International Journal of

Environmental Research
and Public Health

Review
Systematic Review on Mentalization as Key Factor
in Psychotherapy
Jonas Lüdemann *, Sven Rabung and Sylke Andreas

Institute of Psychology, University of Klagenfurt, Universitätsstr. 65-67,

9020 Klagenfurt am Woerthersee, Austria; [email protected] (S.R.); [email protected] (S.A.)
* Correspondence: [email protected]

Abstract: Background: Mentalization processes seem to be of high relevance for social learning and
seem important in all psychotherapies. The exact role of mentalization processes in psychotherapy is
still unknown. The aim of the present systematic review is to investigate whether mentalization is
related to the therapeutic outcome and, if so, whether it has a moderating, mediative, or predictive
function. Method: A systematic review with an electronic database search was conducted. A total of
2567 records were identified, and 10 studies were included in the final synthesis. Results: Psychother-
apy research is still in an initial phase of examining and understanding the impact of mentalization on
psychotherapy outcome. The small number of studies and the executed study designs and statistical
analyses indicate the possible role that mentalization has in psychotherapy. Conclusion: Generally,
strongly elaborated study designs are needed to identify the role of mentalization in psychotherapy.
Mentalization seems to be differently represented in differential treatment approaches. Nevertheless,
it should be noted that the patient’s mentalizing capacity seems to be relevant to the psychotherapy
process. Psychotherapies should be adapted to this.






Keywords: systematic review; mentalization; reflective functioning; psychotherapy; outcome
Citation: Lüdemann, J.; Rabung, S.;
Andreas, S. Systematic Review on
Mentalization as Key Factor in
Psychotherapy. Int. J. Environ. Res.
1. Introduction
Public Health 2021, 18, 9161. https://
doi.org/10.3390/ijerph18179161 In recent years, the new concept of Mentalization Based Psychotherapy (MBT) has
become increasingly popular in psychodynamic psychotherapy and research [1–5]. Mental-
Academic Editor: Paul B. Tchounwou izing is defined as the capacity to understand other people’s intentional or inner mental
states while taking into account one’s own intentional states (e.g., beliefs, thoughts, feelings,
Received: 14 July 2021 desires, goals [6,7].
Accepted: 26 August 2021 The mentalizing approach was developed by Fonagy, Steele, and Steele within the
Published: 31 August 2021 London Parent–Child project [6]. The authors observed that a child’s secure attachment
is not only dependent on the mother’s attachment security [6], but rather results from
Publisher’s Note: MDPI stays neutral the mother’s insight to acquire a psychological understanding of her early childhood
with regard to jurisdictional claims in relationship with her parents [8]. In regard to their predominantly positive attachment
published maps and institutional affil- experiences and the associated secure attachment representations, 79% of these mothers
iations. also had securely attached children compared to the 28% for mothers with an insecure
attachment style. Hence, the assumption could be confirmed that those mothers who
reported more negative childhood experiences during the interview were nevertheless
able to provide coherent statements and coping strategies due to their high reflexive
Copyright: © 2021 by the authors. competence [8].
Licensee MDPI, Basel, Switzerland. The empirical findings of this study, which was conducted Fonagy et al., believe that
This article is an open access article there must be a connection between early attachment experiences [8], self-representation,
distributed under the terms and and affect regulation even in patients with severe personality disorders [4]. Empirical
conditions of the Creative Commons studies have shown that patients with severe personality disorders have lower mentalizing
Attribution (CC BY) license (https:// capacities than normal people [9–13]. In the meantime, this has also been shown for
creativecommons.org/licenses/by/
patients with other mental illnesses such as depression [14–18] or eating disorders [16].
4.0/).

Int. J. Environ. Res. Public Health 2021, 18, 9161. https://doi.org/10.3390/ijerph18179161 https://www.mdpi.com/journal/ijerph
Int. J. Environ. Res. Public Health 2021, 18, 9161 2 of 16

As neuropsychological research shows, mentalization can appear in four different

dimensions [19,20]. The first dimension refers to automatic versus controlled mentalizing.
Controlled mentalization is a serial and relatively slow process that is usually expressed
verbally and requires reflection. In contrast, automatic mentalization is a much faster
process requiring only a little attention, intention, awareness, and effort [21]. In this
dimension, mentalization problems arise when individuals insist exclusively on automatic
assumptions about the mental states of themselves or others. Difficulties also arise when the
situation becomes difficult for the person to adequately apply these automatic assumptions.
The second dimension includes the ability to mentalize oneself or the mental states of
others (self- versus other-mentalizing). This means the ability to understand one’s own
inner mental states of beliefs, desires, needs, and those of others. Mentalization can also
refer to drawing conclusions based on external cues (e.g., facial expressions and gestures).
However, it can also refer to understanding someone’s internal experience from what
a person knows about the person and the situation they find themselves in (internal
versus external mentalizing). The fourth dimension refers to distinguishing whether the
person understands mental states more cognitively or affectively (cognitive versus affective
mentalizing). Cognitive mentalization involves the ability to name, perceive, and find
reasons for inner states, while affective mentalization involves the ability to understand
the feelings of inner states [21]. The dimensions described above are categories in which
mentalization can take place. Disturbances of the mentalization ability are present if one
pole of a dimension (e.g., cognitive mentalization) is too strongly pronounced [21].
MBT was originally developed by Bateman and Fonagy [3] for borderline patients. The
objective of MBT is to improve the mentalization capacity of patients with severe person-
ality disorders throughout the various dimensions [21]. Allen, Fonagy, and Bateman [22]
postulate that the concept of mentalization is represented in all forms of psychotherapy
and that the facilitation of mentalization makes psychotherapies effective [23]. Recently,
Fonagy et al. [24] have expanded the hypothesis from that of an evolutionary and develop-
mental point of view, indicating that improvement in mentalization is needed to maximize
the benefits from social experiences to get along better in the social world. In turn, this is
only possible in a therapeutic relationship that is perceived as secure. Fonagy et al. [24]
postulate that therapeutic change can only occur in a secure therapeutic relationship that
they refer to as epistemic trust. By epistemic trust, we mean the willingness to accept social
messages and messages relevant and helpful to the person within the therapeutic relation-
ship [24]. Epistemic trust provides the basis for social learning and enables individuals to
benefit from their own social environment. It is believed that most mental disorders are due
to epistemic mistrust, where the individual is unable to benefit from social conversations
and interactions. As a result, these individuals experience difficulties updating their own
expectations and beliefs and adapting to changing situations. Epistemic trust allows the
individual to adapt his social imagination to the prevailing social reality through mental-
ization [24]. Therefore, mentalization probably has a more mediating function. A patient
who has mentalization problems will probably also misinterpret the ostensive cues so that
no epistemic trust can be built [24]. That implies that mentalization processes should be of
great importance in any evidence-based psychotherapy [25].

