Paper Outline

O R I G I N A L A R T I C L E

Efficacy of a Problem-Solving Therapy for Depression and Suicide Potential in Adolescents and Young Adults

Mehmet Eskin Æ Kamil Ertekin Æ Hadiye Demir

Published online: 23 November 2007 � Springer Science+Business Media, LLC 2007

Abstract Short-term and structured cognitive behavioral problem-solving therapy (PST) is a developmentally relevant mode of action for the treatment of emotional problems in young

people. This study aimed at testing the efficacy of a problem-solving therapy in treating

depression and suicide potential in adolescents and young adults. A total of 46 self-referred

high school and university students who were randomly assigned to a problem-solving therapy

(n = 27) and a waiting list control (n = 19) conditions completed a controlled cognitive

behavioral problem-solving treatment trial. Participants were administered the measures of

depression, suicide potential, problem solving, self-esteem and assertiveness. Twenty-two of

the 27 participants from the PST condition could be reached after 12-months for follow-up.

Participants completed depression and problem-solving measures at follow-up. Results showed

that post-treatment depression and suicide risk scores of participants within the PST condition

decreased significantly compared to the pre-treatment scores but post-waiting and pre-waiting

depression and suicide risk scores of participants within the WLC condition were unchanged.

Likewise, post-treatment self-esteem and assertiveness scores of participants within the PST

condition increased significantly compared to the pre-treatment scores while post-waiting and

pre-waiting self-esteem and assertiveness scores of participants within the WLC condition

were unchanged. At post-treatment, 77.8% of the participants in the PST but only 15.8% of

those in the WLC condition achieved full or partial recovery according to BDI scores. Sim-

ilarly, 96.3% of participants in the PST but only 21.1% of those in the WLC condition

achieved full or partial recovery according to HDRS scores. The improvements were main-

tained at 12-months follow-up. Therefore, it is concluded that problem-solving therapy should

M. Eskin (&) Department of Psychiatry, School of Medicine, Adnan Menderes University, Aydin, Turkey e-mail: [email protected] URL: http://www.mehmeteskin.com; [email protected]

K. Ertekin NP Hospital, Istanbul, Turkey

H. Demir Middle East Technical University Health Center, Ankara, Turkey

123

Cogn Ther Res (2008) 32:227–245 DOI 10.1007/s10608-007-9172-8

be considered as a viable option for the treatment of depression and suicide potential in

adolescents and young adults.

Keywords Efficacy � Problem-solving therapy � Adolescent � Young adult � Depression � Suicide � Follow-up

Introduction

Adolescence is a developmental period characterized by rapid changes in physical, psycho-

logical and social functioning. Although many young people pass this developmental period

problem free, the changes render some of the adolescents vulnerable to psychological prob-

lems. Studies indicate that, compared to childhood, the frequencies of mental health problems

during adolescence increase (Kim 2003). Compared to the past the data have shown an

increase in emotional problems during adolescence (Collishaw et al. 2004).

One of the most common mental health problems in young people is depression (Hamrin

and Pachler 2005). Although lifetime prevalence rates of depression are less than 3% for

children, the rates of depression for adolescents rise to 14% (Lewinsohn et al. 1998). In a study

with 966 adolescents, Schichor et al. (1994) reported that 22% of the sample felt themselves

frequently depressed. Depression is associated with reduced psychosocial functioning.

Untreated depression in young people is a serious risk factor for mental health problems

(Steinhausen et al. 2006; Wilcox and Anthony 2004) and obesity in adulthood (Franko et al.

2005). Depression is mostly comorbid with anxiety disorders (Ferdinand et al. 2005), and is a

serious risk factor for suicidal behavior in young as well as in adults (Kish et al. 2005;

Thompson et al. 2005).

Self-esteem and assertive social-skills are the two possibly protective factors against

depression in young people. Adolescent psychiatric patients are in general characterized by

low self-esteem compared to controls (Guillon et al. 2003). In early adolescents, MacPhee and

Andrews (2006) found low self-esteem to be the strongest predictor of depression. Following a

group of adolescents over a 6 years period Pelkonen et al. (2003) found that baseline low self-

esteem scores were predictive of depression. Likewise, adolescents with identifiable levels of

psychopathology are characterized by low level of social skills such as inappropriate asser-

tiveness (Landazabal 2006). Chan (1993) with Chinese university undergraduates found

nonassertive responses to correlate with depression. Assertiveness in adolescence promotes the

establishment of socially supportive interpersonal relationships. In a cross-cultural study

involving Swedish and Turkish adolescents, Eskin (2003) showed that more assertive ado-

lescents in both groups reported having more friends and receiving more social support than

their less assertive peers.

A related mental health problem in young people is suicide. Though rare during childhood,

suicide is the leading cause of death in young people and hence constitutes a significant mental

health concern (Brener et al. 2000; Johnson et al. 2000). Not only suicidal deaths but also

nonfatal suicidal behaviors are common in adolescents and young adults. Empirical evidence

shows that lifetime prevalence of suicidal attempts is about 10% while the lifetime prevalence

of suicidal thoughts is approximately 30% in youth (Evans et al. 2005). In one study with

1,262 Turkish university students, Eskin et al. (2005) found that 42% of the sample reported

suicidal ideation during the past 12 months or lifetime, and 7% reported that they attempted to

kill themselves during their lifetime or in the past 12 months. Past suicidal thoughts and

228 Cogn Ther Res (2008) 32:227–245

123

attempts in young people are risk factors for future suicidality (Clark et al. 1989; Joiner et al.

2000; Lewinsohn et al. 1994; Rudd et al. 1996).

As with depression, self-esteem and assertive social skills are the two possibly protective

factors against suicidal behavior in young people. Gröholt et al. (2000), demonstrated that low

self-worth was a predictor of adolescent suicide attempts. Further, the work of Wild et al.

(2004) showed that low self-esteem in the family context was an independent predictor of

adolescent suicide ideation and attempts. In a similar, fashion suicidal young people were

characterized by inadequacies in assertive social skills. With a longitudinal research design

Sourander et al. (2001) showed that suicidal adolescents were characterized by low social

competence. Eskin (1995) found that deficiencies in positive assertion skills were related to

suicide risk scores of Swedish and Turkish high school students. Further et al. (1990) showed

that lack of assertiveness was related to suicidal intent in a group of suicide attempted ado-

lescent psychiatric inpatients.

There is an urgent need for effective treatment methods for depression and suicide potential

in young people. The literature suggests that treatment effectiveness for adolescent and young

adults is far less investigated than it is for adult populations. The use of Selective Serotonin

Reuptake Inhibitors (SSRI, the most widely used pharmacological agents for the treatment of

depression) in children and adolescents for treating depression include safety concerns. For

instance the use of SSRIs was found to be associated with increased risk for suicidal behavior

(Newman 2004; Richmond and Rosen 2005; Wohlfarth et al. 2006). However, the number of

depressed children and adolescents for whom psychotherapy/mental health counseling during

outpatient visits are prescribed decreased significantly over time (Ma et al. 2005). For Ryan

(2005) the debate about the best approach to treat child and adolescent depression continues.

Therefore, there is a need for the determination of effectiveness of short and structured

psychosocial interventions for child and adolescent depression and suicide potential.

A wide variety of research findings indicate deficits in problem-solving ability to have a key

role in the development and maintenance of depression and suicidal behavior. Deficits in

problem-solving ability were shown to be an important predisposing factor for the develop-

ment of depression (Nezu 1986; Nezu and Ronan 1988; Marx et al. 1992; Priester and Clum

1993) and suicidal behavior (Levenson and Neuringer 1971; Schotte and Clum 1987;

Sadowski and Kelley 1993; Reinecke et al. 2001; Chang 2002; Pollock and Williams 2004).

