Psychological Medicine Efficacy of dialectical behavior therapy for

cambridge.org/psm
adolescent self-harm and suicidal ideation:
a systematic review and meta-analysis
Oswald D. Kothgassner1 , Andreas Goreis2,3 , Kealagh Robinson4 ,
Review Article
Mercedes M. Huscsava1 , Christian Schmahl5 and Paul L. Plener1,6
Cite this article: Kothgassner OD, Goreis A,
Robinson K, Huscsava MM, Schmahl C, Plener 1
Department of Child and Adolescent Psychiatry, Medical University of Vienna, Vienna, Austria; 2Department of
PL (2021). Efficacy of dialectical behavior
Clinical and Health Psychology, Faculty of Psychology, University of Vienna, Vienna, Austria; 3Outpatient Unit for
therapy for adolescent self-harm and suicidal
ideation: a systematic review and meta- Research, Teaching and Practice, Faculty of Psychology, University of Vienna, Vienna, Austria; 4School of
analysis. Psychological Medicine 1–11. https:// Psychology, Te Herenga Waka – Victoria University of Wellington, Wellington, New Zealand; 5Department of
doi.org/10.1017/S0033291721001355 Psychosomatic Medicine and Psychotherapy, Central Institute of Mental Health, Medical Faculty Mannheim,
Heidelberg University, Mannheim, Germany and 6Department of Child and Adolescent Psychiatry and
Received: 23 October 2020 Psychotherapy, University of Ulm, Ulm, Germany
Revised: 22 March 2021
Accepted: 26 March 2021
Abstract
Keywords: Background. Given the widespread nature and clinical consequences of self-harm and
Adolescence; borderline personality disorder;
DBT-A; self-harm; self-injury; suicidal ideation
suicidal ideation among adolescents, establishing the efficacy of developmentally appropriate
treatments that reduce both self-harm and suicidal ideation in the context of broader adoles-
Author for correspondence: cent psychopathology is critical.
Oswald D. Kothgassner, Methods. We conducted a systematic review and meta-analysis of the Dialectical Behaviour
E-mail: [email protected]
Therapy for Adolescents (DBT-A) literature on treating self-injury in adolescents (12–19
years). We searched for eligible trials and treatment evaluations published prior to July
2020 in MEDLINE/PubMed, Scopus, Google Scholar, EMBASE, and the Cochrane Library
databases for clinical trials. Twenty-one studies were identified [five randomized-controlled
trials (RCTs), three controlled clinical trials (CCTs), and 13 pre-post evaluations]. We
extracted data for predefined primary (self-harm, suicidal ideation) and secondary outcomes
(borderline personality symptoms; BPD) and calculated treatment effects for RCTs/CCTs and
pre-post evaluations. This meta-analysis was pre-registered with OSF: osf.io/v83e7.
Results. Overall, the studies comprised 1673 adolescents. Compared to control groups, DBT-
A showed small to moderate effects for reducing self-harm (g = −0.44; 95% CI −0.81 to
−0.07) and suicidal ideation (g = −0.31, 95% CI −0.52 to −0.09). Pre-post evaluations sug-
gested large effects for all outcomes (self-harm: g = −0.98, 95% CI −1.15 to −0.81; suicidal
ideation: g = −1.16, 95% CI −1.51 to −0.80; BPD symptoms: g = −0.97, 95% CI −1.31 to
−0.63).
Conclusions. DBT-A appears to be a valuable treatment in reducing both adolescent self-
harm and suicidal ideation. However, evidence that DBT-A reduces BPD symptoms was
only found in pre-post evaluations.

Self-injury, suicidal ideation, and suicide attempts represent major mental health concerns for
adolescents around the world. Suicide is the leading cause of death for female adolescents and
the third highest cause of death for male adolescents in the western world (Collaboration
GBoDP, 2016; Hawton, Saunders, & O’Connor, 2012). In addition, meta-analyses estimate
that 22.9% of adolescents have engaged in self-harm whereby they deliberately and directly
damage their body tissue in the absence of suicidal ideation (Gillies et al., 2018). Although
there is ongoing debate about the nature of the relationship between self-harm and suicidal
thoughts and behaviors (Hamza, Stewart, & Willoughby, 2012), the two are related (Gillies
et al., 2018). Meta-analyses of longitudinal studies estimate that people who engage in self-
harm have 4.27 greater odds of subsequently attempting suicide, and those who engage in
© The Author(s), 2021. Published by deliberate self-harm – regardless of suicidal intent – have 1.51 greater odds of subsequent
Cambridge University Press. This is an Open death by suicide (Ribeiro et al., 2016).
Access article, distributed under the terms of
Adolescence represents a key developmental period for both self-harm and suicidality
the Creative Commons Attribution licence
(http://creativecommons.org/licenses/by/4.0/), (Wyman, 2014). Although suicide is uncommon before the age of 15, the prevalence of suicide
which permits unrestricted re-use, distribution, strongly increases from late adolescence to early adulthood (Bertolote & Fleischmann, 2002).
and reproduction in any medium, provided the Self-harm also tends to begin during early adolescence around 13–15 years, with growing evi-
original work is properly cited. dence to suggest that earlier onset of self-harm increases the risk of a more severe trajectory
(Ammerman, Jacobucci, Kleiman, Uyeji, & McCloskey, 2018; Groschwitz et al., 2015;
Plener, Schumacher, Munz, & Groschwitz, 2015). Self-harm and suicidal ideation typically
present alongside other psychiatric disorders, such as affective and stress-associated disorders
(Nock, Joiner, Gordon, Lloyd-Richardson, & Prinstein, 2006), and among adults the most

Downloaded from https://www.cambridge.org/core. IP address: 103.68.172.94, on 23 May 2021 at 20:43:49, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0033291721001355
 2 Oswald D. Kothgassner et al.