1.1. Operationalization of Mentalization

The assessment of mentalization or reflective functioning, the operationalization of
mentalization, can be conducted via interview assessment, questionnaires, or visual tests.
Up to this point, a variety of instruments that can measure different areas or dimensions of
mentalization capacity are available (e.g., the cognitive-affective dimension, [26]). The most
widespread expert assessment method is the Reflective Functioning Scale (RFS) [26], which
is collected based on adult attachment interviews (AAI) [26]). Based on the narratives
in the AAI, the mental states of the subjects are assessed on a scale from −1 (rejected
reflective functioning) to 9 (very high reflective functioning). The RFS is a reliable and
valid instrument that requires the raters to be specifically trained in its administration.
Int. J. Environ. Res. Public Health 2021, 18, 9161 3 of 16

Slade and Slade et al. combined the RFS with the Parent Development Interview (PDI) [26],
hence developing the Parental Reflective Functioning Coding System (PRF-CS) [26]. The
RFS based on the PDI is a reliable and valid method to measure the reflective functioning
of parents regarding their relationship to their children, their parenthood, and the per-
ception of their children [26]. Since the AAI is a time-consuming procedure, there has
been an increased effort to directly measure reflectivity in specific mental disorders in
recent years. One of these instruments is the Depression-specific Reflective Functioning
Interview (DSRF) [26], which has shown very good results in initial reliability and validity
studies [26]. Considering that the AAI has an average interview duration of 1 to 2 h [26],
Rudden et al. [27] developed a specific interview, the Brief Reflective Functioning Interview
(BRFI; [27]). The questions in the BRFI are intended to provoke reflections on attachment
experiences and were developed based on the AAI. The assessment of reflective functioning
via BRFI follows the same principle as for the AAI, albeit an average value is calculated for
all ten questions. The BRFI thus represents a useful, economical, reliable, and valid method
for measuring reflective functioning.
One recently developed self-assessment instrument is the Reflective Functioning Ques-
tionnaire (RFQ) by Fonagy et al. [28] The instrument uses 12 items on 2 subscales, mea-
suring security and insecurity in reflective functioning. Three studies have demonstrated
both reliability and validation in clinical samples, whereas the scale of insecurity did not
show satisfactory results in a non-clinical sample [26]. Analogous to the PDI, the Parental
Reflective Functioning Questionnaire (PRFQ; [29]) was developed as a self-assessment
measure. The PRFQ measures parental reflective functioning on three dimensions: interest
and curiosity in the mental states of the infant, pre-mentalizing modes, and certainty
about the mental states of the infant. A first study demonstrated reliability and valid-
ity [30]. The methods presented so far are all based on the AAI and the RF scale developed
by Fonagy et al. [28]. Further methods for measuring reflective functioning, primarily
focusing on emotional perception, can be found in Luyten et al. [4].
Mentalization-based psychotherapy has gained increasing importance in recent years.
Moreover, mentalization processes seem to be of high relevance for social learning and
could thus be vital for all psychotherapies. However, it is unknown what role mental-
ization processes exactly have in psychotherapy. Although studies have been conducted
on mentalization processes in psychotherapy (i.e., [31,32]), this is still the case. In this
regard, the existing evidence has not yet been synthesized. Thus, a systematic review of
studies examining the relationship of mentalizing processes in the context of psychother-
apy outcome-relevant processes can provide new insights into mentalization processes for
psychotherapy research and practice. Therefore, the aim of the present systematic review
is to investigate whether mentalization is related to the therapeutic outcome and, if so,
whether it has a moderating, mediative, or predictive function.

1.2. Concepts of Psychotherapy Process Factors

It is necessary to differentiate concepts describing how variables may be associated
with psychotherapy outcome.
A moderator is a statistical baseline variable that has an interactive effect on the
outcome. Moderators suggest different directions or magnitudes in the relationship be-
tween an independent and dependent variable in specific subgroups. Thus, the influence
of an independent variable on a dependent outcome variable is moderated by a third
variable [33–35].
A mediator is a statistical variable intervening in the relationship between the depen-
dent variable (outcome) and the independent variable (treatment). This means that the
independent variable is not only related to outcome, but to the proposed mediator variable
too, which again is related to outcome. Thus, the relationship between treatment and
outcome becomes smaller after controlling for the mediator effect. Even though mediators
alone cannot explain the precise process of change, they may point to possible mechanisms
without necessarily being part of it [33].
Int. J. Environ. Res. Public Health 2021, 18, 9161 4 of 16

A predictor is a statistical baseline variable (not mediator) that predicts outcome

independent of subgroups (not moderator) [35].
As these concepts provide a basis for examining the association of a variable with
psychotherapy outcome, the treatment must be proven effective in changing the outcome
before any further investigation [35]. Moreover, it is impossible to distinguish between pre-
dictors and moderators or mediators of treatment effects without a comparison group [35].
Considering these concepts, the following research questions were investigated in this
systematic review: (I) Is mentalization related to psychotherapeutic treatment outcome?
(II) If mentalizing is related to the outcome of psychotherapeutic treatment, does it act as a
moderating, mediating, or predictive factor?

2. Method
2.1. Literature Search and Study Selection
An electronic database search was conducted using the databases: Ovid MEDLINE(R);
Epub Ahead of Print, In-Process & Other Non-Indexed Citations, Daily and Versions(R)
1946 to 10 September 2019; and PsycINFO 1806 to September Week 2 2019. In order
to identify relevant papers, the following search strategy was executed: “((“reflective
function*” or mentaliz* or mentalis*) and (psychother* or therap* or treatment*)).mp.”.
Following the electronic database search, two raters independently screened the titles
and abstracts considering the inclusion and exclusion criteria. There was no blinding for
journal titles, study authors, or institutions. In a second step, full-text screenings were
conducted for those reports that the screeners had been uncertain about earlier. Finally, the
reference lists of all of the included studies were reviewed. Any publications citing these
studies were identified and were checked for eligible reports. Disagreement was resolved
through discussion.

2.2. Inclusion Criteria

Empirical studies (minimum evidence level III [36]) examining the impact of mental-
ization on the process and outcome of psychotherapy were considered for this systematic
review. Studies published prior to 1991 were not considered, as the concept of mentalization
was introduced during that year [37]. Participants in the study had to be 18 to 65-year-old
patients with a DSM/ICD diagnosis in comprehensive psychotherapy (includes individual
psychotherapy as a substantial part of the intervention; additional group therapy may or
may not be included; at least three months duration). Thus, the psychotherapeutic inter-
vention had to have evidence-based psychotherapy to a substantial part, lasting for at least
twelve sessions or three months. Additional group therapy could be included, but studies
were excluded if the central intervention was group therapy or art therapy. The use of stan-
dardized outcome instruments was required. Additionally, the assessment of mentalizing
capacity had to be reported [4]. Studies had to be published in peer-reviewed journals.

2.3. Data Extraction

Subsequent to the screening process, the following information was extracted from
the included studies: publication (authors, year of first publication, region), type of study,
sample (n, diagnosis, age, gender), treatment context (therapy, treatment duration, num-
ber of therapists), operationalization of mentalization, function of mentalization (predic-
tor/moderator, mediator/outcome), outcome instrument, result (pre/post values: mean
values (M), and standard deviations (SD) if applicable; simple summary data for each
intervention group at least).

