Evidence-Based Psychosocial Treatments for Childand Adolescent Depression

Corinne David-Ferdon and Nadine J. Kaslow

Emory University School of Medicine

The evidence-base of psychosocial treatment outcome studies for depressed youth con-ducted since 1998 is examined. All studies for depressed children meet Nathan andGorman’s (2002) criteria for Type 2 studies whereas the adolescent protocols meetcriteria for both Type 1 and Type 2 studies. Based on the Task Force on the Promotionand Dissemination of Psychological Procedures guidelines, the cognitive-behavioraltherapy (CBT) based specific programs of Penn Prevention Program, Self-Control Ther-apy, and Coping with Depression-Adolescent are probably efficacious. InterpersonalTherapy–Adolescent, which falls under the theoretical category of interpersonal therapy(IPT), also is a probably efficacious treatment. CBT provided through the modalities ofchild group only and child group plus parent components are well-establishedintervention approaches for depressed children. For adolescents, two modalities arewell-established (CBT adolescent only group, IPT individual), and three are probablyefficacious (CBT adolescent group plus parent component, CBT individual, CBTindividual plus parent=family component). From the broad theoretical level, CBT haswell-established efficacy and behavior therapy meets criteria for a probably efficaciousintervention for childhood depression. For adolescent depression, both CBT and IPThave well-established efficacy. Future research directions and best practices are offered.

In 1998, Kaslow and Thompson (Kaslow & Thompson,1998) used the Task Force on Promotion and Dissemi-nation of Psychological Procedures criteria (Chamblesset al., 1996) to examine the psychosocial interventionsfor child and adolescent depression. That review con-cluded that no programs met criteria for well-establishedtreatments. Two approaches met criteria for probablyefficacious interventions: (a) school-based work of Starkand colleagues regarding Self-Control Therapy forchildren with elevated depressive symptoms (Stark,Reynolds, & Kaslow, 1987; Stark, Rouse, & Livingston,1991), which falls under the theoretical rubric of cog-nitive-behavioral therapy (CBT); and (b) AdolescentsCoping with Depression (CWD-A) program, another

CBT approach, conducted by Lewinsohn, Clarke, and col-leagues with high school students with major depressive dis-order (MDD), with dysthymic disorder (DD; Lewinsohn,Clarke, Hops, & Andrews, 1990; Lewinsohn, Clarke,Rohde, Hops, & Seeley, 1996), or who were ‘‘at risk’’ dueto elevated depressive symptom scores (Clarke et al., 1995).

This article first reviews the psychosocial interventionoutcome studies for depressed children and adolescentsconducted since the 1998 Kaslow and Thompson article(Kaslow & Thompson, 1998). Remaining consistentwith the previous review and to underscore the needfor developmentally sensitive interventions tailored fordifferent age groups, the review is divided between stu-dies conducted with children versus adolescents. Thesestudies are analyzed in accord with the guidelines setforth by Nathan and Gorman (2002) in A Guide toTreatments that Work, which delineates criteria for sixtypes of studies ranging from the most methodologicallyrigorous clinical trials (Type 1 studies) to reports withmarginal value (Type 6 studies). Specifically, Type 1 stu-dies refer to double-blind, randomized controlled

Corinne David-Ferdon is currently at the Centers for Disease

Control and Prevention, National Center for Injury Prevention and

Control.

Correspondence should be addressed to Corinne David-Ferdon,

Centers for Disease Control and Prevention, National Center for

Injury Prevention and Control, 4770 Buford Highway NE, MS-F64,

Atlanta, GA 30341. E-mail: cferdon@cdc.gov

Journal of Clinical Child & Adolescent Psychology, 37(1), 62–104, 2008

Copyright # Taylor & Francis Group, LLC

ISSN: 1537-4416 print=1537-4424 online

DOI: 10.1080/15374410701817865

prospective clinical trials that have a clear delineation ofsample characteristics, state-of-the-art assessment anddiagnostic procedures, adequate sample size, compari-son groups, and appropriate data analytic procedures.Type 2 studies are clinical trials that lack some compo-nents of a Type 1 study. Type 3 studies are those withpilot-data and case-control methodologies. Type 4 stu-dies use sophisticated data analytic techniques (e.g.,meta-analysis) to conduct secondary data analysis. Type5 studies are reviews, without secondary data analysis,that overview the literature and give opinions. Type 6studies are case studies, essays, and opinion papers.

Then, the currently reviewed studies, as well as thosefocused on in the Kaslow and Thompson (1998) articleand those investigations inadvertently not included inthe prior review (King & Kirschenbaum, 1990; Vostanis,Feehan, Grattan, & Bickerton, 1996a; Wood, Harrington,& Moore, 1996), are examined for the extent to whicheach conforms to the Task Force on the Promotionand Dissemination of Psychological Procedures criteria(Chambless et al., 1998; Chambless & Hollon, 1998;Chambless et al., 1996; Lonigan, Elbert, & Johnson,1998) for well-established and probably efficaciousinterventions. The Task Force criteria have been debatedand evolved over time, such as the criterion for well-established interventions to be superior to a placebo oralternative treatment rather than no-treatment controlconditions and the criterion for a treatment manual tobe utilized. For this review, the more stringent criteriaare used and are summarized next.

The following are the Task Force’s criteria for evi-dence-based treatments used in this review. Becausethe depression literature does not include a large seriesof single case design experiments, only criteria relatedto between-group design studies are delineated. For anintervention to be deemed well-established, there mustbe at least two well-conducted, between-group designexperiments demonstrating efficacy in one of the followingways: (a) superior to pill or psychological placebo or toanother treatment, or (b) equivalent to an already-estab-lished treatment in experiments with adequate samplesizes. The experiments must be conducted in accordancewith a treatment manual, sample characteristics must bedetailed, and at least two different investigators or investi-gatory teams must demonstrate intervention effects. Foran intervention to be classified probably efficacious, either(a) two experiments must demonstrate that the inter-vention is more effective than a no-treatment controlgroup (e.g., waitlist condition) in improving functioning,or (b) the studies meet all criteria for a well-establishedtreatment except for the requirement that treatment effectsare shown by two different research teams. Probably effi-cacious treatment also must be conducted in accordancewith a treatment manual and have sample characteristicsclearly specified. In this review, we deem interventions as

experimental if they are promising interventions thatdemonstrate significant treatment effects but havereceived limited evaluation through only one rando-mized controlled trial.

After reviewing the findings, which includes a presen-tation of effect sizes, suggestions for future researchdirections and recommendations for best practice areoffered. The concluding section underscores the impor-tance of devising more developmentally, gender, andethnoculturally sensitive interventions that incorporatea broad array of theoretical orientations and treatmentmodalities and that target the unique needs of eachdepressed youth. The importance of conducting studiesthat compare active interventions is noted. Attentionalso is paid to mediators, moderators, and predictorsof treatment outcome and the ways in which these fac-tors may inform treatment outcome.

STUDY INCLUSION AND EXCLUSIONCRITERIA

The review of treatments that work includes only thoseempirical examinations that used a randomized con-trolled trial design and that were published since1998 (Tables 1 and 2). Studies were identified througha comprehensive search of online data bases (e.g.,PsycINFO, Medline) and examination of the referencelists of all located studies. Key search terms includedadolescent depression, youth depression, teen depression,bereaved youth, suicidal youth, depression intervention,and depression treatment. The review separates investi-gations with children (age 12 and younger) and adoles-cents (13 and older). Studies that included childrenand adolescents were categorized based on the age groupof the majority of the participants. The final evaluationand summary of the evidence-based status of the psycho-social interventions include all treatment outcomeresearch, regardless of publication date, and is done sep-arately for children and adolescents (Tables 3 and 4).

For both the examination of treatments that work andthe review of well-established and probably efficaciousinterventions, studies are included if the samples of chil-dren or adolescents met diagnostic criteria for MDD orDD. We also included intervention studies that targeted‘‘at-risk’’ youth (e.g., youth with a depressed parent,bereaved youth, school-referred youth) who had elevatedlevels of depressive symptoms on self-report measuresand the primary outcome of the studies was a reductionin depressive symptoms. The rationale for includingthese youth is that we were considering interventions thattargeted not only clinical cases of depression, but alsodepressive symptoms. Typically, youth who met diag-nostic criteria for a mood disorder were recruited fromclinic settings, and youth with elevated symptom scores

TREATMENTS FOR YOUTH DEPRESSION 63

were recruited from schools. Characteristics of the targetsamples and intervention settings are noted when infor-mation is available. We did not include studies whenyouth were deemed at-risk because of socioculturalfactors (Cardemil, Reivich, Beevers, Seligman, &James, 2007; Cardemil, Reivich, & Seligman, 2002).

A number of projects that demonstrated efficacy inreducing depressive symptoms were excluded becausethey did not use a randomized controlled trial design(Kovacs et al., 2006). In addition, the focus of thisreview is on interventions with the primary purpose toreduce depressive symptoms, so prevention studieswhere the youth were not identified as having depressivesymptoms were excluded (Beardslee, Gladstone,Wright, & Cooper, 2003; Chaplin et al., 2006; Merry,McDowell, Wild, Bir, & Cunliffe, 2004; Pattison &Lynd-Stevenson, 2001; Spence, Sheffield, & Donovan,2003). We also excluded programs targeting youth ofdepressed parents when the youth were not identifiedas depressed or having depressive symptoms. Further,we excluded interventions that targeted suicidal youth,all students regardless of presence of depressive symp-toms or risk factors (i.e., universal prevention pro-grams), or other child=adolescent problems in whichdepression was secondary, except when such an inter-vention was compared to a treatment protocol targetingdepressed youth (Shochet & Ham, 2004). We did notreview studies examining electroconvulsive therapy(American Academy of Child and Adolescent Psy-chiatry, 2004). Finally, pharmacological studies arenot reviewed except when a medication interventionwas compared to and=or paired with a psychosocialtreatment (Treatment for Adolescents with DepressionStudy [TADS] Team, 2004).

TREATMENTS THAT WORK (1998–PRESENT)

Child Studies

Table 1 presents the following information on eachstudy conducted with depressed children: investigators,sample characteristics, assessment tools and source ofinformation, therapists, description of interventionsand settings, results, effect sizes, and follow-up data. Itshould be noted that the effect sizes reported only reflectthe significant differences that emerged in the study. Insome instances there were significant between-group dif-ferences and in other cases there significant within groupdifferences, and the tables clarified where these signifi-cant differences were found. The effect sizes are eitherthose reported by the authors or were calculated usingmeans, standard deviations, and samples sizes. Eachstudy is rated using the criteria for treatments that work(Nathan & Gorman, 2002).

As seen in Table 1, there are seven recently publishedintervention studies with children with elevated depress-ive symptoms (Asarnow, Scott, & Mintz, 2002; DeCuyper, Timbremont, Braet, De Backer, & Wullaert,2004; Gillham, Hamilton, Freres, Patton, & Gallop,2006; Gillham, Reivich et al., 2006; Pfeffer, Jiang,Kakuma, Hwang, & Metsch, 2002; Roberts, Kane,Thomson, Bishop, & Hart, 2003; Yu & Seligman,2002) and three with children who met criteria for adepressive disorder (Muratori, Picchi, Bruni, Patarnello,& Romagnoli, 2003; Nelson, Barnard, & Cain, 2003;Trowell et al., 2007) in accord with the Diagnostic andStatistical Manual (4th ed.; DSM-IV; American Psychi-atric Association, 1994) for a total of 10 new studiessince 1998. These studies included relatively similar pro-portions of males and females in their samples, and chil-dren ranged in age between 6 and 15 years and weredrawn from schools and outpatient mental health set-tings. In general, limited information was provided withregards to the ethnicity of the samples, but for those inwhich such data were available, most of the childrenwere Caucasian, and African American youth wereunderrepresented. There were three unique populationsstudied, namely, Chinese, Dutch, and rural Australianchildren. A broad range of measures was used. The Chil-dren’s Depression Inventory (CDI; Kovacs, 1992) wasthe most often used measure of self-reported depressivesymptoms, whereas the Schedule for Affective Disordersand Schizophrenia for Children (K-SADS; Chamberset al., 1985) was the diagnostic tool most frequentlyused. It is interesting that parents’ perceptions of theirchildren’s functioning were collected in only four stu-dies, with three studies assessing children’s externalizingand internalizing functioning with the Child BehaviorChecklist (CBCL; Achenbach & Edelbrock, 1991) andthe other assessing children’s social adjustment withthe Social Adjustment Inventory for Children and Ado-lescents (John, Gammon, Prusoff, & Warner, 1987).

All studies included one experimental condition and acontrol condition (e.g., waitlist, community services).Seven of the 10 studies included a parent component,which included separate parent education and supportsessions or therapy sessions with the parent and childdyad. One study that included a family therapy con-dition (Trowell et al., 2007) did not specify the scopeof family members included in this treatment. Thus, itappears that none of the studies incorporated standardfamily therapy with all family members present. Inter-ventions were conducted in individual, group, andchild–parent formats, typically for 8 to 16 sessions. In4 studies, little to no information about the therapistswas provided. In the other 6 studies, clinicians variedby discipline (e.g., psychology, psychiatry, nursing,teaching) and level of experience (e.g., bachelor’s levelto graduate students to licensed clinicians).

64 DAVID-FERDON AND KASLOW

TA

BLE

1

Random

ized

Child

hood

Depre

ssio

nT

reatm

ent

Outc

om

eS

tudie

sS

ince

1998

Inve

stig

ato

rsS

am

ple

Sourc

e-M

easu

res

Ther

apis

tT

reatm

ent

Condit

ions

Res

ult

sF

oll

ow

-Up

Nath

an

&

Gorm

an

(2002)

Cri

teri

a

Asa

rno

w,

Sco

tt,

&

Min

tz(2

002)

23

4th

to6th

gra

der

s

(65%

fem

ale

;57%

Cau

casi

an

,17%

His

pan

ic,

13%

Asi

an

,

13%

Afr

ican

Am

eric

an

)w

ith

elev

ate

dd

epre

ssiv

e

sym

pto

ms

Sel

f:A

TQ

,C

DI,

Sel

f-

Rep

ort

Co

pin

gS

cale

Cli

nic

al

psy

cho

logy

gra

du

ate

stu

den

ts

10-s

essi

on

sch

oo

lb

ase

d

Str

ess-

Bu

ster

sC

BT

gro

up

an

dp

are

nt

edu

cati

on

inte

rven

tio

n

over

5w

eek

s,w

ait

list

con

tro

l

Ch

ild

ren

intr

eatm

ent

gro

up

sho

wed

sign

ific

an

tly

gre

ate

rre

du

ctio

ns

in

dep

ress

ive

sym

pto

ms

(wh

eno

utl

ier

was

rem

oved

),n

egati

ve

au

tom

ati

cth

ou

gh

ts,

an

d

inte

rnali

zed

cop

ing

than

con

tro

lacc

ord

ing

to

self

-rep

ort

.E

S:

CD

I¼

.92

No

foll

ow

-up

data

wer

ere

po

rted

.

Typ

e2

Pfe

ffer

,Ji

an

g,

Kak

um

a,

Hw

an

g,

&M

etsc

h

(2002)

75

6-

to15-y

ear-

old

s(5

2

fam

ilie

s)‘‘

at-

risk

’’b

ase

d

on

suff

erin

gth

esu

icid

e

of

ap

are

nt

or

sib

lin

g;

no

eth

nic

ity

data

wer

e

pre

sen

ted

Sel

f:C

DI,

RC

MA

S,

SA

ICA

Pare

nt:

BD

I,S

AIC

A

Cli

nic

ian

:C

PT

SR

I,

K-S

AD

S

No

tsp

ecif

ied

10

90-m

inco

mm

un

ity-

clin

icb

ase

dB

GI

wit

h

psy

cho

edu

cati

on

al

an

d

sup

po

rtiv

eco

mp

on

ents

an

dse

para

tep

are

nt

gro

up

,u

sual

care

an

d

bim

on

thly

ph

on

eca

lls

rece

ived

by

con

tro

l

gro

up

Ch

ild

ren

inB

GI

had

sign

ific

an

tly

gre

ate

r

red

uct

ion

sin

dep

ress

ive

an

dan

xie

tysy

mp

tom

s

than

con

tro

lsacc

ord

ing

tose

lf-r

epo

rt.

Th

ere

was

no

imp

act

on

chil

dre

n’s

PT

SD

sym

pto

ms,

chil

dre

n’s

soci

al

ad

just

men

t,o

rp

are

nt

dep

ress

ion

.E

S:

CD

I¼

.7,

RC

MA

S¼

.9

No

foll

ow

-up

data

wer

ere

po

rted

.

Typ

e2

Yu

&S

elig

man

(2002)

220

8-

to15-y

ear-

old

s(5

5%

male

;100%

Ch

ines

e)

‘‘at-

risk

’’b

ase

do

n

elev

ate

dd

epre

ssiv

e

sym

pto

ms

an

dfa

mil

y

con

flic

t

Sel

f:C

DI,

Co

nfl

ict

sub

scale

so

fF

ES

,C

AS

Q

Tea

cher

sw

ith

40

hr

of

inte

rven

tio

ntr

ain

ing

an

dw

eek

lysu

per

vis

ion

10

120-m

inw

eek

lysc

ho

ol

base

dP

OP

gro

up

,

no

nin

terv

enti

on

con

tro

l

gro

up

.P

OP

iscu

ltu

rall

y

mo

dif

ied

ver

sio

no

fth

e

Pen

nP

reven

tio

n

Pro

gra

m(P

PP

).

Ch

ild

ren

inP

OP

gro

up

sho

wed

sign

ific

an

tly

gre

ate

rd

ecre

ase

sin

dep

ress

ive

sym

pto

ms

an

din

crea

ses

in

op

tim

isti

cex

pla

nato

ry

style

than

con

tro

ls

acc

ord

ing

tose

lf-r

epo

rt.

Mo

reo

pti

mis

tic

exp

lan

ato

ryst

yle

med

iate

dth

ep

reven

tio

n

of

dep

ress

ive

sym

pto

ms.

At

po

stte

st,

41%

of

PO

P

gro

up

an

d46%

of

con

tro

lat

aC

DI

cuto

ff

sco

reo

f15.

ES

:

CD

I¼

.25,

CA

SQ¼

.54

3-

an

d6-m

on

th

foll

ow

-up

;P

OP

gro

up

had

sign

ific

an

t

few

erd

epre

ssiv

e

sym

pto

ms

an

dm

ore

op

tim

isti

cex

pla

nato

ry

style

sth

an

con

tro

ls

acc

ord

ing

tose

lf-r

epo

rt.

ES

3m

on

th:

CD

I¼

.33,

CA

SQ¼

.36.

ES

6m

on

th:

CD

I¼

.39,

CA

SQ¼

.41

Typ

e2

(Co

nti

nu

ed)

65

TA

BLE

1

Continued

Inve

stig

ato

rsS

am

ple

Sourc

e-M

easu

res

Ther

apis

tT

reatm

ent

Condit

ions

Res

ult

sF

oll

ow

-Up

Nath

an

&

Gorm

an

(2002)

Cri

teri

a

Mu

rato

ri,

Pic

chi,

Bru

ni,

Pata

rnel

lo,

&

Ro

magn

oli

(2003)

58

6-

to10-y

ear-

old

s(6

0%

male

)w

ho

met

DS

M-I

V

crit

eria

for

dep

ress

ion

or

an

xie

tyd

iso

rder

;n

o

eth

nic

ity

data

wer

e

pre

sen

ted

Pare

nt:

CB

CL

Cli

nic

ian

:

C-G

AS

,K

-SA

DS

Psy

cho

ther

ap

ists

wit

h

deg

ree

inp

sych

olo

gy

or

chil

dan

d

ad

ole

scen

t

psy

chia

try

11-w

eek

PP

wit

hp

are

nt

an

dch

ild

con

du

cted

ina

ho

spit

al

clin

ic,

usu

al

care

com

mu

nit

y

serv

ices

con

dit

ion

Ch

ild

ren

inb

oth

gro

up

s

sho

wed

sign

ific

an

t

imp

rovem

ents

ino

ver

all

fun

ctio

nin

gan

d

dec

rease

sin

inte

rnali

zin

gacc

ord

ing

top

are

nt

an

dcl

inic

ian

rati

ng

at

6m

on

ths

po

stb

ase

lin

e.P

Pan

d

com

mu

nit

yse

rvic

e

gro

up

sd

idn

ot

sign

ific

an

tly

dif

fer

on

an

ym

easu

reb

etw

een

pre

-an

d

po

stin

terv

enti

on

.

2-y

ear

foll

ow

-up

;P

Ph

ad

sign

ific

an

tly

imp

roved

over

all

fun

ctio

nin

gan

d

less

inte

rnali

zin

gan

d

exte

rnali

zin

gsy

mp

tom

s

than

con

tro

lacc

ord

ing

top

are

nt

an

dcl

inic

ian

rep

ort

.E

S:

CB

CL

-to

tal:

.72;

CB

CL

-ext:

.59;

CB

CL

-in

t¼

.61;

C-G

AS¼

.72

Typ

e2

Nel

son

,B

arn

ard

,&

Cain

(2003)

28

8-

to14-y

ear-

old

s(7

1%

male

;71.4%

Cau

casi

an

,

21.4%

His

pan

ic,

7.2%

Afr

ican

Am

eric

an

)w

ho

met

DS

M-I

Vcr

iter

iafo

r

dep

ress

ion

Sel

f:C

DI

Cli

nic

ian

:K

-SA

DS

-P

No

tsp

ecif

ied

8-w

eek

,p

are

nt–

chil

dfa

ce-

to-f

ace

CB

Tin

com

mu

nit

ycl

inic

,

pare

nt–

chil

d

vid

eoco

nfe

ren

cin

gC

BT

Ch

ild

ren

inth

eb

oth

gro

up

s

sho

wed

dec

rease

sin

dep

ress

ive

sym

pto

ms.

Ch

ild

ren

inth

e

vid

eoco

nfe

ren

cin

ggro

up

sho

wed

sign

ific

an

tly

gre

ate

rre

du

ctio

nin

dep

ress

ion

than

face

-to

-

face

CB

Tb

ase

do

nse

lf-

rep

ort

.A

cro

ssgro

up

s,

82%

rem

issi

on

rate

.E

S:

CD

I¼

.56

No

foll

ow

-up

data

wer

e

rep

ort

ed.

Typ

e2

Ro

ber

ts,

Kan

e,T

ho

mso

n,

Bis

ho

p,

&H

art

(2003)

189

11-

to13-y

ear-

old

7th

gra

der

s(5

0%

fem

ale

;

100%

rura

lA

ust

rali

an

)

‘‘at-

risk

’’b

ase

do

n

elev

ate

dd

epre

ssiv

e

sym

pto

ms

Sel

f:C

DI,

CA

SQ

,R

CM

AS

Pare

nt:

CB

CL

Sch

oo

lp

sych

olo

gis

tsan

d

nu

rses

wit

hb

ach

elo

r-

level

beh

avio

ral

scie

nce

deg

rees

,p

sych

olo

gis

t

had

mast

er’s

deg

ree

12-s

essi

on

sch

oo

l-b

ase

d

CB

Tgro

up

inte

rven

tio

n

fro

mth

eP

PP

,u

sual

care

con

tro

lgro

up

wit

h

mo

nit

ori

ng

of

sym

pto

ms

plu

sh

ealt

hcu

rric

ulu

m

Ch

ild

ren

inb

oth

gro

up

s

sho

wed

dec

rease

sin

dep

ress

ive

sym

pto

ms.

Acc

ord

ing

tose

lf-r

epo

rt,

gro

up

sd

idn

ot

dif

fer

on

dep

ress

ion

,b

ut

CB

T

gro

up

had

less

an

xie

ty.

CB

Tgro

up

had

sign

ific

an

tly

few

er

inte

rnali

zin

gp

rob

lem

s

than

con

tro

lgro

up

base

do

np

are

nt

rep

ort

.

ES

:R

CM

AS¼

.26,

CB

CL¼

.28

6-m

on

thfo

llo

w-u

p;

CB

T

gro

up

had

less

an

xie

ty

than

con

tro

lo

nse

lf-

rep

ort

,b

ut

gro

up

sd

id

no

td

iffe

ro

no

ther

self

-

an

dp

are

nt

mea

sure

s.