well-established link is with borderline personality disorder (BPD; records until 31 July 2020 using the keywords ‘Dialectical
Ferrara et al., 2012; Kaplan et al., 2016). Given the widespread Behaviour Therapy’ OR ‘DBT-A’ and combinations of the key-
nature and clinical consequences of adolescent self-injury, estab- words ‘Self-harm’ OR ‘Self-Injury’, ‘Suicidal Ideation’, OR
lishing developmentally appropriate treatments that reduce both ‘Suicide’ with an age limitation. Studies were included in the
self-harm and suicidal ideation in the context of broader adoles- meta-analysis if they reported an RCT or CCT comparing
cent psychopathology is critical. DBT-A with a control intervention or a pre-post evaluation of
One treatment which has received growing interest is DBT-A and reported outcomes for self-harm and/or suicidal
Dialectical Behaviour Therapy for Adolescents (DBT-A). DBT ideation in adolescents aged 12–19 who had engaged in self-
was initially developed to treat women diagnosed with BPD at injury at least once. We also excluded studies focusing solely on
high-risk for suicide (e.g. Chapman, 2006; Linehan, Heard, & pharmacological treatments. No limitations on language or pub-
Armstrong, 1993) and is widely recommended as an established lication status were invoked, and no other inclusion or exclusion
therapeutic approach for people with BPD, particularly when criteria were applied.
reducing self-injury is a priority (APA, 2006; National We analyzed the frequency of self-harm episodes and suicidal
Collaborating Center for Mental Health, 2009). Subsequently, a ideation as primary outcome measures, with BPD symptoms as a
dialectical behavioral approach was adapted for adolescents, secondary outcome measure. The title, abstract, and main text of
which prioritizes self-harm and suicidal ideation as the primary each study were examined, with the exclusion of documents
targets for therapeutic intervention (Miller, Rathus, & Linehan, occurring at each stage. The initial search generated 932 results.
2017; Rathus & Miller, 2002). DBT-A is a manualized treatment Title and abstracts were screened for eligibility and full-text papers
approach intended for outpatient settings comprised of weekly were obtained where necessary to evaluate inclusion. After screen-
individual therapy with concurrent participation in a skills-group ing, 21 studies – all peer-reviewed journal articles in English –
and which includes parental participation. In particular, DBT-A were identified and included in our meta-analysis.
focuses on developing mindfulness, distress tolerance, interper-
sonal effectiveness, and emotion regulation behavioral skills as
Data extraction and analysis
the main therapeutic tools for overcoming pervasive emotion dys-
regulation and suicidal ideation (Miller et al., 2017; Rathus & Data from included studies were entered into a spreadsheet inde-
Miller, 2015). pendently by two authors (ODK and KR). A third author (AG)
To date, two reviews evaluating the efficacy of psychosocial reviewed and discussed differences until consensus was reached.
treatments for reducing adolescent self-harm and suicidal ideation We coded the sample and intervention characteristics of each
have highlighted DBT-A as a promising treatment (Glenn, study included in the meta-analysis. For analyses of the efficacy
Franklin, & Nock, 2015; Kothgassner, Robinson, Goreis, Ougrin, of DBT-A in RCTs and CCTs, the primary outcome was the stan-
& Plener, 2020). However, previous meta-analysis focused exclu- dardized mean difference (Hedges’ g) between the DBT-A and
sively on a small number (k = 3) of randomized controlled trials control interventions on self-harm and suicidal ideation measured
(RCTs; Kothgassner et al., 2020), and the systematic review of con- post-intervention. The secondary outcome was the standardized
trolled clinical trials (CCTs, trials including a control group, but mean difference (Hedges’ g) for BPD symptoms in the DBT-A
which lack randomization) and pre-post evaluations only included and control interventions measured post-intervention. For
studies published prior to July 2013 (k = 5, no RCTs; Glenn et al., analyses regarding pre-post treatment effects, we computed the
2015). Given both the clinical importance of responding effectively standardized mean difference (Hedges’ g) based on means and
to adolescent self-harm and suicidal ideation, and the limited num- standard deviations (Dunlap et al., 1996) before and after
ber of DBT-A RCTs highlighted in previous reviews, we decided to DBT-A intervention using the formula g = (Mpost-Mpre)/
include all studies across different stages of clinical evaluation in SDpooled, where Mpost is the mean of the measure after the inter-
order to provide the most comprehensive synthesis of the current vention and Mpre the mean before the intervention, with SDpooled
evidence. In addition, although DBT has shown success in treating as the standard deviation for both measurements, defined as
BPD symptoms in adults, the efficacy of DBT-A for treating BPD SDpooled = SQRT(SD2pre + SD2post)/2 (Lakens, 2013).
symptoms among adolescents who self-injure remains to be evalu- Means, standard deviations, and sample sizes were retrieved
ated (Cristea et al., 2017). Thus, in this review, we include RCTs, and inserted into a spreadsheet. If means or standard deviations
CCTs, and pre-post evaluation studies to evaluate the efficacy of were not reported in studies or Supplemental materials, conver-
DBT-A for reducing self-harm, suicidal ideation and BPD symp- sions via Revman Calculator (Cochrane Collaboration, 2014) or
toms among adolescents, and conduct subgroup analyses to com- formulas (Card, 2012) were conducted. If self-harm episodes
pare the results for RCTs with those of less rigorous studies. The were reported as proportions or odds ratios, they were trans-
greater heterogeneity in studies also allows us to assess whether formed to Hedges’ g via the formula provided in Lipsey and
characteristics of the study (e.g. participant age, treatment duration) Wilson (2001). Effect size calculations and meta-analyses were
moderate the meta-analytic effect of DBT-A on outcomes, in order conducted with the metafor package for R (Viechtbauer, 2010).
to better understand the parameters under which DBT-A is most Following established conventions, an effect size of 0.20 was con-
successful. sidered a small effect size, 0.50 a medium effect, and 0.80 a large
effect size (Cohen, 1988). Random-effects models were applied to
estimate aggregated effect sizes. Heterogeneity across study
Method outcomes was reported with I 2 values, where 25% indicates low
heterogeneity, 50% moderate, and 75% high heterogeneity
Search strategy and inclusion criteria
(Higgins, Thompson, Deeks, & Altman, 2003). Moderator ana-
We conducted a search of MEDLINE/PubMed, Scopus, Google lyses (meta-regression) were conducted to test whether treatment
Scholar, EMBASE, and the Cochrane Library databases for clin- duration, gender composition, and participant mean age moder-
ical trials for studies published from the beginning of database ated the effect of the DBT-A on each outcome. Egger’s regressions

Downloaded from https://www.cambridge.org/core. IP address: 103.68.172.94, on 23 May 2021 at 20:43:49, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0033291721001355
 Psychological Medicine 3

Fig. 1. PRISMA flowchart showing the screening, exclusion, and inclusion criteria.

were conducted to estimate publication bias (Sterne & Egger, Results

2005), with adjusted effect sizes calculated using trim-and-fill
In total, 21 studies were identified (see Fig. 1 for the PRISMA flow
analyses and, based on funnel plot asymmetry, numbers of
diagram). Five studies were RCTs, three studies were CCTs, and
imputed missing studies (Duval & Tweedie, 2000). All data and
13 were pre-post evaluation studies (see Table 1 for an overview
analysis code are available on the Open Science Framework
of study characteristics). The study by Rathus and Miller
(doi:10.17605/OSF.IO/YZXPJ).
(2002) – originally included as a CCT – was included as a pre-
post evaluation study, given that data for the control intervention
was unavailable. In total, the 21 studies comprised 1673 adoles-
Risk of bias assessment
cents. Overall, 1063 participants received DBT-A interventions,
Risk of bias for each study was assessed using predefined criteria and 610 received control interventions. A sufficient number of
based on the Agency for Healthcare Research and Quality method studies (k > 1; Pigott, 2012) were identified to calculate aggregate
guide (see Supplement 1; Viswanathan et al., 2018). Each study effect sizes for self-harm and suicidal ideation outcomes in RCTs,
was assessed in regard to randomization, selection and attrition CCTs, and pre-post evaluations. However, only one controlled
bias, confounding bias, measurement bias, and statistical pro- study reported BPD symptoms as an outcome, and so this out-
blems and received a rating of low, moderate or high risk of come was solely assessed among pre-post evaluations. Across
bias. Low risk of bias indicates that the study was judged to be studies, participants tended to be female (M = 82%) and 15.4
valid, moderate risk indicates concerns that probably do not invali- years old (S.D. = 1.3). An average of 7% of participants dropped-
date the study’s results, and high risk of bias indicates significant out across studies (range: 0–40%) and 63% received concurrent
concerns that likely invalidate the study’s results. Two investigators psychopharmacological intervention over the course of the inter-
(KR and ODK) independently assessed the studies and differences vention (k = 11 studies did not provide sufficient data about
were reviewed until consensus was reached. medication).