3. Results
3.1. Study Selection
The electronic database search was conducted on 11 September 2019. In total, 2139 po-
tentially relevant articles were identified, and finally, 10 studies fulfilling the eligibility
criteria were included. The agreement of the two raters was 99.25%. Differences or un-
 3. Results
3.1. Study Selection
Int. J. Environ. Res. Public Health 2021, 18, 9161 5 of 16
The electronic database search was conducted on 11 September 2019. In total, 2139
potentially relevant articles were identified, and finally, 10 studies fulfilling the eligibility
criteria were included. The agreement of the two raters was 99.25%. Differences or unclear
clear allocations
allocations (k =
(k = 16) 16) could
could be resolved
be resolved by consensus.
by consensus. The study
The study selection
selection process
process is
is illus-
illustrated
trated in Figure
in Figure 1. 1.

al. [38]).
Figure 1. PRISMA flow diagram (Moher et al. [38]).

3.2. Study Characteristics

An overview
overview ofof the
the included
includedstudies
studiesand
andtheir
theircharacteristics
characteristicsisissummarized
summarized ininTable
Table1.
The
1. Thestudies
studieswere
wereconducted
conductediningeographical
geographicalEurope
EuropeandandthetheUSA.
USA.HalfHalfofof the
the included
included
studies (originally) were RCTs, and and the
the other
other studies
studies followed
followed aa pre–post
pre–post design.
design. TheThe
sample size ranged between n = 20 and n = 138 patients, respectively, n = 2 and n == 67
therapists. Most of the studies (k = 4) included patients with a major depressive disorder
diagnosis, while
diagnosis, whilethree
threestudies
studiesincluded
included patients with
patients a personality
with disorder
a personality diagnosis.
disorder The
diagnosis.
rest of the studies examined patients with mixed diagnoses (k = 3; with mainly
The rest of the studies examined patients with mixed diagnoses (k = 3; with mainly de- depressive
disorders).
pressive Treatments
disorders). were mainly
Treatments werepsychoanalytic/psychodynamic,
mainly psychoanalytic/psychodynamic, though cognitive
though
behavioralbehavioral
cognitive therapy (CBT)
therapyand other
(CBT) andtreatment approaches
other treatment were examined.
approaches The mean
were examined. The
treatment
mean duration
treatment ranged
duration between
ranged 14 and
between 14 366
and sessions. In all
366 sessions. In of
allthe included
of the included studies,
stud-
a pre–post
ies, improvement
a pre–post improvement of outcome waswas
of outcome measured
measured(significant improvement
(significant improvement in 80% of
in 80%
the studies).
of the studies).
Int. J. Environ. Res. Public Health 2021, 18, 9161 6 of 16

Table 1. Included studies.

Antonsen 2016 Barber 2020 Boldrini 2018 Bressi 2016 Ekeblad 2016
Publication [39–41] [42,43] [44] [45] [46]
(NCT00353470)
Country Norway USA USA Italy Sweden
Study type RCT RCT Pre–post n,d Pre–post n RCT
N patients 37 a 138 a 27 24 85
Mean Age 31.6 (SD 7.7) Not reported 33 (range: 20–70) 44.63 (SD 5.88) 34.2 (SD 10.82)
Female % 75 62.3 48 45.83 68.8
Diagnoses BPD, AvPD PD Mixed e MDD MDD
Psychoanalytic
Intervention Mixed b CBT, PFPP STMBP CBT, IPT
treatment
N Therapist 32 24 8 2 34
19–24 sessions 366 sessions
Duration 65 (SD 60) sessions 40 sessions 14 sessions
(twice-weekly) (range: 120–2836)
Outcome SCL-90-R PDSS GAF, PHI GAF, HAM-D BDI-II
measurement points T0: baseline T0: baseline T1: first 4 sessions T0: baseline Before every session
T1: 8 months T1: week 1 T2: 4 sessions T1: 40 weeks (end)
T2: 18 months T2: week 5 after 1 month T2: 1 year follow-up
T3: 1 month before
T3: 36 months T3: week 10
termination
T4: 72 months T4: at termination T4: last 4 sessions
RFS (AAI-short c ), PSRF RFS, DSRF
Mentalization RFS (AAI) CRF (sessions) RFS (AAI)
(PSRF-I) (AAI-short f )
measurement points T0: baseline TO: baseline T1: first 4 sessions T0: baseline T0: baseline
T1: 36 months T1: week 5 T2: 4 sessions T1: 40 weeks (end)
T2: at termination after 1 month T2: 1 year follow-up
T3: 4 sessions in the
middle phase
T4: 1 month before
termination
T5: last 4 sessions
Int. J. Environ. Res. Public Health 2021, 18, 9161 7 of 16

Table 1. Cont.

Significant improvement
Significant improvement Significant improvement Significant improvement
Outcome Improvement T0–T4 session 1–14 (CBT: d = 1.15;
T0–T3 (d = 1.0) T1–T5 T0–T3
IPT: d = 1.49)
Small, non- significant RFS
RFS did not improv;e
change in therapy (T0-T1);
Mentalization change Not reported PSRF significantly improved No significant CRF change Not examined
significant RFS change at
in PFPP, not in CBT
follow-up (T1-T2)
Early change in RFS is not
Mentalization-Outcome- a. RFS is not a significant Early CRF significantly RFS significantly predicts RFS/DSRF significantly
significantly associated with
relation predictor of outcome predicts outcome change outcome change predicts outcome change
outcome change
b. RFS is a significant
Early change in PSRF was
moderator of treatment
associated with significantly
effects (low RF patients had
greater change in outcome
better outcomes in outpatient
(the association was stronger
individual therapy compared
for CBT)
to control condition)
Fischer-Kern 2015 Karlsson 2006a Karlsson 2006b Müller 2006 Taubner 2015
Publication [9,47–49] [50,51] [50,52] [53] [18,54]
Country Austria, Germany USA USA Germany Germany
Study type RCT RCT (archival) Pre–post (archival) Pre–post Pre–post
N patients 92 64 30 24 20
Mean Age 27.7 (SD 7.3); range: 18–51 35 (SD 8.5) 50 (range: 20–81 years) 28 (SD 10) 39.2 (SD 12.7)
Female % 100 70 66.6 70.5 80
Diagnoses BPD MDD Mixed i Mixed j MDD
Intervention TFP, mixed g CBT, IPT BPDT Mixed k psychoanalytic treatment
N Therapist 67 18 15 not reported 16
Duration at least 1 year h 16.2 (SD 2.5) sessions 15.8 (SD 1.35) sessions 3 months 227,95 (SD 88,48) hours
Outcome STIPO BDI, HSCL-90, HRSD HSCL-90: GSI, BPRS SCL-90-R BDI, SCL-90-R
Int. J. Environ. Res. Public Health 2021, 18, 9161 8 of 16

Table 1. Cont.