ES

:R

CM

AS¼

.24

Typ

e2

66

De

Cu

yp

er,

Tim

bre

mo

nt,

Bra

et,

De

Back

er,

&

Wu

llaer

t(2

004)

20

9-

to11-y

ear-

old

s(7

5%

fem

ale

;100%

Du

tch

)

‘‘at-

risk

’’b

ase

do

n

elev

ate

dd

epre

ssiv

e

sym

pto

ms

Sel

f:C

DI,

SP

PC

,S

TA

IC

Pare

nt:

CB

CL

Cli

nic

ian

:C

AS

Th

erap

ists

of

un

kn

ow

n

train

ing

statu

sw

ho

wer

e

sup

ervis

edb

ytr

ain

ed

CB

Tth

erap

ist

16

60-m

inw

eek

lygro

up

sess

ion

so

fT

ak

ing

Act

ion

at

un

iver

sity

clin

ic,

on

ein

div

idu

al

pare

nt

sess

ion

,b

oo

ster

sess

ion

at

1an

d4

mo

nth

for

po

stin

terv

enti

on

,

wait

list

con

tro

l

Ch

ild

ren

inb

oth

gro

up

s

sho

wed

sign

ific

an

t

imp

rovem

ents

inse

lf-

rep

ort

edd

epre

ssio

nan

d

glo

bal-

self

wo

rth

bu

tn

o

sign

ific

an

tb

etw

een

-

gro

up

dif

fere

nce

s.N

o

sign

ific

an

tw

ith

ino

r

bet

wee

n-g

rou

p

dif

fere

nce

so

nse

lf-

rep

ort

edan

xie

tyo

r

pare

nt

rep

ort

s.

4-

an

d12-m

on

thfo

llo

w-u

p.

At

4-m

on

thtr

eatm

ent

effe

cts

wer

em

ain

tain

ed

exce

pt

con

tro

lgro

up

no

lon

ger

sho

wed

effe

cto

n

self

-rep

ort

edd

epre

ssio

n.

At

12-m

on

thsa

mp

le

incl

ud

edw

ait

list

con

tro

l

wh

ow

ere

pro

vid

ed

trea

tmen

t;si

gn

ific

an

t

imp

rovem

ents

on

self

-

rep

ort

edd

epre

ssio

n,

an

xie

ty,

self

-wo

rth

an

d

on

pare

nt

rep

ort

s

bet

wee

np

retr

eatm

ent

an

dfo

llo

w-u

p.

Typ

e2

Gil

lham

,H

am

ilto

n,

(2006)

271

11-

to12-y

ear-

old

s

(53%

fem

ale

;73%

Cau

casi

an

)‘‘

at-

risk

’’

base

do

nel

evate

d

dep

ress

ive

sym

pto

ms

Sel

f:C

AS

Q,

CD

I

Cli

nic

ian

:D

ICA

-R,

K-S

AD

S,

Dia

gn

ost

ic

info

rmati

on

fro

mH

MO

data

base

Ch

ild

psy

cho

logis

tan

d

chil

dso

cial

wo

rker

wit

h

21–24

yea

rso

f

exp

erie

nce

an

dw

ho

wer

e

sup

ervis

edb

yo

ne

of

the

pro

gra

md

evel

op

ers

12

90-m

inC

BT

gro

up

inte

rven

tio

nse

ssio

ns

fro

mth

eP

RP

in

pri

mary

care

sett

ing;

usu

al

care

con

tro

lgro

up

No

sign

ific

an

tb

etw

een

-

gro

up

dif

fere

nce

sin

the

red

uct

ion

ind

epre

ssiv

e

sym

pto

ms

or

imp

act

on

dep

ress

ive

dis

ord

ers.

6-,

12-,

18-,

an

d24-m

on

th

foll

ow

-up

;P

RP

sho

wed

sign

ific

an

t

imp

rovem

ents

in

exp

lan

ato

ryst

yle

for

po

siti

ve

even

ts.

At

12-

mo

nth

,gir

lsin

PR

P

sho

wed

sign

ific

an

tly

red

uce

dd

epre

ssiv

e

sym

pto

ms.

Over

cou

rse

of

foll

ow

-up

.R

PR

sign

ific

an

tly

pre

ven

ted

dep

ress

ion

,an

xie

ty,

or

ad

just

men

td

iso

rder

s

com

bin

edam

on

gh

igh

sym

pto

mch

ild

ren

.E

S:

CA

SQ¼

.28,

CD

I

(gir

ls)¼

.31

Typ

e2

Gil

lham

,R

eivic

het

al.

(2006)

44

6th

to7th

gra

der

s(7

0%

male

;91%

Cau

casi

an

)

‘‘at-

risk

’’b

ase

do

n

elev

ate

dd

epre

ssiv

ean

d

an

xie

tysy

mp

tom

s

Sel

f:C

DI,

RC

MA

S

Pare

nt:

BA

I,B

DI

Bach

elo

rd

egre

esin

psy

cho

logy

train

edan

d

sup

ervis

edb

yp

rogra

m

dev

elo

per

;p

sych

olo

gis

t

wit

hd

oct

ora

tew

ho

was

ap

rogra

md

evel

op

er

890-m

inw

eek

lyC

BT

gro

up

inte

rven

tio

nse

ssio

ns

fro

mth

eP

RP

plu

s6

90-

min

pare

nt

gro

up

sess

ion

sin

sch

oo

l

sett

ing;

usu

al

care

con

tro

lgro

up

No

sign

ific

an

tb

etw

een

-

gro

up

dif

fere

nce

sin

the

red

uct

ion

ind

epre

ssiv

e

or

an

xie

tysy

mp

tom

s.

6-

an

d12-m

on

thfo

llo

w-u

p;

PR

Pp

lus

pare

nt

had

sign

ific

an

tfe

wer

dep

ress

ive

or

an

xie

ty

sym

pto

ms

at

bo

th

foll

ow

-up

s.E

S:

CD

I

(6m

)¼

.63,

CD

I

(12

m)¼

.46,

RC

MA

S

(6m

)¼

.63,

RC

MA

S

(12

m)¼

.81

Typ

e2

(Co

nti

nu

ed)

67

TA

BLE

1

Continued

Inve

stig

ato

rsS

am

ple

Sourc

e-M

easu

res

Ther

apis

tT

reatm

ent

Condit

ions

Res

ult

sF

oll

ow

-Up

Nath

an

&

Gorm

an

(2002)

Cri

teri

a

Tro

wel

let

al.

(2007)

72

9-

to15-y

ear-

old

s(6

2%

male

;87%

Cau

casi

an

)

wh

om

etD

SM

-IV

crit

eria

for

Majo

r

Dep

ress

ion

or

Dyst

hym

ia

Sel

f:C

DI,

MF

Q

Cli

nic

ian

:K

-SA

DS

,C

-GA

S

Ind

ivid

ual

an

dfa

mil

y

ther

ap

ists

train

edb

y

pro

gra

md

evel

op

ers

8–14

90-m

inS

IFT

sess

ion

s;

16–30

FIP

Pse

ssio

ns

plu

sIn

div

idu

al

Pare

nt

sess

ion

s;in

com

mu

nit

y

an

dh

osp

ital

clin

ics

Ch

ild

ren

inb

oth

gro

up

s

sho

wed

sign

ific

an

t

imp

rovem

ents

on

self

-

rep

ort

edm

easu

res

an

d

clin

icia

nra

ted

fun

ctio

nin

g.

75.7%

of

SIF

Tan

d74.3%

of

FIP

Pn

olo

nger

clin

icall

y

dep

ress

ed.

SIF

T

sign

ific

an

tly

bet

ter

imp

rovem

ent

on

self

-

rep

ort

edm

easu

res,

bu

t

no

gro

up

dif

fere

nce

so

n

clin

icia

nm

easu

res.

ES

:

CD

I¼

.51,

MF

Q¼

.29

6-m

on

thfo

llo

w-u

p;

81%

of

SIF

Tan

d100%

of

FIP

P

no

lon

ger

clin

icall

y

dep

ress

ed.

Bo

thgro

up

s

con

tin

ued

imp

rovem

ents

on

all

mea

sure

s.N

o

sign

ific

an

tgro

up

dif

fere

nce

so

nan

y

mea

sure

s.

Typ

e2

No

te:

Eff

ect

size

(ES

)is

Co

hen

’sd;

AT

Q¼

Au

tom

ati

cT

ho

ug

hts

Qu

esti

on

na

ire–

Rev

ised

;B

AI¼

Bec

kA

nx

iety

Inve

nto

ry;

BD

I¼

Bec

kD

epre

ssio

nIn

ven

tory

;B

GI¼

Ber

eav

emen

tG

rou

p

Inte

rven

tio

n;

CA

S¼

Ch

ild

Ass

essm

ent

Sch

edu

le;

CA

SQ¼

Ch

ild

ren

’sA

ttri

bu

tio

nal

Sty

leQ

ues

tio

nn

air

e;C

BC

L¼

Ch

ild

Beh

av

ior

Ch

eck

list

;C

BT¼

Co

gn

itiv

eB

ehav

iora

lT

her

ap

y;

CD

I¼

Ch

ild

ren

’sD

epre

ssio

nIn

ven

tory

;C

-GA

S¼

Ch

ild

ren

’sG

lob

alA

sses

smen

tS

cale

;C

PT

SR

I¼

Ch

ild

ho

od

Po

sttr

au

mat

icS

tres

sR

eact

ion

Ind

ex;

DIC

A–R¼

Dia

gn

ost

icIn

terv

iew

for

Ch

ild

ren

an

dA

do

lesc

ents

;D

SM

–IV¼

Dia

gnost

icand

Sta

tist

ical

Manual

of

Men

tal

Dis

ord

ers

(4th

ed.)

;ex

t¼

exte

rnal

izin

g;

FE

S¼

Fam

ily

En

vir

on

men

tS

cale

;F

IPP¼

Fo

cuse

dIn

div

idu

al

Psy

cho

-

dy

na

mic

Psy

cho

ther

apy

;in

t¼

inte

rnal

izin

g;

K-S

AD

S-P¼

Sch

edu

lefo

rA

ffec

tive

Dis

ord

ers

an

dS

chiz

op

hre

nia

for

Ch

ild

ren

–P

rese

nt

Ep

iso

de;

MF

Q¼

Mo

od

sa

nd

Fee

lin

gs

Qu

esti

on

na

ire;

PO

P¼

Pen

nO

pti

mis

mP

rogr

am

;P

P¼

Psy

cho

dy

na

mic

Psy

cho

ther

ap

y;

PR

P¼

Pen

nR

esil

ien

cyP

rogr

am

;P

TS

D¼

Po

sttr

au

mat

icS

tres

sD

iso

rder

;R

CM

AS¼

Rev

ised

Ch

ild

ren

’sM

an

ifes

t

An

xiet

yS

cale

;S

AIC

A¼

So

cial

Ad

just

men

tIn

ven

tory

for

Ch

ild

ren

an

dA

do

lesc

ents

;S

IFT¼

Sy

stem

sIn

teg

rati

ve

Fa

mil

yT

her

ap

y;S

PP

C¼

Sel

f-P

erce

pti

on

Pro

file

for

Ch

ild

ren

;S

TA

IC¼

Sta

te-T

rait

An

xiet

yIn

ven

tory

for

Ch

ild

ren

.

68

Most researchers focused on some form of CBT:problem solving, social skills training, attributionretraining, and self-control techniques (Asarnow et al.,2002; Gillham, Reivich et al., 2006; Nelson et al.,2003; Roberts et al., 2003; Yu & Seligman, 2002). How-ever, two studies examined a psychodynamic intervention(Muratori et al., 2003; Trowell et al., 2007), one studyexamined a family system intervention (Trowell et al.,2007), and another program utilized a combination ofpsychoeducation and support (Pfeffer et al., 2002).

Whereas in the past most intervention protocols weredownward extensions of approaches designed for adults,and thus were not guided by a developmental psycho-pathology framework, some of the more recent studieswere developmentally informed. For example, the studyby Pfeffer and colleagues (2002) for depressed, suicidallybereaved children utilized attachment theory (Bowlby,1980) as a guide. The psychodynamic program(Muratori et al., 2003) was influenced by a protocoloriginally developed for infants and toddlers (Cramer& Palacio Espasa, 1993). Further, the CBT was mademore developmentally appropriate by adding compo-nents related to generalization to key environmentalcontexts (Asarnow et al., 2002) and by including parentsand encouraging them to help with the learning and gen-eralization of the skills (Asarnow et al., 2002; Gillham,Reivich et al., 2006; Nelson et al., 2003).

Despite increased attention to developmental issues,attention to cultural and=or ethnic factors in the designand implementation of the interventions remained rare.Some protocols did incorporate assessment tools spe-cific to their population (e.g., Dutch; De Cuyper et al.,2004). In addition, the intervention protocol conductedwith Chinese children (Yu & Seligman, 2002) gave con-sideration to how Western approaches (e.g., assertive-ness training) are contrary to traditional Chinesevalues and implemented a modified protocol that wasmore sensitive and feasible to conduct with the popu-lation of interest.

Overall, results from these studies reveal that a varietyof interventions were more effective in amelioratingdepressive symptoms and disorders in children than werethe control conditions. Interventions improved affectivefunctioning for at-risk children with elevated depressivesymptoms and children who met diagnostic criteria fora depressive disorder. Whereas in the prior review moststudies utilized CBT offered in a group format, there isnow increasing support for interventions built on othertheoretical perspectives (e.g., behavior therapy; nondir-ected support=psychoeducational) and that utilize othertreatment modalities (e.g., parent–child sessions, childgroup plus parent component=intervention). However,the bulk of the published work continues to examinegroup-based CBT; thus these programs have the mostempirical support. Further, no psychosocial intervention

has been established as the superior treatment fordepressed children. The lack of data on the superiorityof one condition over another reflects the limited numberof between-group design investigations that comparedifferent active experimental interventions.

Three of the 10 studies had no follow-up data, and ofthose with available data, the sustainability of inter-vention effects was variable. For instance, one studyfound children in the CBT condition to continue to havefewer depressive symptoms than controls at the 3- and 6-month follow-ups (Yu & Seligman, 2002), whereasanother found posttreatment differences between theCBT and control conditions on internalizing symptomsas rated by parents were no longer present at the6-month follow-up (Roberts et al., 2003). Two studiesfound no group differences in depressive symptomsbetween the intervention and control condition at post-intervention, but at follow-up, youth who received theactive treatment had fewer internalizing symptoms thanthe comparison group (Gillham, Reivich et al., 2006;Muratori et al., 2003). In addition, another study foundyouth participating in individual or family therapy con-tinued to improve following treatment and initial post-treatment between-group differences disappeared atthe 6-month follow-up (Trowell et al., 2007). Becauseof these mixed findings, the long-term benefit for psy-chosocial intervention for depression for childrenremains unclear.

Finally, Table 1 shows that all 10 studies were cate-gorized as Type 2 in accord with the Nathan andGorman (2002) criteria, indicating that although eachstudy had elements of a well-designed and methodologi-cally sound randomized controlled trial, each had designlimitations that precluded it from being categorized as aType 1 intervention. Limitations included small samplesizes, lack of clearly specified inclusion and exclusioncriterion, limited diagnostic information based on thelack of comprehensive assessment batteries, and pro-blems with the randomization protocol.

Adolescent Studies

Studies of psychosocial interventions with depressedadolescents published since 1998 are presented inTable 2 and comparable information to the review ofchild studies in Table 1 is provided. Table 2 illustratesthat since 1998 there have been 18 published treatmentoutcome studies with depressed adolescents. Six studieswere conducted with at-risk adolescents with elevateddepressive symptoms (Ackerson, Scogin, McKendree-Smith, & Lyman, 1998; Asarnow et al., 2005; Clarkeet al., 2001; Kowalenko et al., 2005; Sheffield et al.,2006; Young, Mufson, & Davies, 2006a), and 12 studiesincluded adolescents who met diagnostic criteria forMDD and=or DD (Clarke et al., 2005; Clarke et al.,

TREATMENTS FOR YOUTH DEPRESSION 69

TA

BLE

2

Random

ized

Adole

scent

Depre

ssio

nT

reatm

ent

Outc

om

eS

tudie

sS

ince

1998

Inve

stig

ato

rsS

am

ple

Sourc

e-M

easu

res

Ther

apis

tT

reatm

ent

Condit

ions

Res

ult

sF

oll

ow

-Up

Nath

an

&

Gorm

an

(2002)

Cri

teri

a

Ack

erso

n,

Sco

gin

,

McK

end

ree-

Sm

ith

,&

Lym

an

(1998)

22

7th

to12th

gra

der

s

(64%

fem

ale

;68.2%

Cau

casi

an

,27.3%

Afr

ican

Am

eric

an

,

4.5%

mix

edra

ce)

‘‘at-

risk

’’b

ase

do

n

elev

ate

dd

epre

ssiv

e

sym

pto

ms

Sel

f:C

DI,

AT

Q,

DA

S,

CB

T

Pare

nt:

CB

CL

Cli

nic

ian

:H

RS

D

Sel

f-d

irec

ted

ther

ap

y;

ass

essm

ents

con

du

cted

by

mast

er’s

-lev

el

clin

icia

ns

4-w

eek

,se

lf-d

irec

ted

,in

-

ho

me

CB

T

bib

lio

ther

ap

yw

ith

Fee

ling

Good

(Bu

rns,

1980);

del

ayed

trea

tmen

t

con

tro

lgro

up

;b

oth

con

dit

ion

sre

ceiv

ed

wee

kly

ph

on

eca

lls

fro

m

exp

erim

ente

rb

ut

no

ther

ap

y

Bib

lio

ther

ap

ygro

up

sho

wed

sign

ific

an

t

red

uct

ion

sin

dep

ress

ive

sym

pto

ms

an

d

imp

rovem

ent

in

dysf

un

ctio

nal

thin

kin

g

rela

tive

toco

ntr

ol

acc

ord

ing

tose

lf,

pare

nt,

an

dcl

inic

ian

rep

ort

.E

S:

CD

I¼

1.0

5;

DA

S¼

1.3

2;

CB

CL¼

.45;

HR

SD¼

2.5

7

1m

on

thfo

llo

w-u

p;

no

con

tro

lgro

up

com

pari

son

;

Bib

lio

ther

ap

ygro

up

main

tain

ed

imp

rovem

ents

in

dep

ress

ive

sym

pto

ms

on

self

an

dp

are

nt

rep

ort

.

Co

nti

nu

edim

pro

vem

ent

incl

inic

ian

rep

ort

ed

dep

ress

ion

an

dse

lf-

rep

ort

edd

ysf

un

ctio

nal

thin

kin

g

Typ

e2

Cla

rke,

Ro

hd

e,L

ewin

soh

n,

Ho

ps,

&S

eele

y(1

999);

Lew

inso

hn

,C

lark

e,

Ro

hd

e,H

op

s,&

See

ley

(1996)

123

14-

to18-y

ear-

old

s

(71%

fem

ale

)w

ho

met

DS

M-I

II-R

crit

eria

of

Majo

rD

epre

ssio

no

r

Dyst

hym

ia;

no

eth

nic

ity

data

wer

ep

rese

nte

d

Sel

f:B

DI

Pare

nt:

CB

CL

Cli

nic

ian

:H

AM

-D,

K-S

AD

S-E

,L

IFE

,G

AF

Ad

van

ced

gra

du

ate

psy

cho

logy

or

soci

al

wo

rkst

ud

ents

;m

ast

er’s

-

or

do

cto

ral-

level

clin

icia

ns

8-w

eek

,16

two

-ho

ur

ad

ole

scen

tC

BT

gro

up

sess

ion

so

fC

WD

-A,

ad

ole

scen

tC

BT

gro

up

sess

ion

so

fC

WD

-Aw

ith

sep

ara

tep

are

nt

gro

up

,

wait

list

con

tro

l.A

ctiv

e

trea

tmen

tsra

nd

om

ized

tofo

llo

w-u

pco

nd

itio

ns

of

bo

ost

erse

ssio

nan

d

ass

essm

ent

ever

y4

mo

nth

s,ass

essm

ent

ever

y4

mo

nth

s,o

r

ass

essm

ent

ever

y12

mo

nth

s.C

on

du

cted

in

rese

arc

hcl

inic

CW

D-A

gro

up

s(w

ith

an

d

wit

ho

ut

pare

nt

inte

rven

tio

n)

sho

wed

sign

ific

an

tly

bet

ter

imp

rovem

ents

in

dep

ress

ion

than

con

tro

l

acc

ord

ing

toyo

uth

rep

ort

,b

ut

no

tp

are

nt

an

dm

ixed

fin

din

gs

base

do

ncl

inic

ian

rep

ort

.C

BT

gro

up

had

sign

ific

an

tly

imp

roved

glo

bal

fun

ctio

nin

g

acc

ord

ing

tocl

inic

ian

.

Ad

dit

ion

of

pare

nt

gro

up

had

no

sign

ific

an

t

effe

ct.

At

po

st-

trea

tmen

t,64.9%

of

CW

D-A

on

ly,

68.8%

gro

up

plu

sp

are

nt,

an

d

48.1%

of

con

tro

lw

ere

reco

ver

ed.

ES

:fo

r

CW

D-A

on

ly

BD

I¼

.59,

GA

F¼

.55;

for

CW

D-A

plu

sp

are

nt

BD

I¼

.24,

GA

F¼

.40

Over

24

mo

nth

s;n

oco

ntr

ol

con

dit

ion

.A

cco

rdin

gto

pare

nt,

bo

ost

erþ

ass

essm

ent

had

less

exte

rnali

zin

gsy

mp

tom

s

than

ass

essm

ent

on

ly.

ES

:C

BC

L¼

.63

for

4-

mo

nth

ass

essm

ent

on

ly

an

dC

BC

L¼

.47

for

12-

mo

nth

ass

essm

ent

on

ly.

4-

mo

nth

ass

essm

ent

on

lyh

ad

less

inte

rnali

zin

gsy

m.

rela

tive

to12-m

on

th

ass

essm

ent.

ES

:

CB

CL¼

.16

Typ

e1

70

Mu

fso

n,

Wei

ssm

an

,

Mo

reau

,&

Garf

ink

el

(1999)

48

12-

to18-y

ear-

old

s

(73%

fem

ale

;71%

His

pan

ic)

wh

om

et

DS

M-I

II-R

crit

eria

of

Majo

rD

epre

ssio

n;

no

oth

eret

hn

icit

yd

ata

wer

e

pre

sen

ted

Sel

f:B

DI,

SA

S-S

R,

So

cial

Pro

ble

m-S

olv

ing

Inven

tory

–R

evis

ed

Cli

nic

ian

:C

-GA

S,

DIS

C

2.3

,H

RS

D,

K-S

AD

S-E

Ch

ild

psy

chia

tris

ts,

lice

nse

dcl

inic

al

psy

cho

logis

t,m

ast

er’s

-

level

psy

cho

logis

tw

ith

10

yea

rs’

exp

erie

nce

12

45-m

inw

eek

ly

ind

ivid

ual

sess

ion

so

f

IPT

-A,

clin

ical

mo

nit

ori

ng

con

tro

l.

Co

nd

uct

edin

ho

spit

al

clin

ics

IPT

-Agro

up

had

stro

nger

dec

rease

ind

epre

ssio

n

an

dim

pro

vem

ent

in

soci

al

fun

ctio

nin

gan

d

inte

rper

son

al

pro

ble

m-

solv

ing

than

con

tro

l

base

do

nse

lf-

an

d

clin

icia

nre

po

rt.

Sig

nif

ican

tly

mo

re

(74%

)in

IPT

-Agro

up

reco

ver

edth

an

con

tro

l

(46%

).E

S:

BD

I¼

.66;

HR

SD¼

.66;

SA

S-

frie

nd¼

.55;

SA

S-

dati

ng¼

1.0

;S

AS

-

tota

l¼

.45

No

foll

ow

-up

data

wer

e

rep

ort

ed

Typ

e1

Ro

ssel

lo&

Ber

nal

(1999)

71

13-

to18-y

ears

-old

s

(54%

fem

ale

;100%

Pu

erto

Ric

an

)w

ho

met

DS

M-I

II-R

crit

eria

of

Majo

rD

epre

ssio

n,

Dyst

hym

ia,

or

bo

th

Sel

f:C

DI,

PH

CS

CS

,

SA

SC

A,

FE

ICS

Pare

nt:

CB

CL

Do

cto

ral

clin

ical

psy

cho

logy

stu

den

ts

wit

h3

yea

rs’

exp

erie

nce

12

1-h

rin

div

idu

al

sess

ion

s

over

12

wee

ks

for

bo

th

act

ive

trea

tmen

t

con

dit

ion

so

fin

div

idu

al

CB

Tan

din

div

idu

al

IPT

,w

ait

list

con

tro

l.

Co

nd

uct

edin

ah

osp

ital

clin

ic

CB

Tan

dIP

Th

ad

sign

ific

an

tly

gre

ate

r

red

uct

ion

of

dep

ress

ive

sym

pto

ms

than

con

tro

l.

IPT

als

osu

per

ior

to

con

tro

lin

incr

easi

ng

self

-est

eem

an

dso

cial

ad

ap

tati

on

.G

rou

p

dif

fere

nce

sb

ase

do

n

self

-,n

ot

pare

nt

rep

ort

.