Downloaded from https://www.cambridge.org/core. IP address: 103.68.172.94, on 23 May 2021 at 20:43:49, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0033291721001355
https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0033291721001355
Downloaded from https://www.cambridge.org/core. IP address: 103.68.172.94, on 23 May 2021 at 20:43:49, subject to the Cambridge Core terms of use, available at

4
Table 1. Study characteristics

Treatment
Study Sample Control % Age range duration in Outcome %
Study Year design size Country intervention Female (mean) months Setting Eligibility criteria (measures) Drop-out

Apsche, Bass, and Siv RCT 20 USA Mode deactivation 0% 15–18 (DBT-A: 12 Residential Not stated SI (SIQ) 0%
(2006) therapy 15.9, Control: treatment
16.1)
Berk, Starace, Black, Pre-post 22 USA – 92% 12–17 (15.2) 6 Outpatient Recent history of suicidal and/or SH (past 6-month 8%
and Avina (2020) self-injury behaviors frequency), SI (SIQ/
SIQ-Jr), BPD
(SCID-II)
Buerger et al. (2019) Pre-post 72 Germany – 92% 12–17 (15.7) 6.25 Outpatient 3 + BPD criteria; fluency in SH and SI (SITBI-G), 18%
German BPD (LPI)
Courtney and Pre-post 42 Canada – 93% Adolescents 15 3 Outpatient SIQ score >30, or self-injury in the SH (chart review) 51%
Flament (2015) and above past 4 months BPD (LPI)
(16.5)
Fischer and Peterson Pre-post 7 USA – 100% 14–17 (16.2) 6 Outpatient Binge eating within past 4 weeks; SH (DSHI past 30%
(2015) 1+ suicide attempt or episode of month)
self-injury within past year;
height and weight within or
above typical limits for age and
stage of development
Fleischhaker et al. Pre-post 10 Germany – 100% 13–19 (not 6 Outpatient SH and/or suicidal behavior SH (LPC) 25%
(2011) stated) within past 16 weeks; BPD
diagnosis or 3+ BPD criteria
Geddes, Dziurawiec, Pre-post 6 Australia – 100% 14 years 6 6.5 Outpatient Average cognitive ability and SH and SI (in-house 33%
and Lee (2013) month–15 established reading level; questionnaire)
years 1 month self-injury and/or suicidal
(15.1) ideation within past 12 months; 3
+ BPD criteria
Gillespie, Joyce, Pre-post 84 Ireland – 85% 13–18 (15.7) 4 Outpatient Demonstration of emotional and SH (review of diary 16%
Flynn, and Corcoran behavioral dysregulation; cards), SI (in-house
(2019) persistent self-injury with 1+ questionnaire),
episodes of self-injury or suicidal BPD (BSL-23)
acts within the past 16 weeks or
chronic suicidal ideation;
expressed commitment by both
adolescent and parent/guardian

Oswald D. Kothgassner et al.

Goldstein, Axelson, Pre-post 10 USA – 80% 14–18 (15.8) 6 Outpatient Bipolar diagnosis with an acute SH (K-SADS), SI 10%
Birmaher, and Brent manic, mixed, or depressive (MSSI)
(2007) episode within past 3 months;
engaged in a pharmacotherapy
regimen; 1+ parent/guardian
willing to participate in family
sessions
Goldstein et al. (2015) RCT 20 USA Treatment-as-usual 75% 12–18 (DBT: 12 Outpatient Bipolar disorder diagnosis; an SH (LIFE), SI 2%
15.8, Control: acute manic, mixed, or (SIQ-Jr)
16.8) depressive episode within past 3
months; willingness to engage in
https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0033291721001355
Downloaded from https://www.cambridge.org/core. IP address: 103.68.172.94, on 23 May 2021 at 20:43:49, subject to the Cambridge Core terms of use, available at

pharmacotherapy; 1+ parent/

Psychological Medicine
guardian willing to participate in
family sessions
James, Taylor, Pre-post 16 UK – 100% 15–18 (16.4) 12 Outpatient History of >6 months severe and SH (clinical 13%
Winmill, and persistent self-injury interview), BPD
Alfoadari (2008) (SCID-II)
James, Winmill, Pre-post 18 UK – 88% 13–17 (15.5) 3 Outpatient History of >6 months persistent SH (clinical 28%
Anderson, and self-injury interview), BPD
Alfoadari (2011) (SCID-II)
Katz et al. (2004) CCT 53 Canada Treatment-as-usual 84% 14–17 (15.4) 0.5 Inpatient Admitted to inpatient unit SH (LPC), SI 9%
following a suicide attempt or (SIQ-Jr)
with severe suicidal ideation;
agreed to stay in the hospital for
brief treatment
McCauley et al. (2018) RCT 173 USA Individual and 94% 12–18 (14.9) 6 ED, inpatient, 1+ lifetime suicide attempt; SH (SASII), SI 40%
group outpatient elevated past-month suicidal (SIQ-Jr), BPD
supportive therapy services, and ideation (⩾24 on the SIQ-Jr); ⩾3 (SCID-II)
community lifetime self-injury episodes,
services including 1 episode in the 12
weeks before screening, 3+ BPD
criteria
McDonell et al. (2010) CCT 155 USA Historical control 58% 12–17 (15.54) 12 Inpatient Not stated SH (quality 0%
assurance
database)
Mehlum et al. (2014) RCT 77 Norway Enhanced usual 88% 12–18 (DBT: 4.75 Outpatient 2+ self-injury episodes, with 1+ SH (frequency 0%
care 15.9, Control: within the last 16 weeks; 2+ BPD count), SI (SIQ-Jr),
15.3) criteria (and the self-destructive BPD (BSL-23)
criterion), or, 1+ criterion of BPD
and at least 2 subthreshold-level
criteria; fluency in Norwegian
Perepletchikova et al. Pre-post 11 USA – 55% 8–11 years 6 1.5 Elementary Not stated SI (MFQ) 0%
(2011) months (9.83) school
Rathus and Miller CCT 13 USA Treatment-as-usual 93% Range not 3 Outpatient A suicide attempt with last 16 SI (SSI), BPD (LPI) Defers by
(2002) stated (DBT: weeks or current suicidal group TAU:
16.1, Control: ideation; a BPD diagnosis or 3+ 60%,
15.0) BPD criteria DBT-A:
38%
Santamarina-Perez RCT 35 Spain Treatment-as-usual 89% 12–17 years 11 4 Outpatient Repetitive SH (proposed DSM-V SH (medical 20%
et al. (2020) + group therapy months (DBT: criteria) and/or suicide attempts records), SI (SIQ-Jr)
15.3, Control: over the last 12 months and at
15.2) current high risk of suicide; 1+
parent/guardian willing to
participate in family sessions
Tebbett-Mock et al. CCT 801 USA Historical control 66% 12–17 (DBT): 0.36 Inpatient Not stated SH (medical Not
(2020) 15.7 (1.4) records), SI reported
Control: 15.6 (observation hours
(1.5) for SI)
(Continued )

5
 6 Oswald D. Kothgassner et al.