measurement points T0: pre-treatment T1: session 4 T1: session 1 T0: baseline T0: pre-treatment
T1: 1 year after start of
T2: session 12 T2: session 5 T1: end of treatment T1: 24 months in treatment
therapy
T3: session 14 T2: 36 months in treatment
Mentalization RFS (AAI) RFS (sessions) RFS (sessions) RFS (AAI-shortc) RFS (AAI)
measurement points T0: pre-treatment T1: session 4 T1: session 1 T1: first week in treatment T0: pre-treatment
T1: 1 year after start of
T2: session 12 T2: session 5 T1: 24 months in treatment
therapy
T3: session 14
Significant improvement
Significant improvement Significant improvement Significant improvement
Outcome Improvement T0–T1 session 1–14 (GSI: d = 1.64;
T0–T1 T1–T2 T1–T3
BDI: d = 2.1)
RFS significantly improved in
Significant RFS decrease Significant RFS increase
Mentalization change TFP, but not in the control No significant RFS change Not examined
(T1–T2) (T0–T1)
condition
RFS improvement Process correlates associated RFS significantly predicts
Mentalization-Outcome- RFS is partly related to RFS significantly predicts
significantly predicts with low/high RFS predicted outcome change for BDI, but
relation outcome change outcome change
outcome change poor/good outcome not for GSI
RFS change had no
significant effect on outcome
Note. RCT = randomized controlled trial; BPD = borderline personality disorder; AvPD = avoidant personality disorder; MDD = major depressive disorder; SCL-90-R = SCL-90-R Symptom Checklist-90-Revised;
BDI (II) = Beck Depression Inventory; STIPO = structured interview of personality organization; HSCL-90 = Hopkins Symptom Checklist; HRSD = Hamilton Rating Scale for Depression; RFS (*) = Reflective
Functioning Scale (obtained from * resource); CRF (*) = Computerized Text Analysis measure of Reflective Functioning [55]; DSRF (*) = Depression-Specific Reflective Functioning; AAI = Adult Attachment
Interview; PDSS = Panic Disorder Severity Scale; PSRF = Panic-Specific Reflective Functioning; BPDT = Brief Psychodynamic Therapy; CBT = Cognitive Behavior Therapy; IPT = Interpersonal Psychotherapy;
PFPP = Panic-Focused Psychodynamic Psychotherapy, a 24-session, twice-weekly (12 weeks), manualized psychoanalytic psychotherapy [56]; STMBP = short-term psychodynamic psychotherapy with
mentalization-based techniques; TFP = Transference-Focused Psychotherapy; n naturalistc; a sub-sample of the RCT fulfilling the eligibility criteria; b mainly psychoanalytic/psychodynamic background,
further interventions with cognitive and systemic elements; c AAI demand questions; d archival data measured from end of 1960s to 2011; e depressive disorder, personality disorders, sexual disorder; f AAI
questions 1–11; g 36.5% psychoanalysis; 34.6% behaviour therapy; 7.7% client-centered therapy, 7.7% systemic psychotherapy, 1.9% gestalt psychotherapy; h 40.4% of the randomized patients continued therapy;
I depression, dysthymia, and generalized anxiety disorder; j eating disorders and depressive disorders; k integrative psychodynamically oriented treatment methods.
Int. J. Environ. Res. Public Health 2021, 18, 9161 9 of 16

3.3. The Mentalization Variable in the Psychotherapy Process

Most of the included studies used the RFS for the assessment of mentalization. Further
measurements that were used were the Computerized Text Analysis measure of Reflective
Functioning (CRF; [55]), the DSRF [26], and the Panic-Specific Reflective Functioning
(PSRF; [27]). The original application of the RFS on the AAI was executed in four studies.
Three studies used a short form of the AAI, and two applied the RFS on transcribed
psychotherapy sessions. CRF was also applied during psychotherapy sessions, DSRF on
an AAI short form, and the PSRF was used with an associated interview. In two studies,
mentalization was only measured at baseline or within the first week of treatment, whereas
the other studies (80%) had at least one other second measurement point to assess the
capacity of mentalization.

3.4. Mentalization as Predictor

Five studies investigated the possible role of mentalization as a predictor of psychother-
apy outcome via the RFS. In the study by Antonsen et al. [39], linear mixed modeling (LMM)
analyses with the parameters for RFS and time were without significant longitudinal effects
predicting the improvement of outcome variables (with the GSI for symptom distress) [39].
The six-year follow-up findings were in line with the 3-year follow-up analysis of the same
sample by Gullestad et al. [40]. Bressi et al. [45] reported a significant HAM-D improvement
predicted by the RFS score at baseline (β = 0.75, p = 0.025), explaining 13.8% of the variance
in the HAM-D change [45]. Ekeblad et al. [46] reported a significant prediction of the initial
RFS score on BDI-II improvement over 14 sessions with a medium-sized effect of β = 0.35
(standardized regression coefficient). In a three-month inpatient psychotherapy, RFS also
significantly predicted overall improvement (r = −0.37 (p < 0.05) for RFS and the GSI at
termination, partialling out the effect of GSI at onset and the effect of the overall structural
level according to the operationalized psychodynamic diagnostics (OPD), as examined
by Müller et al. [53]. Mixed results were found by Taubner et al. [54]. For a period of
36 months, Taubner et al. [54] found a significant BDI improvement predicted by the RFS
score at baseline (r = −0.48, p < 0.05), explaining 23.04% of the variance in BDI-change,
while there was no significant effect on the initial RFS score on GSI change (r = −0.318,
p = 0.17) [54]. Previous examinations of the investigated sample after 8 months and after
15 months of treatment could not find significant effects of RFS scores on BDI/GSI scores
and changes [17,18].
In two studies, a different measurement of mentalization method was used to inves-
tigate the possible role as a predictor. Besides the RFS, Ekeblad et al. [46] also applied
the DSRF [26,46]. They found a significant prediction on BDI-II improvement over 14 ses-
sions (β = 0.41). Boldrini et al. [44] examined therapy sessions with the CRF and found
that CRF significantly predicted the final personality health index (PHI; [57]) and GAF
scores as well as the pre–post changes in these measures [44]. However, for the analysis,
Boldrini et al. [44] mixed initial scores with scores after one month of treatment; thus, the
results do not represent a bona fide predictor analysis [44].

3.5. Mentalization as Moderator

The examination of mentalization as a possible moderator was only reported by
Antonsen et al. [39]. Regarding the treatment type, Antonsen et al. found significant
moderator effects of RFS on GSI depending on the patients’ mentalization capacity [39].
Patients with a medium RFS at baseline (RFS of 3 or above) had smaller effect sizes on all
of the outcome variables in the outpatient treatment than patients in the control treatment.
The comparison in patients with a low RFS at baseline (RFS below 3), on the other hand,
resulted in greater effect sizes in outpatient treatment.

3.6. Mentalization as Mediator

None of the included studies conducted a state-of-the-art mediator analysis exam-
ining an overall treatment effect (outcome change) with a control condition additionally
Int. J. Environ. Res. Public Health 2021, 18, 9161 10 of 16

examining the treatment effect on mentalization (change), and if applicable, the effect of
the change in the mentalization capacity on outcome change. While a possible mediating
effect of mentalization in the psychotherapeutic process was not examined, half of the
included studies investigated the change of mentalization in treatment and its impact on
the outcome.