Recently, Speckens and Hawton (2005) reviewed the studies investigating the relationship

between problem solving and suicidal behavior in young people. They conclude that inef-

fective problem solving is an important vulnerability factor for suicidal behavior in youth.

Moreover, McAuliffe et al. (2002) found repeaters of parasuicide to exhibit more deficits in

problem solving than nonrepeaters.

Thus, one can presume that a problem-solving approach is an important intervention

strategy for the treatment of depression and suicide risk in young people. In line with this,

problem solving was identified as a prevention strategy for adolescent emotional problems

(Heppner et al. 1984; Spence et al. 2003). Empirical evidence suggests that problem-solving

therapy (PST) is an effective treatment for depression (Arean et al. 1993; Biggam and Power

2002; Dowrick et al. 2000; Nezu 1986) and suicidal problems (Lerner and Clum 1990; Sal-

kovskis et al. (1990; Townsend et al. 2001) in adults and to some extent young adults. Most

recently, Malouff et al. (2007) conducted a meta-analysis of 31 studies involving mostly adult

participants (n = 2,895) on the effectiveness of problem-solving therapy. They showed that

PST was significantly more effective than no treatment, treatment as usual and attention

placebo in reducing mental and physical health problems.

Adolescence and young adulthood are the life periods characterized by changes in bio-

logical, psychological and social domains. The young individual has to cope with these

Cogn Ther Res (2008) 32:227–245 229

123

multiple domain challenges/changes. Therefore, young people need effective coping strategies

or skills in order to gain resilience against multiple domain changes/challenges. Problem

solving is an important life skill for coping and tackling with life difficulties and challenges

during these stages of life. Thus, the overall objective of this study was to assess the efficacy of

a problem-solving therapy in treating depression and suicide proneness in adolescents and

young adults. A second objective was to evaluate whether or not PST leads to improvements in

self-appraised problem-solving ability. A third objective was to investigate the impact of PST

on protective factors such as self-esteem and assertiveness.

Methods

Design

A randomized, controlled trial was carried out to compare a problem-solving treatment for

major depression and suicide potential to a waiting list control condition in a group of high

school and university students.

Recruitment

Participants were recruited through announcements describing the symptoms of major

depression according to DSM-IV (APA 1994). The announcements were placed on the boards

of 10 high schools in the city of Aydin and a university campus area in Ankara.

Exclusion Criteria

Students who did not meet the DSM-IV criteria for major depression, students who were

currently under medical treatment, those who were psychotic and students with bipolar illness,

and students whose parents did not consent were excluded from the study.

Site of the Study

The part of the study involving high school students was conducted at the psychiatry

department of Adnan Menderes University School of Medicine in Aydin. The part involving

the university students was done at the Health Center of the Middle East Technical University

in Ankara.

Participants

A total of 54 high school and 28 university students (n = 82) responded to the announcements.

Assessment with the Structured Clinical Interview-Clinical Version (SCID-I/CV) (First et al.

1997) for DSM-IV Axis I diagnoses revealed that all university and 25 of the high school

students (n = 53) received a diagnosis for major depression. Twenty-nine students not

receiving a depression diagnosis were excluded. Of participants who received a depression

diagnosis, 27 (high school = 13; university = 14) were randomly assigned to a PST condition

and 26 (high school = 12; university = 14) were randomly assigned to the WLC condition.

One high school and eight university students who were assigned to the WLC condition

dropped out. At the end, all (high school = 13; university = 14) participants who were

230 Cogn Ther Res (2008) 32:227–245

123

assigned to the PST condition and 19 (high school = 11; university = 8) who were assigned to

the WLC condition (n = 46) completed the study. To see if attrition in the WLC condition

might have affected the findings, participants who dropped out of the WLC condition were

compared with those who stayed in the WLC condition on demographics and pre-treatment

scores and were found to be similar. Twenty-two participants (81.5%) from the PST condition

(high school = 10; university = 12) could be reached for follow-up after 12-months.

The demographic characteristics of participants are presented in Table 1. As the table

shows, a majority of the participants were of urban background, female and perceived their

family income as medium and had on average two siblings. Of participants who obtained a

diagnosis for major depression, two also obtained a diagnosis for social phobia, one for

posttraumatic stress disorder, one for dysthymia and another one obtained a diagnosis for

specific phobia.

Participants in the treatment and control conditions were found to be similar in terms of

group (high school vs. university), sex, background (urban vs. rural), perceived family income,

number of siblings, age, and paternal education, but mothers of participants in the problem-

solving condition had greater number of school years (mean = 10.1 years, SD = 4.3) than the

mothers of those in the control condition (mean = 6.6 years, SD = 5.1), t(44) = 2.5, P \ 0.05. Similarly, the two groups (high school and university) were similar in terms of sex,

background (urban vs. rural), condition, perceived family income, number of siblings, and

Table 1 Demographic characteristics of participants

Variables Group

Problem-solving therapy

Waiting list control Total

n % M SD n % M SD n % M SD

N 27 58.7 19 41.3 46 100

Sex

Male 7 25.9 7 36.8 14 30.4

Female 20 74.1 12 63.2 32 69.6

School

High school 13 48.1 11 57.9 24 52.2

University 14 51.9 8 42.1 22 47.8

Background

Urban 22 81.5 13 68.4 35 76.1

Rural 5 18.5 6 31.6 11 23.9

Perceived family income

Low 1 3.7 3 15.8 4 8.7

Medium 26 96.3 15 78.9 41 89.1

High 0 0.0 1 5.3 1 2.2

Number of siblings 1.7 1.8 2.3 2.6 2.0 2.1

Age 19.0 3.2 19.3 3.7 19.1 3.4

Maternal education (number of school years) 10.1 4.3 6.6 5.1 8.7 4.9

Paternal education (number of school years) 11.5 4.5 9.3 5.0 10.6 4.8

Cogn Ther Res (2008) 32:227–245 231

123

maternal education, but fathers of university group had greater number of school years

(mean = 12.8 years, SD = 3.5) than the fathers of high school group (mean = 8.5 years,

SD = 5.0), t(44) = 3.3, P \ 0.005. The mean age of the high school group was 16.4 years (SD = 1.1) and it was 22.1 years (SD = 2.3) for the university group.

Measures

Hamilton Depression Rating Scale

The 17-item clinician-administered Hamilton Depression Rating Scale (HDRS) was used to

assess severity of the depressive symptoms (Hamilton 1960). The validity and reliability of the

Turkish version of the HDRS was well established (Akdemir et al. 1996). The test–retest

reliability coefficient of the Turkish HDRS was 0.85 and the internal consistency reliability

coefficient was 0.75. The Turkish HDRS had a correlation coefficient of 0.48 with Beck

Depression Inventory. The total HDRS scores range from 0 to 52. Higher scores indicate

greater depression. Patients with HDRS scores of 7 or less were considered to have clinically

recovered; patients with scores of 8–12 were regarded as partially recovered; and patients with

scores of 13 or more were seen as not recovered (Frank et al. 1991).

Beck Depression Inventory

A 21-item Beck Depression Inventory (BDI 1978 version) (Beck et al. 1979) is used to assess

self-rated depressive symptoms. The BDI is the most widely used self-report measure of

depression. Hisli (1988) translated the BDI into Turkish and assessed its psychometric prop-

erties. The respondent rates the frequency and the intensity of symptoms on a four-point scale.