Th

ere

wer

en

o

dif

fere

nce

sb

etw

een

IPT

an

dC

BT

.A

t

po

sttr

eatm

ent,

82%

in

IPT

an

d59%

inC

BT

had

CD

Isc

ore

of

17

or

less

.E

Sfo

rC

BT

:

CD

I¼

.35.

ES

for

IPT

:

CD

I¼

.76;

PH

CS

CS¼

.46;

SA

SC

A¼

.74

3-m

on

thfo

llo

wu

p;

no

-

con

tro

lgro

up

com

pari

son

;n

o

sign

ific

an

td

iffe

ren

ces

bet

wee

nC

BT

an

dIP

T

acc

ord

ing

tose

lf-

an

d

pare

nt

rep

ort

Typ

e1

Cla

rke

etal.

(2001)

94

13-

to18-y

ear-

old

s

(59%

fem

ale

;11%

no

nw

hit

e)‘‘

at-

risk

’’

base

do

nh

avin

g

dep

ress

edp

are

nts

an

d

elev

ate

dd

epre

ssiv

e

sym

pto

ms;

no

oth

er

eth

nic

ity

data

wer

e

pre

sen

ted

Sel

f:C

ES

-D

Pare

nt:

CB

CL

Cli

nic

ian

:F

-SA

DS

GA

F,

HA

M-D

,K

-SA

DS

-E

Mast

er’s

-lev

elth

erap

ists

15

1-h

rgro

up

sess

ion

so

f

CW

D-A

plu

sp

are

nt

com

po

nen

tan

dH

MO

care

,u

sual

HM

Oca

re

on

ly.

HM

Ose

ttin

g

CB

Tgro

up

sho

wed

sign

ific

an

tly

bet

ter

imp

rovem

ents

in

dep

ress

ive

sym

pto

ms

than

con

tro

lacc

ord

ing

tose

lf-r

epo

rtan

d

inte

rvie

wra

tin

gs,

bu

t

no

tp

are

nt

rep

ort

.

Acc

ord

ing

tocl

inic

ian

,

CB

Tgro

up

had

sign

ific

an

tly

few

er

suic

ide

item

san

db

ette

r

glo

bal

fun

ctio

nin

g.

ES

:

CE

S-D¼

.47;

HA

M-

D¼

.31;

K-

SA

DS¼

.10;

CA

F¼

.04

12-

an

d24-m

on

thfo

llo

w-

up

;si

gn

ific

an

tly

few

er

pro

spec

tive

case

so

f

majo

rd

epre

ssio

nin

CB

Tgro

up

than

con

tro

l

at

12-m

on

thfo

llo

w-u

p;

CB

Tgro

up

pre

ven

tive

effe

ctp

ersi

sted

at

24-

mo

nth

foll

ow

-up

bu

tat

ad

ecli

ned

level

Typ

e2

(Co

nti

nu

ed)

71

TA

BLE

2

Continued

Inve

stig

ato

rsS

am

ple

Sourc

e-M

easu

res

Ther

apis

tT

reatm

ent

Condit

ions

Res

ult

sF

oll

ow

-Up

Nath

an

&

Gorm

an

(2002)

Cri

teri

a

Cla

rke

etal.

(2002)

88

13-

to18-y

ear-

old

s

(69%

fem

ale

;9%

min

ori

ty)

wh

om

et

DS

M-I

II-R

crit

eria

of

Majo

rD

epre

ssio

n

an

d=

or

Dyst

hym

ia;

no

oth

eret

hn

icit

yd

ata

wer

e

pre

sen

ted

Sel

f:C

ES

-D

Pare

nt:

CB

CL

Cli

nic

ian

:F

-SA

DS

,

HA

M-D

,K

-SA

DS

,G

AF

Mast

er’s

-lev

elth

erap

ists

16

2-h

rgro

up

sess

ion

so

f

CW

D-A

plu

sp

are

nt

com

po

nen

tan

dH

MO

care

,u

sual

HM

Oca

re

on

ly.

Co

nd

uct

edin

HM

Ose

ttin

g

No

sign

ific

an

tad

van

tage

of

CB

To

ver

usu

al

HM

Oca

reacc

ord

ing

to

self

-,p

are

nt,

an

d

clin

icia

nre

po

rt

12-

an

d24-m

on

thfo

llo

w-

up

;n

osi

gn

ific

an

tgro

up

dif

fere

nce

sacc

ord

ing

to

self

-,p

are

nt,

an

d

clin

icia

nre

po

rt

Typ

e2

Dia

mo

nd

,R

eis,

Dia

mo

nd

,

Siq

uel

an

d,

&Is

aacs

(2002)

32

13-

to17-y

ear-

old

s

(78%

fem

ale

;69%

Afr

ican

Am

eric

an

,31%

Cau

casi

an

;69%

low

inco

me)

wh

om

et

DS

M-I

II-R

crit

eria

of

Majo

rD

epre

ssio

n

Sel

f:B

DI,

BH

S,

YS

R,

SR

FF

,S

TA

IC,

Su

icid

al

Idea

tio

n

Pare

nt:

CB

CL

Cli

nic

ian

:H

AM

-D

K-S

AD

S-P

Do

cto

ral

an

dm

ast

er’s

-lev

el

ther

ap

ists

12-w

eek

AB

FT

incl

ud

ing

ad

ole

scen

tan

dp

are

nt

con

du

cted

ina

chil

dre

n’s

ho

spit

al=

chil

dgu

idan

ce

clin

ic,

6-w

eek

wait

list

con

tro

lth

at

incl

ud

ed

wee

kly

15-m

inp

ho

ne

call

sre

stri

cted

to

mo

nit

ori

ng

AB

FT

gro

up

had

sign

ific

an

tly

low

erle

vel

s

of

dep

ress

ive

sym

pto

ms

than

con

tro

lsacc

ord

ing

tocl

inic

ian

s,an

dlo

wer

level

so

fan

xie

tyan

d

fam

ily

con

flic

to

nse

lf-

rep

ort

.A

tth

een

do

f

trea

tmen

t,81%

of

AB

FT

gro

up

did

no

t

mee

tcr

iter

iafo

rM

DD

vs.

47%

of

con

tro

l

gro

up

.E

S:

HA

M-

D¼

.64;

ST

AIC¼

1.0

5;

SR

FF¼

1.3

6-m

on

thfo

llo

w-u

p;

no

-

con

tro

lgro

up

.O

f15

AB

FT

gro

up

ad

ole

scen

tsin

foll

ow

-up

sam

ple

,87%

con

tin

ued

ton

ot

mee

tcr

iter

iafo

r

Majo

rD

epre

ssio

n

Typ

e2

Mu

fso

net

al.

(2004);

Yo

un

g,

Mu

fso

n,

&

Davie

s(2

006)

63

12-

to18-y

ears

-old

s

(84.1%

fem

ale

;71%

Lati

na)

wh

om

et

DS

M-I

Vcr

iter

iafo

ra

dep

ress

ive

dis

ord

er;

no

oth

eret

hn

icit

y

data

wer

ep

rese

nte

d

Sel

f:B

DI,

SA

S-S

R

Cli

nic

ian

:H

AM

D,

C-G

AS

,C

GI

Sch

oo

l-b

ase

dso

cial

wo

rker

san

dd

oct

ora

l-

level

psy

cho

logis

ts

12

30-m

inin

div

idu

al

sch

oo

l-b

ase

dse

ssio

no

f

IPT

-Ao

ver

12

wee

ks,

ind

ivid

ual

trea

tmen

tas

usu

al

con

tro

l

(su

pp

ort

ive

ther

ap

y)

IPT

-Ah

ad

sign

ific

an

tly

few

erd

epre

ssiv

e

sym

pto

ms

an

db

ette

r

glo

bal

fun

ctio

nin

gth

an

con

tro

lacc

ord

ing

to

clin

icia

nra

tin

gs.

IPT

-A

had

sign

ific

an

tly

few

er

dep

ress

ive

sym

pto

ms

an

db

ette

rgen

eral

an

d

soci

al

fun

ctio

nin

gth

an

con

tro

lacc

ord

ing

to

self

-rep

ort

.N

o

sign

ific

an

tgro

up

dif

fere

nce

sin

fun

ctio

nin

gfo

ryo

uth

wit

hco

mo

rbid

an

xie

ty.

ES

:H

AM

D¼

.50;

C-G

AS¼

.54;

CG

I¼

.48;

SA

S-

dati

ng¼

.43;

SA

S-t

ota

l¼

.55

16-w

eek

foll

ow

-up

.

Acc

ord

ing

tocl

inic

ian

rati

ngs,

IPT

-Agro

up

had

sign

ific

an

tly

few

er

dep

ress

ive

sym

pto

ms

than

con

tro

l.E

S:

HA

MD¼

.51

Typ

e1

72

TA

DS

Tea

m(2

004)

439

12-

to17-y

ear-

old

s

(54.4%

fem

ale

;73.8%

Cau

casi

an

,12.5%

Afr

ican

Am

eric

an

,

8.9%

His

pan

ic)

wit

ha

pri

mary

DS

M-I

Vd

iagn

osi

s

of

curr

ent

Majo

r

Dep

ress

ive

Dis

ord

er,

IQ>

80

an

dn

ot

tak

ing

an

an

tid

epre

ssan

t

pri

or

toth

ein

itia

tio

n

of

the

stu

dy

Sel

f:R

AD

S,

SIQ

Cli

nic

ian

:C

DR

S-R

,C

GI

Ad

ole

scen

tp

sych

iatr

ist

an

dC

BT

ther

ap

ists

train

edan

dsu

per

vis

ed

by

pro

gra

md

evel

op

ers

15

sess

ion

so

f50–60

min

of

CB

Talo

ne

over

a

12-w

eek

per

iod

plu

s

two

pare

nt-

on

ly

sess

ion

s;fl

uo

xet

ine

alo

ne

(mo

nit

ori

ng

an

d

enco

ura

gem

ent

du

rin

g

620-

to30-m

invis

its

acr

oss

12

wee

ks,

start

ing

do

seo

f

10

mg=

day

incr

ease

dto

20m

g=

day,

maxim

um

do

seo

f40m

g=

day);

CB

Tp

lus

flu

oxet

ine;

pla

ceb

o(s

am

ed

osa

ges

as

flu

oxet

ine

plu

s

mo

nit

ori

ng).

Inte

rven

tio

ns

for

the

mu

ltic

ente

rtr

ial

wer

eco

nd

uct

edin

aca

dem

ich

ealt

h

scie

nce

sce

nte

rs

Co

mb

ined

trea

tmen

t

was

mo

reef

fect

ive

than

bo

thth

ep

lace

bo

an

dei

ther

inte

rven

tio

n

alo

ne

inre

du

cin

g

dep

ress

ive

sym

pto

ms

an

do

ver

all

psy

cho

logic

al

dif

ficu

ltie

s

acc

ord

ing

tocl

inic

ian

rati

ngs

an

dsu

icid

al

idea

tio

nacc

ord

ing

to

the

yo

uth

.F

luo

xet

ine

alo

ne

was

bet

ter

than

CB

Talo

ne.

ES

calc

ula

ted

rela

tive

to

pla

ceb

o.

ES

for

com

bin

ed:

CD

RS

-

R¼

.98;

CG

I¼

.84;

SIQ¼

.28;

RA

DS¼

.57.

ES

for

CB

Talo

ne:

CD

RS

-

R¼

.03;

CG

I¼

.20;

SIQ¼

.33;

RA

DS¼

.10.

ES

for

med

alo

ne:

CD

RS

-

R¼

.68;

CG

I¼

.58;

SIQ¼

.05;

RA

DS¼

.10

Fo

llo

w-u

pd

ata

are

fort

hco

min

g

Typ

e1

Asa

rno

wet

al.

(2005)

418

13-

to21-y

ear-

old

s

(78%

fem

ale

;13.4%

Afr

ican

Am

eric

an

,1.2%

Asi

an

,56%

His

pan

ic,

13.6%

Mix

ed,

12.7%

Cau

casi

an

,3.1%

oth

er;

64%

spo

ke

lan

gu

age

oth

erth

an

En

gli

shat

ho

me)

‘‘at-

risk

’’b

ase

d

up

on

elev

ate

dd

epre

ssiv

e

sym

pto

ms

Sel

f:C

ICI-

12,

CE

S-D

,

MC

S-1

2,

MH

I-5,

Sati

sfact

ion

surv

ey

Cli

nic

ian

:D

ISC

Do

cto

ral

an

dm

ast

er’s

-

level

ther

ap

ists

or

nu

rses

train

edan

dp

rovid

ed

on

go

ing

con

sult

ati

on

on

CB

Tth

erap

y,

pri

mary

care

ph

ysi

cian

san

d

nu

rses

Qu

ali

tyim

pro

ved

trea

tmen

tin

clu

ded

sup

po

rt=tr

ain

ing

of

pri

mary

care

clin

icia

ns

an

dtr

eatm

ent

op

tio

ns

of

CB

T,

CB

Tan

d

med

icati

on

,ca

rem

an

ger

foll

ow

-up

,o

rre

ferr

al.

Pati

ents

cou

ldch

an

ge

trea

tmen

t.C

BT

was

34-s

essi

on

sm

od

ule

s

con

du

cted

wee

kly

for

50-m

in.

Ind

ivid

ual

vs.

fam

ily

sess

ion

sn

ot

spec

ifie

d.

Tre

atm

ent

as

usu

al

was

usu

al

inte

ract

ion

wit

hp

rim

ary

care

clin

icia

ns

wh

oh

ad

train

ing

an

ded

uca

tio

n

mate

rials

ab

ou

t

dep

ress

ion

bu

tn

ot

CB

T.

Co

nd

uct

edin

pri

mary

care

sett

ing

Qu

ali

tyim

pro

ved

trea

tmen

th

ad

sign

ific

an

tly

less

dep

ress

ive

sym

pto

ms,

hig

her

qu

ali

tyo

fli

fe,

an

dgre

ate

rsa

tisf

act

ion

wit

hm

enta

lh

ealt

h

imp

rovem

ents

than

con

tro

lacc

ord

ing

to

self

rep

ort

.E

S:

CE

S-

D¼

.19;

MC

S-1

2¼

.15;

Sati

sfact

ion¼

.32

No

foll

ow

-up

data

wer

ere

po

rted

.

Typ

e1

(Co

nti

nu

ed)

73

TA

BLE

2

Continued

Inve

stig

ato

rsS

am

ple

Sourc

e-M

easu

res

Ther

apis

tT

reatm

ent

Condit

ions

Res

ult

sF

oll

ow

-Up

Nath

an

&

Gorm

an

(2002)

Cri

teri

a

Cla

rke

etal.

(2005)

152

12-

to18-y

ear-

old

s

(78%

fem

ale

;14%

min

ori

ty)

wh

om

et

DS

M-I

Vcr

iter

iafo

r

Majo

rD

epre

ssio

nan

d

wer

ep

resc

rib

edan

SS

RI

by

ap

rim

ary

care

ph

ysi

cian

Sel

f:C

ES

-D,

HR

SD

,Y

SR

Pare

nt:

CB

CL

Cli

nic

ian

:K

-SA

DS

,

C-G

AS

,S

AS

Y;

Sh

ort

Fo

rm-1

2

Mast

er’s

-lev

elth

erap

ists

5–9

60-m

inin

div

idu

al

CB

Tp

lus

SS

RI,

SS

RI

edu

cati

on

,te

lep

ho

ne

con

tact

at

1,

2,

3,

5,

7,

an

d9

mo

nth

saft

erin

per

son

sess

ion

s,an

d

sep

ara

tem

on

thly

pare

nt

edu

cati

on

sess

ion

s;H

MO

trea

tmen

tas

usu

al

plu

sS

SR

I.C

on

du

cted

ina

HM

Ose

ttin

g

No

sign

ific

an

tad

van

tage

of

CB

Tp

lus

SS

RI

over

usu

al

HM

Op

lus

SS

RI

care

acc

ord

ing

tose

lf-,

pare

nt,

an

dcl

inic

ian

rep

ort

6-,

12-,

26-,

an

d52-w

eek

foll

ow

-up

s.N

o

sign

ific

an

tgro

up

dif

fere

nce

sin

dep

.

reco

ver

yat

an

yfo

llo

w-

up

.D

ep.

recu

rren

ce

occ

urr

edeq

uall

yin

the

gro

up

sb

yW

eek

52.

CB

Tsi

gn

ific

an

tly

gre

ate

rim

pro

vem

ent

on

Sh

ort

-Fo

rm12,

nu

mb

er

of

vis

its,

an

dm

edic

ati

on

days

Typ

e2

Ko

wale

nk

oet

al.

(2005)

82

13-

to16-y

ear-

old

s

(100%

fem

ale

;70%

Au

stra

lia

or

New

Zea

lan

d)

‘‘at-

risk

’’

base

do

nsc

ori

ng

ab

ove

18

on

CD

I

Sel

f:C

DI,

AC

S,

CA

TS

Sch

oo

lco

un

selo

ran

d

com

mu

nit

yad

ole

scen

t

men

tal

hea

lth

wo

rker

train

edan

dsu

per

vis

ed

by

pro

gra

md

evel

op

er

890-m

inw

eek

lyC

BT

AC

E

gro

up

;w

ait

list

con

tro

l.

Co

nd

uct

edat

sch

oo

l

AC

Efe

male

sh

ad

sign

ific

an

tly

low

er

dep

ress

ion

an

d

imp

roved

cop

ing

than

con

tro

l.A

CE

fem

ale

sign

ific

an

tly

less

lik

ely

to

be

ab

ove

clin

ical

cuto

ff

on

CD

Ith

an

con

tro

l

(50%

vs.

81.5%

).E

S:

CD

I¼

.55,

AC

S¼

.38–

.57,

CA

TS¼

.52–

.53

6-m

on

thfo

llo

w-u

p;

no

-

con

tro

lgro

up

.R

elati

ve

top

retr

eatm

ent,

AC

E

fem

ale

ssh

ow

sign

ific

an

t

imp

rovem

ents

in

dep

ress

ion

an

dco

pin

gat

foll

ow

-up

Typ

e2

Ro

hd

e,C

lark

e,M

ace

,

Jorg

ense

n,

&S

eele

y

(2004);

Kau

fman

,

Ro

hd

e,S

eele

y,

Cla

rke,

&S

tice

(2005)

93

13-

to17-y

ears

-old

s

(48%

fem

ale

;80.6%

Cau

casi

an

,1.1%

Afr

ican

Am

eric

an

,

3.2%

Nati

ve

Am

eric

an

,1.1%

Asi

an

,

4.3%

His

pan

ic,

9.7%

oth

eret

hn

icit

y)

wh

om

et

DS

M-I

Vcr

iter

iafo

r

Majo

rD

epre

ssio

nan

d

Co

nd

uct

Dis

ord

er

Sel

f:B

DI-

II,

SA

S-S

R,

AT

Q,

PE

S,

Issu

e

Ch

eck

list

Pare

nt:

CB

CL

Cli

nic

ian

:H

DR

S,

K-S

AD

S-E

,L

IFE

CW

D-A

incl

ud

edat

least

mast

er’s

-lev

elm

enta

l

hea

lth

pro

vid

eran

d

coll

ege=

hig

hsc

ho

ol

stu

den

tass

ista

nt;

LS

uti

lize

dh

igh

sch

oo

l

teach

ers

an

dad

ult

ass

ista

nts

16

2-h

rC

WD

-Ap

lus

pare

nt

com

po

nen

to

ver

8

wee

ks,

LS

.C

on

du

cted

in

rese

arc

hcl

inic

Dep

ress

ion

reco

ver

yra

tes

(39%

vs.

19%

)an

d

soci

al

ad

just

men

tw

as

sign

ific

an

tly

bet

ter

wit

h

CW

D-A

than

LS

base

d

on

self

-,b

ut

no

tp

are

nt

rep

ort

.C

lin

icia

nre

sult

s

wer

em

ixed

.C

han

ge

in

neg

ati

ve

thin

kin

g

med

iate

dth

ere

du

ctio

n

of

dep

ress

ive

sym

pto

ms.

Gro

up

sd

idn

ot

dif

fer

in

Co

nd

uct

dis

ord

er

reco

ver

yra

tes.

ES

:

BD

I¼

.17;

HD

RS¼

.39;

SA

S-S

R¼

.30

6-

an

d12-m

on

thfo

llo

w-u

p;

no

sign

ific

an

tgro

up

dif

fere

nce

so

n

dep

ress

ion

or

con

du

ct

acc

ord

ing

tose

lf-,

pare

nt,

an

dcl

inic

ian

rep

ort

Typ

e1

74

Mel

vin

etal.

(2006)

73

12-

to18-y

ear-

old

s

(66%

fem

ale

;90%

Au

stra

lian

;69%

com

orb

idd

iagn

osi

s)

wh

om

etD

SM

-IV

crit

eria

for

Majo

r

Dep

ress

ion

,D

yst

hym

ia,

or

Dep

ress

ion

NO

S

Sel

f:R

AD

S,

RC

MA

S,

SIQ

,S

EQ

,F

AD

Pare

nt:

CB

CL

,F

AD

Cli

nic

ian

:K

-SA

DS

,G

AF

,

GA

RF

S

Reg

iste

red

psy

cho

logis

ts,

sup

ervis

edp

rob

ati

on

ary

psy

cho

logis

t,gen

eral

med

ical

pra

ctit

ion

ers,

an

dso

cial

wo

rkw

ith

1–5

yea

rsex

per

ien

cein

pro

vid

ing

CB

T.

Wee

kly

tob

iwee

kly

sup

ervis

ion

wit

hex

per

tth

erap

ist

an

dw

eek

lyp

eer

sup

ervis

ion

.

12

50

min

ute

s,in

div

idu

al

wee

kly

CB

Tse

ssio

ns

wit

hse

para

tep

are

nt

sess

ion

savail

ab

lean

d

two

pare

nt-

ad

ole

scen

t

sess

ion

s;se

rtra

lin

ealo

ne

(ed

uca

tio

nan

d

mo

nit

ori

ng

acr

oss

12

wee

ks,

start

ing

do

seo

f

25

mg=

day,

maxim

um

do

seo

f100

mg=d

ay);

CB

Tp

lus

sert

rali

ne.

3

mo

nth

lyp

ost

trea

tmen

t

bo

ost

erse

ssio

ns

for

each

con

dit

ion

.In

terv

enti

on

s

for

the

mu

lti-

cen

ter

tria

l

wer

eco

nd

uct

edin

com

mu

nit

y-b

ase

dcl

inic

Fo

ro

dd

so

fa

dep

ress

ive

dis

ord

er,

CB

Talo

ne

had

sign

ific

an

tly

low

ero

dd

s

than

med

icati

on

alo

ne,

an

dco

mb

ined

gro

up

did

no

tsi

gn

ific

an

tly

dif

fer

fro

mC

BT

alo

ne

or

med

icati

on

alo

ne.

No

bet

wee

n-g

rou

p

dif

fere

nce

so

nfu

ll

rem

issi

on

rate

s.A

ll

gro

up

sd

emo

nst

rate

d

sign

ific

an

t

imp

rovem

ents

on

self

-

rep

ort

edd

epre

ssio

n,

an

xie

ty,

an

dsu

icid

al

idea

tio

n,

on

pare

nt

rep

ort

edin

tern

ali

zin

g,

an

do

ncl

inic

ian

rep

ort

of

gen

eral

fun

ctio

nin

g;

ho

wev

er,

no

sign

ific

an

t

bet

wee

n-g

rou

p

dif

fere

nce

so

nm

easu

res

6-m

on

thfo

llo

w-u

p;

wit

hin

gro

up

dif

fere

nce

s

main

tain

edo

nse

lf-

rep

ort

mea

sure

s;n

o

bet

wee

n-g

rou

p

dif

fere

nce

sin

od

ds

of

dep

ress

ion

dis

ord

er,

rem

issi

on

rate

s,o

r

con

tin

uo

us

mea

sure

s

Typ

e1

San

ford

etal.

(2006)

31

13-

to18-y

ear

old

s

(64.5%

fem

ale

;100%

fro

mH

am

ilto

n,

On

tari

o;

hig

hra

teo

f

com

orb

idd

iso

rder

s)

wh

om

etD

SM

-IV

crit

eria

for

Majo

r

Dep

ress

ion

an

dre

sid

ed

wit

hp

are

nt;

no

oth

er

cult

ura

lo

ret

hn

icit

yd

ata

wer

ep

rese

nte

d

Sel

f:A

CL

,C

SQ

,F

AD

,

RA

DS

,S

SA

I,

sub

stan

ceu

se

Pare

nt:

CS

Q,

FA

D,

SS

AI

Cli

nic

ian

:C

-GA

S,

K-S

AD

S-P

No

ted

no

tto

be

del

iver

ed

by

no

vic

ecl

inic

ian

sb

ut

do

no

tre

po

rto

nsp

ecif

ic

qu

ali

fica

tio

ns

of

ther

ap

ist

12

90-m

inF

PE

incl

ud

ing

all

fam

ily

mem

ber

s,

plu

so

ne

bo

ost

erse

ssio

n,

an

du

sual

trea

tmen

t

of

ind

ivid

ual

or

gro

up

ther

ap

yan

d=o

r

med

icati

on

;u

sual

trea

tmen

tco

ntr

ol;

all

part

icip

an

tsre

ceiv

ed

han

do

uts

ab

ou

t

dep

ress

ion

.F

PE

con

du

cted

infa

mil

y

ho

me.