3.7. Mentalization Change

Fischer-Kern et al. [47] examined the change in mentalization in two treatment arms
(transference-focused psychotherapy, TCP, and treatment by experienced community ther-
apists, ECP). They found significant improvements in mentalization after one year of
TFP, but no significant improvements in ECP. They could also show a correlation of im-
proved mentalization with better outcome scores, but they did not examine the directional
causation of mentalization change on outcome change.
Barber et al. [42] examined early mentalization change with the RFS and PSRF in two
comprehensive psychotherapy approaches (CBT and Panic-Focused Psychodynamic Psy-
chotherapy; PFPP). They examined the association of mentalization change from intake to
week five of the treatment on the subsequent change in panic severity until the termination
of treatment. For both conditions, Barber et al. [42] found no improvement in the RFS.
Additionally, the test on the association of early RFS change on change in outcome (PDSS)
was not significant in both treatments. The application of the panic specific mentalization
instrument PSRF led to different results. Early PSRF change was significant in PFPP but not
in the CBT condition. Nevertheless, early PSRF change was associated with a significantly
greater change in outcome (PDSS) in both interventions and was (not significantly) stronger
for CBT. Due to the missing control condition, it is unclear if the change of symptom specific
mentalization was caused by (a specific) treatment or not. Besides reporting that there is
no early change in RFS (without a significant effect on outcome change) and only an early
change in PSRF in PFPP, the results of Barber et al. [42] indicate that the magnitude of early
mentalization change has a higher impact on outcome change than the mean probability of
occurrence of mentalization change.
In the study by Taubner et al. [54] the RFS change examination is only reported for the
therapy condition and not for the control condition (baseline RFS scores were reported for
both conditions in Taubner et al. [18]). Due to the missing control group, it remains unclear
if the RFS change is an effect of the psychotherapeutic treatment. Taubner et al. [54] found a
significant improvement in the RFS from baseline to the second assessment 24 months later
(with a medium effect of d = 0.61, p = 0.039). Further, Taubner et al. [54] examined a possible
correlation between the change of RFS and at symptomatic change between baseline and
the 36-month assessment. They found neither a significant correlation between RFS change
and BDI change nor between RFS change and GSI change. It is still a question of whether
the non-significant correlations are explicable based on the treatment effect. Thus, the
results only allow very limited interpretations regarding the role of mentalization change
in treatment.
In their first study, Karlsson and Kermott [50] found a significant RFS (measured on
session transcripts) decrease from session 4 to session 12 for IPT, while they did not find any
significant RFS change for CBT. In a post hoc analysis, they examined significant correlates
between high versus low RFS with PQS items. A subsequent analysis with the relevant
PQS items and the outcome showed that, in general, the correlation of PQS items with high
RFS correlated with good outcome scores and that PQS items correlating with low RFS
correlated with poor outcome. No intervention comparison was reported for the post hoc
analysis. Based on that and the missing analysis of the direct impact of mentalization on
the outcome, the influence of mentalization change in treatment can scarcely be evaluated.
The second study by Karlsson and Kermott [50] was an analogous examination of
a BPDT treatment (sessions 1, 5, and 14). They found no significant RFS change during
treatment. The same post hoc analysis was executed as in their first study, and they found
correlations between high versus low RFS and good versus poor outcome. Nevertheless,
Int. J. Environ. Res. Public Health 2021, 18, 9161 11 of 16

many correlations did not achieve significance due to low statistical power, as reported by
Karlsson and Kermott [50]. The interpretation of the results is comparably limited, as is
also the case for their first study.
As mentioned in the section on the predicting role of mentalization above, Boldrini et al. [44]
examined the prediction of an early period CRF score (which was a mean cumulation
of the first four sessions of the treatment and four sessions after one month of treatment)
instead of the prediction of an initial CRF score. Boldrini et al. [44] examined sessions
from three treatment phases (early, middle, late) and found no significant CRF change. An
analysis of CRF course impact on outcome was not reported.

4. Discussion
The role of mentalization as a predictor is not clear in terms of longitudinal effects—
Antonsen et al. [39] seem to negate it, while the results of Taubner et al. [18,54] suggest
a longitudinal effect, and the results of Boldrini et al. [44] are not comparable due to the
different design of the data-analysis. The effect seems to be more consistent in shorter in-
vestigations [45,46,53], all being in favor of a positive prediction of psychotherapy outcome
by the initial capacity of mentalization.
The role of mentalization as a moderator was only examined by Antonsen et al. [39].
The study could be an indication that patients with different levels of mentalization require
different treatment types.
For the role of mentalization as a mediator, no state-of-the-art mediator analysis was
examined. Yet, half of the included studies give an indication for further research on
that topic. Regarding that, it is interesting whether mentalization changes during the
psychotherapy process for the examination of mentalization as a mediator (Table 2).

Table 2. Mentalization scores and changes.

RFS T1 T2 T1 Control T1 Control

Antonsen 2016 3.5 (SD 1.7) n/r 3.0 (SD 1.5) a n/r
Bressi 2016 4.00 (SD 2.09) 4.13 (SD 1.80)
Ekeblad 2016 2.62 (SD 1.22) n/r
Fischer-Kern 2015 2.82 (SD 1.29) 3.32 (SD 0.99) 2.80 (SD 0.96) b 2.92 (SD 1.00)
Karlsson 2006a 5.13 (SD 1.37) 3.99 (SD 1.35) 3.79 (SD 1.29) c 3.41 (SD 1.26)
Karlsson 2006b 4.62 (SD 1.39) 4.37 (SD 0.82)
Müller 2006 median: 3 (range: 1–5) n/r
Taubner 2015 3.85 (SD 0.94) 4.38 (SD 0.93)
Barber 2020 4.04 (SD 1.23) 4.42 (SD 1.27) 4.39 (SD 1.32) d 4.37 (SD 1.13)
CRF T1 T2
Boldrini 2018 29.09 (SD 7.75) 27.25 (SD 9.00)
DSRF T1
Ekeblad 2016 2.37 (SD 0.98)
PSRF T1 T2 T1 Control T1 Control
Barber 2020 3.50 (SD 1.19) 4.43 (SD 1.38) 3.68 (SD 1.21) d 3.68 (SD 1.10)
Note. a outpatient treatment vs. step-down treatment; b TFP vs. ECT; c IPT vs. CBT; d PFPP vs. CBT; n/r: not reported.

The capacity of mentalization may change in the psychotherapy process but does
not always change in psychotherapy. Two studies found a significant pre–post increase
of RFS [47,54], one found a significant increase of PSRF (while RFS did not change signif-
icantly in the same study) [42], one examined a significant RFS change in the follow-up
assessment (while RFS did not change significantly in the same study) [45], and one found
a significant decrease of RFS (study 1) [50]. Two studies reported no significant change
in mentalization (study 2) [44,50], and three studies did not examine or report a change
of mentalization [39,46,53]. The differential results reveal that the change of mentaliza-
tion in the psychotherapy process is not consistent. Studies examining mentalization
change with at least two treatment conditions indicate that mentalization change differs
Int. J. Environ. Res. Public Health 2021, 18, 9161 12 of 16

depending on treatment (for RFS: [47,50]; for PSRF: [42]). The link between mentalization
change and outcome change remains unclear regarding this point. While Barber et al. [42]
and Taubner et al. [54] did not find an association of RFS change with outcome change,
Fischer-Kern et al. [47] did, and Barber et al. [42] found an association for PSRF.
Instead of an explicit distinction, the results indicate that mentalization may hold
several roles in the psychotherapy process at the same time. One possible explanation may
be found in the conceptualization of mentalization. This integrates different dimensions
(e.g., self-other, cognitive-affective, internal-external, automatic-controlled [19,20]), which
may but do not have to simultaneously emerge and (differently) affect the psychother-
apy process.
The review of the current status of research on the role of mentalization in the psy-
chotherapy process reveals that psychotherapy research is still in an initial phase of exam-
ining and understanding the impact of mentalization on psychotherapy outcome. Besides
the small number of studies examining the associations between mentalization (change)
and outcome (change), the executed study designs and statistical analyses indicate that, at
maximum, the systematic summary of the results only permits indications of the possible
role that mentalization has in psychotherapy.
The data availability does not allow for more than one to report observations made
in the systematic rework of the included studies. One of these observations is that the in
the studies that lasted for one year, changes in mentalization were found. The changes
in mentalization may take place but not before a particular time in treatment. Neverthe-
less, the included studies supporting the assumption of predictive effects indicate that
attention should also be paid to the initial capacity of mentalization. However, these
significant results are not definite in terms of the exact role of mentalization due to missing
control conditions.