The Turkish BDI had an internal consistency reliability of 0.80 and a split-half reliability of

0.74. It had a correlation coefficient of 0.50 with the MMPI depression subscale. The total BDI

scores range from 0 to 63. Higher scores indicate greater depression. Patients with BDI scores

of 9 or less were considered to have clinically recovered; patients with scores of 10–15 were

regarded as partially recovered; and patients with scores of 16 or more were considered as not

recovered (Shaw et al. 1985).

Suicide Probability Scale

The Turkish version of the Suicide Probability Scale (SPS) (Cull and Gill 1988) was used to

assess suicide potential. The SPS is a 36-item self-report measure of suicide risk to be used

with adolescents and adults. Participants rate each item on a 4-point Likert scale according to

the frequency with which they experience a specific emotion or behavior by selecting scale

anchors ranging from ‘‘None or a little of the time’’ to ‘‘Most or all of the time.’’ It consists of

four empirically derived subscales based on current theories of suicidal behavior. The sub-

scales are: (1) Hopelessness; (2) Suicide ideation: (3) Negative self-evaluation; and (4)

Hostility. The SPS was previously translated into Turkish and was found to be a reliable and a

valid instrument by Eskin (1993). The Turkish SPS had a test–retest reliability coefficient of

0.95 over a 48.7 days period and it had an internal consistency coefficient of 0.89. It correlated

inversely with perceived social support from family -0.60 and perceived social support from

friends -0.75. The total SPS scores range from 36 to 144. Higher scores indicate greater

suicide proneness.

232 Cogn Ther Res (2008) 32:227–245

123

Problem-Solving Inventory

The Problem-Solving Inventory (PSI) developed by Heppner and Petersen (1982) was used to

measure self-appraised problem-solving ability. It has 32 items which the respondents rate for

the frequency of engaging in specific problem-solving behaviors on a 6-point Likert scale

ranging from ‘‘Always (1)’’ to ‘‘Never (6).’’ The PSI scores range from 32 to 192 and higher

scores indicate lower self-appraised problem-solving ability. The PSI was adapted into Turkish

by Şahin et al. (1993) and found to be a reliable and valid instrument. The internal consistency

coefficient for the Turkish PSI was 0.88 and the split-half reliability coefficient was 0.81. The

Turkish PSI had a correlation coefficient of 0.33 with BDI and it had a correlation coefficient

of 0.45 with the trait anxiety scores on the STAI (Şahin et al. 1993).

Scale for Interpersonal Behavior

Assertiveness was measured by Scale for Interpersonal Behavior (SIB) developed by Arrindell

and van der Ende (1985). The SIB is a 50-item multidimensional measure of assertiveness. The

respondent rates each item on a 5-point Likert scale for the frequency of engaging in a specific

assertive behavior. Forty-five of the 50 items are classified into four factorially derived cat-

egories of assertive behavior. The four SIB dimensions are: (1) Display of negative feelings;

(2) Expression of and dealing with personal limitations; (3) Initiating assertiveness; and (4)

Positive assertion. In addition to these subscales a total assertiveness score is computed by

summing the 50 items. Psychometric properties of the Turkish version of the SIB were

assessed and found to be highly reliable and valid (Eskin 1993). The test–retest reliability of

the SIB was 0.71 and its internal consistency reliability was 0.90. It had a correlation of 0.35

with perceived social support from friends. The SIB scores range from 50 to 250 with higher

scores indicating higher levels of assertiveness.

Rosenberg Self-Esteem Scale

The 10-item Rosenberg Self-Esteem Scale was used to measure self-esteem (Rosenberg 1965).

It is a global measure of self worth scored on a 4-point Likert scale. It has been adapted into

Turkish by Cuhadaroglu (1996). The scores range from 10 to 40, with higher scores repre-

senting higher self-esteem.

Therapeutic Alliance Scale

A 7-item short therapeutic alliance scale (TAS) designed by the first author was used to

measure the extent to which participants perceived being supported and understood by their

therapists. The TAS was designed to meet the specific requirements of the problem-solving

treatment approach used in this study. The items of this scale were as follows:

1. I feel that my therapist understands my problems.

2. I feel that there is a warm relationship based on mutual trust between my therapist and me.

3. I feel that my therapist has the right approach for the resolution of my problems.

4. I feel that my therapist and I are in agreement about what my problems and distresses are.

5. I feel that my therapist and I are in agreement about the causes of my problems.

6. I feel that my therapist accepts me the way I am.

7. I feel that my therapist works in cooperation with me during my treatment.

Cogn Ther Res (2008) 32:227–245 233

123

Participants rated these items on 5-point Likert scales for how much they agree with the

content of each item. Response alternatives ranged from ‘‘totally agree (5)’’ to ‘‘totally

disagree (1).’’ The responses to seven items are summed to obtain a therapeutic alliance score.

The scores range from 7, indicating minimum alliance between the therapist and the patient, to

35 indicating maximum alliance. The internal consistency coefficient (Cronbach’s apha) for

the TAS with 27 participants was 0.80.

Procedure

Announcements describing symptoms of major depression according to DSM-IV were placed

on the announcement boards of high schools and a university to recruit the participants.

Permissions from the local branch of the Ministry of Education and the governing body of the

university were obtained to place the announcements. The announcements asked students to

apply for the treatment project free of charge.

All participants and the parents of the high school students were asked to sign an informed

consent form. Then, they were given an appointment for a SCID-I interview. Participants

meeting the criteria for a major depression diagnosis were administered the Hamilton

Depression Rating Scale and were also asked to fill in a questionnaire that included the above-

mentioned measures as well as socio-demographics. Participants obtaining a diagnosis for

major depression were designated as the study group. Then they were assigned randomly to a

problem-solving therapy and a waiting list control conditions. For ethical reasons, the students

not obtaining a major depression diagnosis were offered 2–3 sessions of counseling. In a

similar fashion, participants who were assigned to the WLC condition were also offered six

sessions of PST after the waiting period. Participants in the PST condition filled in the

Therapeutic Alliance Scale (TAS) after the completion of the third session. Therapists

recorded the session lengths and problems identified during the first sessions.

Treatment

An individual PST treatment using a manual consisting of 6 sessions was used. The PST used

in this study was modeled according to the PST approach developed by D’Zurilla and

Goldfried (1971) and D’Zurilla and Nezu (1999). Like the one used in the primary care by

Gath and Mynors-Wallis (2000) the PST used in this study did not include a problem-orien-

tation component. Two graduate students in clinical psychology (the second and third authors)

received education and supervision on the PST from the first author. The manual developed by

the first author included a brief introduction to the problem-solving treatment of psychological

conditions. Then the manual described six sessions of PST that correspond to the six stages of

problem solving. The treatment lasted for 6 weeks with weekly scheduled sessions. The

sessions were in brief as follows:

Session 1: Definition of problems: The rationale behind the problem-solving treatment was explained to the participants. Then the therapist and the patient worked together to define a

problem. Emotional symptoms of the patients were identified. The manual instructed

therapists to relate emotional symptoms to problems and to define problems in behavioral

terms.

Session 2: Goal setting: In this session, the therapeutic task is to set goals. Goals are set in collaboration. The manual instructs that the goals should be attainable, objective and

234 Cogn Ther Res (2008) 32:227–245

123

realistic. It is also necessary to identify the strengths and resources of the patient at this

session.

