 1 1 0 0 2 0 9

Krischer et al. (2013) ? 0 0 ? 0 0 0 0 ? ? ? ? ? ?

51
Laezer (2015) 1 0 0 1 0 0 1 1 1 1 1 0 0 7

Milrod et al (2014) 1 0 0 1 0 0 0 0 1 1 0 0 0 4

Rossouw & Fonagy (2012). 1 1 1 1 1 1 1 1 1 1 0 2 1 13

Ryynänen et al. (2015) 1 0 1 0 0 0 1 1 0 0 0 2 0 6

Salzer, Cropp & Streeck-Fischer 1 0 0 1 1 0 0 1 1 0 0 2 0 8

(2014)

Salzer et al. (2013) 1 1 1 1 1 0 1 1 1 1 0 2 0 11

Seiffge-Krenke & Nitzko (2011) 1 0 0 1 0 0 0 0 0 0 1 0 0 3

Slonim et al. (2011) 1 0 0 0 0 0 0 1 0 0 0 0 1 3

Slonim et al. (2013) 1 0 0 0 0 0 0 1 0 0 0 0 0 2

Stefini et al. (2013) 1 0 0 1 0 0 0 1 1 0 1 2 0 7

Sugar & Berkovitz (2011a) 0 0 0 0 0 0 0 0 0 0 0 0 0 0

Sugar & Berkovitz (2011b) 0 0 0 0 0 0 0 0 0 0 0 0 0 0

Weitkamp et al. (2014) 1 0 0 1 0 0 0 1 0 1 1 0 0 5

52
Note: Total score ranges from 0-14, with higher scores reflecting higher study quality.

(1) Objective: 1 = Specific objectives and hypotheses. Hypotheses are amenable to explicit statistical evaluation; 0 = Objectives

or hypotheses not explicitly established.

(2) Sample size: 1 = Process for determining the sample size discussed along with any interim analyses and stopping rules; 0 =

Determination of sample size not discussed.

(3) Power: 1 = Study is sufficiently powered to detect differences between treatment groups (e.g. at least 71 subjects per

condition with active comparison, 27 subjects per condition with passive comparison); 0 = Study is not sufficiently powered.

(4) Outcome: 1 = Established primary & secondary outcome measure. Primary outcome is specified as outcome of greatest

importance; 0 = Primary or secondary outcome measures are not specified.

(5) Sequence generation: 1 = Process for generating a random sequence described with sufficient detail to confirm that each

participant had an unpredictable, independent chance of receiving each intervention; 0 = Process was not purely random,

unspecified.

(6) Allocation concealment: 1 = Process of assigning participants to groups described with sufficient detail to confirm that

investigators recruiting and conducting the initial assessment could not discern the participant's treatment group; 0 =

Process was not concealed, unspecified.

53
(7) Active comparison: 1 = At least one active comparison (e.g. alternative model, treatment as usual); 0 = All comparison

conditions were passive (e.g. waitlist, no-treatment control).

(8) Baseline data: 1 = Baseline demographic and clinical characteristics reported by condition; 0 = Baseline demographic or

clinical characteristics not reported.

(9) Manualized treatment: 1 = At least one treatment condition was guided by a manual; 0 = None of the treatments were guided

by a manual, unspecified.

(10) Treatment adherence rating: 1 = Treatment adherence monitored with scales, checklists, or rating forms completed

by therapist, supervisor, independent observer, and/or patient; 0 = Treatment adherence was not monitored using rating

forms, unspecified.

(11) Collateral report: 1 = At least one outcome is a collateral report (e.g. parent, caregiver, teacher); 0 = No collateral

report.

(12) Intention-to-treat (ITT) analysis: 2 = ITT analysis. All subjects analyzed in groups to which they were assigned; 1 =

Available case analysis. Only subjects who completed one of more research assessments were analysed; 0 = Treated case

analysis. Only subjects who completed a portion of the treatment were analyzed.

54
 (13) Blind assessment: 1 = Follow-up assessments completed by treatment-blind evaluator; 0 = Follow-up not completed

by blind evaluator, unspecified.

i
The term 'psychodynamic therapy' will be used in this paper to cover both psychoanalytic and psyychodynamic approaches, although where specific studies refer to one or
the other term, we follow the authors' own terminology

55