First, we considered the overall efficacy of DBT-A, compared

Children; LIFE, Longitudinal Interval Follow-Up Evaluation Self-Injurious/Suicidal Behaviour Scale; LPC, Lifetime Parasuicide Count; LPI, The Life Problems Inventory; MFQ, Mood and Feeling Questionnaire; MSSI, Modified Scale for Suicidal Ideation;
SASII, Suicide Attempt Self-Injury Interview; SCID-II, Structured Clinical Interview for DSM-IV Axis II; SITBI-G, Self-Injurious Thoughts and Behaviours Interview (German version); SIQ, Suicidal Ideation Questionnaire; SIQ-Jr, Suicidal Ideation Questionnaire
Outcomes: BPD, borderline personality disorder; SH, self-harm; SI, suicidal ideation. Measures: BSL-23, Borderline Symptom List; DSHI, Deliberate Self-Harm Inventory; K-SADS, Schedule for Affective Disorders and Schizophrenia for School-Age
Drop-out
to control interventions, for reducing adolescent self-harm.
%

37% Seven effect sizes encompassing 1314 participants (DBT-A: n =

714, Control: n = 600, k = 7) were extracted. Meta-analysis
revealed a significant difference between groups (g = −0.44, 95%
CI −0.81 to −0.07, p = 0.021) with a high heterogeneity between
SH and SI (TSCC)

studies (I 2 = 80.13%). That is, DBT-A interventions showed a

(measures)
Outcome

small-to-medium improvement in reducing self-harm compared

to control interventions. See Fig. 2 for a comparison of DBT-A
(relative to control interventions) by study type (RCT or CCT)
for reducing self-harm.
willing to commit to the 15-week

Next, we considered the overall efficacy of DBT-A, compared

adolescent and parent/guardian
unstable affect or relationships
self-injury, and/or intense and

to control interventions, for reducing adolescent suicidal ideation.

within the past 3–6 months;
History of suicide attempts,

Five effect sizes encompassing 1159 participants (DBT-A: n = 604,

Eligibility criteria

Control: n = 555, k = 6) were extracted. Meta-analysis revealed a

significant difference between groups (g = −0.31, 95% CI −0.52
to −0.09, p = 0.006), with moderate heterogeneity between studies
(I 2 = 44.05%). That is, DBT-A was moderately more effective at
treatment

reducing suicidal ideation than control interventions. See Fig. 3

for a comparison of DBT-A (relative to control interventions)
by study type (RCT or CCT) for reducing suicidal ideation.
In terms of BPD symptoms, Mehlum et al. (2014) conducted
the only RCT which assessed the efficacy of DBT-A in reducing
services, and
Setting

community
outpatient,

BPD symptoms. They reported that both DBT-A (n = 39) and

Inpatient,

schools

‘enhanced usual care’ (non-manualized standard care provided

at least once weekly for the purpose of the trial; n = 38) reduced
BPD symptoms, but found no significant group difference.
duration in
Treatment

Given that only one RCT was eligible for inclusion in this review,
months

3.75

we did not conduct meta-analysis of the effect of DBT-A and con-

trol interventions on BPD symptoms.
Moving beyond group comparisons of treatment and control
interventions, we next considered within-subject changes in self-
Age range

13–18 (16.0)

harm, suicidal ideation, and BPD symptoms following DBT-A

(mean)

intervention. Table 2 shows the effects of DBT-A across outcome

measures. Among participants who received DBT-A, pre-post
comparisons indicate large reductions in self-harm (g = −0.98),
suicidal ideation (g = −1.16), and BPD symptoms (g = −0.97).
Female

82%
%

All effects were statistically significant, with suicidal ideation

(I 2 = 54.58) and BPD symptoms (I 2 = 43.51) showing moderate
heterogeneity, and self-harm showing low-to-moderate hetero-
geneity (I 2 = 0.00) across studies.
Junior; SSI, Scale for Suicidal Ideation; TSCC, Trauma Symptom Checklist for Children.
intervention
Control

Next, in order to better understand the parameters in which

DBT-A is most effective, we assessed whether study characteristics
moderated the efficacy of DBT-A in reducing adolescent
self-harm and suicidal ideation. We conducted multiple meta-
–

regression to assess whether treatment duration (in months),

Country

age (in years), or proportion of young women in the overall

USA

sample (compared to young men) moderated the size of

meta-analytic effects between DBT-A and control interventions
Sample

(see Supplement 2 for all coefficients). Treatment duration was

size

28

negatively associated with the change in effect sizes for suicidal

ideation (b = −0.08, p = 0.012), but not self-harm (b = −0.06,
Pre-post
design
Study

p = 0.379). That is, a longer duration of DBT-A led to a larger

reduction in suicidal ideation when compared to control interven-
tions. In contrast, gender composition and age of the sample
did not influence the effect size of the difference between
Table 1. (Continued.)

DBT-A and control interventions for self-harm or suicide idea-

Popenoe (2008)
Woodberry and

tion (all ps > 0.05).

Study Year

We repeated these meta-regressions for the assessments of

effect sizes in pre-post evaluations (see Supplement 2 for all coef-
ficients). Again, treatment duration was negatively associated
with larger effect sizes for reducing BPD symptoms (b = −0.29,

Downloaded from https://www.cambridge.org/core. IP address: 103.68.172.94, on 23 May 2021 at 20:43:49, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0033291721001355
 Psychological Medicine 7

Fig. 2. Forest plot of trials comparing the effect of DBT-A and controls on symptoms of self-harm.

Fig. 3. Forest plot of trials comparing the effect of DBT-A and controls on symptoms of suicidal ideation.

p = 0.016), such that the longer the DBT-A treatment the larger distributed. That is, we found no evidence for publication bias
the reduction in symptoms from pre to post. No other moderating in the studies assessing self-harm, suicidal ideation, or BPD symp-
effects of treatment duration were found. Similarly, gender com- toms included in the present meta-analyses. In addition, Egger’s
position and age of the sample did not influence any of the out- regression found no evidence for funnel plot asymmetry in the
comes in pre-post analyses. analyses we conducted (all zs < 1.64, all ps > 0.100). Given that
We next consider the potential impact of publication bias in no indication of publication bias was found, no adjustments
the studies examined in these meta-analyses. For each outcome according to trim-and-fill analysis were conducted in any of the
(both in the effect as compared to control interventions and in analyses.
pre-post comparisons), visual inspection of funnel plots suggested Finally, we evaluate the quality of the studies included in the
that across all four outcomes, studies were symmetrically current review using predefined criteria based on the Agency

Downloaded from https://www.cambridge.org/core. IP address: 103.68.172.94, on 23 May 2021 at 20:43:49, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0033291721001355
 8 Oswald D. Kothgassner et al.