Session 3: Generating alternative solutions: At this session the participant is encouraged to produce alternative solutions to the problem defined during the first session. ‘‘Brain-

storming’’ and ‘‘advice to a friend who has the same problem’’ methods are applied to

facilitate generating more alternative solutions. Therapist instructs the participant to be

nonjudgemental and produce as many solutions as possible.

Session 4: Decision making: The therapeutic task in this session is to choose the best solution to the problem. The best solution should be applicable by the patient and should

reach the goal set during the second session.

Session 5: Solution implementation: In this stage the solution should be implemented. In order to implement the chosen solution, the patient is taught necessary skills. To counteract

the skill deficit, the therapist may use techniques such as role-playing and the like to teach

the participant necessary skills for the implementation of the selected solution.

Session 6: Assessment and verification: In the last session, the solution implementation is assessed and verified.

Statistical Analyses

The data were analyzed by SPSS-9.0 for Windows. Repeated measures Analysis of Variance

(ANOVA) procedures were used to analyze the results, with condition and group as the

grouping factor, and time being the ‘within groups’ factor. A total of six conditions (PST and

WLC) by group (high school and university) by time (pre-/post-treatment/waiting) ANOVAs

were performed to compare the groups on five main outcome measures. Since condition by

time interaction effect tests the treatment efficacy, it is presented first in the results. Pre-

treatment/waiting and post-treatment/waiting means and standard deviations of measures

according to condition are presented in Table 2.

To further determine and quantify the efficacy of PST, controlled and uncontrolled (Feske

and Chambless 1995) effect sizes (Cohen’s d; Cohen 1988) were calculated. A controlled

effect size was calculated by subtracting the post-treatment mean of the treatment group from

the post-waiting mean of the control group divided by the standard deviation of the control

group. An uncontrolled effect size was calculated by subtracting the post-treatment mean of

Table 2 Means and standard deviations of outcome measures according to condition

Measures Problem-solving therapy Waiting list control

Pre-treatment Post-treatment Pre-waiting Post-waiting

M SD M SD M SD M SD

Depression

Beck Depression Inventory 26.7 9.4 10.7 10.4 28.0 9.0 22.0 5.5

Hamilton Depression Rating Scale 16.1 6.6 4.3 3.3 17.8 5.5 16.6 5.7

Suicide potential 84.1 15.1 71.9 15.5 77.9 16.4 75.2 14.9

Assertiveness 152.9 19.0 167.6 24.0 157.1 18.4 158.0 21.9

Problem solving 111.6 16.6 100.9 20.6 107.2 18.7 104.6 20.3

Self-esteem 23.8 4.6 28.0 5.2 23.3 4.8 23.4 5.1

Cogn Ther Res (2008) 32:227–245 235

123

the treatment group from its pre-treatment mean divided by the standard deviation of the

treatment group (Butler et al. 2006).

Nonparametric tests were used to compare the scores of groups involving fewer than thirty

participants. In this case one cannot assume normal distribution of data. The Mann–Whitney

U-test which uses median rather than mean was used to compare the scores of two independent groups. Scores of two related groups were compared by means of Wilcoxon Signed Ranks Test

procedure. Proportions of participants who fully and partially recovered and not recovered

were computed. Chi-square tests were used to test the association between two dichotomous

variables. Pearson product-moment correlation coefficients were calculated between the PSI

difference scores (differences between baseline and follow-up scores) and depression scores.

Results

Number and Length of Sessions

All participants in the PST condition received six sessions of problem-solving therapy. Session

lengths ranged from 30 to 60 min with an average session length of 37.6 min (SD = 6.9).

Session lengths were longer in the university (mean = 42.6 min, SD = 5.3) than in the high

school group (mean = 32.1 min, SD = 3.3), Z = 4.2, P \ 0.0001.

Therapeutic Alliance

Mean therapeutic alliance scale score was 30.9 (SD = 2.5). High school students scored sig-

nificantly higher (mean = 32.1, SD = 1.9) than the university students (mean = 29.8,

SD = 2.2) on the therapeutic alliance scale, Z = 2.8, P \ 0.01.

Depression

BDI

The means and standard deviations for depression, suicide potential, problem solving, self-

esteem and assertiveness are given in Table 2. The ANOVA produced a significant condition

by time interaction effect, F(1, 42) = 10.3, P \ 0.01. The ANOVA revealed also a main effect for time, F(1, 42) = 43.8, P \ 0.0001, a main effect for condition, F(1, 42) = 9.7, P \ 0.01, and a main effect for group, F(1, 42) = 20.3, P \ 0.0001.

Although post-treatment/waiting BDI scores were smaller than the pre-treatment/waiting

scores, baseline BDI scores of participants within the PST and WCL conditions were similar

(Z = 0.6, P [ 0.05) but post-treatment BDI scores of participants within the PST condition were lower than the post-waiting BDI scores of participants within the WLC condition

(Z = 5.3, P \ 0.0001). Concerning the main effect for time, baseline BDI scores were higher (mean = 27.0) than post-treatment/waiting scores (mean = 16.3). Regarding the main effect

for condition, mean BDI scores of students in the PST condition were lower (mean = 18.8)

than the scores of students in the WLC (mean = 24.4). The group main effect showed that BDI

scores of high school students were higher (mean = 25.7) than the university students’ scores

(mean = 17.6).

236 Cogn Ther Res (2008) 32:227–245

123

HDRS

The ANOVA gave a condition by time interaction effect, F(1, 42) = 37.7, P \ 0.0001. There was a main effect for time, F(1, 42) = 55.7, P \ 0.0001, a main effect for condition, F(1, 42) = 39.3, P \ 0.0001, and a main effect for group, F(1, 42) = 9.3, P \ 0.01.

Post-treatment HDRS scores were significantly lower than the pre-treatment scores within

the PST condition (Z = 4.5, P \ 0.0001) but pre- and post-waiting HDRS scores within the WLC condition were similar (Z = 0.4, P [ 0.05). Time main effect showed that the baseline HDRS scores were higher (mean = 17.2) than the post-treatment/waiting scores (mean =

10.2). Regarding the main effect for condition the HDRS scores of students in PST were

lower (mean = 10.2) than the scores of students in the WLC (mean = 17.8). The group main

effect showed that the HDRS scores of high school students were lower (mean = 12.1) than the

scores of university students (mean = 15.8).

An uncontrolled effect size from pre-treatment to post-treatment for HDRS was 2.4 and it

was 1.6 for BDI. A controlled effect size between PST and WLC conditions was 2.2 for HDRS

and it was 1.6 for BDI.

Suicide Potential

The ANOVA gave a condition by time interaction effect, F(1, 42) = 7.3, P \ 0.05, a main effect for time, F(1, 42) = 16.3, P \ 0.0001, and a main effect for group, F(1, 42) = 16.7, P \ 0.0001.

Post-treatment SPS scores were significantly lower than the pre-treatment SPS scores

within the PST condition (Z = 3.5, P \ 0.0001) but pre- and post-waiting SPS scores within the WLC condition were similar (Z = 1.4, P [ 0.05). The mean of the baseline SPS total scale scores was higher (mean = 80.4) than the mean of the post-treatment/waiting scores

(mean = 73.1). The mean of the SPS total scale scores of university students was lower

(mean = 69.1) than the mean scores of high school students (mean = 84.3).

An uncontrolled effect size from pre-treatment to post-treatment for SPS was 0.80. A

controlled effect size between PST and WLC conditions was 0.21.