Usu

al

trea

tmen

t

con

du

cted

ino

utp

ati

ent

clin

ics

Bo

thF

PE

an

dco

ntr

ol

had

dec

lin

esin

self

-rep

ort

ed

dep

ress

ion

,b

ut

gro

up

s

did

no

tsi

gn

ific

an

tly

dif

fer.

21%

of

FP

Ean

d

50%

of

con

tro

ls

con

tin

ued

tom

etcr

iter

ia

for

Majo

rD

epre

ssio

n.

FP

Eh

ad

sign

ific

an

tly

gre

ate

rim

pro

vem

ents

in

soci

al

fun

ctio

nin

gan

d

ad

ole

scen

t-p

are

nt

rela

tio

nsh

ipacc

ord

ing

toad

ole

scen

tan

dp

are

nt

rep

ort

san

dgre

ate

r

pare

nta

ltr

eatm

ent

sati

sfact

ion

.E

S:

SS

AI

ad

ol¼

.93;

SS

AI

pare

nt¼

.96;

AC

L¼

.52–.9

2

9-m

on

thfo

llo

w-u

p;

gro

up

dif

fere

nce

sfa

vo

rin

g

FP

Ew

ere

main

tain

ed.

25%

of

FP

Ean

d53%

of

con

tro

lsco

nti

nu

edto

mee

tcr

iter

iafo

rM

ajo

r

Dep

ress

ion

.E

S:

SS

AI

ad

ol¼

1.1

4;

SS

AI

pare

nt¼

1.1

7;

AC

L¼

.63–1.1

Typ

e2

(Co

nti

nu

ed)

75

TA

BLE

2

Continued

Inve

stig

ato

rsS

am

ple

Sourc

e-M

easu

res

Ther

apis

tT

reatm

ent

Condit

ions

Res

ult

sF

oll

ow

-Up

Nath

an

&

Gorm

an

(2002)

Cri

teri

a

Sh

effi

eld

etal.

(2006)

521

13-

to15-y

ear-

old

s

(69%

fem

ale

;88%

Au

stra

lian

bo

rn)

‘‘at-

risk

’’b

ase

dsc

ori

ng

in

the

top

20%

on

two

self

-rep

ort

dep

ress

ive

mea

sure

s

no

oth

ercu

ltu

ral

or

eth

nic

ity

data

wer

e

pre

sen

ted

2,4

79

13-

to15-y

ear-

old

s

(54%

fem

ale

;87%

Au

stra

lian

bo

rn)

sele

cted

for

un

iver

sal

com

po

nen

tw

ith

ou

t

regard

tosy

mp

tom

so

f

dep

ress

ion

Sel

f:B

HS

,C

AS

AF

S,

CA

TS

,C

DI,

CE

S-D

,

NP

O,

SC

AS

,S

PS

I-R

,

sub

stan

ceu

se,

YS

R

Cli

nic

ian

:A

DIS

-C,

LIF

E

Ind

icate

dco

nd

itio

nb

y

sch

oo

lco

un

selo

rs,

com

mu

nit

yo

rm

enta

l

hea

lth

pro

fess

ion

s.

Un

iver

sal

con

dit

ion

by

teach

ers

Ind

icate

dfo

r‘‘

at-

risk

’’:

890-m

ingro

up

CB

T

sess

ion

so

ver

8w

eek

s;

Un

iver

sal:

845–50-m

in

CB

Tcl

ass

sess

ion

so

ver

8w

eek

s;U

niv

ersa

lp

lus

ind

icate

dfo

r‘‘

at-

risk

’’:

8-w

eek

un

iver

sal

com

po

nen

tfo

llo

wed

by

8-w

eek

ind

icate

d

com

po

nen

t;n

o

inte

rven

tio

nco

ntr

ol.

Co

nd

uct

edin

sch

oo

l

sett

ing

Acr

oss

con

dit

ion

s,‘‘

at-

risk

’’

sho

wed

sign

ific

an

t

red

uct

ion

sin

dep

ress

ion

,

ho

pel

essn

ess,

neg

ati

ve

thin

kin

g,

an

xie

ty,

exte

rnali

zin

gan

d

neg

ati

ve

pro

ble

m

solv

ing

an

d

imp

rovem

ents

inso

cial

fun

ctio

nin

gacc

ord

ing

to

self

-rep

ort

.20%

of

‘‘at-

risk

’’h

ad

dep

ress

ive

epis

od

eo

ver

stu

dy.

No

ne

of

the

con

dit

ion

s

sign

ific

an

tly

dif

fere

do

n

self

-rep

ort

edo

utc

om

e

mea

sure

s

12-m

on

thfo

llo

w-u

p;

no

ne

of

the

con

dit

ion

s

sign

ific

an

tly

dif

fere

do

n

ou

tco

me

mea

sure

s

Typ

e1

Yo

un

g,

Mu

fso

n,

&D

avie

s(2

006)

41

11-

to16-y

ear-

old

s

(85%

fem

ale

;93%

His

pan

ic)

‘‘at-

risk

’’

base

do

nC

ES

-D,

K-S

AD

S,

C-G

AS

;yo

uth

wit

hcl

inic

al

dis

ord

ers

excl

ud

ed

Sel

f:C

ES

-D

Cli

nic

ian

:K

-SA

DS

,

C-G

AS

Ind

icate

dco

nd

itio

nb

y

do

cto

rate

an

dm

ast

er

level

ther

ap

ists

an

d

soci

al

wo

rker

s.C

on

tro

l

con

dit

ion

by

sch

oo

l

gu

idan

ceco

un

selo

rsan

d

soci

al

wo

rker

s

2in

div

idu

al

an

d8

90

min

ute

sw

eek

lygro

up

sess

ion

so

fIP

T-A

ST

;

0–7

ind

ivid

ual

or

gro

up

SC

sess

ion

s.C

on

du

cted

insc

ho

ol

sett

ing

IPT

-AS

Th

ad

sign

ific

an

tly

few

erd

epre

ssiv

e

sym

pto

ms

an

db

ette

r

over

all

fun

ctio

nin

gth

an

SC

acc

ord

ing

tose

lf-

an

dcl

inic

al

rep

ort

.

No

IPT

-AS

Tyo

uth

an

d3

SC

ad

ole

scen

t

had

clin

ical

dep

ress

ive

dis

ord

er.

ES

:C

ES

-D¼

1.5

2,

C-G

AS¼

.96

3-

an

d6-m

on

thfo

llo

w-u

p;

IPT

-AS

Th

ad

sign

ific

an

tly

few

er

dep

ress

ive

sym

pto

ms

an

db

ette

ro

ver

all

fun

ctio

nin

gth

an

SC

.

Acr

oss

foll

ow

-up

,3.6%

IPT

-AS

Tan

d28.6%

of

SC

dev

elo

ped

dep

ress

ion

dia

gn

osi

s.E

S:

CE

S-D

(3m

)¼

1.1

0,

CE

S-D

(6m

)¼

1.0

9,

C-G

AS

(3m

)¼

.82,

C-G

AS

(6m

)¼

1.2

1

Typ

e2

76

Go

od

yer

etal.

(2007)

208

11-

to17-y

ear-

old

s

(74%

fem

ale

;72%

Man

ches

ter,

En

gla

nd

;

28%

Cam

bri

dge,

En

gla

nd

)w

ho

met

DS

M-I

Vcr

iter

iafo

r

Majo

ro

rp

rob

ab

le

dep

ress

ion

an

d

mo

der

ate

tose

rve

dif

ficu

ltie

so

n

Ho

NO

SC

A

Sel

f=P

are

nt:

Mo

od

an

d

feel

ings

qu

esti

on

nair

e,

CD

RS

-R,

C-G

AS

,C

GI

Cli

nic

ian

:K-S

AD

S,

Ho

NO

SC

A

Psy

chia

tris

tsth

erap

ists

train

edan

dex

per

ien

ced

wit

hC

BT

19

55-m

in,

ind

ivid

ual

CB

T

sess

ion

so

ver

28

wee

ks

(wee

kly

for

12

wee

ks,

bim

on

thly

for

12

wee

ks)

wit

hp

are

nt

part

icip

ati

on

at

end

of

each

sess

ion

plu

sS

SR

I;

SS

RI

on

ly(f

luo

xet

ine

pri

mary

oth

erS

SR

Is

all

ow

ed,

start

ing

do

seo

f

10

mg=

day,

maxim

um

do

seo

f60

mg=

day).

Inte

rven

tio

nat

ou

tpati

ent

men

tal

hea

lth

clin

ics.

Bo

thgro

up

ssh

ow

ed

imp

rovem

ent

in

dep

ress

ion

,m

oo

d,

fun

ctio

nin

g,

an

d

suic

idali

ty.

No

sign

ific

an

tad

van

tage

of

CB

Tp

lus

SS

RI

over

SS

RI

care

acc

ord

ing

to

self=p

are

nt

an

dcl

inic

ian

rep

ort

.

No

foll

ow

-up

data

.T

yp

e1

No

te:

Eff

ect

size

(ES

)is

Co

hen

’sd.

AB

FT¼

Att

ach

men

t-B

ase

dF

amil

yT

her

ap

y;

AC

E¼

Ad

ole

scen

tsC

op

ing

wit

hE

mo

tio

ns;

AC

L¼

Ad

ject

ive

Ch

eck

list

;A

CS¼

Ad

ole

scen

tC

op

ing

Sca

le;

AD

IS-C¼

An

xiet

yD

iso

rder

sIn

terv

iew

Sch

edu

lefo

rC

hil

dre

n;

AT

Q¼

Au

tom

ati

cT

ho

ug

hts

Qu

esti

on

na

ire;

BD

I¼

Bec

kD

epre

ssio

nIn

ven

tory

;B

HS¼

Bec

kH

op

eles

snes

sS

cale

;C

AS

AF

S¼

Ch

ild

an

dA

do

lesc

ent

So

cial

an

dA

da

pti

veF

un

ctio

nin

gS

cale

;C

AT

S¼

Ch

ild

ren

’sA

uto

ma

tic

Th

ou

gh

tS

cale

;C

BC

L¼

Ch

ild

Beh

av

ior

Ch

eck

list

;C

BT¼

Co

gn

itiv

eB

ehav

iora

lT

her

ap

y;

CB

T¼

Co

gn

itiv

eB

ibli

oth

erap

yT

est;

CD

I¼

Ch

ild

Dep

ress

ion

Inv

ento

ry;

CD

RS

-R¼

Ch

ild

ren

’sD

epre

ssio

nR

ati

ng

Sca

le-R

evis

ed;

CE

S-D¼

Cen

ter

for

Ep

idem

iolo

gic

al

Stu

die

s-D

epre

ssio

n

Sca

le;

C-G

AS¼

Ch

ild

ren

’sG

lob

alA

sses

smen

tS

cale

;C

GI¼

Cli

nic

al

Glo

bal

Imp

ress

ion

s;C

IDI-

12¼

Co

mp

osi

teIn

tern

ati

on

al

Dia

gn

ost

icIn

terv

iew

,V

ersi

on

2.1

;C

SQ¼

Cli

ent

Sa

tisf

act

ion

Qu

esti

on

na

ire;

CW

D–

A¼

CB

TA

do

lesc

ents

Co

pin

gw

ith

Dep

ress

ion

Co

urs

e;D

AS¼

Dy

sfu

nct

ion

al

Att

itu

de

Sca

le;

DIS

C2

.3¼

Dia

gn

ost

icIn

terv

iew

Sch

edu

lefo

rC

hil

dre

nV

ersi

on

2.3

;

DS

M-I

II-R¼

Dia

gnost

icand

Sta

tist

ical

Manual

of

Men

tal

Dis

ord

ers

(3rd

ed.,

Rev

.);

FA

D¼

Fa

mil

yA

sses

smen

tD

evic

e;F

EIC

S¼

Fa

mil

yE

mo

tio

na

lIn

volv

emen

ta

nd

Cri

tici

smS

cale

;

FP

E¼

Fa

mil

yP

sych

oed

uca

tio

n;

F-S

AD

S¼

Fam

ily

Sch

edu

leo

fA

ffec

tive

Dis

ord

ers

an

dS

chiz

op

hre

nia

;G

AF¼

Glo

bal

Ass

essm

ent

of

Fu

nct

ion

ing

Sca

le;

GA

RF

S¼

Glo

bal

Ass

essm

ent

of

Rel

ati

on

al

Fu

nct

ion

ing

Sca

le;

HA

M-D¼

Ha

mil

ton

Dep

ress

ion

Ra

tin

gS

cale

;H

DR

S¼

Dep

ress

ion

Ra

tin

gS

cale

;H

oN

OS

CA¼

Hea

lth

of

the

Nati

on

Ou

tco

me

Sca

les

for

Ch

ild

ren

an

dA

do

-

lesc

ents

;H

RS

D¼

Ha

mil

ton

Ra

tin

gS

cale

for

Dep

ress

ion

;IP

T-A¼

Inte

rper

son

alP

sych

oth

era

py

for

Dep

ress

edA

do

lesc

ents

;IP

T-A

ST¼

Inte

rper

son

al

Psy

cho

ther

ap

y-A

do

lesc

ent

Sk

ills

Tra

in-

ing

;K

-SA

DS

-E¼

Sch

edu

lefo

rA

ffec

tive

Dis

ord

ers

an

dS

chiz

op

hre

nia

for

Ch

ild

ren

-Ep

idem

iolo

gic

Ver

sio

n;

LIF

E¼

Lo

ng

itu

din

al

Inte

rva

lF

oll

ow

-up

Ev

alu

atio

n;

LS¼

Lif

eS

kil

ls=T

uto

rin

g;

MC

S-1

2¼

Men

tal

Hea

lth

Su

mm

ary

Sco

re;

MH

I-5¼

Men

tal

Hea

lth

Inv

ento

ry5

;N

OS¼

No

tO

ther

wis

eS

pec

ifie

d;

NP

O¼

Neg

ati

ve

Pro

ble

mO

rien

tati

on

;P

ES¼

Ple

asa

nt

Ev

ents

Sch

edu

le;

PH

CS

CS¼

Pie

rs-H

arr

isC

hil

dre

n’s

Sel

f-C

on

cep

tS

cale

;R

AD

S¼

Rey

no

lds

Ad

ole

scen

tD

epre

ssio

nS

cale

;R

CM

AS¼

Rev

ised

Ch

ild

ren

’sM

an

ifes

tA

nx

iety

Sca

le;

SA

SC

A¼

So

cial

Ad

just

men

t

Sca

lefo

rC

hil

dre

na

nd

Ad

ole

scen

ts;

SA

S-S

R¼

So

cia

lA

dju

stm

ent

Sca

le–

Sel

f-R

epo

rtv

ersi

on

;S

AS

Y¼

So

cial

Ad

just

men

tS

cale

for

Yo

uth

;S

CA

S¼

Sp

ence

Ch

ild

ren

’sA

nxi

ety

Sca

le;

SC¼

Sch

oo

lC

ou

nse

lin

g;

SE

Q¼

Sel

f-E

ffic

acy

Qu

esti

on

na

ire

for

Dep

ress

edA

do

lesc

ents

;S

IQ¼

Su

icid

al

Idea

tio

nQ

ues

tio

n;

SP

SI-

R¼

So

cia

lP

rob

lem

-So

lvin

gIn

ven

tory

–R

evis

ed;

SR

FF¼

Sel

f-R

epo

rto

fF

amil

yF

un

ctio

nin

g;S

SA

I¼

Str

uct

ure

dS

oci

al

Ad

just

men

tIn

terv

iew

;S

TA

IC¼

Sta

te-T

rait

An

xie

tyIn

ven

tory

for

Ch

ild

ren

;Y

SR¼

Yo

uth

Sel

f-R

epo

rt.

77

2002; Clarke, Rohde, Lewinsohn, Hops, & Seeley, 1999;Diamond, Reis, Diamond, Siqueland, & Isaacs, 2002;Goodyer et al., 2007; Melvin et al., 2006; Mufson,Dorta, Wickramaratne et al., 2004; Mufson, Weissman,Moreau, & Garfinkel, 1999; Rohde, Clarke, Mace,Jorgensen, & Seeley, 2004; Rossello & Bernal, 1999;Sanford et al., 2006; TADS Team, 2004) in accord withthe DSM (American Psychiatric Association, 1987, 1994).

Relative to the child domain, there were morerecently published adolescent studies, and these focusedmore on clinically referred individuals who met diagnos-tic criteria for a depressive disorder than on school-based interventions with youth with elevated depressivesymptoms. Male and female adolescents were includedin these protocols, although female participants tendedto be overrepresented. This gender pattern is consistentwith the higher rates of depression for female adoles-cents than male (American Academy of Child andAdolescent Psychiatry, 1998). Participants typicallyranged in age from 12 to 18 years, but one studyincluded those as old as 21 (Asarnow et al., 2005).

As observed with the child studies, information aboutthe ethnicity of the adolescent samples was limited, butavailable data indicated a varied sample population withthe inclusion of several typically understudied popula-tions. Some studies included predominantly Caucasianadolescents (Ackerson et al., 1998; Clarke et al., 2001;Rohde et al., 2004; TADS Team, 2004), whereas othersincluded predominantly African American (Diamondet al., 2002), Latino (Asarnow et al., 2005; Mufson,Dorta, Wickramaratne et al., 2004; Mufson et al.,1999; Young et al., 2006a), Puerto Rican (Rossello &Bernal, 1999), or Australian (Melvin et al., 2006) adoles-cents. A broad range of assessment tools were used withthe Beck Depression Inventory (Beck, Steer, & Garbin,1988; Beck, Ward, Mendelson, Mock, & Erbaugh, 1961)being the self-report measure of depressive symptomsmost often used. The Externalizing and InternalizingSymptom subscales and an Extracted Depression sub-scale of the CBCL were the most common measures ofparental perceptions of their adolescents’ functioning.Based on adolescent and=or parent interviews, the K-SADS served as the diagnostic tool most frequentlycompleted by a clinician.

All investigations included one to two experimentalconditions and=or a control condition (e.g., waitlist,treatment as usual). Consistent with the child studies,only 10 of the 18 adolescent studies included aparent=family intervention component (Clarke et al.,2005; Clarke et al., 2001; Clarke et al., 2002; Clarkeet al., 1999; Diamond et al., 2002; Goodyer et al.,2007; Melvin et al., 2006; Rohde et al., 2004; Sanfordet al., 2006; TADS Team, 2004), and 1 additional studydiscussed the importance of interventionists being sensi-tive to cultural values about the authority and respect of

parents even when parental involvement was limited tosymptom assessment (Rossello & Bernal, 1999). Whenparents were included, involvement ranged from partici-pation in a separate parent course where they reviewedwhat the adolescents were learning (Clarke et al., 2001;Clarke et al., 2002; Clarke et al., 1999; Rohde et al.,2004) to attendance at parent–adolescent sessions(Diamond et al., 2002). One study evaluated the efficacyof a family psychoeducation component added to a usualcare condition that was conducted in the adolescent’shome with all family members (Sanford et al., 2006).Interventions, conducted in individual, group, and ado-lescent–parent formats, typically ranged in length from 8to 16 sessions, although 1 study examined a brief CBTprogram that had an average of 5 completed sessions(Clarke et al., 2005). Clinicians represented various disci-plines (e.g., psychology, child psychiatry, social workers),and their experience level was broad (layperson to gradu-ate students to licensed clinicians).

Similar to the child studies, most interventions withadolescents were consistent with CBT and taught moodmonitoring, affect regulation, pleasant activities sched-uling, cognitive restructuring, social skills, communi-cation, and conflict resolution (Ackerson et al., 1998;Asarnow et al., 2005; Clarke et al., 2001; Clarke et al.,2002; Clarke et al., 1999; Goodyer et al., 2007;Kowalenko et al., 2005; Rohde et al., 2004; Rossello &Bernal, 1999; TADS Team, 2004). In addition, there isa growing literature on interpersonal psychotherapy(IPT) for adolescents and with this theoretical approach,the following issues are addressed: grief, interpersonaldispute, role transitions, interpersonal deficits, andother family and relational problems (e.g., single-parentfamilies; Mufson, Dorta, Wickramaratne et al., 2004;Mufson et al., 1999; Rossello & Bernal, 1999; Younget al., 2006a). One recently published family interventionutilized an attachment-based model to guide the processof repairing relational ruptures and rebuilding trust-worthy relationships (Diamond et al., 2002), andanother integrated medical, patient education, stress,communication, and coping models (Sanford et al.,2006). Intervention tasks in the family treatmentsinclude increasing knowledge, relational reframing,improving communication, building alliances with theadolescent and parent, addressing attachment failure,and building competency and coping.

As shown in Table 2, greater reductions in depressivesymptoms and disorders were observed in depressed andat-risk youth who participated in CBT, IPT, or attach-ment-based family therapy relative to those in treatmentcontrol conditions. Taking together all the research todate, there is some limited support for family interven-tions for depressed youth, but CBT and IPT appear tobe the most promising psychosocial interventions fordepressed adolescents at the present time. As is the case

78 DAVID-FERDON AND KASLOW

with the child intervention literature, no psychosocialintervention is clearly superior for depressed adoles-cents. In the one study that compared CBT and IPT,no significant between-group differences emerged(Rossello & Bernal, 1999).

Fourteen studies reported follow-up data with peri-ods ranging from 1 to 24 months postintervention(Ackerson et al., 1998; Clarke et al., 2005; Clarke et al.,2001; Clarke et al., 2002; Clarke et al., 1999; Diamondet al., 2002; Kowalenko et al., 2005; Melvin et al.,2006; Mufson, Dorta, Wickramaratne et al., 2004;Rohde et al., 2004; Rossello & Bernal, 1999; Sanfordet al., 2006; Sheffield et al., 2006; Young et al., 2006a).Relative to the child studies, slightly more informationis available on the long-term efficacy of treatments foradolescents. However, for most studies with follow-updata, the control group was not available. Thus, it wasonly possible to ascertain if treatment gains were main-tained, and the continuation of between-group differ-ences could not be examined. When both treatmentand comparisons groups were available at follow-up,some studies found that between-group differences atpostintervention were not maintained over time (Boriset al., 2000; Rohde et al., 2004). However, there wereexceptions to these results, such as findings by threeinvestigatory teams that youth who participated in theactive interventions continued to be less depressed atfollow-up than youth in the control group (Clarkeet al., 2001; Mufson, Dorta, Wickramaratne et al.,2004; Young et al., 2006a). Another study examiningthe benefits of booster sessions found that youth whowere still depressed at the end of the active treatmentphase experienced continued recovery when randomlyassigned to receive booster sessions (Clarke et al.,1999). Unfortunately, however, the booster sessionswere not associated with reduced rates of reoccurrenceof a unipolar depressive episode. These findings suggestthat for youth who do not respond fully to short-termpsychosocial interventions, continued periodic treat-ment contact may be helpful in further reducing depress-ive symptoms.

Finally, Table 2 indicates that 10 of the 18 studiesmeet the Nathan and Gorman (2002) criteria for a Type1 study, as they were methodologically sophisticated,randomized controlled trials with reasonable samplesizes. These research designs included clearly articulatedinclusion and exclusion criterion and used appropriatediagnostic information gleaned from state-of-the-artassessment batteries. The other adolescent studies wereType 2 and had similar limitations to those noted pre-viously for the child studies. The fact that 56% of theadolescent studies were Type 1 is in stark contrast tothe child studies, in which all of the studies were classi-fied as Type 2 secondary to some minor methodologicalflaws.

EVIDENCE-BASED TREATMENTS

To ascertain the evidence-based status of child and ado-lescent psychosocial treatments for depression accordingto the Task Force on Promotion and Dissemination ofPsychological Procedures criteria (Chambless et al.,1998; Chambless & Hollon, 1998; Chambless et al.,1996; Lonigan et al., 1998), we included all of the studiespublished to date in our evaluation, and summaryreviews are presented in Tables 3 and 4. Based on theguidelines, we did not include (a) studies that failed todemonstrate that an active intervention was more likelyto be associated with improvements in depression than acontrol or comparison group, even if the interventionbeing examined reduced depressive symptoms (Butler,Miezitis, Friedman, & Cole, 1980; Clarke et al., 2002;Goodyer et al., 2007; Liddle & Spence, 1990; Vostanis,Feehan, Grattan, & Bickerton, 1996b); (b) open trials,which are available for both CBT with children(Hannan, Rapee, & Hudson, 2000; Kerfott, Harrington,Harrington, Rogers, & Verduyn, 2004; Szigethy et al.,2004) and IPT-A with adolescents (Santor & Kusumaker,2001) and offer promising results; (c) programs that selec-ted for children with depressive symptoms but where theintervention was not targeted toward the ameliorationof these symptoms (Fristad, Arnett, & Gavazzi, 1998;Fristad, Gavazzi, & Soldano, 1998; Fristad, Goldberg-Arnold, & Gavazzi, 2003); and (d) interventions that didnot demonstrate treatment effects immediately followingthe completion of active treatment, even if group differ-ences were found at long-term follow-up (Gillham,Reivich et al., 2006; Muratori et al., 2003).