Table 2. Pre-post treatment effects (Hedges’ g) and heterogeneity indices of DBT-A

Outcome k n Hedges’ g 95% CI I2

Reduction at treatment completion

Self-harm 16 498 −0.98* −1.15 to −0.81 0.00
Suicidal ideation 11 299 −1.16* −1.51 to −0.80 54.58
Borderline personality disorder symptoms 5 218 −0.97* −1.31 to −0.63 43.51
Hedges’ g indicates change from pre to post-intervention such that a negative effect size indicates a reduction in that outcome following DBT-A.
*Indicates effect sizes that are statistically significant ( p < 0.001).

for Healthcare Research and Quality method guide (Viswanathan which further BPD symptoms are addressed in later stages of
et al., 2018). Overall, statistical problems were common with therapy, after an initial focus on establishing sufficient behavior
57.1% of studies rated at ‘High’ risk of bias due to low sample control. Since the therapeutic relationship can be considered as
sizes. High risk of measurement bias (38.1%) and confounding a critical reinforcement for people with BPD (Bedics, Atkins,
bias (33.3%) was also common, due to reliance on unvalidated Harned, & Linehan, 2015), a longer duration may mean a more
instruments or inadequately addressing potential confounds. effective use of the therapeutic relationship in terms of contin-
Taken together, quality assessments highlight the need for wide- gency management (Miller et al., 2017).
spread adoption of standardized measurement instruments and We also found larger effect sizes for self-harm and suicidal
well-powered replication studies (see Supplement 1). ideation in RCTs than in CCTs. This difference might be
explained by the fact that these study types differ by recruitment
setting. RCTs predominantly recruited adolescents receiving out-
Discussion
patient care (Goldstein et al., Mehlum et al., 2014;
Given the scarcity of studies investigating the efficacy of DBT-A in Santamarina-Perez et al., 2020), except for McCauley et al.
reducing adolescent self-harm and suicidal ideation, particularly (2018) who included adolescents recruited from both inpatient
in the context of comorbid psychopathology such as BPD, we and outpatient settings, whereas all CCTs reporting self-harm
conducted a systematic review of controlled trials to inform best- and suicidal ideation outcomes consisted of participants recruited
practice clinical decision making. Our meta-analysis included 21 from inpatient settings (Katz, Cox, Gunasekara, & Miller, 2004;
studies comprised of 1673 participants and provides evidence to McDonell et al., 2010; Tebbett-Mock, Saito, McGee, Woloszyn,
support the efficacy of DBT-A (compared to control interven- & Venuti, 2020). Another potential explanation may lie in the dif-
tions) for reducing self-harm and suicidal ideation as primary ferent methodological quality of RCT and CCT studies.
outcomes. The effect size for self-harm reduction in favor of No study accounted for the combined effect of pharmacological
DBT-A was large for RCTs and small-to-medium when CCTs treatment with DBT-A, despite the fact that psychopharmaco-
were included. The effect sizes for suicidal ideation reduction in logical treatment for adolescents with BPD in general is common
favor of DBT-A were small-to-medium for both RCTs and (Cailhol et al., 2013), and over half of participants treated in the
when all controlled studies were included in the analysis. The cur- included trials received additional psychopharmacological treat-
rent review identified an insufficient number of studies to evaluate ment. Further, some studies reported reduction of medication or
the efficacy of DBT-A in BPD symptoms. adherence as an outcome variable (Katz et al., 2004; McDonell
A growing body of research demonstrates that therapeutic et al., 2010; Tebbett-Mock et al., 2020). However, to date the effi-
interventions for self-harm and suicidal ideation in general cacy of a combined therapy approach remains unclear.
show limited efficacy (Fox et al., 2020; Kothgassner et al., 2020), Critically, DBT-A targets both the adolescent and their family.
highlighting the importance of isolating specific therapies which Typically, adolescents continue to live in the environment where
show promise for further development. In a review of all pub- they acquired their dysfunctional patterns and so families are inte-
lished RCTs targeted at reducing suicidal thoughts and behaviors, grated into therapy in order to address invalidating behaviors
DBT showed a small treatment effect for self-harm, but had no within the family context. In terms of contingencies, this holistic
effect on suicidal ideation (Fox et al., 2020). Similarly, in a review approach reinforces skills and helps to decrease maladaptive beha-
of controlled trials investigating the efficacy of DBT among adult viors by addressing both the adolescent’s and parent’s behavioral
samples found a small effect in favor of DBT for reducing self- and communicative repertoire. Preliminary research provides tenta-
injury, but no effect on suicidal ideation (DeCou, Comtois, & tive support for these mechanisms of change. In a non-randomized
Landes, 2019). Focusing specifically on children and adolescents, pilot of DBT-A among a small sample of ethnic minority adoles-
our results reveal promising effects of DBT-A for both self-harm cents, adaptive coping at pretreatment predicted subsequent
and suicidal ideation for both RCTs and CCTs. Given that adoles- increased use of DBT skills at post-treatment (Yeo et al., 2020).
cence is a key developmental period for both self-harm and sui- Secondary analysis of McCauley et al. (2018) revealed that adoles-
cidal ideation (Wyman, 2014), future research is needed to cents who reported higher emotion dysregulation at baseline, and
understand the underlying mechanism(s) of how DBT-A works whose parents reported greater psychopathology and emotion dys-
to improve self-harm and suicidal ideation. regulation demonstrated greater reduction in self-harm following
Across all studies, our findings indicate that longer duration of 6-months of DBT-A treatment (Adrian et al., 2019). In a longitu-
DBT-A may be crucial for greater efficacy, particularly for suicidal dinal study of adults with a recent suicide attempt who received
ideation. Additionally, longer treatment duration was associated DBT treatment, participants with higher problem-focused coping
with greater reductions in BPD symptoms in pre-post evaluations. and poorer access to emotion regulation strategies were more likely
These findings correspond to the DBT-A treatment hierarchy in to reattempt suicide over the course of 2 years (Kuehn, King,

Downloaded from https://www.cambridge.org/core. IP address: 103.68.172.94, on 23 May 2021 at 20:43:49, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0033291721001355
 Psychological Medicine 9