Assertiveness

The ANOVA revealed a significant condition by time interaction effect, F(1, 42) = 7.5,

P \ 0.01, and a main effect for time, F(1, 42) = 10.0, P \ 0.01, on the SIB total scale scores. Post-treatment SIB scores were significantly higher than the pre-treatment SIB scores

within the PST condition (Z = 3.2, P \ 0.01) but pre- and post-waiting SIB scores were similar (Z = 0.5, P [ 0.05) within the WLC condition. Regarding the main effect for time, post-treatment/waiting SIB scores were higher (mean = 162.2) than pre-treatment/waiting SIB

scores (mean = 154.4).

An uncontrolled effect size from pre-treatment to post-treatment for SIB was 0.68. A

controlled effect size between PST and WLC conditions was 0.48.

Problem Solving

The ANOVA produced a nonsignificant condition by time interaction effect, F(1, 42) = 2.2,

P [ 0.05, but it gave a statistically significant main effect for time, F(1, 42) = 6.4, P \ 0.05,

Cogn Ther Res (2008) 32:227–245 237

123

on the PSI total scale scores. Pre-treatment/waiting PSI scores were higher (mean = 109.4)

than post-treatment/waiting PSI scores (mean = 102.5).

An uncontrolled effect size from pre-treatment to post-treatment for PSI was 0.58. A

controlled effect size between PST and WLC conditions was 0.19.

Self-Esteem

The ANOVA gave a statistically significant condition by time interaction effect, F(1,

42) = 7.1, P \ 0.05, and a main effect for time, F(1, 42) = 7.3, P \ 0.05. A marginally significant main effect for condition, F(1, 42) = 4.0, P \ 0.10, was also detected.

Post-treatment self-esteem scale scores were significantly higher than the pre-treatment

self-esteem scale scores within the PST condition (Z = 3.5, P \ 0.01) but pre- and post- waiting self-esteem scores were similar (Z = 0.2, P [ 0.05) within the WLC condition. The main effect for time showed that post-treatment/waiting self-esteem scale scores were higher

(mean = 25.6) than pre-treatment/waiting self-esteem scale scores (mean = 23.5). Marginally

significant main effect for condition showed that self-esteem scale scores of participants in the

PST condition tended to be higher (mean = 25.9) than the self-esteem scale scores of students

in the WLC condition (mean = 23.3).

An uncontrolled effect size from pre-treatment to post-treatment for self-esteem was 0.89.

A controlled effect size between PST and WLC conditions was 0.93.

Post-Treatment Recovery

BDI

Table 3 presents the recovery rates according to post-treatment/waiting BDI and HDRS scores

by condition. A chi-square test between condition and recovery categories indicated that

participants in the two conditions differed significantly from one another in relation to per-

centages of recovery according to BDI, v2 = 19.3, d.f. = 2, P \ 0.0001. As the table shows, 77.8% of participants within the PST condition achieved full or partial recovery but only

15.8% did so in the WLC condition.

Table 3 Recovery rates according to post-treatment/waiting BDI and HRSD scores by condition

Recovery Problem-solving therapy Waiting list control

n % n %

Beck Depression Inventory

Recovered (BDI, 0–9) 14 51.9 0 0.0

Partially recovered (BDI, 10–15) 7 25.9 3 15.8

Not recovered (BDI, 16 and above) 6 22.2 16 84.2

Total 27 100.0 19 100.0

Hamilton Depression Rating Scale

Recovered (HDRS, 0–7) 22 81.5 1 5.3

Partially recovered (HDRS, 8–12) 4 14.8 3 15.8

Not recovered (HDRS, 13 and above) 1 3.7 15 78.9

Total 27 100.0 19 100.0

238 Cogn Ther Res (2008) 32:227–245

123

HDRS

A chi-square test between condition and recovery showed that there was a significant asso-

ciation between the two variables, v2 = 31.1, d.f. = 2, P \ 0.0001. As it is seen in the table, 96.3% of the participants in the PST condition achieved full or partial recovery according to

post-treatment HDRS scores, while only 21.1% did so in the control condition.

Follow-up

Depression

Mean follow-up BDI scores were found to be 7.6 (SD = 7.3). Follow-up BDI scores were

statistically significantly lower than pre-treatment BDI scores, Z = 4.1, P \ 0.0001, but similar to post-treatment BDI scores, Z = 1.6, P [ 0.05. Mean follow-up BDI scores of uni- versity students were significantly lower (mean = 9.9, SD = 7.0) than the mean BDI scores of

high school students (mean = 5.6, SD = 7.4), Z = 2.0, P \ 0.05. Mean follow-up HDRS scores were 3.7 (SD = 1.9). Follow-up HDRS scores were statis-

tically significantly lower than pre-treatment HDRS scores, Z = 4.1, P \ 0.0001, but similar to post-treatment HDRS scores, Z = 0.1, P [ 0.05 (for pre- and post-treatment means see Table 2). Mean follow-up HDRS scores of university students were significantly lower

(mean = 2.6, SD = 1.5) than the mean follow-up HDRS scores of high school students

(mean = 4.9, SD = 1.7), Z = 2.6, P \ 0.05. According to predetermined criteria for remission, all the 22 patients (100%) achieved full

remission on the basis of follow-up HDRS scores (scores ranged from 1 to 7). Considering the

follow-up BDI scores, 17 participants (77.3%) achieved full remission, 2 (9.1%) achieved

partial remission, and 3 (13.6%) were still depressed.

Problem Solving

Mean follow-up PSI scores were 88.6 (SD = 15.6). Follow-up PSI scores were significantly

lower than pre-treatment, Z = 3.7, P \ 0.0001) and post-treatment PSI scores Z = 2.0, P \ 0.05 (for pre- and post-treatment means see Table 2). Follow-up PSI scores of university (mean = 90.4, SD = 8.2) and high school (mean = 86.1, SD = 18.7) students were similar,

Z = 1.1, P [ 0.05. The correlation coefficient between PSI difference scores and follow-up BDI scores was r = -0.41, n = 22, P \ 0.05 (one-tailed), and with HDRS it was r = -0.26, n = 22, P [ 0.05 (one-tailed).

Discussion

This study tested the efficacy of problem-solving therapy in treating depression and suicide

potential in adolescents and young adults. Forty-six self-referred high school and university

students who were randomly assigned to a problem-solving therapy and waiting list control

conditions participated in the study. A manual-based PST was used. Participants assigned to

the experimental condition received six sessions of PST for an average length of approxi-

mately 38 min per session. Participants in the problem-solving condition perceived their

therapeutic alliance as being highly satisfactory. The results obtained from the study indicate

Cogn Ther Res (2008) 32:227–245 239

123

that problem-solving therapy is an effective and acceptable treatment method for emotional

problems in adolescents and young adults.

Depression is common in young people and causes considerable impairment in physical,

psychological, social, academic and vocational functioning. The potential of a problem-solving

approach for the treatment of depression and suicide proneness was anticipated in the litera-

ture. In line with findings from Nezu (1986), the results obtained in this study supported this

anticipation. Post-treatment depression scores of participants who received PST were found to

be statistically significantly lower than their pre-treatment depression scores while pre- and

post-treatment depression scores of participants in the WLC condition were unchanged. High

effect sizes (Cohen 1988) were observed between treatment and no treatment conditions, and

between pre- and post-treatment for depression. The effect sizes observed in this study make a

strong case for the use of problem-solving therapy in treating depression in young people. The

results of this empirical investigation are in line with findings from Arean et al. (1993),

Dowrick et al. (2000), and Nezu (1986) with adults. Further, the results of this study replicate

and extend the findings from Lerner and Clum 1990 with young adults and Biggam and Power

(2002) with incarcerated young offenders to a sample that included high school students.

