Midgley Narrative - Review - 5.12.16
Midgley Narrative - Review - 5.12.16
Midgley Narrative - Review - 5.12.16
5SD
1
Psychodynamic psychotherapy for children and adolescents: an
Abstract
Introduction
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
sense. Their roots lie in the internal world of the child that has been built up
which formally evaluated therapy outcomes for children aged 3-18. Of these
studies, nine were randomised controlled trials (RCTs), three had a quasi-
2
delivered in a group or dyadic format, or with children below the age of 3, was
With regard to the practice and clinical implications of the 2011 review,
the following conclusions were cautiously drawn: [Probably need to revise this
research instruments.
3
children with emotional disorders, giving some evidence of a possible
However there are also studies that suggest that older children and
disruptive/externalising disorders
4
When children present with more marked difficulties e.g. with conduct
may be important
functioning and that more intensive work could, in some cases, add to
this field were draw in the previous review paper (Midgley & Kennedy, 2011).
The key points were that studies tended to have heterogonous clinical
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
development of cumulative knowledge about the evidence base for this type
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
5
(e.g. Hoffmann et al., 2015; Cregeen et al., 2016) and in addition, two further
because it reviews many of the same studies as our 2011 paper, yet reaches
based on the same studies. Palmer et al. aimed to identify, describe, and
review studies published before May 2012, evaluating the efficacy and/ or
relevant databases. Unlike our 2011 review, Palmer et al. included studies
justified this broader inclusion on the basis that the distinction between
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
Based on their review, Palmer et al. (2013) drew the following key
conclusions (p.175):
There is some evidence that effects tend to increase following the end
of treatment.
the evidence is stronger for younger children, where parents are almost
especially helpful.
evidence base for CBT in severe childhood disorders, further research into
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
In the same year as the Palmer et al. review was published, Abbass
analytic approach, the authors were able to pool results from a range of
smaller number of studies (11), these were all randomised controlled trials,
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
Robust (g= 1.07, 95% CI: 0.80–1.34) within group effect sizes were
increased gains.
8
STPP did not separate from what were mostly robust treatment
As with the other reviews discussed here, Abbass et al. noted that
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
Carol Holliday, Clarke, & Ilie, 2013), although based primarily on a review of
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
aggregate findings, to draw any firm conclusions about the efficacy and
al., 2013; Palmer et al., 2013). However, our 2011 paper concluded by noting
9
by decade since the 1970s, and that at the time of writing the 2011 review
therapy for children with specific diagnostic groups, and that these studies
Given the rapid developments in this field, the aim of this review paper
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:
therapeutic effectiveness.
adolescents.
(4) To assess the degree to which these more recent studies support,
(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
covering the period from March 2011 to November 2016. Key psychology and
psychiatry databases were searched, using the same search terms as the
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.
The inclusion and exclusion criteria for this review followed that of the earlier
old, and no participant was over the age of 25. Studies that focused on
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
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.
Data extraction
Studies that met inclusion criteria for this review were summarised and are
study was undertaken. Studies were assessed for quality using a checklist
attributes which are judged as being met (1) or not met or unclear (0). One
higher scores reflecting higher study quality (Becker & Curry, 2008). The
second author rated the studies, and the studies were also double rated by a
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
report on.
Mixed diagnoses
The only study of mixed diagnoses to use an RCT design was carried
effectiveness RCT”, with the aim of drawing on the strengths of both RCT’s
naturalistic setting and with few exclusion criteria). The treatment group had a
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
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
can shift clients’ attachment towards a secure style. Participants with both
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
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
sizes, as measured by the CGAS (d = 1.54) and the GAF (d = 2.02), as well
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.
between 2002 and 2009. However, it is important to note the limitations of this
beyond the end of treatment, and they excluded participants from the analysis
as measured by the CGAS, with a large effect size (d = 1.35). The authors
CGAS. Effect sizes were large for all diagnostic groups (anxiety; attention-
15
showed larger improvements in general functioning at the end of treatment
than older children (10-12 years old). The study was naturalistic, conducted
study, with no control group or follow up, limiting the extent to which any firm
somatic, mental and social impairment. The strengths of this study are that
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
those in the treatment group became less rigid in their interpersonal patterns
Tishby, 2011). The second published study from this research focused on
16
treatment, as the treatment group increased in “close and supportive
Shefler, Slonim, & Tishby, 2013). The limitations of these studies were the
psychiatry setting examined the outcomes of all children below the age of 16
comprised of 118 children (aged 4-15), and a comparison group of 118 age
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
not measured. A further limitation of this study is that long-term follow up was
whether clients may have sought treatment elsewhere or if they had future
17
A small scale study sought to examine the feasibility and clinical use of
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
the GBOM (Emanuel, Catty, Anscombe, Cantle, & Muller, 2014). The mean
important to note that this study was not specifically aiming to report 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
18
psychodynamic psychotherapy, and their parents received fortnightly support
sessions alongside their child’s therapy (Gatta, Sisti, Sudati, Miscioscia, &
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
One study carried out a long-term follow-up with eight males who had
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
Depression
19
specialist therapies, Short-Term Psychoanalytic Psychotherapy (STPP) and
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
self-injury episodes.