Information about the evidence-based treatmentclassification (e.g., probably efficacious, well-estab-lished) of the interventions is presented in three forms:(a) evidence-based treatment classification for specificintervention programs within a given theoretical orien-tation (specific program), (b) evidence-based classi-fication for overall modality within a given theoreticalorientation (modality), and (c) evidence-based classi-fication for all treatment protocols that fall under therubric of a given theoretical orientation or approach(theoretical orientation). We selected to classify interven-tions by their specific program, as well as their modalityand theoretical orientation, to enhance the clinical rel-evance of this work and to provide youth, families,and therapists a broader perspective on the evidence-based support of different programs and approachesto better meet varying individual and setting needs.The specific program, modality, and theoretical orien-tation categories utilized were independently judged bythe two authors of this review, and a consensus wasreached on each determination. The authors formedtheir categorizations based on information presented

TREATMENTS FOR YOUTH DEPRESSION 79

TA

BLE

3

Sum

mary

of

Evid

ence-B

ased

Tre

atm

ent

Sta

tus

of

Child

hood

Depre

ssio

nT

reatm

ent

Stu

die

sby

Theore

ticalO

rienta

tion,

Modalit

y,and

Speci

fic

Pro

gra

m

Inte

rven

tion

Tre

atm

ent

Stu

die

sS

upport

ing

Sig

nif

icant

Ben

efit

Rel

ati

veto

Alt

ernat

ive

Tre

atm

ent

or

No

-Tre

atm

ent

Co

ntr

ol

Evi

den

ce-B

ase

dS

tatu

s

Co

gn

itiv

e-B

eha

vio

ral

Tre

atm

ent

Sta

rk,

Rey

no

lds,

&K

asl

ow

(19

87)

;K

ah

n,

Keh

le,

Jen

son

,&

Cla

rk(1

99

0);

Sta

rk,

Ro

use

,&

Liv

ingst

on

(19

91

);Ja

yco

x,

Rei

vich

,G

illh

am

,&

Sel

igm

an

(19

94

);G

illh

am

,R

eiv

ich

,Ja

yco

x,

&S

elig

ma

n(1

99

5);

Wei

sz,

Th

urb

er,

Sw

een

ey,

Pro

ffit

t,&

LeG

agn

ou

x(1

99

7);

Asa

rno

w,

Sco

tt,

&M

intz

(20

02)

;

Yu

&S

elig

man

(2002)

;N

elso

n,

Barn

ard

,&

Cain

(2003)

;R

ob

erts

,K

an

e,

Th

om

son

,B

ish

op

,&

Ha

rt(2

00

3)

Wel

l-E

sta

bli

shed

Gro

up

,ch

ild

on

lyS

tark

,R

eyn

old

s,&

Ka

slo

w(1

98

7);

Ka

hn

,K

ehle

,Je

nso

n,

&C

lark

(19

90

);

Jayc

ox

,R

eiv

ich

,G

illh

am

,&

Sel

igm

an

(19

94

);G

illh

am

,R

eivi

ch,

Jay

cox

,&

Sel

igm

an

(19

95)

;W

eisz

,T

hu

rber

,S

wee

ney

,P

roff

itt,

&L

eGa

gn

ou

x(1

99

7);

Yu

&S

elig

ma

n(2

00

2);

Ro

ber

ts,

Ka

ne,

Th

om

son

,B

ish

op

,&

Ha

rt(2

00

3)

Wel

l-E

sta

bli

shed

Ch

ild

gro

up

plu

sp

are

nt

com

po

nen

tS

tark

,R

ou

se,

&L

ivin

gst

on

(19

91

);A

sarn

ow

,S

cott

,&

Min

tz(2

00

2)W

ell-

Est

ab

lish

ed

Pare

nt-

chil

dN

elso

n,

Barn

ard

,&

Cain

(2003)

Ex

per

imen

tal

Indiv

idual

video

self

-monit

ori

ng

Ka

hn

,K

ehle

,Je

nso

n,

&C

lark

(19

90)

Ex

per

imen

tal

Co

pin

gw

ith

Dep

ress

ion

Ka

hn

,K

ehle

,Je

nso

n,

&C

lark

(19

90)

Ex

per

imen

tal

Pen

nP

reve

nti

on

Pro

gra

min

cludin

gcu

ltur

all

yre

leva

nt

modif

icati

ons

as

seen

inth

eP

enn

Opti

mis

mP

rogra

m

Jayc

ox

,R

eivi

ch,

Gil

lha

m,

&S

elig

ma

n(1

99

4);

Gil

lha

m,

Rei

vich

,Ja

yco

x,

&

Sel

igm

an

(19

95)

;Y

ua

nd

Sel

igm

an

(20

02)

;R

ob

erts

,K

an

e,T

ho

mso

n,

Bis

ho

p,

&H

art

(20

03)

Pro

babl

yE

ffic

aci

ous

Pri

ma

rya

nd

Sec

on

dar

yC

on

tro

lE

nha

nce

men

tT

rain

ing

Pro

gra

m

Wei

sz,

Th

urb

er,

Sw

een

ey,

Pro

ffit

t,&

LeG

agn

ou

x(1

99

7)

Ex

per

imen

tal

Sel

f-C

on

tro

lT

hera

py

Sta

rk,

Rey

no

lds,

&K

asl

ow

(19

87)

;S

tark

,R

ou

se,

&L

ivin

gst

on

(19

91

)P

robabl

yE

ffic

aci

ous

Str

ess-

Bu

ster

sA

sarn

ow

,S

cott

,&

Min

tz(2

00

2)E

xp

erim

enta

l

No

ndir

ecte

dS

up

po

rt=P

sych

oed

uca

tio

na

lP

feff

er,

Jia

ng

,K

ak

um

a,

Hw

an

g,

&M

etsc

h(2

00

2)E

xp

erim

enta

l

Ch

ild

gro

up

plu

sp

are

nt

inte

rven

tion

Pfe

ffer

,Ji

an

g,

Ka

ku

ma

,H

wa

ng

,&

Met

sch

(20

02)

Ex

per

imen

tal

Ber

eave

men

tG

roup

Inte

rven

tion

Pfe

ffer

,Ji

an

g,

Ka

ku

ma

,H

wa

ng

,&

Met

sch

(20

02)

Ex

per

imen

tal

Fam

ily

Sy

stem

sT

row

ell,

Joff

e,C

am

pb

ell,

Cle

men

te,

Alm

qv

ist,

So

inin

enet

al.

(20

07)

Ex

per

imen

tal

Fam

ily

inte

rven

tio

nT

row

ell,

Joff

e,C

am

pb

ell,

Cle

men

te,

Alm

qv

ist,

So

inin

enet

al.

(20

07)

Ex

per

imen

tal

Sys

tem

sIn

teg

rati

veF

am

ily

The

rap

yT

row

ell,

Joff

e,C

am

pb

ell,

Cle

men

te,

Alm

qv

ist,

So

inin

enet

al.

(20

07)

Ex

per

imen

tal

Beh

avi

or

Ther

ap

yK

ah

n,

Keh

le,

Jen

son

,&

Cla

rk(1

99

0);

Kin

g&

Kir

sch

enb

aum

(19

90)

Pro

bably

Eff

icaci

ous

Gro

up

,ch

ild

on

ly,

rela

xa

tion

tra

inin

gK

ah

n,

Keh

le,

Jen

son

,&

Cla

rk(1

99

0)E

xp

erim

enta

l

Ch

ild

gro

up

plu

sp

are

nt=

tea

cher

consu

lta

tio

n,

soci

al

skil

ls

tra

inin

g

Kin

g&

Kir

sch

enb

au

m(1

99

0)E

xp

erim

enta

l

Wis

con

sin

Ear

lyIn

terv

enti

on

Kin

g&

Kir

sch

enb

au

m(1

99

0)E

xp

erim

enta

l

No

te:

Bo

ldan

dU

nd

erli

ned

isth

eore

tica

lo

rien

tati

on

;b

old

ism

od

ali

ty;

itali

cize

dis

spec

ific

inte

rven

tio

n.

Stu

die

sare

list

edch

ron

olo

gic

all

y.

80

in the studies, manuals, and=or communications withinvestigators.

A program was identified as ‘‘specific’’ when theinvestigators either provided sufficient information inthe study about the session-by-session content andintervention procedures and=or provided references tothe specific manualized intervention they implementedand evaluated. If a specific program was evaluated inmore than one study, it is noted in this review anymodifications in content and=or implementation proce-dures the investigators reported using. In instanceswhen it was unclear whether a specific interventionevaluated by one set of investigators differed from aspecific intervention implemented by another set ofinvestigators (Mufson, Dorta, Wickramaratne et al.,2004; Mufson et al., 1999; Rossello & Bernal, 1999),the investigators were contacted to clarify the speci-ficity of their program.

The intervention modality and theoretical orientationtypically were clearly stated by investigators. For inter-ventions where the treatment modality only includedindividual or group sessions with children or adolescentsand no parent component was offered (i.e., identified asGroup, child only; Group, adolescent only; Individual),investigators usually clearly stated that parents wereinvolved only to provide consent for participation, toreceive information about the length and format of theintervention, and=or to participate in the assessment-only portion of the study. In such studies, it is possiblethat investigators did not always report the presence oflimited informal contact or occasional individual meet-ings between the interventionists and parents that wereoutside the treatment protocol but clinically necessary.The theoretical frameworks for interventions examinedin this review were also typically directly stated by inves-tigators. Investigators usually presented this infor-mation by directly stating they were, for instance,evaluating a cognitive behavioral intervention and=orproviding references that clearly indicated the theoreti-cal underpinnings of their intervention.

In Tables 3 and 4, child (Kahn, Kehle, Jenson, &Clark, 1990; Nelson et al., 2003) and adolescent(Asarnow et al., 2005; Brent et al., 1997; Fine & Carlson,1991; Reynolds & Coats, 1986; Rossello & Bernal, 1999;TADS Team, 2004) studies may be listed under amodality and theoretical orientation indicating that theactive treatment was found to have a significant benefitrelative to the control condition, but the study may notbe listed with a specific program. This omission occurswhen the intervention’s treatment modality and theor-etical approach are clearly identified but the specificprogram is not. When investigators report that theirunnamed specific program was developed based in parton another program that is identifiable, this informationis presented next.

For a specific program to be deemed well-establishedor probably efficacious, it must meet the Task Force cri-teria previously summarized. For our review, a modalityor theoretical orientation is deemed well-established if itmeets the Task Force criteria for well-established plus atleast one of the specific programs under the rubric isclassified as at least probably efficacious. For ourreview, for a modality or theoretical orientation to bedeemed probably efficacious, it must meet the TaskForce criteria for probably efficacious, without ouradditional requirement that at least one of the specificprograms under the rubric is classified as at least prob-ably efficacious. Interventions are deemed experimentalif they demonstrate significant improvements in youth’sfunctioning relative to a treatment control condition butare examined by only one study.

Child Studies

Table 3 presents the evidence-based treatment status ofthe child protocols.

Specific program classification schema. Consistentwith the 1998 review (Kaslow & Thompson, 1998), fromthe present review we conclude that the Self-ControlTherapy of Stark and colleagues (Stark et al., 1987; Starket al., 1991) meets criteria for a probably efficaciouspsychosocial intervention for depressed children. Inaddition, the Penn Prevention Program, which includesculturally relevant modifications associated with thePenn Optimism Program, can be deemed probably effi-cacious (Cardemil et al., 2002; Gillham & Reivich,1999; Gillham, Reivich, Jaycox, & Seligman, 1995;Jaycox, Reivich, Gillham, & Seligman, 1994; Robertset al., 2003; Yu & Seligman, 2002). A brief review ofthese two specific protocols and the evidence-basedcriteria met by these programs is presented next.

First, Self-Control Therapy is a school-basedintervention based on a cognitive-behavioral model andteaches self-management skills (e.g., self-monitoring,self-evaluating, self-consequating, causal attributions).Stark et al. (1987) weighed the relative efficacy of a12-session group intervention involving Self-ControlTherapy, behavioral problem-solving therapy, and awaitlist control for 29 fourth through sixth graders withelevated depressive symptoms. The behavioral problem-solving therapy consisted of education, self-monitoringof pleasant events, and problem solving directed towardimproving social behavior. Children in the waitlist con-trol condition were required to wait the 5-week durationof the active interventions before being able to receivethose services; however, during this waiting period, theyhad access to services as usual.

Postintervention within-group analyses found thatchildren in both active interventions self-reported fewer

TREATMENTS FOR YOUTH DEPRESSION 81

symptoms of depression and anxiety, whereas thecontrol group reported minimal change. Between-groupanalyses revealed that children in the self-control con-dition reported significantly fewer depressive symptomson the CDI than children in the waitlist control con-dition at posttesting. Furthermore, group comparisonsof pre- and posttreatment responses on the ReynoldsChildren Depression Scale (Reynolds, 1989) and onthe Children’s Depression Rating Scale–Revised(Poznanski et al., 1984) approached significance in favorof the self-control condition, but none of these between-group differences were significant. Significant between-group differences did not merge on mother ratings.

At 8-week follow-up, 88% of the children in the self-control condition and 67% of those in the behavioralproblem-solving condition obtained CDI scores belowthe cutoff for depression, and none of the participantsin either experimental group met criteria for clinicallysignificant levels of depressive symptoms according totheir responses on the Children’s Depression RatingScale–Revised. At follow-up, children in the self-controlcondition were significantly less depressed on the Chil-dren’s Depression Rating Scale–Revised and reportedsignificantly better self-concepts on the CoopersmithSelf-Esteem Inventory (Coopersmith, 1967, 1975) thanchildren in the behavioral problem-solving group. Incontrast, the behavioral problem-solving group wasrated by mothers on the CBCL as significantly lesssocially withdrawn and as having fewer internalizingsymptoms than the self-control group. No between-group differences in depression levels were noted onthe CDI or on children’s reports of anxiety. Takentogether, results reveal that both experimental interven-tions were relatively successful in reducing symptomsof depression in a school setting. For the most part,findings regarding the comparison of the self-controland behavioral problem-solving group interventionswere equivocal. However, the pattern suggested thatthe self-control intervention was more beneficial to thechildren.

Based on results from the aforementioned study,Stark et al. (1991) evaluated an expanded version ofSelf-Control Therapy with 26 fourth through seventhgraders who endorsed high levels of depressive symp-toms. This research, also conducted in a school setting,offers only a partial replication of Stark’s earlier work(Stark et al., 1987), as the self-control therapies testedin the two different treatment outcome studies weresimilar but not identical, and there was no behavioralproblem-solving condition in this second study. Theexperimental intervention in this second study was a24- to 26-session CBT program that consisted of self-control and social skills training, assertiveness training,relaxation training and imagery, and cognitive restruc-turing. This treatment was compared to a traditional

counseling condition designed to control for nonspecificelements of the intervention. Monthly family meetingsfor the CBT group encouraged parents to assist theirchildren in applying their new skills and to increasethe frequency of positive family activities. Monthlyfamily sessions for the traditional counseling conditionaddressed improving communication and increasingpleasant family events. Postintervention and 7-monthfollow-up assessments revealed significant decreases inself-reported depressive symptoms for both groups ofchildren. At postintervention, youth in the CBT con-dition self-reported significantly fewer depressive symp-toms on a semistructured interview and endorsedsignificantly fewer depressive cognitions than childrenin the control condition.

In sum, these two outcome studies (Stark et al., 1987;Stark et al., 1991) suggest that Self-Control Therapymeets the criteria for a probably efficacious treatment,as client characteristics were specified, a treatment man-ual was used, and two adequate group design studiesdemonstrated that a self-control program was superiorto a placebo psychological treatment (i.e., treatment asusual; traditional counseling). However, the self-controlinterventions used in the two studies were not identicalin terms of length of treatment or material covered. Inaddition, the examination of this intervention has beenlimited to one research group. With these qualifications,this specific intervention is best classified as a probablyefficacious evidence-based treatment.

The second specific psychosocial intervention we wantto review is the Penn Prevention Program (Gillham &Reivich, 1999; Gillham, Reivich et al., 2006; Gillhamet al., 1995; Jaycox et al., 1994; Roberts et al., 2003),and its culturally relevant modification the Penn Opti-mism Program (Yu & Seligman, 2002). The programhas also been referred to as the Penn Resiliency Program.This specific intervention is based on a cognitive-behavioral model and is designed to address depressivesymptoms among at-risk 10- to 15-year-olds in schoolsettings. This group intervention has two components:cognitive and social problem solving. The cognitivecomponent teaches children to identify negative beliefs,to evaluate the evidence for them, and to generatemore realistic alternatives. It also involves explanatorytraining (i.e., attribution retraining), in which youngpeople are taught to identify pessimistic explanationsand generate alternative more optimistic and realisticexplanations. The social problem-solving componentfocuses on goal setting, perspective taking, informationgathering, generating alternatives for action, decisionmaking, and self-instruction. The program educates chil-dren about strategies for effectively dealing with familyconflict and other stressors, such as decatastrophizing,distancing and distraction, relaxation training, andenhancing their social support network.

82 DAVID-FERDON AND KASLOW

Jaycox et al. (1994) compared at-risk 10- to 13-year-olds assigned randomly to the 12 session Penn Preven-tion Program or to a no-treatment control condition.Random assignment was based on school rather thanchild. Children were identified as at risk if they hadelevated self-reported depressive symptoms on the CDIand elevated parental conflict scores on the Child’sPerception Questionnaire (Emery & O’Leary, 1982).

At postintervention, those in the experimental con-dition self-reported significantly greater reductions indepressive symptoms on the CDI and ReynoldsChildren Depression Scale than those in the controlcondition. In a related vein, fewer children in the experi-mental group as compared to the control group self-reported moderate or more severe depressive symptomsat postintervention. These group differences based onself-report were maintained at 6-month follow-up. How-ever, no significant between-group differences emergedwith regards to the major treatment target of explana-tory style as assessed by the Children’s AttributionalStyle Questionnaire (CASQ; Seligman, Peterson,Kaslow, Tanenbaum, & Abramson, 1984). There wereno significant between-group differences on parents’reports of internalizing and externalizing problems onthe CBCL at postintervention, but at 6-month follow-up children who received the experimental treatmenthad significantly fewer conduct, but not internalizing,problems than their peers in the control group. Further,compared to children in the control group, those in theexperimental group manifested significantly greaterimprovements in classroom behavior at postinterven-tion. No follow-up teacher data were available. Inter-vention effects were maintained at the 2-year (Gillhamet al., 1995), but not the 3-year (Gillham & Reivich,1999), follow-ups.

Yu and Seligman (2002), using a culturally relevantmodification of the Penn Prevention Program for chil-dren in China, also implemented the intervention inschools. Modifications included a reduction in thenumber of sessions (from 12 to 10 sessions), culturallyrelevant alterations to the assertiveness training compo-nent, and protocol implementation by teachers ratherthan mental health professionals. The program wascompared to a nonintervention control group. The sam-ple consisted of 8- to 15-year-olds in China selectedbased on elevated levels of depressive symptoms onthe CDI and of family conflict on the Cohesion andConflict subscales of the Family Environment Scale(Moos & Moos, 1981). At posttreatment, compared tochildren in the control condition, those in the activetreatment showed significantly greater reductions indepressive symptoms on the CDI and greater increasesin their optimistic explanatory styles on the CASQ.Treatment effects were maintained at 3- and 6-monthfollow-up assessments.

Roberts et al. (2003) also used the Penn PreventionProgram in a school setting and compared it to a usualcare control condition that included symptom monitor-ing plus a regular health curriculum. Participants were11- to 13-year-olds in rural Australia who had elevateddepressive symptoms on the CDI. Although childrenin both conditions self-reported decreases in depressivesymptoms at postintervention, there were no significantbetween-group differences. In contrast, parents of youthin the experimental group reported on the CBCL thattheir children had significantly fewer internalizingproblems at postintervention when compared to theparents’ reports of children in the control group. At6-month follow-up, there were no significant groupdifferences according to child or parent reports.

The Penn Resiliency Program also has been evaluatedfor its effectiveness in reducing depressive symptomswhen delivered in a primary care setting (Gillham,Hamilton et al., 2006). Participants were recruited froman HMO patient care directory and announcements inpediatrician offices. Children who were 11- to 12-year-olds and at risk due to elevated depressive symptomson the CDI were assigned randomly to the 12 sessiongroup intervention or to a usual care control group.At postintervention, children in the two conditions didnot differ significantly in their depressive symptoms,and no intervention effect was found for depressivedisorders. At follow-up, children in the interventioncondition showed significant improvements in theirexplanatory style for positive events on the CASQ,and girls in the intervention condition showed signifi-cantly fewer depressive symptoms on the CDI.

Recognizing parental depression as a risk factor foryouth, the effect of the Penn Resiliency Program hasbeen examined when it includes a supplemental parentgroup that teaches parents the same skills their childrenare learning in the school groups (Gillham, Reivich et al.,2006). Children, deemed at risk because of elevateddepressive and anxiety symptoms, and their parentswere recruited from schools. Families were randomlyassigned to the group treatment and a usual care controlgroup. At the conclusion of the active intervention, chil-dren in the Penn Resiliency Program condition did notdiffer significantly in their depressive or anxiety symp-toms relative to their peers in the control condition.However, intervention effects became evident at the6- and 12-month follow-ups when youth in the activetreatment did show significantly fewer symptoms onthe CDI and RCMAS than control children.

Overall, outcome studies evaluating the Penn Preven-tion Program (Gillham, Reivich et al., 2006; Gillhamet al., 1995; Jaycox et al., 1994; Roberts et al.,2003; Yu & Seligman, 2002) provide sufficient infor-mation to deem this intervention a probably efficacioustreatment in reducing depressive symptoms, as client

TREATMENTS FOR YOUTH DEPRESSION 83

characteristics were specified, a treatment manual wasused, and at least two adequate group design studiesdemonstrated that the program was more efficaciousthan comparison conditions. Although this programhas been evaluated by different investigatory teams, itfalls short of being a well-established interventionbecause only one study has demonstrated it superiorto a psychological placebo or alternative treatment con-dition and not the required two studies. The other stu-dies have found the intervention more effective than ano-treatment control group. An effort has been madeto fill this gap by designing a study to compare the PennResiliency Program to another active treatment (PennEnhancement Program); however, because of recruit-ment challenges, this study included all consented chil-dren rather than an at-risk sample (Gillham et al.,2007). At postintervention, the universal Penn Resili-ency Program was not more effective than the otheractive intervention in reducing depressive symptoms,but its potential benefit became more evident over time,and there were some differences between schools.

Other specific psychosocial interventions (i.e., Copingwith Depression, Primary and Secondary ControlEnhancement Training Program, Stress-Busters,Bereavement Group Intervention, Systems IntegrativeFamily Therapy; Wisconsin Early Intervention) havebeen found to be efficacious in reducing depressivesymptoms relative to control conditions (Asarnowet al., 2002; Kahn et al., 1990; King & Kirschenbaum,1990; Pfeffer et al., 2002; Trowell et al., 2007; Weisz,Thurber, Sweeney, Proffitt, & LeGagnoux, 1997). How-ever, these specific interventions are deemed experi-mental because of the limited extent to which theyhave been reviewed to date.

Modality classification schema. Given the numberof protocols that fall under the CBT rubric, the studieswere divided by modality (group–child only, child groupplus parent, parent–child, individual video self-monitor-ing). Table 3 reveals that CBT modalities of the group–child only (Gillham et al., 1995; Jaycox et al., 1994;Kahn et al., 1990; Roberts et al., 2003; Stark et al.,1987; Weisz et al., 1997; Yu & Seligman, 2002) and childgroup plus parent component (Asarnow et al., 2002;Stark et al., 1991) are the only modalities that arewell-established interventions.