5. Conclusions
Overall, the included studies cannot directly support the postulate by Allen et al. [22],
who stated that mentalization is represented in all forms of psychotherapy and makes
them effective by means of facilitation of mentalization [22]. While it remains unclear
if mentalization improves during psychotherapy, mentalization seems to be represented
differently in various treatment approaches. Surprisingly, MBT was not examined in any
of the studies. Therefore, the role of mentalization is still not explicit for this mentalization
specific treatment.
All assumptions and postulates on the mediating role of mentalization remain without
empirical support. Future examinations need complex and comprehensive study designs
to control for the real effects of mediation. A more (time) efficient operationalization
of mentalization is needed to trace mentalization in the psychotherapy process. For
example, this is necessary to demonstrate a timeline of change for the proposed mediator
variable occurring before outcome changes [34]. For this reason, instruments such as self-
report measures or computer text analysis represent suitable approaches. However, these
measures need to be further elaborated, so the time-consuming rating measures still need
to be utilized momentarily. Furthermore, at the current state of mentalization research,
the comparability of the different instruments should be repeatedly verified despite the
existing construct validation. In this regard, the advantages and disadvantages for the
use and development of disorder-specific measures versus a general operationalization
of mentalization could also be further explored. Generally, strongly elaborated study
designs are needed to identify paths of clinical change mechanisms that have an impact
on mentalization or that are influenced by mentalization (change) [34]. By using control
groups, future studies may further examine the role of mentalization by focusing more on
different treatment approaches and patient groups.
While there is still much research needed to empirically understand and define the
role of mentalization in the psychotherapy process, the results of this systematic review
Int. J. Environ. Res. Public Health 2021, 18, 9161 13 of 16

have at least one implication for practice: the patient’s mentalizing capacity matters, and
the psychotherapeutic treatment should (also) be adapted to this.

6. Limitations
This review critically investigated the current psychotherapy research on the role
of mentalization in the process of outcome improvement. One limitation was that the
research question itself. The review examined how mentalization explains outcome change;
thus, it did not examine variables that have a positive, a negative, or a compensatory
effect on mentalization itself. Because of that, it remains unclear as to what determines
mentalization change. The narrow inclusion criteria excluded studies examining the
process of mentalization without investigating the association with psychotherapy outcome.
Those studies should be taken into account for an update of this review to establish
profound evidence on possible mentalization impact.
As in any systematic review, this systematic review is restricted by the time period of
the literature search and the study selection process. To ensure that the detections are not
outdated, an update of the literature search and study selection process was conducted
by two reviewers in August 2021. The electronic database search was repeated for the
period after the original database search (k = 724 record identified), and publications from
this period that cited the included studies were also screened. Finally, one additional
study fulfilled the eligibility criteria [58]. This study investigated the relationship between
mentalization, psychotherapeutic alliance, and treatment outcome in psychoanalytic psy-
chotherapy (24 months) and CBT (5 months) in bulimia nervosa patients. The n = 70
patients were randomized, and the mentalization and treatment outcomes were assessed
at three time points (baseline, after 5 months, after 24 months). Mentalization was mea-
sured with the RFS. No correlation between the baseline RFS score and outcome could be
found. It was found that the RFS scores increased more in patients in the psychoanalytic
psychotherapy condition than in the CBT condition. A significant relationship between
RFS change and symptom change in the psychoanalytic psychotherapy condition can be
seen as a further indication of the role of mentalization as a mediator in the psychotherapy
process. Furthermore, the treatment type and duration had an impact on how mentaliza-
tion changed and affected the treatment outcome [58]. The results fit with the findings of
this systematic review.
Similar to the included studies, the review itself could not grasp the complexity of
mentalization in the psychotherapy process. In future examinations, possible predictors,
moderators, and mediators of mentalization should also be considered. Step by step, an
empirically based mentalization process model could be derived from this.

Author Contributions: Conceptualization, J.L., S.R. and S.A.; methodology, J.L., S.R. and S.A.;
investigation, J.L., S.R. and S.A.; project administration, J.L. and S.A.; supervision, S.R. and S.A.;
writing—original draft, J.L. and S.A.; writing—review and editing, S.R. and S.A.; data curation, J.L.;
visualization, J.L. All authors have read and agreed to the published version of the manuscript.
Funding: This research received no external funding.
Data Availability Statement: The data supporting this systematic review are from previously re-
ported studies, which have been cited. The processed data are available from the corresponding
author upon request.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Bateman, A.; Fonagy, P. Effectiveness of partial hospitalization in the treatment of borderline personality disorder: A randomized
controlled trial. Am. J. Psychiatry 1999, 156, 1563–1569. [CrossRef]
2. Bateman, A.; Fonagy, P. Randomized Controlled Trial of Outpatient Mentalization-Based Treatment Versus Structured Clinical
Management for Borderline Personality Disorder. Am. J. Psychiatry 2010, 8, 55–65. [CrossRef]
Int. J. Environ. Res. Public Health 2021, 18, 9161 14 of 16