Young people in all three arms of the study were found to have
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
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
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
Nevertheless, these findings are the strongest support to date for the long-
The strengths of this study were the large sample, long-term follow up, the
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
secondary analyses on data from Trowell et al’s (2007) study was published.
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
21
adjustment with friends improved more for those who received
children and young people (3-21 years old) who took part in a quasi-
group, with a large effect size based on child (d = 0.81) and parent-report (d =
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
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
Taken together with the studies reported in our 2011 review, these
additional studies support the view that young people with moderate to severe
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
controlled trial limits the confidence with which we can conclude that STPP is
Self-harm
The only study in this review to investigate treatment for reducing self-
RCT (Rossouw & Fonagy, 2012). The authors found significant reductions in
self-harm and risk taking behaviours in both groups (as measured by the Risk
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
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
studies are required to strengthen the evidence base for the treatment of self-
harm.
Anxiety disorders
of which were pilot studies with small sample sizes. The first of these studies
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
follow up. However, the study was limited by a very small sample size and
years old who met criteria for an anxiety disorder were recruited. 18 were
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
diagnostic criteria for any anxiety disorder, whereas in the waitlist group, no
participants had remitted by the end of the waitlist. Parent and teacher
Measelle, Ablow, Cowan, & Cowan, 1998). At six-month follow-up, the effects
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
allocation, which poses the risk of bias in the study. While these studies offer
disorders.
25
were recruited into this controlled observational study (Laezer, 2015).
groups. An interesting finding in this study was that the majority of children in
takes longer than any other form of treatment, whereas in fact medication was
relatively small sample size and allocated to the treatment arms was
Personality disorders
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
The second study reported on three females who were treated for
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
any firm conclusions from this uncontrolled study because of the small
(Clausen, Ruff, Wiederhold, & Heineman, 2012). These findings present some
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
Physical health
27
idiopathic headache (Balottin et al., 2014). Participants were randomly
authors reported statistically significant greater gains for the treatment group
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
problems, yet this area remains understudied, with this being the only study of
The majority of the studies (78%) met the attribute for having explicit
outcomes. Despite this, only four (24%) reported the process for determining
sample size and four studies were sufficiently powered to detect differences
Six of the studies (26%) were randomized, thus fulfilling the criteria that
28
detail to determine that treatment allocation of participants was not identifiable
which were treatment as usual (Balottin et al., 2014; Rossouw & Fonagy,
2012; Salzer et al., 2013), while the other four compared psychoanalytic
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
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
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
was the use of blind assessors, which were only reported in three (13%) of
the studies.
Discussion
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
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
as a treatment option for children and young people with a variety of clinical
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
30
severe difficulties (self-harm and depression; and adolescent with co-morbid
treatment that might benefit young people with such complex conditions. Of
studies with a comparison group. Such study designs limit the conclusions
that can be made regarding the effectiveness of the intervention. Yet there
intervention for children and young people with a range of clinical diagnoses.
The majority of studies focused on children and young people with mixed
The studies included in this review were assessed for study quality,
and the highest quality studies were the RCT’s, which had investigated
31
approaches (Müller et al., 2015; Rothschild-Yakar, Lacoua, & Stein, 2013).
Such studies make it less straightforward to review the evidence fort the
approaches.
therapies (1.96% of total funding compared to 27.55% for CBT) and this
RCT it has been found to be at least as clinically and cost effective as other
Conclusion
References
Abbass, A. a., Rabung, S., Leichsenring, F., Refseth, J. S., & Midgley, N.
32
American Academy of Child and Adolescent Psychiatry, 52(8), 863–875.
Atzil-Slonim, D., Shefler, G., Gvirsman, S. D., & Tishby, O. (2011). Changes
Atzil-Slonim, D., Shefler, G., Slonim, N., & Tishby, O. (2013). Adolescents in
Balottin, U., Ferri, M., Racca, M., Rossi, M., Rossi, G., Beghi, E., … Termine,
Barlow, J., Bennett, C., Midgley, N., Larkin, S., & Wei, Y. (2015). Parent-infant
Baruch, G., Fearon, P., & Gerber, A. (1998). Evaluating the outcome of a
33
Clinical Psychology, 76(4), 531–543.
Chanen, A. M., Jackson, H. J., McCutcheon, L. K., Jovev, M., Dudgeon, P.,
Emanuel, R., Catty, J., Anscombe, E., Cantle, A., & Muller, H. (2014).