Linehan, & Harned, 2020). However, future research is needed to Adrian, M., McCauley, E., Berk, M. S., Asarnow, J. R., Korslund, K., Avina,
better establish mechanisms of therapeutic change, as well as iden- C. & Linehan, M. M. (2019). Predictors and moderators of recurring self-
tify who stands to benefit most from DBT-A. harm in adolescents participating in a comparative treatment trial of psy-
chological interventions. Journal of child psychology and psychiatry, 60
Our review has several limitations. First, only one RCT
(10), 1123–1132. https://doi.org/10.1111/jcpp.13099.
(Mehlum et al., 2014) assessed the impact of DBT-A (compared
American Psychiatric Association (2006). American Psychiatric Association
to control interventions) on BPD symptoms, preventing us practice guidelines for the treatment of psychiatric disorders. Washington
from meta-analytically considering change in BPD symptoms as DC: American Psychiatric Pub.
a secondary outcome. Second, we reported a moderately high het- Ammerman, B. A., Jacobucci, R., Kleiman, E. M., Uyeji, L. L., & McCloskey, M.
erogeneity among studies assessing the efficacy of DBT-A on S. (2018). The relationship between nonsuicidal self-injury age of onset and
self-harm. This heterogeneity might be due to the use of different severity of self-harm. Suicide Life-Threatening Behavior, 48(1), 31–37.
control interventions, with some studies using specific psychothera- doi:10.1111/sltb.12330.
peutic control interventions (McCauley et al., 2018) and providing *Apsche, J. A., Bass, C. K., & Siv, A. (2006). A treatment study of mode deacti-
enhanced usual care or additional propositions (Mehlum et al., vation therapy in an out patient community setting. International Journal of
Behavioral Consultation and Therapy, 2(2), 277–285. doi: 10.1037/
2014; Santamarina-Perez et al., 2020). Alternatively, this heterogen-
h0100782.
eity may reflect differences in how self-harm was assessed; 58.8% (k
Bedics, J. D., Atkins, D. C., Harned, M. S., & Linehan, M. M. (2015). The
= 7) of studies used unvalidated clinical interviews, medical records, therapeutic alliance as a predictor of outcome in dialectical behavior
daily diary cards, or instruments developed in-house to assess self- therapy versus nonbehavioral psychotherapy by experts for borderline
harm. Moreover, our meta-analysis is limited by the low number of personality disorder. Psychotherapy, 52(1), 67. doi:10.1037/a0038457.
effects available to be included which contributes to the wide con- *Berk, M. S., Starace, N. K., Black, V. P., & Avina, C. (2020). Implementation of
fidence intervals of some of the estimates. Finally, young women dialectical behavior therapy with suicidal and self-harming adolescents in a
made up 75% or more of the sample in most studies (81.0%, k = community clinic. Archives of Suicide Research, 24(1), 64–81. doi:10.1080/
17) included in this review. Thus, questions remain regarding the 13811118.2018.1509750.
efficacy of DBT-A in reducing self-harm and suicidal ideation Bertolote, J. M., & Fleischmann, A. (2002). Suicide and psychiatric diagnosis:
A worldwide perspective. World Psychiatry, 1(3), 181–185.
among young men and gender-diverse young people.
*Buerger, A., Fischer-Waldschmidt, G., Hammerle, F., Auer, K. V., Parzer, P.,
Despite these limitations, this review provides the most com-
& Kaess, M. (2019). Differential change of borderline personality disorder
prehensive analysis to date of the available evidence for the effi- traits during dialectical behavior therapy for adolescents. Journal of
cacy of DBT-A in reducing self-harm and suicidal ideation Personality Disorders, 33(1), 119–134. doi:10.1521/pedi_2018_32_334.
among adolescents. Current evidence indicates that DBT-A is Cailhol, L., Jeannot, M., Rodgers, R., Guelfi, J. D., Perez-Diaz, F.,
superior to control interventions in reducing both self-harm Pham-Scottez, A., … Speranza, M. (2013). Borderline personality disorder
and suicidal ideation among adolescents, with limited evidence and mental healthcare service use among adolescents. Journal of Personality
of efficacy for reducing BPD symptoms. Future research should Disorders, 27(2), 252–259. doi:10.1521/pedi.2013.27.2.252.
focus on improving the quality of evidence for the efficacy of Card, N. A. (2012). Applied meta-analysis for social science research. New York,
DBT-A for underrepresented populations such as young men NY: Guilford.
Chapman, A. L. (2006). Dialectical behavior therapy: Current indications and
and gender diverse people, and for reducing BPD symptoms
unique elements. Psychiatry, 3(9), 62–68
among adolescents in general. Greater investigation is also needed
Cochrane Collaboration (2014). Review manager (RevMan) [computer pro-
to understand the combined effect of DBT-A and psychophar- gram]. version 5.3. Copenhagen: The Nordic Cochrane Centre.
macological interventions, as well as to establish the feasibility Cohen, J. (1988). Statistical power analysis for the behavioral sciences.
and efficacy of DBT-A in teletherapeutic contexts in order to New York, NY: Routledge Academic.
improve accessibility for young people who require specialized Collaboration GBoDP (2016). Global and national burden of diseases and
care for reducing their self-harm behavior and suicidal ideation. injuries among children and adolescents between 1990 and 2013:
Findings from the global burden of diseases and injuries among children
Supplementary material. The supplementary material for this article can and adolescents between 1990 and 2013: Findings from the global burden
be found at https://doi.org/10.1017/S0033291721001355. of disease 2013 study. JAMA Pediatrics, 170(3), 267–287. doi:10.1001/
jamapediatrics.2015.4276.
Author contributions. ODK, AG, KR, and MH wrote the first draft of the *Courtney, D. B., & Flament, M. F. (2015). Adapted dialectical behavior ther-
manuscript. ODK and KR conducted the literature search, coded the studies, apy for adolescents with self-injurious thoughts and behaviors. The Journal
and conducted the risk of bias assessment. ODK and AG prepared the statistical of Nervous and Mental Disease, 203(7), 537–544. doi: 10.1097/
procedures and AG analyzed the data and made final figures. CS and PLP con- NMD.0000000000000324.
tributed extensively to the first draft. All authors approved the final manuscript. Cristea, I. A., Gentili, C., Cotet, C. D., Palomba, D., Barbui, C., & Cuijpers, P.
(2017). Efficacy of psychotherapies for borderline personality disorder:
Financial support. This research received no specific grant from any fund-
A systematic review and meta-analysis. JAMA Psychiatry, 74(4), 319–328.
ing agency, commercial or not-for-profit sectors.
doi:10.1001/jamapsychiatry.2016.4287
Conflict of interest. None of the authors declare any conflict of interest with DeCou, C. R., Comtois, K. A., & Landes, S. J. (2019). Dialectical behavior ther-
regard to this manuscript. Paul Plener has received research funding from the apy is effective for the treatment of suicidal behavior: A meta-analysis.
German Federal Institute for Drugs and Medical Devices (BfArM), German Behavior Therapy, 50(1), 60–72. doi:10.1016/j.beth.2018.03.009.
Federal Ministry of Education and Research (BMBF), VW-Foundation, Dunlap, W. P., Cortina, J. M., Vaslow, J. B., & Burke, M. J. (1996). Meta-ana-
Baden-Württemberg Foundation, Lundbeck, and Servier. He received a speak- lysis of experiments with matched groups or repeated measures designs.
er’s honorarium from Shire. Psychological Methods, 1(2), 170–177. https://doi.org/10.1037/1082-
989X.1.2.170.
Duval, S., & Tweedie, R. (2000). A nonparametric ‘trim and fill’ method of
accounting for publication bias in meta-analysis. Journal of the American
References
Statistical Association, 95(449), 89–98.
Studies included in the meta-analysis are denoted with an Ferrara, M., Terrinoni, A., & Williams, R. (2012). Non-suicidal self-injury
asterisk. (Nssi) in adolescent inpatients: assessing personality features and attitude