At least two well-conducted studies from differentinvestigative teams found these modalities more effectivein reducing depressive symptoms than a psychologicalplacebo or alternative treatment, and at least one specificintervention (Penn Prevention Program, Self-ControlTherapy) included in each modality meets criteria for aprobably efficacious treatment. Because only one CBTstudy utilized the parent–child modality (Nelson et al.,2003) and only one CBT study examined individual

video self-monitoring (Kahn et al., 1990), theseapproaches are deemed experimental at the present time.The individual video self-monitoring modality includedvideo taping participants behaving in a nondepressedway and then having them watch the video duringtreatment sessions. For the other theoretical orientations(e.g., nondirected support=psychoeducational, systemstherapy, behavior therapy), different modalities havebeen examined on a limited basis; accordingly, they arealso deemed experimental.

Theoretical orientation classification schema. Usingtheoretical orientation broadly as the classificationrubric when considering the quality and quantity ofpsychosocial interventions for depressed youth, weconclude that CBT is a well-established intervention(Asarnow et al., 2002; Gillham et al., 1995; Jaycoxet al., 1994; Kahn et al., 1990; Nelson et al., 2003;Roberts et al., 2003; Stark et al., 1987; Stark et al.,1991; Weisz et al., 1997; Yu & Seligman, 2002). Thisdecision is based on the fact that there are more thantwo studies demonstrating that CBT is more effectivethan psychological placebo or alternative treatmentcontrols in improving children’s affective functioning.In addition, the studies have been conducted by differ-ent investigators and at least two of the studies haveadequate sample size, specify sample characteristicsclearly, and use treatment manuals. Finally, at leastone of the specific program protocols that falls underCBT was at least probably efficacious (i.e., Penn Preven-tion Program, Self-Control Therapy).

Although CBT interventions have received the mostattention, other broad theoretical approaches aregaining support. The broad theoretical classification ofbehavior therapy meets criteria for a probably effi-cacious intervention (Kahn et al., 1990; King &Kirschenbaum, 1990; Stark et al., 1987). Behavior ther-apy in these studies included education, self-monitoringof pleasant events, problem solving to improve socialbehavior, and progressive relaxation. This theoreticalapproach has been examined by at least two differentinvestigative teams who have found children participat-ing in behavior therapy have greater reductions indepressive symptoms than children in control con-ditions, and the studies under this rubric specify clientcharacteristics and use treatment manuals. However,because each specific behavior therapy interventionsis experimental, it is premature to view this theoreticalapproach as well-established. One recent study usingthe theoretical approach of nondirected support=psychoeducational (Pfeffer et al., 2002) found greaterimprovements among children in the active interventionrelative to the control condition, but this approach isexperimental because of its limited examination to date.An examination of an intervention based on systems

84 DAVID-FERDON AND KASLOW

TA

BLE

4

Sum

mary

of

Evi

dence-B

ased

Tre

atm

ent

Sta

tus

of

Adole

scent

Depre

ssio

nT

reatm

ent

Stu

die

sby

Theore

ticalO

rienta

tion,

Modalit

y,

and

Specific

Pro

gra

m

Inte

rven

tio

n

Tre

atm

ent

Stu

die

sS

uppor

ting

Sig

nif

icant

Ben

efit

Rel

ati

veto

Alt

ernat

ive

Tre

atm

ent

or

No

-Tre

atm

ent

Co

ntr

ol

Evi

den

ce-B

ase

dS

tatu

s

Beh

avi

or

Th

era

py

Rey

no

lds

&C

oa

ts(1

98

6)E

xp

erim

enta

l

Gro

up

,a

do

lesc

ent

on

ly,

rela

xa

tion

tra

inin

gR

eyn

old

s&

Co

ats

(19

86)

Ex

per

imen

tal

Co

gn

itiv

e-B

ehavi

ora

lT

rea

tmen

tR

eyn

old

s&

Co

ats

(19

86)

;L

ewin

soh

n,

Cla

rke,

Ho

ps,

&A

nd

rew

s(1

99

0);

Cla

rke

eta

l.(1

99

5);

Lew

inso

hn

,C

lark

e,R

oh

de,

Ho

ps,

&S

eele

y(1

99

6);

Wo

od

,H

arr

ing

ton

,&

Mo

ore

(19

96)

;B

ren

t

eta

l.(1

99

7);

Ack

erso

n,

Sco

gin

,M

cKen

dre

e-S

mit

h,

&L

yma

n(1

99

8);

Cla

rke,

Ro

hd

e,

Lew

inso

hn

,H

op

s,&

See

ley

(19

99

);R

oss

ello

&B

ern

al

(19

99)

;C

lark

eet

al.

(20

01)

;R

oh

de,

Cla

rke,

Ma

ce,

Jorg

ense

n,

&S

eele

y(2

00

4);

TA

DS

Tea

m(2

00

4);

Asa

rno

wet

al.

(20

05)

;

Ko

wale

nk

oet

al.

(2005)

;M

elvin

etal.

(2006)

Wel

l-E

sta

bli

shed

Gro

up

,a

do

lesc

ent

on

lyR

eyn

old

s&

Co

ats

(19

86)

;L

ewin

soh

n,

Cla

rke,

Ho

ps,

&A

nd

rew

s(1

99

0);

Cla

rke

eta

l.(1

99

5);

Lew

inso

hn

,C

lark

e,R

oh

de,

Ho

ps,

&S

eele

y(1

996)

;C

lark

e,R

oh

de,

Lew

inso

hn

,H

op

s,&

See

ley

(19

99

);K

ow

ale

nk

oet

al.

(20

05)

Wel

l-E

sta

bli

shed

Ad

ole

scen

tg

roup

plu

sp

are

nt

com

po

nen

tL

ewin

soh

n,

Cla

rke,

Ho

ps,

&A

nd

rew

s(1

990)

;L

ewin

soh

n,

Cla

rke,

Ro

hd

e,H

op

s,&

See

ley

(1996);

Cla

rke,

Ro

hd

e,L

ewin

soh

n,

Ho

ps,

&S

eele

y(1

99

9);

Cla

rke

eta

l.(2

00

1);

Ro

hd

e,C

lark

e,M

ace

,

Jorg

ense

n,

&S

eele

y(2

00

4)

Pro

bably

Eff

icaci

ous

Indiv

idual

Wo

od

,H

arr

ingt

on

,&

Mo

ore

(1996);

Ro

ssel

lo&

Ber

nal

(1999)

Pro

bably

Eff

icaci

ous

Indiv

idual

plu

spare

nt=

fam

ily

com

po

nen

tB

ren

tet

al.

(1997)

;T

AD

ST

eam

(2004)

;M

elvin

etal.

(2006)

Pro

bably

Eff

icaci

ous

Sel

f-d

irec

ted

bib

lioth

era

py

Ack

erso

n,

Sco

gin

,M

cKen

dre

e-S

mit

h,

&L

ym

an

(19

98

)E

xp

erim

enta

l

CB

Ten

ha

nce

dp

rim

ary

care

serv

ices

Asa

rno

wet

al.

(20

05)

Ex

per

imen

tal

Ad

ole

scen

tsC

op

ing

wit

hD

epre

ssio

nL

ewin

soh

n,

Cla

rke,

Ho

ps,

&A

nd

rew

s(1

990)

;C

lark

eet

al.

,(1

995);

Lew

inso

hn

,C

lark

e,R

oh

de,

Ho

ps,

&S

eele

y(1

996)

;C

lark

e,R

oh

de,

Lew

inso

hn

,H

op

s,&

See

ley

(1999)

;C

lark

eet

al.

(2001);

Ro

hd

e,C

lark

e,M

ace

,Jo

rgen

sen

,&

See

ley

(20

04)

Pro

bably

Eff

icaci

ous

Ad

ole

scen

tsC

op

ing

wit

hE

mo

tio

ns

Ko

wa

len

ko

eta

l.(2

00

5)

Ex

per

imen

tal

Dep

ress

ion

Tre

atm

ent

Pro

gra

mm

eW

oo

d,

Ha

rrin

gto

n,

&M

oo

re(1

99

6)

Ex

per

imen

tal

Fee

lin

gG

oo

d(

Bu

rns,

19

80)

Bib

lio

ther

ap

yA

cker

son

,S

cog

in,

McK

end

ree-

Sm

ith

,&

Ly

ma

n(1

99

8)

Ex

per

imen

tal

Tim

efo

ra

Fu

ture

-Ad

ole

scen

tD

epre

ssio

n

Pro

gra

m

Mel

vin

etal.

(2006a

)E

xp

erim

enta

l

Inte

rper

son

al

Psy

cho

ther

ap

yM

ufs

on

,W

eiss

ma

n,

Mo

rea

u,

&G

arf

ink

el(1

99

9);

Ro

ssel

lo&

Ber

na

l(1

99

9);

Mu

fso

net

al.

(20

04

);

Yo

un

g,

Mu

fso

n,

&D

av

ies

(20

06a

)

Wel

l-E

sta

bli

shed

Indiv

idual

Mu

fso

n,

Wei

ssm

an

,M

ore

au

,&

Garf

ink

el(1

999)

;R

oss

ello

&B

ern

al

(1999);

Mu

fso

net

al.

(2004)

Wel

l-E

sta

bli

shed

Gro

up

,a

do

lesc

ent

on

lyY

ou

ng

,M

ufs

on

,&

Da

vie

s(2

00

6a)

Ex

per

imen

tal

Inte

rper

sonal

Psy

chot

her

apy

for

Dep

ress

ed

Ad

ole

scen

ts

Mu

fso

n,

Wei

ssm

an

,M

ore

au

,&

Garf

ink

el(1

999)

;M

ufs

on

etal.

(2004)

Pro

bably

Eff

icaci

ous

Inte

rper

sonal

Psy

choth

erap

y-A

dol

esce

nt

Sk

ills

Tra

inin

g

Yo

un

g,

Mu

fso

n,

&D

av

ies

(20

06a

)E

xp

erim

enta

l

No

ndir

ecte

dS

up

port

Fin

e,F

ort

h,

Gil

ber

t,&

Ha

ley

(19

91)

Ex

per

imen

tal

Gro

up

,a

do

lesc

ent

on

lyF

ine,

Fo

rth

,G

ilb

ert,

&H

ale

y(1

99

1)

Ex

per

imen

tal

Fa

mil

yS

yst

ems

Theo

ryD

iam

on

d,

Rei

s,D

iam

on

d,

Siq

uel

an

d,

&Is

aa

cs(2

00

2)E

xp

erim

enta

l

Pare

nt-

adole

scen

tD

iam

on

d,

Rei

s,D

iam

on

d,

Siq

uel

an

d,

&Is

aa

cs(2

00

2)E

xp

erim

enta

l

Att

ach

men

t-B

ase

dF

amil

yT

her

apy

Dia

mo

nd

,R

eis,

Dia

mo

nd

,S

iqu

ela

nd

,&

Isa

acs

(20

02)

Ex

per

imen

tal

No

te:

Bo

lda

nd

Un

der

lin

edis

theo

reti

cal

ori

enta

tio

n;

bo

ldis

mo

dal

ity

;it

ali

cize

dis

spec

ific

inte

rven

tio

n.

Stu

die

sa

reli

sted

chro

no

log

ica

lly

.

85

therapy, also deemed experimental at this time, demon-strated that children in this treatment conditionimproved more on self-rated depression and mood andon clinician ratings of functioning relative to a individ-ual, psychodynamic treatment condition (Trowell et al.,2007); however, both groups demonstrated postinter-vention improvements and between-group differenceswere no longer present at follow-up.

Adolescent Studies

Table 4 presents the evidence-based treatment status ofthe adolescent interventions.

Specific program classification schema. Althoughthe literature examining psychosocial interventions fordepressed adolescents has continued to grow since the1998 review by Kaslow and Thompson (1998), no spe-cific psychosocial intervention for depressed adolescentshas emerged as well-established. Consistent with the pre-vious review, CWD-A (Clarke et al., 1995; Clarke et al.,2001; Clarke et al., 2002; Clarke et al., 1999; Kaufman,Rohde, Seeley, Clarke, & Stice, 2005; Lewinsohn et al.,1990; Lewinsohn et al., 1996; Rohde et al., 2004) meetscriteria for a probably efficacious intervention. Ourreview presented here also reveals that IPT-A (Mufson,Dorta, Wickramaratne et al., 2004; Mufson et al., 1999)is a probably efficacious intervention for adolescentdepression. The following is a brief review of thesetwo specific protocols.

The prior review revealed that the CWD-A course metcriteria for probably efficacious (Clarke et al., 1995;Lewinsohn et al., 1990; Lewinsohn et al., 1996), andmore recent studies further support the efficacy of thisintervention (Clarke et al., 2001; Clarke et al., 1999;Kaufman et al., 2005; Rohde et al., 2004). A series of stu-dies have examined the CWD-A program (Clarke et al.,1995; Clarke et al., 2001; Clarke et al., 2002; Clarke et al.,1999; Kaufman et al., 2005; Lewinsohn et al., 1990;Lewinsohn et al., 1996; Rohde et al., 2004). All but oneof these studies (Clarke et al., 2002), which was conduc-ted in an HMO setting, found this program to be effi-cacious. The CWD-A program is an adolescent-versionof an earlier program designed for depressed adults.The adolescent version has been modified over the years,but in all versions, adolescents are taught relaxation,cognitive restructuring, pleasant activity scheduling,communication, and conflict-reduction techniques. Theprogram is typically 15 to 16 sessions, and sessions rangefrom 45 min to 2 hr. For some of the protocols, there areseparate parent groups, during which time parents areinformed about the general topics being addressed inthe CWD-A program, the skills being taught, and therationale for their use.

CWD-A has been examined in a number of studies.For instance, Lewinsohn et al. (1990) randomly assigned59 adolescent students who met diagnostic criteria forMajor Depression or for minor or intermittentdepression to one of three conditions: CWD-A ado-lescent group only, CWD-A adolescent group plus par-ent, or a waitlist control. Relative to youth in the controlcondition, youth in both treatment groups showed sig-nificantly greater declines in depression based on self-reports and clinician interviews but not parent reports.In the two treatment groups there was a significantdecline in adolescents meeting diagnostic criteria fordepression at posttreatment relative to the controlgroup, but there were no significant differences betweenthe two treatment groups. Effects were maintained atthe 2-year follow-up, but information was available onlyfor the two treatment groups.

Clarke et al. (1995) reported findings from a school-based intervention study with 150 at-risk adolescentswho had elevated but subdiagnostic levels of depressivesymptoms on the Center for Epidemiologic Studies–Depression Scale (CES-D; Radloff, 1977) and theK-SADS. Youth were randomly assigned to eitherCWD-A or usual care (i.e., free to continue with anypre-existing intervention or seek new assistance). Rela-tive to the usual care condition, at postinterventionCWD-A was associated with significantly betterimprovements in depressive symptomatology as assessedby the CES-D and in global psychological functioning,but between-group differences were lessened at the12-month follow-up. In the CWD-A group, there weresignificantly fewer cases of MDD and=or DD at fol-low-up than the usual care group. No between-groupdifferences were noted for nonaffective disorders.

A replication study (Clarke et al., 1999; Lewinsohnet al., 1996) of the original examination of CWD-A(Lewinsohn et al., 1990) further supported the benefitsof this specific program. The replication study had twomodifications: skills training was presented throughoutthe course and booster sessions were added. This studyincluded 123 adolescents with MDD or DD recruitedfrom the community (announcements to health profes-sionals and school counselors, media). Youth wererandomly assigned to CWD-A, CWD-A plus parent,or a waitlist control condition (i.e., received no treatmentduring the time frame of the experimental interventionsbut offered an experiment treatment at the conclusionof the study). Participants also were randomized to fol-low-up conditions of booster sessions plus assessmentor assessment only conditions. Youth in both the activeinterventions showed significantly higher depressionrecovery rates and significantly greater improvementsin self-reported depressive symptoms on the BDI atpostintervention when compared to their peers in thewaitlist control condition. No between-group differences

86 DAVID-FERDON AND KASLOW

emerged according to parent reports, and differencesaccording to interviewer report were mixed. The additionof the parent group had no significant effect. Theaddition of the booster sessions was associated withaccelerated rates of recovery of youth still depressed atthe end of the active treatment, but they had no impacton the rates of reoccurrence.

In a study with 94 at-risk adolescents (i.e., youth withelevated depressive symptoms and depressed parents),Clarke et al. (2001) compared CWD-A plus usualHMO care to usual HMO care only. Youth in bothconditions were permitted to initiate or continue anynon-study-related mental health services conducted inan HMO clinic. The CWD-A plus HMO care groupshowed significantly greater improvements in overallpsychological functioning and depressive symptoms atpost-intervention relative to their counterparts in theHMO care only condition. These differences were notedby the adolescents on the CES-D and interviewers whorated the adolescents using the Hamilton DepressionRating Scale (Endicott, Cohen, Nee, Fleiss, &Sarantakos, 1981) but not by parents on the CBCL.At the 12-month follow-up, there were significantlyfewer cases of depressive disorders for youth in theactive intervention as compared to the control group.This finding continued at the 18- and 24-monthfollow-ups but at a declined level.

In a hybrid efficacy-effectiveness study, Rohde et al.(2004) randomly assigned 93 youth who met criteriafor both MDD and conduct disorder to CWD-A or alife skills=tutoring control group. In this evaluation,the CWD-A course was modified slightly to includetwo group leaders to assist with behavior managementand assignment completion, a behavioral point rewardsystem, and two optional group meetings of parentsonly to provide information and teach problem-solvingskills. Youth in both conditions were allowed to receiveno-research treatment as usual. All of the youth werereferred by the Department of Corrections and wereunder the supervision of an intake, probation, or paroleofficer. At posttreatment, depression recovery rates weresignificantly better for teens assigned to the CWD-Acondition as compared to those in the control group,but recovery rates for conduct disorder did not differby intervention. These differences were most pro-nounced according to the youth’s self-report and leastevident based on parent report. Interviewer data weremixed. Despite the between-group differences noted atpostintervention, at the 6- and 12-month follow-ups,there were no between-group differences with regardsto depression or conduct disorder.

Overall, this series of outcome studies on CWD-Asuggest that this specific intervention meets the criteriafor a probably efficacious treatment, as client character-istics were specified, a treatment manual was used, and

more than two adequate group design studies demon-strated that CWD-A, with or without a parent compo-nent, was superior to an alternative or no-treatmentcontrol condition. However, this specific program failsto meet the full criteria for a well-established treatmentas CWD-A has been evaluated by only one interrelatedresearch group.

CWD-A has guided the development of treatmentprotocols by other investigatory teams (Asarnow et al.,2005; TADS Team, 2004) that found the active treat-ment more effective in improving adolescent functioningthan control conditions. CWD-A also has been utilizedas a basis for an acute, brief intervention (Clarke et al.,2005), but this brief treatment was not found to be moreefficacious than the treatment as usual condition inreducing depression symptoms. None of these studiesutilized the specific CWD-A treatment program.

IPT-A also is a probably efficacious specific psycho-social treatment for adolescent depression. Two rando-mized controlled studies that have examined theefficacy of IPT-A (Mufson, Dorta, Wickramaratneet al., 2004; Mufson et al., 1999) in reducing depressivesymptoms and improving interpersonal functioningfound this specific program to be more efficacious inimproving functioning than control conditions. IPT-Ais an adolescent version of an earlier program designedfor depressed adults. Similar to the adult program,IPT-A addresses a specific area of interpersonal dif-ficulty but shifts the focus to interpersonal issues specificto adolescence, such as changes in the parent–adolescentrelationship due to shifts in closeness and authority. Theprogram supports adolescents in relating their difficultiesto one of four primary problem areas (i.e., grief, role dis-putes, role transitions, interpersonal deficits) and indeveloping effective strategies (e.g., improved communi-cation, expression of affect related to changes in relation-ship, development of a new and effective social supportsystem) to deal with their problem area. IPT-A consistsof 12 individual sessions, 30 to 60 min in length. Sessionsare conducted during a 12- to 16-week period, and onestudy evaluating this program (Mufson et al., 1999) alsoincluded a weekly phone contact between the therapistand adolescent for the first 4 weeks of the intervention.

To evaluate the efficacy of IPT-A, Mufson et al.(1999) randomly assigned 48 adolescents who met cri-teria for any depressive disorder to the IPT-A programor a clinical monitoring control condition. Participantswere recruited from hospital and school clinics or werefamily or self-referred. Youth in the control conditionwere assigned a therapist with whom they could meetmonthly, with an option for a second session withinthe month, and this therapist monitored their symptomsbut refrained from advice giving or skill training. Ado-lescents in the IPT-A treatment showed significantlygreater improvements in their depressive symptoms,

TREATMENTS FOR YOUTH DEPRESSION 87

social functioning, and interpersonal problem solvingthan adolescents in the control condition accordingto self-report measures and clinical interviews. Thestudy did not include a follow-up assessment to helpdetermine whether the benefits of IPT-A were main-tained over time.

In an examination of the same version of the IPT-Aprogram delivered in a school-based setting, 63 adoles-cents who met criteria for a depressive disorder wererandomly assigned to IPT-A or an individual therapycontrol condition that resembled supportive therapyand that was the customary treatment offered in theschool health clinic (Mufson, Dorta, Wickramaratneet al., 2004). Engagement in both conditions led toreductions in depressive symptoms, but IPT-A parti-cipants evidenced a significantly greater reduction intheir depression and significant improvements in theirglobal and social functioning according to self-reportand clinician ratings relative to youth in the control con-dition. Teens were re-evaluated at a 16-week follow-upwith a clinician assessment, and the significantbetween-group differences were maintained.

Taken together, outcome studies evaluating IPT-A(Mufson, Dorta, Wickramaratne et al., 2004; Mufsonet al., 1999) provide sufficient information to deemIPT-A a probably efficacious treatment for adolescentdepression, as client characteristics were specified, atreatment manual was used, and two adequate groupdesign studies demonstrate that IPT-A was superior totreatment controls. Rossello and Bernal (1999) alsoevaluated the efficacy of an IPT treatment with PuertoRican adolescents and found youth in the IPT conditionhad significantly greater reductions in depressive symp-toms and significantly greater increases in self-esteemand social adaptation when compared to youth in thecontrol group. However, the IPT treatment protocol uti-lized by Rossello and Bernal is not the same specific pro-gram as IPT-A examined by Mufson and colleagues(Mufson, Dorta, Wickramaratne et al., 2004; Mufsonet al., 1999). Accordingly, as is found for CWD-A,IPT-A also falls short of being a well-established specifictreatment for depressed adolescents as it has only beenevaluated by one investigatory team.

Other specific psychosocial interventions (i.e., Ado-lescents Coping with Emotions, Attachment-BasedFamily Therapy, Depression Treatment Program, Feel-ing Good, Interpersonal Psychotherapy–AdolescentSkills Training, Time for a Future–AdolescentDepression Program) have been found to be efficaciousin reducing depressive symptoms (Ackerson et al., 1998;Diamond et al., 2002; Kowalenko et al., 2005; Woodet al., 1996) and rates of depression diagnosis (Melvinet al., 2006) relative to control conditions. However,these specific interventions are deemed experimentalbecause of their limited examination.

Modality classification schema. The adolescenttreatment studies also utilized a variety of modalities,and their status as an experimental, probably effi-cacious, or well-established modality of treatment ispresented in Table 4. Several intervention protocolsfor depressed adolescents have been examined thatfall under the CBT rubric, and these CBT studies weredivided by modality (i.e., adolescent-only group, ado-lescent group plus parent component, individual, indi-vidual plus parent=family component, self-directedbibliotherapy, CBT-enhanced primary care services).The modality of CBT group, adolescent only (Clarkeet al., 1995; Clarke et al., 1999; Kowalenko et al.,2005; Lewinsohn et al., 1990; Lewinsohn et al., 1996;Reynolds & Coats, 1986) can be considered well-estab-lished because it includes at least two well-conductedstudies by different investigatory teams that found thisapproach more effective than a psychological placeboor alternative treatment control condition, specifiedclient characteristics, had adequate sample sizes, andutilized treatment manuals. In addition, one specificintervention (i.e., CWD-A) included in this modalitymeets criteria for probably efficacious.

The modalities under CBT of individual (Rossello &Bernal, 1999; Wood et al., 1996) and individual plusparent=family component (Brent et al., 1997; TADSTeam, 2004) are deemed probably efficacious, becausethey do not include a specific CBT program that is atleast probably efficacious. The adolescent group plusparent component modality (Clarke et al., 1995; Clarkeet al., 2001; Clarke et al., 1999; Lewinsohn et al., 1990;Lewinsohn et al., 1996; Rohde et al., 2004) does includea specific CBT program (i.e., CWD-A) that is probablyefficacious; however, it falls short of being well-estab-lished as it has only been examined by one interrelatedinvestigatory team.

Finally, the CBT modalities of self-directed bib-liotherapy (Ackerson et al., 1998) and enhanced primarycare services (Asarnow et al., 2005) are presently experi-mental because of the limited extent to which they havebeen evaluated. Self-directed bibliotherapy involvesadolescents independently reading Feeling Good (Burns,1980), which is based on Beck’s (1970) cognitive theoryof depression, and completing the book’s exercises.Enhanced primary care services includes education andsupport of primary care clinicians in evaluating andmanaging adolescents’ depression, care managerstrained in manualized CBT for depression, coordinationof treatment between care managers and primary careclinicians, and patient and clinician choice on treatmentapproach.