Predetermined criteria for remission according to BDI and HDRS scores indicated that the

number of participants in the PST condition who recovered were greater than the number of

participants who did so in the WLC condition. As Table 3 shows, 77.8% of participants in the

PST condition achieved full or partial recovery while only 15.8% achieved only partial

recovery in the WLC condition according to BDI scores. According to HDRS scores, 96.3% of

participants in the PST condition achieved full or partial recovery while only 21.1% did so in

the WLC condition. At follow-up, 86.4% were still in full or partial remission according to

BDI scores whereas 100.0% were in full remission according to HDRS scores. The findings

from the study showed that the improvements in depression were maintained over a 12-month

follow-up period.

Suicidal behavior is a major mental health concern among adolescents and young adults.

Therefore, effective psychosocial treatment alternatives for the treatment of suicide problem

are needed in these populations. The results obtained in the present study are encouraging.

Post-treatment suicide risk scores (as measured by the SPS) of participants who received PST

were found to be statistically significantly lower than their pre-treatment suicide potential

scores while pre- and post-waiting suicide risk scores of participants in the WLC condition

were unchanged. The effect size between treatment and no treatment obtained in the study was

low but the effect size between pre- and post-treatment suicide risk was high. The present

results are consistent with findings from Townsend et al. (2001) and Salkovskis et al. (1990)

with adults. Comparing problem-solving therapy with supportive therapy for treating suicidal

ideation of university students, Lerner and Clum (1990) found that the two treatments were

similar. But problem-solving therapy was more effective at reducing depression, hopelessness

and loneliness than supportive therapy. Unlike findings from Lerner and Clum (1990), current

results indicate that problem-solving therapy is effective at reducing suicide potential. Thus,

the results from this study suggest that problem-solving therapy can be taken as a viable

treatment alternative for suicide problems in adolescents and young adults.

Traditional psychotherapy outcome studies have usually been conducted to investigate the

efficacy of a given intervention in reducing the levels of psychopathology (i.e., depression,

etc.). Very few studies were designed to see if a given intervention works in increasing the

levels of protective factors such as self-esteem and assertiveness. These protective factors are

important in a developmental period like adolescence and young adulthood during which

global changes take place. Thus, the present study aimed also at testing if PST increases the

levels of self-esteem and assertiveness while decreasing the levels of psychopathology. The

240 Cogn Ther Res (2008) 32:227–245

123

data showed indeed that this is the case. As with scores of depression and suicide potential,

post-treatment self-esteem and assertiveness scores of participants in the PST condition were

found to be significantly higher than their pre-treatment self-esteem and assertiveness scores

while pre- and post-waiting self-esteem and assertiveness scores of participants in the WLC

condition were unchanged. The PST leading to increases in assertiveness is not surprising

because, according to therapist records, the problems identified to work with during the

treatment were mainly of interpersonal nature.

Kazdin and Nock (2003) discussed the importance of mechanisms of therapeutic change in

child and adolescent psychotherapy. Problem-solving therapy offers unique mechanism of

therapeutic change that fit the developmental needs of young people. One might have antic-

ipated an improvement in the post-treatment self-appraised problem-solving ability of

participants who received problem-solving therapy compared to self-appraised problem-

solving ability of participants in the WLC condition. Unlike findings from Nezu (1986) and

Biggam and Power (2002), the current results did not confirm this anticipation. However,

improvement in self-appraised problem-solving ability from baseline to follow-up was asso-

ciated with lower depression (e.g., Dixon 2000) suggesting a causal relationship between

problem-solving ability and depression. Although, post-treatment PSI scores of participants in

the PST condition did not improve compared to the pre-treatment scores, follow-up PSI scores

were statistically significantly lower than the pre-treatment scores. But this is valid only for

participants within the PST condition. Lack of support for the above anticipation may be due to

several reasons. First, PST may reduce psychopathology through other mechanisms of change

than the development of problem-solving ability. Second, PST applied in this study may not be

able to significantly improve global problem-solving ability as measured by the PSI. PSI scores

reflect problem orientation variables (i.e., attitudes toward problems and one’s own problem-

solving ability) as well as problem-solving skills (e.g., problem definition, generation of

alternative solutions). However, the present PST program focused only on problem-solving

skills. It did not address problem orientation. Nevertheless, following treatment, an

improvement in problem-solving performance that may have resulted from the PST program

may have eventually resulted in an improvement in problem orientation (see D’Zurilla and

Nezu 1999, 2007), which may be reflected in the significant improvement in the PSI scores of

the PST participants from post-treatment to follow-up. Third, the measure of problem solving

used in this study may not be an adequate instrument to show the development of problem-

solving skills. Fourth, since participants in the control condition were placed on a waiting list

there might have been a placebo effect.

Taken together, the findings from this study make a strong case for the use of problem-

solving therapy for the treatment of depression and suicide potential in adolescents and young

adults. The findings demonstrated that PST was able to reduce levels of psychopathology

(depression and suicide potential) and to increase levels of protective factors (assertiveness and

self-esteem). The improvements in depression were maintained over a 12-month follow-up

period. However, the present findings should be approached with caution when generalizing to

other samples for several reasons. First, therapists themselves administered the HDRS to

participants. Since they were not blind to the conditions this may have caused bias. Second, the

participants were a group of self-referred high school and university students. Therefore,

further investigations are needed to clarify to what extent our findings apply to clinically

referred adolescents and young adults. One outcome study on PST for depression in adults

found that a PST program that included a problem orientation was significantly more effective

than a PST program that focused only on problem-solving skills (Nezu and Perri 1989). It is

possible that a similar effect might be found with PST for depression and suicide potential in

adolescents and young adults. Third, a problem-orientation component was not part of the PST

Cogn Ther Res (2008) 32:227–245 241

123

offered in the study. Studies employing PST with a problem-orientation component are

therefore warranted.

Acknowledgements The authors would like to offer their deep appreciation to Dr. Rick E. Ingram (Editor) and the two anonymous reviewers for their valuable comments and suggestions.

References

Akdemir, A., Örsel, S., Dağ, I., Türkçapar, M. H., Işcan, N., & Özbay, H. (1996). Hamilton depresyon derecelendirme ölçeği (HDDÖ)’nin geçerliği, güvenirliği ve klinikte kullanımı [Validity, reliability and the clinical use of Hamilton Depression Rating Scale, in Turkish]. Journal of Psychiatry, Psychology and Psychopharmacology, 4, 251–259.

American Psychiatric Association (1994). Diagnostic and statistical manual of mental disorders (DSM-IV). Washington, DC: American Psychiatric Association.

Arean, P. A., Perri, M. G., Nezu, A. M., Schein, R. L., Christopher, F., & Joseph, T. X. (1993). Comparative effectiveness of social problem-solving therapy and reminiscence therapy as treatment for depression in older adults. Journal of Consulting and Clinical Psychology, 61, 1003–1010.

Arrindell, W. A., & van der Ende, J. (1985). Cross-sample invariance of the structure self-reported distress and difficulty in assertiveness. Advances in Behavior Research and Therapy, 7, 205–243.

Beck, A. T., Rush, A. J., Shaw, B. F., & Emery, G. (1979). Cognitive therapy of depression. New York: Guilford Press.

Biggam, F. H., & Power, K. G. (2002). A controlled, problem-solving, group-based intervention with vulnerable incarcerated young offenders. International Journal of Offender Therapy and Comparative Criminology, 46, 678–698.