Downloaded from https://www.cambridge.org/core. IP address: 103.68.172.94, on 23 May 2021 at 20:43:49, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0033291721001355
 10 Oswald D. Kothgassner et al.

toward death. Child and Adolescent Psychiatry and Mental Health, 6(1), child abuse on non-suicidal self-injury and suicidality. Comprehensive
1–8. https://doi.org/10.1186/1753-2000-6-12. Psychiatry, 7, 86–94. doi:10.1016/j.comppsych.2016.08.016.
*Fischer, S., & Peterson, C. (2015). Dialectical behavior therapy for adolescent *Katz, L. Y., Cox, B. J., Gunasekara, S., & Miller, A. L. (2004). Feasibility of dia-
binge eating, purging, suicidal behavior, and non-suicidal self-injury: lectical behavior therapy for suicidal adolescent inpatients. Journal of the
A pilot study. Psychotherapy, 52(1), 78–92. doi:10.1037/a0036065. American Academy of Child & Adolescent Psychiatry, 43(3), 276–282.
*Fleischhaker, C., Böhme, R., Sixt, B., Brück, C., Schneider, C., & Schulz, E. doi:10.1097/00004583-200403000-00008.
(2011). Dialectical behavioral therapy for adolescents (DBT-A): A clinical Kothgassner, O. D., Robinson, K., Goreis, A., Ougrin, D., & Plener, P. L.
trial for patients with suicidal and self-injurious behavior and borderline (2020). Does treatment method matter? A meta-analysis of the past 20
symptoms with a one-year follow-up. Child and Adolescent Psychiatry years of research on therapeutic interventions for self-harm and suicidal
and Mental Health, 5(1), 3. doi:10.1186/1753-2000-5-3. ideation in adolescents. Borderline Personality Disorder and Emotion
Fox, K. R., Huang, X., Guzm�n, E. M., Funsch, K. M., Cha, C. B., Ribeiro, J. Dysregulation, 7, 9. doi:10.1186/s40479-020-00123-9.
D., & Franklin, J. C. (2020). Interventions for suicide and self-injury: A Kuehn, K. S., King, K. M., Linehan, M. M., & Harned, M. S. (2020). Modeling
meta-analysis of randomized controlled trials across nearly 50 years of the suicidal behavior cycle: Understanding repeated suicide attempts among
research. Psychological Bulletin, 146(12), 1117–1145. https://doi.org/ individuals with borderline personality disorder and a history of attempting
10.1037/bul0000305. suicide. Journal of Consulting and Clinical Psychology, 88, 570–581. doi:
*Geddes, K., Dziurawiec, S., & Lee, C. W. (2013). Dialectical behaviour therapy 10.1037/ccp0000496.
for the treatment of emotion dysregulation and trauma symptoms in self- Lakens, D. (2013). Calculating and reporting effect sizes to facilitate cumulative
injurious and suicidal adolescent females: A pilot programme within a science: A practical primer for t tests and ANOVAs. Frontiers in Psychology,
community-based child and adolescent mental health service. Psychiatry 4, 863. doi: 10.3389/fpsyg.2013.00863
Journal, 2013, 145219. doi:10.1155/2013/145219. Linehan, M. M., Heard, H. L., & Armstrong, H. E. (1993). Naturalistic
*Gillespie, C., Joyce, M., Flynn, D., & Corcoran, P. (2019). Dialectical behav- follow-up of a behavioral treatment for chronically parasuicidal borderline
iour therapy for adolescents: A comparison of 16-week and 24-week pro- patients. Archives of General Psychiatry, 50(12), 971–974. doi: 10.1001/
grammes delivered in a public community setting. Child and Adolescent archpsyc.1993.01820240055007
Mental Health, 24(3), 266–273. doi:10.1111/camh.12325. Lipsey, M. W., & Wilson, D. B. (2001). Practical meta-analysis. Thousand
Gillies, D., Christou, M. A., Dixon, A. C., Featherston, O. J., Rapti, I., Oaks, CA: Sage.
Garcia-Anguita, A., … Christou, P. A. (2018). Prevalence and characteristics *McCauley, E., Berk, M. S., Asarnow, J. R., Adrian, M., Cohen, J., Korslund, K.,
of self-harm in adolescents: Meta-analyses of community-based studies … Linehan, M. M. (2018). Efficacy of dialectical behavior therapy for ado-
1990–2015. Journal of the American Academy of Child & Adolescent lescents at high risk for suicide: A randomized clinical trial. JAMA
Psychiatry, 57(10), 733–741. doi:10.1016/j.jaac.2018.06.018. Psychiatry, 75(8), 777–785. doi: 10.1001/jamapsychiatry.2018.1109.
Glenn, C. R., Franklin, J. C., & Nock, M. K. (2015). Evidence-based psycho- *McDonell, M. G., Tarantino, J., Dubose, A. P., Matestic, P., Steinmetz, K.,
social treatments for self-injurious thoughts and behaviors in youth. Galbreath, H., & McClellan, J. M. (2010). A pilot evaluation of dialectical
Journal of Clinical Child & Adolescent Psychology, 44(1), 1–29. behavioural therapy in adolescent long-term inpatient care. Child and
doi:10.1080/15374416.2014.945211 Adolescent Mental Health, 15(4), 193–196. doi:10.1111/j.1475-3588.2010.
Goldstein, T. R., Axelson, D. A., Birmaher, B., & Brent, D. A. (2007). 00569.x.
Dialectical behavior therapy for adolescents with bipolar disorder: *Mehlum, L., Tørmoen, A. J., Ramberg, M., Haga, E., Diep, L. M., Laberg, S.,
A 1-year open trial. Journal of the American Academy of Child & … Grøholt, B. (2014). Dialectical behavior therapy for adolescents with
Adolescent Psychiatry, 46(7), 820–830. doi:10.1097/chi.0b013e31805c1613. repeated suicidal and self-harming behavior: A randomized trial. Journal
*Goldstein, T. R., Fersch-Podrat, R. K., Rivera, M., Axelson, D. A., Merranko, of the American Academy of Child & Adolescent Psychiatry, 53(10),
J., Yu, H., … Birmaher, B. (2015). Dialectical behavior therapy for adoles- 1082–1091. doi:10.1016/j.jaac.2014.07.003.
cents with bipolar disorder: Results from a pilot randomized trial. Journal of Miller, A. L., Rathus, J. H., & Linehan, M. M. (2017). Dialectical behavior ther-
Child and Adolescent Psychopharmacology, 25(2), 140–149. doi:10.1089/ apy with suicidal adolescents. New York: Guilford Press.
cap.2013.0145. National Collaborating Center for Mental Health (2009). Borderline personal-
Groschwitz, R. C., Plener, P. L., Kaess, M., Schumacher, T., Stoehr, R., & Boege, ity disorder: The NICE guideline on treatment and management. Leicester/
I. (2015). The situation of former adolescent self-injurers as young adults: London: The British Psychological Society & The Royal College of
A follow-up study. BMC Psychiatry, 15(1), 1–9. doi:10.1186/s12888- Psychiatrists; [updated July 2018. Available from: https://www.nice.org.uk/
015-0555-1. guidance/cg78.].
Hamza, C. A., Stewart, S. L., & Willoughby, T. (2012). Examining the link Nock, M. K., Joiner Jr T. E., Gordon, K. H., Lloyd-Richardson, E., & Prinstein,
between nonsuicidal self-injury and suicidal behavior: A review of the litera- M. J. (2006). Non-suicidal self-injury among adolescents: Diagnostic corre-
ture and an integrated model. Clinical Psychology Review, 32(6), 482–495. lates and relation to suicide attempts. Psychiatry Research, 144(1), 65–72.
doi:10.1016/j.cpr.2012.05.003. doi:10.1016/j.psychres.2006.05.010.
Hawton, K., Saunders, K. E., & O’Connor, R. C. (2012). Self-harm and suicide *Perepletchikova, F., Axelrod, S. R., Kaufman, J., Rounsaville, B. J.,
in adolescents. The Lancet, 379(9834), 2373–2382. doi:10.1016/S0140-6736 Douglas-Palumberi, H., & Miller, A. L. (2011). Adapting dialectical
(12)60322-5. behaviour therapy for children: Towards a new research agenda for paedi-
Higgins, J. P., Thompson, S. G., Deeks, J. J., & Altman, D. G. (2003). atric suicidal and non-suicidal self-injurious behaviours. Child and
Measuring inconsistency in meta-analyses. BMJ, 327(7414), 557–560. Adolescent Mental Health, 16(2), 116–121. doi:10.1111/j.1475-3588.2010.
doi:10.1136/bmj.327.7414.557. 00583.x.
*James, A. C., Taylor, A., Winmill, L., & Alfoadari, K. (2008). A preliminary Pigott, T. D. (2012). Advances in meta-analysis. New York, NY: Springer.
community study of dialectical behaviour therapy (DBT) with adolescent Plener, P. L., Schumacher, T. S., Munz, L. M., & Groschwitz, R. C. (2015).
females demonstrating persistent, deliberate self-harm (DSH). Child and The longitudinal course of non-suicidal self-injury and deliberate self-harm:
Adolescent Mental Health, 13(3), 148–152. doi:10.1111/j.1475-3588.2007. A systematic review of the literature. Borderline Personal Disorder and
00470.x. Emotion Dysregulation, 2, 2. doi:10.1186/s40479-014-0024-3.
*James, A. C., Winmill, L., Anderson, C., & Alfoadari, K. (2011). A preliminary *Rathus, J. H., & Miller, A. L. (2002). Dialectical behavior therapy adapted for
study of an extension of a community dialectic behaviour therapy (DBT) suicidal adolescents. Suicide and Life-Threatening Behavior, 32(2), 146–157.
programme to adolescents in the looked after care system. Child and Rathus, J. H., & Miller, A. L. (2015). DBT skills manual for adolescents.
Adolescent Mental Health, 16(1), 9–13. doi:10.1111/j.1475-3588.2010. New York: Guilford Publications.
00571.x. Ribeiro, J. D., Franklin, J. C., Fox, K. R., Bentley, K. H., Kleiman, E. M., Chang,
Kaplan, C., Tarlow, N., Stewart, J. G., Aguirre, B., Galen, G., & Auerbach, R. P. B. P., & Nock, M. K. (2016). Self-injurious thoughts and behaviors as risk
(2016). Borderline personality disorder in youth: The prospective impact of factors for future suicide ideation, attempts, and death: A meta-analysis