The modality of individual treatment under thetheoretical orientation of IPT also is a well-establishedapproach for adolescent depression. This modality hasbeen used in more than two well-conducted studies from

88 DAVID-FERDON AND KASLOW

different investigative teams that have found it moreeffective than a psychological placebo or alternativetreatment control condition. Studies using this modalityspecified client characteristics, had adequate samplessizes, and used a treatment manual (Mufson, Dorta,Wickramaratne, et al., 2004; Mufson et al., 1999;Rossello & Bernal, 1999). Further, one specific inter-vention (i.e., IPT-A) included in this modality is at leastprobably efficacious. Group treatment under the theor-etical IPT approach is experimental because of its lim-ited examination to date (Young et al., 2006a).

Other treatment modalities for adolescent depressionalso have been examined and are promising. Theseinclude nondirected support provided through anadolescent only group (Fine, Forth, Gilbert, & Haley,1991), as well as a family-systems-oriented approachoffered with the parent–adolescent subsystem (Diamondet al., 2002). These approaches have been examined infew studies and thus are classified as experimental atthe present time.

Theoretical orientation classification schema. Thestatus of interventions for treating adolescent depressionalso was examined from the broader theoretical orien-tation perspective. Based on the published studies,CBT (Ackerson et al., 1998; Asarnow et al., 2005; Brentet al., 1997; Clarke et al., 1995; Clarke et al., 2001;Clarke et al., 1999; Kowalenko et al., 2005; Lewinsohnet al., 1990; Lewinsohn et al., 1996; Reynolds & Coats,1986; Rohde et al., 2004; Rossello & Bernal, 1999;TADS Team, 2004; Wood et al., 1996) and IPT(Mufson, Dorta, Wickramaratne, et al., 2004; Mufsonet al., 1999; Rossello & Bernal, 1999; Young et al.,2006a) meet criteria for well-established treatments.Numerous studies demonstrate that CBT and IPT aremore efficacious than alternative treatments or treat-ment controls; studies have been conducted by differentinvestigators; and at least two studies have adequatesample sizes, specify sample characteristics, and usetreatment manuals. One of the specific protocols forCBT (i.e., CWD-A) and IPT (i.e., IPT-A) also meetscriteria for at least probably efficacious treatments foradolescent depression.

The theoretical approaches of nondirected support(Fine et al., 1991), family systems theory (Diamondet al., 2002), and behavior therapy (Reynolds & Coats,1986) are viewed as experimental even though theyhave been found more effective than control con-ditions, used treatment manuals, and specified samplecharacteristics. These theoretical approaches havereceived limited study, and none include a specificintervention that is at least probably efficacious, whichlimits the status of these overall theoretical orientationsto experimental.

CONCLUDING COMMENTS

In this section, we offer some final evaluative commentsabout the published literature on psychosocial treat-ments for depressed youth and highlight some strengthsand weaknesses of the literature. We provide researchsuggestions that would support the continued advance-ment of this field. Specific attention is paid to the grow-ing interest and use of antidepressant mediations withyouth. Finally, recommendations for best practice areprovided.

Review Summary

The field has made remarkable progress in expandingupon the breadth and quality of published evaluationsof psychosocial interventions of youth since the 1998Kaslow and Thompson review: More interventionsnow meet criteria for probably efficacious, unique treat-ment modalities (e.g., videoconferencing) are now beingevaluated, and a greater range of theoretical approachesare being examined (e.g., psychodynamic, family sys-tems). Several psychosocial interventions for child andadolescent depression, the majority of which are basedon a cognitive-behavioral model or interpersonal model,reduce depressive symptoms and=or alleviating depres-sive disorders in clinical and nonclinical samples. Thisassertion is based on a growing number of studies thatuse varying treatment approaches and evaluationdesigns but generally are well-designed, methodologi-cally sound randomized controlled trials. According toguidelines by Nathan and Gorman (2002), the childstudies typically met criteria for Type 2 trials due to aseries of minor methodological weaknesses. In contrast,the studies with adolescents were better designed andthus more met criteria for Type 1 trials.

The continued growth of well-designed, randomizedcontrolled trials examining the efficacy of psychosocialinterventions for depressed youth has bolstered theevidence-based status of specific programs, modalitiesof treatment, and broad theoretical approaches. Ourreview indicates that the specific interventions of Self-Control Therapy and Penn Prevention Program for chil-dren and of CWD-A and IPT-A for adolescents areprobably efficacious. These programs fell short of beingwell-established specific interventions because they havebeen examined by only one investigatory team or theirefficacy relative to a psychological placebo or alternativetreatment has received limited study. However, the qual-ity of the trials and at times breadth of investigationsacross a myriad of settings and patient populations indi-cate that these are very promising treatments.

Further, this review demonstrates that positive treat-ment effects are found regardless of treatment modality(group, individual, or family therapy) or the nature or

TREATMENTS FOR YOUTH DEPRESSION 89

extent of parental involvement. However, for children,there is the most evidence to support the modalities ofgroup-child only and child group plus parent compo-nent under the theoretical rubric of CBT, resulting inthese being well-established modalities. For adolescents,the modalities of group adolescent under CBT and indi-vidual treatment under the broad framework of IPThave the most support and are deemed well-established.Other promising modalities under CBT for adolescentsare adolescent group plus parent component, individual,and individual plus parent=family component, and theseapproaches meet criteria for probably efficacious. Directcomparisons of treatment modalities are limited andhave not demonstrated any modality to be superior(Clarke et al., 1999; Lewinsohn et al., 1996; Spielmans,Pasek, & McFall, 2007).

From a broad theoretical perspective, CBT for bothchildren and adolescents and IPT for adolescents appearto be well-established. These theoretical approaches havethe greatest empirical support to guide the developmentand implementation of interventions with depressedyouth, and the effectiveness of these approaches is sup-ported by a recent meta-analytic review (Watanabe,Hunot, Omori, Churchill, & Furukawa, 2007). Forchildren, the theoretical approach of behavior therapyalso shows some promise, and available data suggestthat this approach is probably efficacious.

Studies with both children and adolescents acrossmodalities and theoretical orientations reveal positivefindings regardless of the setting in which the inter-vention was conducted. Youth showed improvementsin their levels of depressive symptoms, reductions inrates of depressive disorders, and progress in terms ofother related psychological symptoms and psychosocialadjustment markers when they received active interven-tions in school settings, community clinics, hospital-based clinics, research clinics, primary care settings, orHMOs.

In addition, despite the broad range of control con-ditions, ranging from no-treatment or waitlist controlson one end of the continuum to treatment as usual, clini-cal monitoring or traditional counseling on the otherend of the continuum, findings suggest that manualizedand structured interventions are associated with greaterintervention gains. However, treatment gains wereendorsed primarily by the youth themselves. In general,mixed efficacy was found for clinician ratings based oninput from youth and=or their parents, and parentsfrequently did not rate their youth in the active inter-vention as showing greater gains than those in the con-trol groups.

As reported in Tables 1 and 2, even within informant,effect sizes ranged from small to large treatment gainsfor children (e.g., .25–.92 on the CDI) and adolescents(e.g., .17–.66 on the BDI). These effect size results are

consistent with variability indicated by prior reviews(M effect size ¼ .34, Weisz, McCarty, & Valeri, 2006;M effect size ¼ .72, Michael & Crowley, 2002), and withthe general finding in these reviews that psychosocialinterventions with depressed children and adolescentsproduce moderate treatment gains.

Future Research Directions

Relative to other problem areas for children and adoles-cents that are targeted by psychosocial interventions,the availability of studies examining interventionsfor depressed youth is currently a small but developingliterature. Accordingly, there are many avenues forfuture research to consider and explore. In the followingsection, we offer recommendations of factors to considerwhen developing and evaluating psychosocial interven-tions for depressed youth. For instance, few investigatorshave examined variables that moderate (i.e., factors thatinfluence the direction and=or strength of the inter-vention) and=or mediate (i.e., the mechanism by whichan intervention is efficacious) treatment responses(Baron & Kenny, 1986; Holmbeck, 1997, 2002; Kraemer,Wilson, Fairburn, & Agras, 2002). Researchers whoexamine depression interventions increasingly assess the-orized change mechanisms, but tests of mediation oftenare not conducted (Weersing & Weisz, 2002). We discusssome factors that should be considered as well as waysthe field can be advanced by examining understudiedtheoretical approaches and modalities, comparing activeinterventions, and conducting effectiveness research.This line of empirical inquiry is crucial to further under-standing of how and for which youth interventions aremost likely to be beneficial.

Development. With regards to demographic fac-tors, the limited data suggest that psychosocial interven-tions are more effective for adolescents than for childrenand for younger adolescents than older adolescents, butresults are mixed regarding the moderating effect of ageof onset of the first depressive episode (Clarke et al.,1992; Jayson, Wood, Kroll, Fraser, & Harrington,1998; Michael & Crowley, 2002). Recent treatmentsfor children have demonstrated increased attention todevelopmental issues (Asarnow et al., 2002; Muratoriet al., 2003; Nelson et al., 2003; Pfeffer et al., 2002),and similar focus with adolescent interventions iswarranted.

The design and implementation of future interventionprograms must be developmentally informed. How thecognitive, biological, and socioemotional maturity ofyouth influence the development and presentation ofdepressive symptoms and how these factors may impacta youth’s responsiveness to an intervention protocolneed to be considered (Cicchetti & Toth, 1998). Data

90 DAVID-FERDON AND KASLOW

suggests that more rational thinking is associated withbetter treatment response and short depressive episodes(Clarke, DeBar, & Lewinsohn, 2003). Because of thestronger cognitive maturity of adolescents relative tochildren, adolescents may respond better to interventioncomponents that seek to address their cognitive skills(Durlak, Fuhrman, & Lampman, 1991).

Data, though, are mixed with regards to cognitivefactors as treatment mediators. Kolko, Brent, Baugher,Bridge, and Birmaher (2000) found that changes inadolescents’ cognitive distortions did not mediate theeffects of a CBT intervention on depressive symptoms.In contrast, others have found changes in dysfunctionalthinking patterns and automatic thoughts and in opti-mistic explanatory style have been found to mediatetreatment outcomes with both children and adolescents(Ackerson et al., 1998; Gillham et al., 1995; Kaufmanet al., 2005; Yu & Seligman, 2002). Continued examin-ation of the effect of cognitive change components ofinterventions is needed, and investigators need to beattentive to the cognitive development of youth whendesigning these elements to be sensitive to developmentaldifferences.

Because of the varying cognitive, biological, andsocioemotional needs of children and adolescents, futuretreatment developers should follow the lead of morerecent interventions and consider using developmentallyinformed theories to base their protocols on (Pfefferet al., 2002). In addition, attention should be given tothe increased involvement of key individuals, such asparents, to help support the development of more adapt-ive skills and to foster generalization beyond the therapyenvironment (Asarnow et al., 2002; Gillham, Reivich,et al., 2006; Nelson et al., 2003). Although Kolko et al.(2000) did not find family variables to mediate or mod-erate treatment outcome of adolescents, studies thathave examined family factors have shown them to besignificant predictors of youth’s affective functioningand suggests they should be considered in futureresearch. For example, greater parental involvement isassociated with improved outcomes (Clarke et al.,2003), the positive effects of CBT are weakened for ado-lescents whose mothers are depressed (Weersing &Brent, 2003), and family factors predict youth’sdepression recovery and reoccurrence (Birmaher et al.,2000; Rohde, Seeley, Kaufman, Clarke, & Stice, 2006).

Further, the increased involvement and influence ofpeers during the adolescent stage of development sug-gest that the inclusion of peer components also may bean important developmental consideration when design-ing effective interventions for adolescents. Because ofthe limited attention to date of the potential mediatingor moderating role of parents, peers, or other influentialindividuals and data not showing a clear benefitfor involving others in psychosocial interventions

(Clarke et al., 1999; Lewinsohn et al., 1996; Sanfordet al., 2006), additional evaluation is needed. Thisresearch would ascertain whether this developmentalconsideration impacts the direction and=or the magni-tude of the intervention effect and what modality isthe most likely to lead to enhanced impact (e.g., separateparent education group vs. integrated family therapy).

Attention to such developmental considerationsappears in the ACTION treatment program that iscurrently being evaluated by Stark and colleagues (Starket al., in press; Stark et al., 2006). This group inter-vention program includes developmentally appropriateexperiential exercises and parent training and teacherconsultation components. The evaluation of theseelements will be informative to the next wave ofintervention development.

Ethnocultural factors. Other patient characteristics,such as ethnocultural factors, may moderate or mediatetreatment efficacy. One study that has examined themoderating role of ethnicity (Rohde et al., 2006) foundthat a CBT intervention was more effective fordepressed Caucasian youth than a life-skills controlcondition, whereas non-White adolescents had similarrecovery rates across conditions. In addition, there isevidence from a prevention of depression study(Cardemil et al., 2007; Cardemil et al., 2002) that theprogram may not be equally efficacious for childrenfrom different ethnic and racial backgrounds. Attentionto ethnocultural factors in the design and implemen-tation of interventions is developing but remains rare.Only two investigatory teams describe offering cultu-rally competent care in treatments that were found tobe efficacious (Rossello & Bernal, 1996, 1999; Yu &Seligman, 2002), and only a few others have discussedculturally informed intervention programs (Bernal &Scharron-del-Rio, 2001; Griffith, Zucker, Bliss, Foster,& Kaslow, 2001; McClure, Connell, Zucker, Griffith,& Kaslow, 2005). There is some evidence for the valueof culture-specific interventions for depressed youthand their families (Breland-Noble, Bell, & Nicolas,2006; Griffith et al., 2001; McClure et al., 2005) andthe benefit that such approaches have for facilitatingtreatment engagement (Breland-Noble et al., 2006).Accordingly, the efficacy of psychosocial treatment fordepression with cultural=ethnic minority youth isunknown and deserves further study.

Clinicians and researchers agree that the field andpatients would benefit from the design and evaluationof ethnoculturally relevant interventions (AmericanPsychological Association, 2003; Bernal & Scharron-del-Rio, 2001; Kleinman & Good, 1985). Interventionscould be either created to be specific to a cultural groupor adapted from existing evidence-based interventionsto be culturally sensitive. With either approach, how

TREATMENTS FOR YOUTH DEPRESSION 91

depression is commonly manifested and viewed by aparticular community (e.g., stigmatized issue, familyproblem) must be taken into account (American Psychi-atric Association, 1994; Marsella & Kaplan, 2002). Inaddition, potential ethnicity differences in recovery timeshould also be considered (Rohde et al., 2006).

Intervention developers must have knowledge of thehistory, customs, and traditions of the target population.Key members from the target community should beinvolved in the intervention’s development to ensurethe appropriateness and compatibility of an interven-tion’s components (e.g., increasing assertiveness ofyouth vs. supporting authority of parents, emotionalexpression vs. reserve when discussing affect) andmodality (e.g., individual, family, group) with the cultureof the target population. Constructs particularly relevantto a targeted population should be incorporated and beexamined as potential moderators, predictors, and=oroutcomes when evaluating the intervention. Whendesigning or selecting an intervention, it would be impor-tant to take into account a youth’s ethnocultural back-ground as well as their level of ethnocultural identity(Marsella & Kaplan, 2002). With the rapid growth ofmultiracial and multicultural families, how related ident-ity issues may influence the acceptability and impact of aprogram should be examined (McDowell et al., 2005).

Treatments also need to be sensitive to languagepreferences and boundary issues that may develop asa result of using a translator or merely by the therapistbeing an outsider to the family and discussing a child’smental health needs. Intervention developers shouldconsider the potential utility and=or ethnoculturalnecessity to have therapists from the same communitybeing served and who fluently speak the dominant lan-guage of the targeted family. At a minimum, therapistsmust be aware of their own ethnocultural knowledge,attitudes, and biases and address these areas to enhancethe level of trust and the efficacy of the intervention(American Psychological Association, 2003). Finally,treatment developers should adequately describe the cul-tural and ethnic makeup of their participants and shoulddiscuss the generalizability of their programs to otherethnocultural groups.

One of the inherent challenges of evaluating the effi-cacy of a treatment for varying ethnocultural groups isthe limited availability of evaluation instruments thathave appropriate psychometric properties for a widerange of unique population groups. To move culturallysensitive intervention research forward, instrumentsneed to be developed and=or adapted for a broaderrange of ethnocultural groups (American PsychologicalAssociation, 2003; Bernal & Scharron-del-Rio, 2001;Marsella & Kaplan, 2002). These instruments need tobe sensitive to the dialect and linguistic characteristicsof the target group. The evaluation tools also need to

be tested to ensure the global construct (e.g., depression)as well as indicators (e.g., depressive symptoms) arevalid in the target population. When interpreting datafrom these measures, sensitivity is needed to the uniquebackground of the individual being evaluated, as well asthe attitudes and biases the interpreter of the data maybe bringing to bear on the data.

Finally, when assessing the mental health of youthand the impact of interventions, multiple informants(e.g., youth, parent, clinician) are typically employed,and as this review demonstrates, the apparent efficacyof an intervention often varies depending on the inform-ant. Although discordance in reports of the affectivefunctioning of youth is not rare (Achenbach & Rescorla,2001), researchers should attend to how ethnoculturalfactors may play a role in an informant’s awareness ofa youth’s mental health functioning and=or willingnessto report positively or negatively on the functioning ofthe youth (Roberts, Alegria, Roberts, & Chen, 2005).

Gender. The moderating role of gender alsoremains uncertain because of limited examination todate and inconsistent findings. For instance, genderhas been found to have no effect on treatment outcome(Jayson et al., 1998; Kolko et al., 2000), whereas otherresearchers suggest that interventions for depressionare more efficacious for females (Michael & Crowley,2002) and still others highlight trends for increasedepression recovery for males (Clarke et al., 1999). Aswith other potential moderators of intervention efficacy,the question related to gender differences is how datarelated to these variations may suggest gender specificintervention components and modalities. Interventionsshould attend to data suggesting that female childrenare more likely to have risk factors for depression thantheir male peers even though these vulnerability factorsmay only manifest as depression after facing the biopsy-chosocial challenges of adolescence (Nolen-Hoeksema &Girgus, 1994), and this heighten vulnerability for femaleadolescents is supported cross-culturally (Galambos,Leadbeater, & Barker, 2004; Wade, Cairney, & Pevalin,2002). Female youth may benefit more from interper-sonally oriented treatments whereas their male peersmay respond better to cognitive behaviorally orientedapproaches, however the dearth of data make it prema-ture to draw any conclusions in this regard (Garber,2006).

In addition, data suggest that female and male youthexperience different correlates for depression and copewith depression differently. When designing either entireintervention protocols specific for a particular gendergroup or in designing components of treatment thataddress the unique needs of males and females, suchgender differences should be considered. For instance,peer difficulties appear to be more strongly correlated

92 DAVID-FERDON AND KASLOW

with depressive symptoms in females than males (Nolen-Hoeksema, 2002), which suggests that intervention com-ponents that target the enhancement of social relation-ships may be particularly valuable for female youth.When developing intervention components, consider-ation should be given to findings that females tend tobe more internally focused and cope using ruminationstrategies, whereas males tend to be more externallyfocused (Nolen-Hoeksema, Larson, & Grayson, 1999;Sethi & Nolen-Hoeksema, 1997). Interventions forfemales may increase their impact if they specifically tar-get the negative thought cycle and the rumination thatexacerbates and maintains their depressive symptoms,whereas programs for males may be more efficaciousif they focus on how male youth perceive and interpretexternal cues. Finally, gender specific interventionsshould consider socialization differences that tend topromote females to be more emotionally expressiveand gender differences regarding who youth seek helpfrom to deal with a difficulty (Wintre, Hicks, McVey,& Fox, 1988). Females may be more likely to attendthe intervention and experience positive treatmentrelated impact (e.g., reductions in hopelessness) whentheir fellow group participants includes only otherfemales (Chaplin et al., 2006).

The ACTION treatment program that is currentlybeing evaluated by Stark and colleagues (Stark et al.,in press; Stark et al., 2006), is a recently developedgender-sensitive intervention. The intervention format,activities used, coping skills emphasized, and inter-personal focus are specific for young females. Prelimi-nary results suggest the ACTION Program is effectiveand associated with a 70% recovery rate. The descrip-tions of this program highlight important factors toconsider in the development and implementation of agender-sensitive protocol, which are extremely usefulto the future development and evaluation of suchprograms.

Diagnosis=comorbidity. The moderating role ofdiagnostic factors, such as symptom severity and comor-bidity, has received limited attention. One study foundthat youth with recurrent major depression evidencedsignificantly faster depression recovery after participat-ing in a CBT intervention relative to a life-skills controlcondition, whereas adolescents experiencing their firstdepressive episode faired equally well across treatmentconditions (Rohde et al., 2006). Trowell et al. (2007)found that individual psychodynamic therapy and fam-ily therapy were associated with decreases in the preva-lence of MDD, DD, and both MDD and DD as well asother comorbid conditions. Available research alsohighlights a predictive association between many diag-nostic factors and changes in depression over time andsuggests that the examination of the moderating role

of such factors in psychosocial interventions would beimportant in future research. For instance, youth parti-cipating in CBT have been found to have significantreductions in their depressive symptoms when theyentered the trial with a greater number of past psychi-atric diagnoses, lower levels of depression, and suicidalideation (Barbe, Bridge, Birmaher, Kolko, & Brent,2004; Brent et al., 1998; Clarke et al., 1992; Jaysonet al., 1998; Rohde, Clarke, Lewinsohn, Seeley, &Kaufman, 2001; Weersing & Brent, 2003).

Whereas some studies have revealed that comorbidanxiety disorders are associated with greater improve-ments in depression (Rohde et al., 2001; Weersing &Brent, 2003), others have found that lower levels of anxi-ety are associated with greater reductions in depressivesymptoms (Clarke et al., 2002; Young, Mufson, &Davies, 2006b). In addition, lifetime substanceuse=dependence appears to be associated with slowertime to recovery (Rohde et al., 2001), and comorbidattention deficit disorder and disruptive behavior disor-ders appear to be linked with longer recovery time and agreater risk for depression recurrence postintervention(Rohde et al., 2001). Investigators are beginning toascertain the effectiveness of interventions with youthwith comorbid conditions (Curry, Wells, Lochman,Craighead, & Nagy, 2003; Rohde et al., 2004; Trowellet al., 2007; Young et al., 2006b). Future research thatincludes more clinically referred youth will allow fora broader examination of the impact that symptomseverity and comorbidity (e.g., MDD plus DD) haveon treatment efficacy and effectiveness.

Examination of understudied theoreticalapproaches and modalities. This review demonstratesthat there is growing attention to and support of theor-etical approaches (e.g., behavior therapy, family systemstheory, psychodynamic psychotherapy) other than CBTto addressing child and adolescent depression. A contin-ued development and evaluation of such protocolswould benefit the field by allowing for a greater rangeof evidence-based treatment options of youth, families,and therapists to better meet their individual needsand perspectives.

Increased attention to family-based interventions inparticular is needed. Depression in youth often is asso-ciated with problems in family functioning (Kaslow,Deering, & Ash, 1996; Kaslow, Deering, & Racusin,1994), and at long-term follow-up participation in familyinterventions is associated more strongly with reductionsin family conflict and with improvements in parent–childrelationships than is involvement in CBT (Kolko et al.,2000). In addition, family interventions with depressedyouth have been found to enhance family members’knowledge about depression, increase their utilizationof appropriate services, and improve family interactions

TREATMENTS FOR YOUTH DEPRESSION 93

and adolescent’s social functioning (Fristad, Arnett, &Gavazzi, 1998; Fristad, Gavazzi, & Soldano, 1998;Fristad et al., 2003). Family approaches also may beparticularly important to building protective factors foryouth when a parent is depressed (Beardslee et al., 2003).

Although studies with a parent component in thisreview did not demonstrate that parental involvementadded to the benefits of the treatment protocol, onlyone study made clear that it evaluated a standard familytherapy component with all family members present. Inthe reviewed programs that involved any family mem-bers in the treatment (Asarnow et al., 2002; Clarkeet al., 2005; Clarke et al., 2001; Clarke et al., 2002;Clarke et al., 1999; Diamond et al., 2002; Gillham,Reivich, et al., 2006; Muratori et al., 2003; Nelsonet al., 2003; Pfeffer et al., 2002; Rohde et al., 2004;Sanford et al., 2006; Trowell et al., 2007), typically aparent was the only family member involved, and fre-quently the parent had separate sessions.