Brener, N. D., Krug, E. G., & Simon, T. R. (2000). Trends in suicide ideation and suicidal behavior among high school students in the United States, 1991–1997. Suicide and Life-Threatening Behavior, 30, 304–312.

Brent, D. A., Kolko, D. J., Allan, M. J., & Brown R. V. (1990). Suicidality in affectively disordered adolescent inpatients. Journal for the American Academy of Child and Adolescent Psychiatry 586–593.

Butler, A. C., Chapman, J. E., Forman, E. M., & Beck, A. T. (2006). The empirical status of cognitive-behavioral therapy: A review of meta-analyses. Clinical Psychology, Review, 26, 17–31.

Chan, D. W. (1993). Components of assertiveness: Their relationships with assertive rights and depressed mood among Chinese college students in Hong Kong. Behavior Research and Therapy, 31, 529–238.

Chang, E. C. (2002). Predicting suicide ideation in an adolescent population: Examining the role of social problem solving as a moderator and a mediator. Personality and Individual Differences, 32, 1279–1291.

Clark, D. C., Gibbons, R. D., Fawcett, J., & Scheftner, W. A. (1989). What is the mechanism by which suicide attempts predispose to later suicide attempts? A mathemathical model. Journal of Abnormal Psychology, 98, 42–49.

Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Erlbaum. Collishaw, S., Maughan, B., Goodman, R., & Pickles, A. (2004). Time trends in adolescent mental health.

Journal of Child Psychology and Psychiatry, 45, 1350–1362. Cuhadaroglu, F. (1986). Üniversite öğrencilerinde psikiyatrik semptom dağılımı [Symptom distribution in

university students, in Turkish]. In R. Saygılı (Ed.), Proceedings of the XXII national congress of psychiatric and neurological sciences (pp. 35–39) _Izmir: Ege University Press.

Cull, J. G., & Gill, W. S. (1988). Suicide Probability Scale (SPS) manual. Los Angels: Western Psychological Services.

Dixon, W. A. (2000). Problem-solving appraisal and depression: Evidence for a recovery model. Journal of Counseling & Development, 78, 87–91.

Dowrick, C., Dunn, G., Ayuso-Mateos, J. L., Dalgard, O. S., Page, H., Lehtinen, V., et al. (2000). Problem solving treatment and group psychoeducation for depression: Multicentre randomized controlled trial. British Medical Journal, 321, 1450–1454.

D’Zurilla, T. J., & Goldfried, M. R. (1971). Problem solving and behavior modification. Journal of Abnormal Psychology, 78, 107–126.

D’Zurilla, T. J., & Nezu, A. M. (1999). Problem-solving therapy: A social competence approach to clinical intervention (2nd ed.). New York: Springer.

D’Zurilla, T. J., & Nezu, A. M. (2007). Problem-solving therapy: A positive approach to clinical intervention (3rd ed.). New York: Springer.

242 Cogn Ther Res (2008) 32:227–245

123

Eskin, M. (1993). Reliability of the Turkish version of the perceived social support from friends and family scales, scale for interpersonal behavior, and the suicide probability scale. Journal of Clinical Psychology, 49, 515–522.

Eskin, M. (1995). Suicidal behavior as related to social support and assertiveness among Swedish and Turkish high school students: A cross-cultural investigation. Journal of Clinical Psychology, 51, 158–172.

Eskin, M. (2003). Self-reported assertiveness in Swedish and Turkish adolescents: A cross-cultural comparison. Scandinavian Journal of Psychology, 44, 7–12.

Eskin, M., Kaynak-Demir, H., & Demir, S. (2005). Same-sex sexual orientation, childhood sexual abuse, and suicidal behavior in university students in Turkey. Archives of Sexual Behavior, 34, 185–195.

Evans, E., Hawton, K., Rodham, K., & Deeks, J. (2005). The prevalence of suicidal phenomenon in adolescents: A systematic review of population-based studies. Suicide and Life-Threatening Behavior, 35, 239–250.

Ferdinand, R. F., Nijs, P. F. A., Lier, P., & Verhults, F. C. (2005). Latent class analysis of anxiety and depressive symptoms in referred adolescents. Journal of Affective Disorders, 88, 299–306.

Feske, U., & Chambless, D. L. (1995). Cognitive versus exposure only treatment for social phobia: A meta analysis. Behavior Therapy, 26, 695–720.

First, M. B., Spitzer, R. L., Gibbon, M., & Williams, J. B. W. (1997). Structured Clinical Interview for DSM-IV Axis I Disorders (SCID-I), clinical version [Turkish translation by A. Çorapçıoğlu]. Ankara: Hekimler Yayın Birliği.

Frank, E., Prien, R. F., Jarrett, R. B., Keller, M. B., Kupfer, D. J., Lavori, P. W., et al. (1991). Conceptualization and rationale for consensus definitions of terms in major depressive disorder. Remission, recovery, relapse, and recurrence. Archives of General Psychiatry, 48, 851–855.

Franko, D. L., Striegel-Moore, R. H., Thompson, D., Schreiber, G. B., & Daniels, S. R. (2005). Does adolescent depression predict obesity in black and white young adult women? Psychological Medicine, 35, 1505– 1513.

Gath, D., & Mynors-Wallis, L. (2000). Problem-solving treatment in primary care. In D. M. Clark & C. G. Fairburn (Eds.), Science and practice of cognitive behaviour therapy (pp. 415–431). Oxford: Oxford University Press.

Gröholt, B., Ekeberg, Ö., Wichström, L., & Haldorsen, T. (2000). Young suicide attempters: A comparison between a clinical and an epidemiological sample. Journal of the American Academy of Child and Adolescent Psychiatry, 39, 868–875.

Guillon, M. S., Crocq, M. A., & Bailey, P. E. (2003). The relationship between self-esteem and psychiatric disorders in adolescents. European Psychiatry 59–62.

Hamilton, H. (1960). A rating scale for depression. Journal of Neurological and Neurosurgical Psychiatry, 33, 56–62.

Hamrin, V., & Pachler, M. C. (2005). Child and adolescent depression: Review of the latest evidence-based treatments. Journal of Psychosocial Nursing & Mental Health Services, 43, 54–63.

Heppner, P. P., Neal, G. W., & Larson, L. M. (1984). Problem-solving training as prevention with college students. The Personnel and Guidance Journal, 62, 514–519.

Heppner, P. P., & Petersen, C. H. (1982). The development and implications of a personal problem solving inventory. Journal of Counseling Psychology, 29, 66–75.

Hisli, N. (1988). Beck Depresyon Envanteri’nin geçerliği üzerine bir çalışma [A study on the validity of Beck Depression Inventory, in Turkish]. Turkish Journal of Psychology, 7, 3–13.

Johnson, G. R., Krug, E. G., & Potter, L. B. (2000). Suicide among adolescents and young adults: A cross- national comparison of 34 countries. Suicide and Life-Threatening Behavior, 30, 74–82.

Joiner, T. E., Rudd, M. D., Rouleau, M. R., & Wagner, K. D. (2000). Parameters of suicidal crises vary as a function of previous suicide attempts in youth inpatients. Journal of the American Academy of Child and Adolescent Psychiatry, 39, 876–880.

Kazdin, A. E., & Nock, M. K. (2003). Delineating mechanisms of change in child and adolescent therapy: Methodological issues and research recommendations. Journal of Child Psychology and Psychiatry, 44, 1116–1129.

Kim, H. Y. (2003). Correlation of mental health problems with psychological constructs in adolescence: Final results from a 2-year study. International Journal of Nursing Studies, 40, 115–124.