Downloaded from https://www.cambridge.org/core. IP address: 103.68.172.94, on 23 May 2021 at 20:43:49, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0033291721001355
 Psychological Medicine 11

of longitudinal studies. Psychological Medicine, 46(2), 225–236. doi:10.1017/ Viswanathan, M., Patnode, C. D., Berkman, N. D., Bass, E. B., Chang, S.,
S0033291715001804 Hartling, L., … Kane, R. L. (2018). Recommendations for assessing
*Santamarina-Perez, P., Mendez, I., Singh, M. K., Berk, M., Picado, M., Font, the risk of bias in systematic reviews of health-care interventions.
E., … Cosi, A. (2020). Adapted dialectical behavior therapy for adolescents Journal of Clinical Epidemiology, 97, 26–34. doi: 10.1016/j.jclinepi.2017.
with a high risk of suicide in a community clinic: A pragmatic randomized 12.004.
controlled trial. Suicide and Life-Threatening Behavior, 50(3), 652–667. *Woodberry, K. A., & Popenoe, E. J. (2008). Implementing dialectical behavior
10.1111/sltb.12612. therapy with adolescents and their families in a community outpatient
Sterne, J. A., & Egger, M. (2005). Regression methods to detect publication and clinic. Cognitive and Behavioral Practice, 15(3), 277–286. doi:10.1016/
other bias in meta-analysis. Publication bias in Meta-analysis: Prevention, j.cbpra.2007.08.004.
Assessment and Adjustments, 54(10), 99–110. doi:10.1002/0470870168.ch6. Wyman, P. A. (2014). Developmental approach to prevent adolescent suicides:
*Tebbett-Mock, A. A., Saito, E., McGee, M., Woloszyn, P., & Venuti, M. Research pathways to effective upstream preventive interventions. American
(2020). Efficacy of dialectical behavior therapy versus treatment as usual Journal of Preventive Medicine, 47(3), S251–S256. doi:10.1016/
for acute-care inpatient adolescents. Journal of the American Academy of j.amepre.2014.05.039.
Child & Adolescent Psychiatry, 59(1), 149–156. doi:10.1016/j.jaac.2019.01.020. Yeo, A. J., Germán, M., Wheeler, L. A., Camacho, K., Hirsch, E., & Miller, A.
Viechtbauer, W. (2010). Conducting meta-analyses in R with the metafor (2020). Self-harm and self-regulation in urban ethnic minority youth: A
package. Journal of Statistical Software, 36(3), 1–48. doi:10.18637/ pilot application of dialectical behavior therapy for adolescents. Child and
jss.v036.i03 Adolescent Mental Health, 25(3), 127–134. 10.1111/camh.12374.

Downloaded from https://www.cambridge.org/core. IP address: 103.68.172.94, on 23 May 2021 at 20:43:49, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0033291721001355