34
Implementing an aim-based outcome measure in a psychoanalytic child
Fonagy, P., & Target, M. (1994). The efficacy of psychoanalysis for children
55.
77.
Gatta, M., Sisti, M., Sudati, L., Miscioscia, M., & Simonelli, A. (2016). The
Gerber, A. J., Kocsis, J. H., Milrod, B. L., Roose, S. P., Barber, J. P., Thase,
35
586.
Goodyer, I. M., Reynolds, S., Barrett, B., Byford, S., Dubicka, B., Hill, J., …
Göttken, T., White, L. O., Klein, A. M., & von Klitzing, K. (2014). Short-term
Heede, T., Runge, H., Storebø, O. J., Rowley, E., & Hansen, K. G. (2009).
Horn, H., Geiser-Elze, A., Reck, C., Hartmann, M., Stefini, A., Victor, D., …
36
Young People.
Kolaitis, G., Giannakopoulos, G., Tomaras, V., Christogiorgos, S., Pomini, V.,
Krischer, M. K., Trautmann-Voigt, S., Kaspers, S., Voigt, B., Flechtner, H.-H.,
Kronmüller, K., Postelnicu, I., Hartmann, M., Stefini, A., Geiser-Elze, A.,
37
559—577. article.
Kronmüller, K., Stefani, A., Gesier-Elze, A., Horn, H., Hartmann, M., &
Lanyardo, M., & Horne, A. (2009). The Handbook of Child and Adolescent
Lock, J., Le Grange, D., Agras, W. S., Moye, A., Bryson, S. W., & Jo, B.
Lush, D., Boston, M., Morgan, J., & Kolvin, I. (1998). Psychoanalytic
51–69.
McLaughlin, C., Carol Holliday, Clarke, B., & Ilie, S. (2013). Research on
38
2003-2011. Leics: BACP.
Measelle, J. R., Ablow, J. C., Cowan, P. A., & Cowan, C. P. (1998). Assessing
Milrod, B., Shapiro, T., Gross, C., Silver, G., Preter, S., Libow, A., & Leon, A.
67(4), 359–366.
Funding. Mq.
Müller, J. M., Averbeck-Holocher, M., Romer, G., Fürniss, T., Achtergarde, S.,
Muratori, F., Picchi, L., Apicella, F., Salvadori, F., Espasa, F. P., Ferretti, D., &
Muratori, F., Picchi, L., Bruni, G., Patarnello, M., & Romagnoli, G. (2003). A
39
two- year follow up of psychodynamic psychotherapy for internalizing
Muratori, F., Picchi, L., Casella, C., Tancredi, R., Milone, A., & Patarnello, M.
Odhammar, F., Sundin, E. C., Jonson, M., & Carlberg, G. (2011). Children in
Palmer, R., Nascimento, L. N., & Fonagy, P. (2013). The State of the
22(2), 149–214.
Ryynänen, T., Alen, M., Koivumaa-Honkanen, H., Joskitt, L., & Ebeling, H.
40
Salzer, S., Cropp, C., Jaeger, U., Masuhr, O., & Streeck-Fischer, A. (2013).
Salzer, S., Cropp, C., & Streeck-Fischer, A. (2014). Early Intervention for
368–382.
Stefini, A., Hildegard, H., Winkelmann, K., Geiser-Elze, A., Hartmann, M., &
41
psychoanalytic psychotherapy for children and adolescents.
1(1), 6–19.
Szapocznik, J., Rio, A., Murray, E., Cohen, R., Scopetta, M., Rivas-Vazquez,
Target, M., & Fonagy, P. (1994). The efficacy of psychoanalysis for children:
Trowell, J., Joffe, I., Campbell, J., Clemente, C., Almqvist, F., Soininen, M., …
42
Trowell, J., Kolvin, I., Weeramanthri, T., Sadowski, H., Berelowitz, M., Glaser,
234–247.
Trowell, J., Rhode, M., Miles, G., & Sherwood, I. (2003). Childhood
Vrouva, I., Fonagy, P., Fearon, P. R. M., & Roussow, T. (2010). The risk-
Weitkamp, K., Daniels, J. K., Hofmann, H., Timmermann, H., Romer, G., &
Winkelmann, K., Stefini, A., Hartmann, M., Geiser-Elze, A., Kronmüller, A.,
43
Table 1. Table of studies from 2011 to 2016
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
(2014)
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
(2014)
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
(2) Sample size: 1 = Process for determining the sample size discussed along with any interim analyses and stopping rules; 0 =
(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
(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 =
53
(7) Active comparison: 1 = At least one active comparison (e.g. alternative model, treatment as usual); 0 = All comparison
(8) Baseline data: 1 = Baseline demographic and clinical characteristics reported by condition; 0 = Baseline demographic or
(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
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