Kisch, J., Leino, V., & Silverman, M. M. (2005). Aspects of suicidal behavior, depression, and treatment in college students: Results from the Spring 2000 National College Health Assessment Survey. Suicide and Life-Threatening Behavior, 35, 3–13.

Landazabal, M. G. (2006). Psychopathological symptoms, social skills, and personality traits: A study with adolescents. Spanish Journal Psychology, 9, 182–192.

Lerner, M. S., & Clum, G. A. (1990). Treatment of suicide ideators: A problem-solving approach. Behavior Therapy, 21, 403–411.

Cogn Ther Res (2008) 32:227–245 243

123

Levenson, M., & Neuringer, C. (1971). Problem-solving behavior in suicidal adolescents. Journal of Consulting and Clinical Psychology, 37, 433–436.

Lewinsohn, P. M., Rohde, P., & Seeley J. R. (1994). Psychosocial risk factors for future adolescent suicide attempts. Journal of Consulting and Clinical Psychology, 62, 297–305.

Lewinsohn, P. M., Rohde, P., & Seeley J. R. (1998). Major depressive disorders in older adolescents: Prevalence, risk factors, and clinical implications. Clinical Psychology Review, 18, 765–794.

Ma, J., Lee, K. V., & Stafford, R. S. (2005). Depression treatment during outpatient visits by U.S. children and adolescents. Journal of Adolescent Health, 37, 434–442.

MacPhee, A. R., & Andrews, J. J. (2006). Risk factors for depression in early adolescence. Adolescence, 41, 435–466.

Malouff, J. M., Thorsteinsson, E. B., & Schutte, N. S. (2007). The efficacy of problem-solving therapy in reducing mental and physical health problems: A meta-analysis. Clinical Psychology Review, 27, 46–57.

Marx, E. M., Williams, J. M. G., & Claridge, G. C. (1992). Depression and social problem solving. Journal of Abnormal Psychology, 101, 78–86.

McAuliffe, C., Keeley, H. S., & Corcoran, P. (2002). Problem solving and repetition of parasuicide. Behavioural and Cognitive Psychotherapy, 30, 385–397.

Newman, T. B. (2004). A black-box warning for antidepressants in children? New England Journal of Medicine, 35, 1595–1598.

Nezu, A. M. (1986). Efficacy of a social problem-solving therapy approach for unipolar depression. Journal of Consulting and Clinical Psychology, 54, 196–202.

Nezu, A. M., & Perri, M. G. (1989). Social problem-solving therapy for unipolar depression: An initial dismantling investigation. Journal of Consulting and Clinical Psychology, 57, 408–413.

Nezu, A. M., & Ronan, G. F. (1988). Social problem solving as a moderator of stress-related depressive symptoms: A prospective analysis. Journal of Counseling Psychology, 35, 134–138.

Pelkonen, M., Marttunen, M., & Aro, H. (2003). Risk for depression: A 6-year follow-up of Finnish adolescents. Journal of Affective Disorders, 77, 41–51.

Pollock, L. R., & Williams, J. M. G. (2004). Problem-solving in suicide attempters. Psychological Medicine, 34, 163–167.

Priester, M. J., & Clum, G. A. (1993). Perceived problem-solving ability as a predictor of depression, hopelessness, and suicide ideation in a college population. Journal of Counseling Psychology, 40, 79–85.

Reinecke, M. A., DuBois, D. L., & Schultz, T. M. (2001). Social problem solving, mood, and suicidality among inpatient adolescents. Cognitive Therapy and Research, 25, 743–756.

Richmond, T. K., & Rosen, D. S. (2005). The treatment of adolescent depression in the era of the black box warning. Current Opinion in Pediatrics, 17, 466–472.

Rosenberg, M. (1965). Society and the adolescent self-image. New Jersey: Princeton University Press. Rudd, M. D., Joiner, T., & Rajab, M. H. (1996). Relationship among suicide ideators, attempts, and multiple

attempters in a young-adult sample. Journal of Abnormal Psychology, 105, 541–550. Ryan, N. (2005). Treatment of depression in children and adolescents. The Lancet, 366, 933–940. Sadowski, C., & Kelley, M. L. (1993). Social problem solving in suicidal adolescents. Journal of Consulting and

Clinical Psychology, 61, 121–127. Şahin, N., Şahin, N. H., & Heppner, P. P. (1993). Psychometric properties of the problem solving inventory in a

group of Turkish university students. Cognitive Therapy and Research, 17, 379–396. Salkovskis, P. M., Atha, C., & Storer, D. (1990). Cognitive-behavioural problem solving in the treatment

of patients who repeatedly attempt suicide: A controlled trial. British Journal of Psychiatry, 157, 871–876.

Schichor, A., Bernstein, B., & King, S. (1994). Self-reported depressive symptoms in inner-city adolescents seeking routine health care. Adolescence, 29, 379–388.

Schotte, D. E., & Clum, G. A. (1987). Problem-solving skills in suicidal psychiatric patients. Journal of Consulting and Clinical Psychology, 55, 49–54.

Shaw, B., Vallis, M., & McCabe, S. (1985). The assessment of the severity and symptom patterns in depression. In E. Beckham & W. Leber (Eds.), Handbook of Depression. Homewood, IL: Dorsey Press.

Sourander, A., Helstelä, L., Haavisto, A., & Bergroth, L. (2001). Suicidal thoughts and attempts among adolescents: A longitudinal 8-year follow-up study. Journal of Affective Disorders, 63, 59–66.

Speckens, A. E. M., & Hawton, K. (2005). Social problem solving in adolescents with suicidal behavior: A systematic review. Suicide and Life-Threatening Behavior, 35, 365–387.

Spence, S. H., Sheffield, J. K., & Donovan, C. L. (2003). Preventing adolescent depression: An evaluation of the problem solving for life program. Journal of Consulting and Clinical Psychology, 71, 3–13.

Steinhausen, H. C., Haslimeier, C., & Winkler-Metzke, C. (2006). The outcome of episodic versus persistent adolescent depression in young adulthood. Journal of Affective Disorders, 96, 49–57.

244 Cogn Ther Res (2008) 32:227–245

123

Thompson, E. A., Mazza, J. J., Herting, J. R., Randell, B. P., & Eggert, L. L. (2005). The mediating roles of anxiety, depression, and hopelessness on adolescent suicidal behaviors. Suicide and Life-Threatening Behavior, 35, 14–34.

Townsend, E., Hawton, K., Altman, D. G., Arensman, E., Gunnell, D., Hazell, P., et al. (2001). The efficacy of problem-solving treatments after deliberate self-harm: Meta-analysis of randomized controlled trials with respect to depression, hopelessness and improvements in problems. Psychological Medicine, 31, 979–988.

Wilcox, H. C., & Anthony, J. C. (2004). Child and adolescent clinical features as forerunners of adult-onset major depressive disorder: Retrospective evidence from an epidemiological sample. Journal of Affective Disorders, 82, 9–20.

Wild, L. G., Flisher, A. J., & Lombard, C. (2004). Suicidal ideation and attempts in adolescents: Associations with depression and six domains of self-esteem. Journal of Adolescence, 27, 611–624.

Wohlfarth, T. D., Zwieten, B. J., Lekkerkerker, F. J., Wied, C. C. G., Ruis, J. R., Elferink, A. J. A., et al. (2006). Antidepressants use in children and adolescents and the risk of suicide. European Neuropsychopharma- cology, 16, 79–83.

Cogn Ther Res (2008) 32:227–245 245

123