3. Bateman, A.; Fonagy, P. Mentalization Based Treatment for Personality Disorders: A Practical Guide, 2nd ed.; OUP Oxford: Oxford,
UK, 2016; ISBN 9780191669910.
4. Luyten, P.; Malcorps, S.; Fonagy, P.; Ensink, K. Assessment of Mentalizing. In Handbook of Mentalizing in Mental Health Practice,
2nd ed.; Bateman, A., Fonagy, P., Eds.; American Psychiatric Association Publishing: Washington, DC, USA, 2019; pp. 37–62.
ISBN 9781615372508.
5. Fonagy, P. Attachment theory approach to treatment of the difficult patient. Bull. Menn. Clin. 1998, 62, 147.
6. Fonagy, P.; Steele, H.; Steele, M. Maternal representations of attachment during pregnancy predict the organization of infant-
mother attachment at one year of age. Child Dev. 1991, 62, 891–905. [CrossRef]
7. Fonagy, P.; Bateman, A. The development of borderline personality disorder—A mentalizing model. J. Pers. Disord. 2008, 22, 4–21.
[CrossRef] [PubMed]
8. Fonagy, P.; Steele, M.; Steele, H.; Moran, G.S.; Higgitt, A.C. The capacity for understanding mental states: The reflective self in
parent and child and its significance for security of attachment. Infant Ment. Health J. 1991, 12, 201–218. [CrossRef]
9. Fischer-Kern, M.; Buchheim, A.; Hörz, S.; Schuster, P.; Doering, S.; Kapusta, N.; Taubner, S.; Tmej, A.; Rentrop, M.; Buchheim,
P.; et al. The relationship between personality organization, reflective functioning, and psychiatric classification in borderline
disorder. Psychoanal. Psychol. 2010, 395–409. [CrossRef]
10. Fonagy, P.; Target, M. Playing with reality: III. The persistence of dual psychic reality in borderline patients. Int. J. Psychoanal.
2000, 81, 853–873. [CrossRef]
11. Skårderud, F. Eating one’s words, part I: ‘Concretised metaphors’ and reflective function in anorexia nervosa—An interview
study. Eur. Eat. Disord. Rev. 2007, 15, 163–174. [CrossRef] [PubMed]
12. Skårderud, F. Eating one’s words, part II: The embodied mind and reflective function in anorexia nervosa—Theory. Eur. Eat.
Disord. Rev. 2007, 15, 243–252. [CrossRef]
13. Skårderud, F. Eating one’s words: Part III. Mentalisation-based psychotherapy for anorexia nervosa—An outline for a treatment
and training manual. Eur. Eat. Disord. Rev. 2007, 15, 323–339. [CrossRef] [PubMed]
14. Lemma, A.; Target, M.; Fonagy, P. The development of a brief psychodynamic intervention (dynamic interpersonal therapy) and
its application to depression: A pilot study. Psychiatry 2011, 74, 41–48. [CrossRef] [PubMed]
15. Markowitz, J.C.; Meehan, K.B. Does depressive state influence reported attachment status? J. Clin. Psychiatry 2009, 70, 1726.
[CrossRef] [PubMed]
16. Robinson, P.; Hellier, J.; Barrett, B.; Barzdaitiene, D.; Bateman, A.; Bogaardt, A.; Clare, A.; Somers, N.; O’Callaghan, A.; Goldsmith,
K.; et al. The NOURISHED randomised controlled trial comparing mentalisation-based treatment for eating disorders (MBT-ED)
with specialist supportive clinical management (SSCM-ED) for patients with eating disorders and symptoms of borderline
personality disorder. Trials 2016, 17, 549. [CrossRef] [PubMed]
17. Staun, L.; Kessler, H.; Buchheim, A.; Kächele, H.; Taubner, S. Mentalisierung und chronische Depression. Psychotherapeut 2010, 55,
299–305. [CrossRef]
18. Taubner, S.; Kessler, H.; Buchheim, A.; Kächele, H.; Staun, L. The role of mentalization in the psychoanalytic treatment of chronic
depression. Psychiatry 2011, 74, 49–57. [CrossRef]
19. Fonagy, P.; Luyten, P. A developmental, mentalization-based approach to the understanding and treatment of borderline
personality disorder. Dev. Psychopathol. 2009, 21, 1355–1381. [CrossRef]
20. Luyten, P.; Fonagy, P.; Lowyck, B.; Vermote, R. Assessment of mentalization. In Handbook of Mentalizing in Mental Health Practice,
1st ed.; Bateman, A., Fonagy, P., Eds.; American Psychiatric Publishing: Washington, DC, USA, 2012; pp. 43–65.
21. Fonagy, P.; Bateman, A. Introduction. In Handbook of Mentalizing in Mental Health Practice, 2nd ed.; Bateman, A., Fonagy, P., Eds.;
American Psychiatric Association Publishing: Washington, DC, USA, 2019; pp. 3–20. ISBN 9781615372508.
22. Allen, J.G.; Fonagy, P.; Bateman, A. Mentalizing in Clinical Practice, 1st ed.; American Psychiatric Pub: Washington, DC, USA, 2008;
ISBN 9781585628360.
23. Allen, J.G. Restoring Mentalizing in Attachment Relationships: Treating Trauma with Plain Old Therapy, 1st ed.; American Psychiatric
Pub: Arlington, VA, USA, 2012; ISBN 9781585624188.
24. Fonagy, P.; Campbell, C.; Allison, E. Therapeutic Models. In Handbook of Mentalizing in Mental Health Practice, 2nd ed.; Bateman,
A., Fonagy, P., Eds.; American Psychiatric Association Publishing: Washington, DC, USA, 2019; pp. 169–180. ISBN 9781615372508.
25. Allen, J.G.; Fonagy, P. The Handbook of Mentalization-Based Treatment; John Wiley & Sons: West Sussex, UK, 2006.
26. Khorashad, B.S.; Baron-Cohen, S.; Roshan, G.M.; Kazemian, M.; Khazai, L.; Aghili, Z.; Talaei, A.; Afkhamizadeh, M. The “Reading
the Mind in the Eyes” Test: Investigation of Psychometric Properties and Test-Retest Reliability of the Persian Version. J. Autism
Dev. Disord. 2015, 45, 2651–2666. [CrossRef]
27. Rudden, M.; Milrod, B.; Target, M.; Ackerman, S.; Graf, E. Reflective functioning in panic disorder patients: A pilot study. J. Am.
Psychoanal. Assoc. 2006, 54, 1339–1343. [CrossRef] [PubMed]
28. Fonagy, P.; Luyten, P.; Moulton-Perkins, A.; Lee, Y.-W.; Warren, F.; Howard, S.; Ghinai, R.; Fearon, P.; Lowyck, B. Development
and Validation of a Self-Report Measure of Mentalizing: The Reflective Functioning Questionnaire. PLoS ONE 2016, 11, e0158678.
[CrossRef] [PubMed]
Int. J. Environ. Res. Public Health 2021, 18, 9161 15 of 16

29. Luyten, P.; Mayes, L.C.; Nijssens, L.; Fonagy, P. The parental reflective functioning questionnaire: Development and preliminary
validation. PLoS ONE 2017, 12, e0176218. [CrossRef] [PubMed]
30. Rutherford, H.J.V.; Booth, C.R.; Luyten, P.; Bridgett, D.J.; Mayes, L.C. Investigating the association between parental reflective
functioning and distress tolerance in motherhood. Infant Behav. Dev. 2015, 40, 54–63. [CrossRef] [PubMed]
31. Hayden, M.C.; Müllauer, P.K.; Gaugeler, R.; Senft, B.; Andreas, S. Improvements in mentalization predict improvements in
interpersonal distress in patients with mental disorders. J. Clin. Psychol. 2018, 74, 2276–2286. [CrossRef]
32. Kuipers, G.S.; van Loenhout, Z.; van der Ark, L.A.; Bekker, M.H.J. Is reduction of symptoms in eating disorder patients after 1
year of treatment related to attachment security and mentalization? Eat. Disord. 2018, 26, 263–269. [CrossRef]
33. Kazdin, A.E. Mediators and mechanisms of change in psychotherapy research. Annu. Rev. Clin. Psychol. 2007, 3, 1–27. [CrossRef]
[PubMed]
34. Kazdin, A.E. Understanding how and why psychotherapy leads to change. Psychother. Res. 2009, 19, 418–428. [CrossRef]
35. Kraemer, H.C.; Wilson, G.T.; Fairburn, C.G.; Agras, W.S. Mediators and moderators of treatment effects in randomized clinical
trials. Arch. Gen. Psychiatry 2002, 59, 877–883. [CrossRef]
36. Chambless, D.L.; Hollon, S.D. Defining empirically supported therapies. J. Consult. Clin. Psychol. 1998, 66, 7–18. [CrossRef]
37. Fonagy, P. Thinking about thinking: Some clinical and theoretical considerations in the treatment of a borderline patient. Int. J.
Psychoanal. 1991, 72, 639–656. [PubMed]
38. Moher, D.; Liberati, A.; Tetzlaff, J.; Altman, D.G. Preferred reporting items for systematic reviews and meta-analyses: The PRISMA
statement. PLoS Med. 2009, 6, e1000097. [CrossRef]
39. Antonsen, B.T.; Johansen, M.S.; Rø, F.G.; Kvarstein, E.H.; Wilberg, T. Is reflective functioning associated with clinical symptoms
and long-term course in patients with personality disorders? Compr. Psychiatry 2016, 64, 46–58. [CrossRef]
40. Gullestad, F.S.; Johansen, M.S.; Høglend, P.; Karterud, S.; Wilberg, T. Mentalization as a moderator of treatment effects: Findings
from a randomized clinical trial for personality disorders. Psychother. Res. 2013, 23, 674–689. [CrossRef]
41. Arnevik, E.; Wilberg, T.; Urnes, O.; Johansen, M.; Monsen, J.T.; Karterud, S. Psychotherapy for personality disorders: Short-term
day hospital psychotherapy versus outpatient individual therapy—A randomized controlled study. Eur. Psychiatry 2009, 24,
71–78. [CrossRef]
42. Barber, J.P.; Milrod, B.; Gallop, R.; Solomonov, N.; Rudden, M.G.; McCarthy, K.S.; Chambless, D.L. Processes of therapeutic change:
Results from the Cornell-Penn Study of Psychotherapies for Panic Disorder. J. Couns. Psychol. 2020, 67, 222–231. [CrossRef]
43. Solomonov, N.; Aafjes van-Doorn, K.; Lipner, L.M.; Gorman, B.S.; Milrod, B.; Rudden, M.G.; Chambless, D.L.; Barber, J.P.
Panic-Focused Reflective Functioning and Comorbid Borderline Traits as Predictors of Change in Quality of Object Relations in
Panic Disorder Treatments. J. Contemp. Psychother. 2019, 49, 255–264. [CrossRef]
44. Boldrini, T.; Nazzaro, M.P.; Damiani, R.; Genova, F.; Gazzillo, F.; Lingiardi, V. Mentalization as a predictor of psychoanalytic
outcome: An empirical study of transcribed psychoanalytic sessions through the lenses of a computerized text analysis measure
of reflective functioning. Psychoanal. Psychol. 2018, 35, 196–204. [CrossRef]
45. Bressi, C.; Fronza, S.; Minacapelli, E.; Nocito, E.P.; Dipasquale, E.; Magri, L.; Lionetti, F.; Barone, L. Short-term psychodynamic
psychotherapy with mentalization-based techniques in major depressive disorder patients: Relationship among alexithymia,
reflective functioning, and outcome variables—A Pilot study. Psychol. Psychother. 2016, 90, 299–313. [CrossRef]
46. Ekeblad, A.; Falkenström, F.; Holmqvist, R. Reflective functioning as predictor of working alliance and outcome in the treatment
of depression. J. Consult. Clin. Psychol. 2016, 84, 67–78. [CrossRef]
47. Fischer-Kern, M.; Doering, S.; Taubner, S.; Hörz, S.; Zimmermann, J.; Rentrop, M.; Schuster, P.; Buchheim, P.; Buchheim, A.
Transference-focused psychotherapy for borderline personality disorder: Change in reflective function. Br. J. Psychiatry 2015, 207,
173–174. [CrossRef] [PubMed]
48. Tmej, A.; Fischer-Kern, M.; Doering, S.; Alexopoulos, J.; Buchheim, A. Changes in Attachment Representation in Psychotherapy:
Is Reflective Functioning the Crucial Factor? Z. Psychosom. Med. Psychother. 2018, 64, 222–236. [CrossRef] [PubMed]
49. Doering, S.; Hörz, S.; Rentrop, M.; Fischer-Kern, M.; Schuster, P.; Benecke, C.; Buchheim, A.; Martius, P.; Buchheim, P. Transference-
focused psychotherapy v. treatment by community psychotherapists for borderline personality disorder: Randomised controlled
trial. Br. J. Psychiatry 2010, 196, 389–395. [CrossRef] [PubMed]
50. Karlsson, R.; Kermott, A. Reflective-functioning during the process in brief psychotherapies. Psychother. Theory Res. Pract. 2006,
43, 65–84. [CrossRef]
51. Elkin, I.; Shea, M.T.; Watkins, J.T.; Imber, S.D.; Sotsky, S.M.; Collins, J.F.; Glass, D.R.; Pilkonis, P.A.; Leber, W.R.; Docherty, J.P.
National Institute of Mental Health Treatment of Depression Collaborative Research Program. General effectiveness of treatments.
Arch. Gen. Psychiatry 1989, 46, 971–983. [CrossRef] [PubMed]
52. Jones, E.E.; Pulos, S.M. Comparing the process in psychodynamic and cognitive-behavioral therapies. J. Consult. Clin. Psychol.
1993, 61, 306–316. [CrossRef] [PubMed]
53. Müller, C.; Kaufhold, J.; Overbeck, G.; Grabhorn, R. The importance of reflective functioning to the diagnosis of psychic structure.
Psychol. Psychother. 2006, 79, 485–494. [CrossRef] [PubMed]
54. Taubner, S.; Schulze, C.I.; Kessler, H.; Buchheim, A.; Kächele, H.; Staun, L. Veränderungen der mentalisierten Affektivität nach 24
Monaten analytischer Psychotherapie bei Patienten mit chronischer Depression. Psychother. Forum 2015, 20, 20–28. [CrossRef]
55. Fertuck, E.A.; Mergenthaler, E.; Target, M.; Levy, K.N.; Clarkin, J.F. Development and criterion validity of a computerized text
analysis measure of reflective functioning. Psychother. Res. 2012, 22, 298–305. [CrossRef]
Int. J. Environ. Res. Public Health 2021, 18, 9161 16 of 16

56. Milrod, B.L. Manual of Panic-Focused Psychodynamic Psychotherapy; American Psychiatric Press: Washington, DC, USA, 1997;
ISBN 0-88048-871-9.
57. Waldron, S.; Moscovitz, S.; Lundin, J.; Helm, F.; Jemerin, J.; Gorman, B. Evaluating the outcomes of psychotherapies: The
Personality Health Index. Psychoanal. Psychol. 2011, 28, 363–388. [CrossRef]
58. Katznelson, H.; Falkenström, F.; Daniel, S.I.F.; Lunn, S.; Folke, S.; Pedersen, S.H.; Poulsen, S. Reflective functioning, psy-
chotherapeutic alliance, and outcome in two psychotherapies for bulimia nervosa. Psychotherapy 2020, 57, 129–140. [CrossRef]
[PubMed]