The one study (Sanford et al., 2006) that evaluated afamily psychoeducation component that involved allfamily members living in the adolescent’s home didnot find that it led to significant changes in depressivesymptoms among adolescent participants relative tothe control condition. However, it was associated withimprovements in the adolescent’s social functioningand the adolescent–parent relationship. Accordingly, itbehooves future investigators to examine various formsof family therapy for depressed youth and to considerthe inclusion of the entire family constellation ratherthan only a single parent (Cottrell, 2003; Kaslow,Mintzer, Meadows, & Grabill, 2000). The incorporationof family services may be substantial and include mul-tiple sessions or it may be limited to a couple of sessionsto review treatment progress, answer the family’s ques-tions, and=or reinforce learning principles taught inthe program the target youth in participating in. Ineither case, it is important of investigators to consist-ently report in their studies the addition, frequency,and scope of family contact to broaden the understand-ing of the benefit of this modality.

Comparison of active interventions. Few between-group design studies that compare a psychosocial inter-vention to a pill or psychological placebo condition oranother active treatment have been conducted, whichprecludes us from concluding that any specific inter-vention approach is more efficacious in reducingdepression. Many psychosocial interventions are foundto be efficacious relative to no-treatment control con-ditions (Clarke et al., 1999; Jaycox et al., 1994; Yu &Seligman, 2002), but the lack of an alternative treatmentcontrol group comparison limits the ability to rule outthat positive results may be due to nonspecific factorssuch as attention.

The few studies that have compared different activeinterventions (Brent et al., 1997; Rossello & Bernal,1999; TADS Team, 2004; Trowell et al., 2007) yieldedmixed results across informants and outcome measures.With the growing evidence supporting various specificinterventions (e.g., Penn Prevention Program, Self-Control Therapy, CWD-A, IPT-A), between-groupcomparisons of active treatments are more likely to beconducted. It is imperative that future investigationsbe conducted in a culturally competent fashion, toexamine a broader array of treatments including novelapproaches (e.g., videoconferencing, Web-based treat-ments) and the involvement of parents and other familymembers, and to compare active interventions. In thesefuture investigations, it also behooves researchers toreport the percentage of youth who are fully recoveredat the end of treatment across active treatments; thisinformation is not consistently reported and previousreviews have suggested a large range of effectivenessby treatment (Weisz, Jensen, & McLeod, 2005).

Replication by unrelated investigatory teams. Thisreview also indicates that one of the main stumblingblock for gaining the necessary evidence-based supportfor a specific program (e.g., Self-Control Therapy,CWD-A, and IPT-A) to be deemed well-established isthat intervention protocols tend to be evaluated byone investigatory team. The potential problems ofallegiance effects (i.e., intervention effects tend to belarger when evaluated by the group of researchers whoprefer and developed the intervention protocol) havebeen debated (Robinson, Berman, & Neimeyer, 1990).One strategy to address this difficulty would be to relaxthe evidence-based criteria, but such a strategy wouldremove the importance of replication. Replication isnecessary to demonstrate that the treatment protocolis contributing to change, and that change is not onlythe result of expertise. In addition, replication isrequired to demonstrate that the positive impact of anintervention will generalize to different therapists,settings, and patients.

Future research is needed that involves independentevaluation of probably efficacious intervention. To sup-port replication, it is helpful for investigators to reportconsistently even minor changes in manualized treat-ments from study to study for new investigatory teamsto capitalize on the progress and lessons learned bythe original developers of an intervention. Little atten-tion has been paid to treatment adherence variables ortherapist variables that may impact treatment outcome(Clarke et al., 2003; Kolko et al., 2000). Thus, the fieldis ripe for a more systematic and comprehensive evalu-ation of a broad array of variables that may influencetreatment outcome.

94 DAVID-FERDON AND KASLOW

Design and evaluation of long-term interventioneffects. The long-term efficacy of the reviewed inter-ventions remains unclear because of the limited avail-ability of follow-up data and mixed results whenavailable. A recent meta-analysis also questions thelong-term benefit of psychotherapy with depressedyouth relative to control conditions (Watanabe et al.,2007). Of the limited studies with follow-up data thatwe reviewed, a number found that between-group differ-ences immediately following the active intervention wereno longer evident at follow-up (Birmaher et al., 2000;Roberts et al., 2003; Rohde et al., 2004). In contrast,two interventions did not demonstrate treatment effectsimmediately following the completion of active treat-ment, but group differences in favor of the interventionwere found at long-term follow-up (Gillham, Reivichet al., 2006; Muratori et al., 2003). Another study foundthe benefits of one treatment over another that wereapparent at the end of the active phase of treatment dis-appeared at follow-up because both groups continued toimprove following treatment and became relativelyequal in their effect by the 6-month follow-up (Trowellet al., 2007). In addition, the limited inclusion of boostersessions (Clarke et al., 1999) and the low frequency withwhich participants use booster sessions when offered(Melvin et al., 2006) make it unclear whether periodictherapeutic sessions following the completion of themain treatment intervention could help to maintainand=or extend recovery.

Because depression is a recurrent disorder in youngpeople that is associated with a myriad of psychosocialsequelae, it is imperative that future research includesfollow-up evaluations, that longer term treatments beconsidered, and that more attention is paid to the utilityof booster sessions. There is an inherent challenge to thisrecommendation in that youth initially assigned to a con-trol condition often receive the active treatment after thepostintervention assessment, and thus it is not practicalto compare the treatment and control groups over time.With the likely conduct of studies that compare activeinterventions, this challenge may be less problematic.

Prevention research. This article focused on inter-vention trials with children and adolescents that hadas their primary purpose to reduce depressive symp-tomatology, and prevention studies where youth werenot identified as having depressive symptoms wereexcluded (Barrett, Farrell, Ollendick, & Dadds, 2006;Barrett & Turner, 2001; Beardslee et al., 2003; Chaplinet al., 2006; Merry et al., 2004; Pattison & Lynd-Stevenson, 2001; Possel, Baldus, Horn, Groen, &Hautzinger, 2005; Possel, Horn, Groen, & Hautzinger,2004; Spence et al., 2003). Several of the reviewedstudies treated youth with clinically significant levelsof depression (Clarke et al., 1999; Nelson et al., 2003;

Rossello & Bernal, 1999), whereas others targeted at-risk youth who were deemed at risk because they hadelevated depressive symptoms but were subclinical (Ack-erson et al., 1998; Asarnow et al., 2005; Pfeffer et al.,2002). Studies targeting at-risk youth inconsistentlyself-identified themselves as prevention studies, but mostthat did self-identify as prevention had as their primaryfocus a treatment and reduction of depressive symptomsrather than the prevention of developing an affectivedisorder (Roberts et al., 2003; Yu & Seligman, 2002).For instance, Roberts et al. (2003) titled their study asa prevention trial but stated that the primary purposeof their project was to reduce depressive symptoms,and they did not evaluate at either postintervention orfollow-up whether their program prevented the develop-ment of a depressive disorder.

Some studies examined whether at-risk youth whoparticipated in an intervention program were less likelyto have a depressive disorder at follow-up; thus, theyrepresent true prevention trials. One protocol foundadolescents in the active treatment experienced a signifi-cant preventative effect for a major depressive episode(Clarke et al., 2001). Another study also found adoles-cents in an active treatment were less likely than peersin a control condition to develop a depressive disorder(3.7% vs. 28.6%, respectively) although the differencewas not statistically significant (Young et al., 2006a).In contrast, Gillham, Hamilton, et al. (2006) evaluatinga prevention program for children did not find a signifi-cant preventative effect for depressive disorders but didsupport a significant preventative effect for high-symp-tom children for depression, anxiety, and adjustmentdisorders combined. Another study evaluated whethera targeted and=or universal intervention would have apreventative effect for at-risk youth and any participat-ing adolescent and found that none of the interventionsdemonstrated a significant prevention effect relative to ano-treatment control condition, and despite improve-ments for most participating at-risk youth, 20% experi-enced a depressive episode over the course of the study(Sheffield et al., 2006).

One could argue that an intervention targeting subcli-nical youth for the purpose of reducing their depressivesymptomatology is a prevention trial. However, for astudy to be a prevention trial, it needs to have the pre-vention of the development of an affective disorder asone of its key objectives and hypotheses, and it mustinclude an evaluation of clinical diagnosis at postinter-vention and=or follow-up. Prevention trials would alsobe strengthened by including an active comparison con-dition and having adequate sample sizes to detectchange (Merry et al., 2004). Because depression hasmultiple negative outcomes for youth (Birmaher, Ryan,Williamson et al., 1996), the prevention of subclinicaland clinically significantly levels of depression is

TREATMENTS FOR YOUTH DEPRESSION 95

extremely important (Birmaher, Ryan, & Williamson,1996; Farrell & Barrett, 2007). Furthermore, preventionresearch would be strengthened by including a morecomplete evaluation of an intervention’s preventativeimpact as well as utilizing more rigorous designs (e.g.,randomized control trial). Such research would alsobe bolstered by considering developing preventions thatare specifically tailored to youth most at risk for devel-oping a depressive disorder and attend to developmen-tal, cultural, and gender uniquenesses (Horowitz &Garber, 2006).

Efficacy and effectiveness. Most outcome researchon the intervention of child and adolescent depressioninvolves demonstrations of efficacy rather than evalua-tions of effectiveness. Because the majority of studieswith children were conducted in schools with nonre-ferred youth with depressive symptoms and used rela-tively inexperienced clinicians, the generalizability ofthe findings across populations, settings, and clinician-experience level remains unclear. In addition, moststudies focus predominantly on Caucasian youth andfemales, which further limit the practice inferences thatcan be drawn from available studies. Efficacy trials arecritical first steps to understanding whether a particularintervention works in reducing depressive symptoms.However, as the evidence builds regarding the efficacyof various treatments, investigators are beginning tomove toward effectiveness trials that can evaluate thegeneralizability of positive effects (Asarnow et al.,2005; Rohde et al., 2004). Mufson, Dorta, Olfson,Weissman, and Hoagwood (2004) provided a helpfuldiscussion of their process, challenges, and strategiesto adapting an evidence-based intervention (IPT-A)initially designed and implemented in a university set-ting for a school setting that others may drawn uponwhen considering how to transport their intervention.

To support the evaluation of treatment effectivenessand dissemination of psychosocial treatments to broaderpopulations of youth, attention is needed to methodolo-gical differences between efficacy and effectivenessstudies (Schoenwald & Hoagwood, 2001). The develop-ment of researcher-community collaborations are alsoneeded to support the design and implementation ofinterventions that are appropriate and acceptable tothe selected setting’s characteristics and population(Weisz, Southam-Gerow, Gordis, & Connor-Smith,2003).

A series of evaluation studies also are needed toidentify an intervention’s necessary and sufficient treat-ment components and evaluate successive modificationsto the treatment protocol to ensure the treatment isappropriate for the clinically referred and treated youthin terms of their gender, ethnicity, and developmentalstage, whereas core treatment elements are maintained

(Weisz et al., 2005). It would be optimal if these evalu-ation studies were performed in a broad range of treat-ment settings, and with youth who range in their clinicalpresentation from depressive symptoms to clinicallydepressed to clinically depressed with comorbidconditions.

Further, treatments need to be assessed as to whetherthey are having the desired impact on the targeted cogni-tions and behaviors they are designed to have based inpart on the theoretical foundation of the treatmentand previous support of mediators of treatment efficacy(Zeiss, Lewinsohn, & Munoz, 1979), especially in lightof findings that not all interventions do (Sheffieldet al., 2006). To support such effectiveness trials and dis-semination, attention also should be given to addressingbarriers (e.g., poor detection of depressive symptoms inyouth, lack of insurance, limited availability of services)that limit youth from receiving necessary interventionsand to integrating mental health services into existingservices, such as primary medical care, to reach a largerpopulation of youth (Asarnow et al., 2005; Olfson,Gameroff, Marcus, & Waslick, 2003; Wells, Kataoka,& Asarnow, 2001).

Pharmacological Interventions

Although this review focuses on psychosocial treatments,many depressed young people receive antidepressantmedications (Debar, Clarke, O’Connor, & Nichols,2001; Delate, Gelenberg, Simmons, & Motheral, 2004;Olfson et al., 2003; Wagner & Ambrosini, 2001) eitheralone or in combination with psychosocial treatments.There is a burgeoning literature of randomizedcontrolled trials documenting the efficacy of variouspharmacological interventions with depressed youth,particularly adolescents. Positive effects have been forfluoxetine (Prozac; Emslie et al., 2002; Emslie & Mayes,2001; Emslie et al., 1997; TADS Team, 2004), paroxetine(Paxil; Keller et al., 2001), sertraline (Zoloft; Wagneret al., 2003), and escitalopram (Lexapro; Wagner, Jonas,Findling, Ventura, & Saikali, 2006; Wagner et al., 2004).There is an algorithm for pharmacological interventionsfor depressed youth (Hughes et al., 1999) that has beenfound to be feasible and effective (Emslie et al., 2004)and has recently been revised (Hughes et al., 2007). Atpresent, only fluoxetine is approved by the Food andDrug Administration (FDA) for use with children andadolescents with MDD.

One of the most recent and compelling studies sup-porting the efficacy of a selective serotonin reuptakeinhibitor (i.e., fluoxetine) in the treatment of depressedadolescents comes from the National Institutes ofMental Health–sponsored multisite trial. This trial isthe TADS, which has been described in several recentpublications (Curry et al., 2006; Emslie et al., 2006;

96 DAVID-FERDON AND KASLOW

Kennard et al., 2006; Kratochvil et al., 2006; March,Silva, Vitiello, & the TADS Team, 2006; May et al.,2007; The Treatment for Adolescents with DepressionStudy Team, 2003, 2005; TADS Team, 2004; Vitielloet al., 2006), and information about the study is pre-sented in Table 2. TADS is a randomized, masked effec-tiveness trial that evaluates the short-term (12-week) andlong-term (36-week) effectiveness of four interventionsfor adolescents who meet diagnostic criteria for MDD.The four interventions are CBT alone, fluoxetine alone,a combination of both medication and CBT (combined),and placebo pill alone.

The CBT intervention is a 15-session interventionover the course of 12 weeks that allows for flexibilityin its modality to integrate parent and family sessionswith individual sessions as needed. The CBT inter-vention is based on the CWD-A intervention developedand evaluated by Clarke, Lewinsohn, and Hops (1990)and the individual CBT intervention developed byBrent and Poling (1997). The CBT program is a skills-based treatment built on the premise that depression iscaused by the maintenance of depressive thought pat-terns and a lack of positively reinforcing behavioral pat-terns. Sessions include psychoeducation aboutdepression and its causes, goal setting, mood monitor-ing, increasing pleasant activities, social problem solv-ing, cognitive restructuring, and enhancing socialskills. There are two parent-only sessions that providepsychoeducation about depression and then one to threeconjoint sessions to address parent and adolescent con-cerns. The fluoxetine condition uses a flexible dosingschedule starting at 10 mg per day and increasing to20 mg per day at the end of the 1st week. Youth mayreceive up to 40 mg per day of fluoxetine by Week 8.Youth in the fluoxetine alone condition meet for sixmedication visits over 12 weeks with a pharmacothera-pist, who provides clinical and medication monitoringand general encouragement about the effectiveness ofpharmacotherapy. Youth in the combined treatmentreceive all of the components of both CBT alone andmedication alone; however, the two interventions arefunctionally independent. Youth in the placebo groupreceive a sugar pill following the same dose patternand monitoring pattern as the fluoxetine alone group.

Current results from TADS are based on a sample of439 adolescents with a diagnosis of MDD and revealthat the combined treatment was more effective thaneither treatment alone or the pill placebo in reducingdepressive symptoms according to clinician ratings andin alleviating suicidal ideation according to the youth’sreport. Further, fluoxetine alone was superior to CBTalone, and CBT was not more effective than placebo.Suicidal events were twice as common in adolescentstreated with fluoxetine alone than in youth in the com-bined or CBT-alone groups, potentially indicating that

CBT protects against suicidal behavior; however, abenefit of CBT has not been supported by others(Goodyer et al., 2007). Following 12 weeks of acutetreatment, 71% of teens across groups no longer metdiagnostic criteria, but 50% had residual symptoms.The combined treatment group also had other positiveoutcomes, such as faster onset of benefit and stabilityof response relative to the placebo and CBT groups,greater improvements in global functioning and globalhealth than fluoxetine alone and better effectiveness intreating mild to moderate depression than CBT aloneor fluoxetine alone. These results represent the first ofa series of outcome reports that will be published, andTADS will include a 1-year follow-up.

A program of research similar to the TADS trial thatfound a different pattern of results is the Time for aFuture–Adolescent Depression Program (Melvin et al.,2006). This program evaluates the effectiveness of threeinterventions for adolescents who met diagnostic criteriafor MDD, DD, or Depression NOS and had high ratesof comorbidity (e.g., anxiety, conduct disorder). Thethree interventions are CBT alone, sertraline alone,and a combination of both medication and CBT. Youthin all interventions showed within group improvementsin functioning. However, contrary to the TADS results,combined treatment was not superior to either CBTalone or medication alone. Adolescents in the CBTcondition had a significantly superior response thanadolescents in the medication-alone condition on ratesof depressive disorders at postintervention. Theresearchers express caution in interpreting and general-izing their results because of the medication titrationschedule they employed.

An additional multicenter trial currently underway isentitled the Treatment of Resistant Depression in Ado-lescents (http://www.wpic.pitt.edu/research/tordia/).This study targets 12- to 18-year-olds currently beingtreated for depression and whose depression has notresponded adequately to treatment. Youth are assignedrandomly to one of three medications: fluoxetine(Prozac), citalopram (Celexa), and venlafaxine (EffexorXR). These medications are administered either aloneor in combination with CBT. Results from this trial willoffer useful information about the relative efficacy ofvarious medication=psychosocial interventions, for arelatively understudied population, namely adolescentswith refractory depression.

Despite efficacy data, many parents and mentalhealth professionals are concerned about the use of anti-depressants in children and adolescents, in part becauseof recent warnings put out by the FDA (http://www.fda.gov/cder/drug/antidepressants=default.htm).In October 2004, the FDA requested that manufacturersof all antidepressant medications include in their label-ing a boxed warning (black box) and Patient Education

TREATMENTS FOR YOUTH DEPRESSION 97

Guide alerting consumers about the increased risk ofsuicidal thinking and behavior in youth treated withthese agents. The warnings include the following:

1. Youth with MDD or other psychiatric disorders whotake an antidepressant are at increased risk for suici-dal thinking or behavior.

2. When considering an antidepressant for a child oradolescent, it is important to weigh the increased riskof suicidality with possible benefits of the medi-cation.

3. Young people who initiate a trial of an antidepres-sant must be closely monitored for clinical worsen-ing, suicidality, or unusual behavior changes.

4. Family members and other caregivers must closelyobserve the youth for increased symptoms orworsening of functioning and immediately communi-cate with the provider about any such observations.

5. A statement regarding whether the particular medi-cation is approved for pediatric populations.

These FDA recommendations stem from a pooling ofdata across 24 short-term placebo-controlled antidepres-sant trials of more than 4,400 participants with MDD,obsessive–compulsive disorder, and other psychiatricdisorders. The overall results from these studies indi-cated an increased risk of suicidal thinking or behaviorin depressed youth (4% on active drug vs. 2% onplacebo; Hammad, Laughren, & Racoosin, 2006). Therewas no evaluation of the risk for untreated youth,long-term trials, or the combination of medication andpsychosocial interventions. There was no increase in sui-cidality reported for individuals with anxiety disorder,and there were no completed suicides in any studies.The Columbia Classification Algorithm of SuicideAssessment categorized suicide-related behavior in theFDA’s pediatric risk analysis, and this methodologyyielded more suicidal events but fewer suicide attempts(Posner, Oquendo, Gould, Stanley, & Davies, 2007).However, a recent matched case-control study of high-risk patients (e.g., severely depressed who required inpa-tient treatment) demonstrated children and adolescentstreated with antidepressants were significantly morelikely to attempt and complete suicide than youth whowere not treated with antidepressants (Olfson, Marcus,& Shaffer, 2006).

The use of pharmacological interventions, eitheralone or in combination with psychosocial interventions,with depressed youth is only beginning to be understood.Some studies have found a benefit of combined psycho-therapy treatment to be more effective than medicationalone (Clarke et al., 2005; TADS Team, 2004), whereasothers have not (Goodyer et al., 2007; Melvin et al.,2006), and the risk of youth using antidepressants is

uncertain. Clearly, more well-designed, randomizedcontrolled trials of these interventions are needed tonot only understand for whom these interventions maybe efficacious but also to understand under what con-ditions such interventions can be safely applied (Cheung,Emslie, & Mayes, 2005).

Recommendations for Best Practice

This review identifies a number of specific treatmentprograms and theoretical and modality approaches thatare efficacious in the treatment of depressed youth.However, no single intervention has emerged as the mostbeneficial, and effectiveness trials and examinations ofmediators and moderators of treatment are only begin-ning to emerge. CBT for children and CBT and IPTfor adolescents appear to be the most promising to basean intervention on, but this review also reveals that othertreatments may be effective and deserve consideration.To identify the most appropriate approach for a specificyouth and his or her family and to establish the youth’sbaseline level of functioning to help monitor treatmentprogress, treatment should begin with a thorough evalu-ation of a youth’s functioning. The evaluation shouldinclude information from multiple informants; providea comprehensive view of the patient’s strengths andweaknesses; and incorporate assessment tools appropri-ate for the youth’s developmental stage and, if possible,gender and cultural background.

Strengths and deficit areas identified by the evalu-ation will help determine the most appropriate treat-ment approach. For instance, if a depressed youth isfound to have dysfunctional thinking patterns, a thera-pist would begin with a CBT intervention that teachesthe patient to identify negative beliefs, evaluate the evi-dence for them, and generate more realistic alternatives.At the conclusion of the CBT course of treatment, are-evaluation of the youth’s functioning would deter-mine whether treatment can be terminated, whether arepeated course or a booster session of the CBT-basedintervention is needed, or whether another deficit area(e.g., interpersonal) is present that should be addressedthrough a different intervention activity (e.g., IPT-basedimproved communication intervention). Interventionsshould be applied sequentially and re-evaluation of thepatient’s needs should be done at each stage of treat-ment. In the selection of intervention components,therapists should consider how to capitalize on a youth’sinterests and areas of strength to support and enhancetreatment impact and increase the likelihood of engage-ment. For instance, a youth with strong language andreading skills may find the use of a journal and readingsthat supplement session by session CBT exercisesappealing and beneficial. Because depressed youth maybenefit from antidepressant medication, either alone or

98 DAVID-FERDON AND KASLOW

when provided in combination with psychosocial treat-ments (TADS Team, 2004), the possible benefit androle of medication should be part of the initial and ongoingevaluation of progress. Medications should be consideredin cases of moderate or severe depression (National Insti-tute for Health and Clinical Excellence, 2005). If medica-tions are provided, monitoring of functioning in accordwith FDA recommendations is necessary.

At each stage of intervention, therapists shouldmaintain the integrity of the treatment manual of theevidence-based intervention they selected. However,because of the reality of individual differences, thera-pists need to tailor the approach (e.g., frequency ofsessions, speed=intensity of session) to the needs andtreatment progress of the child, family, or group. Also,the modality that an intervention is conducted (e.g.,group vs. individual, inclusion of parent component)would depend on a number of factors (e.g., age of thepatient, treatment setting). Parental and=or entire familyinvolvement may be essential to support the generaliza-tion of treatment effects for young people and to effec-tively treat depressed youth with particular culturalbackgrounds (Tharp, 1991) even if not part of the orig-inal intervention protocol. Sensitivity to such issues mayoccur at simply the assessment phase (Rossello &Bernal, 1996, 1999) or be integrated into specifictreatment approaches (Yu & Seligman, 2002). Unfortu-nately, any of these modifications may negatively impactthe demonstrated efficacy of the intervention, andresearch is limited with regard to how to implement pro-grams found to be efficacious with fidelity to supportthe generalization of positive outcomes to a new treat-ment group or setting. Therapists should consult recom-mendations presented by a number of researchers andclinicians regarding making such adjustments (Bernal& Scharron-del-Rio, 2001; Nagayama-Hall & Okazaki,2002; Tharp, 1991). As the development and evaluationof psychosocial treatments for depressed youth progressforward, attention must be given to how evidence-basedinterventions can be implemented successfully whileallowing for variations in service delivery processes,level of therapist training, contextual factors, and indi-vidual needs of patients (Fagan & Mihalic, 2003; Filene,Lutzker, Hecht, & Silovsky, 2005).

REFERENCES

Achenbach, T. M., & Edelbrock, C. S. (1991). Manual for the child

behavior checklists=4-18 and 1991 profile. Burlington: University of

Vermont.

Achenbach, T. M., & Rescorla, L. A. (2001). Manual for the ASEBA

school-age forms and profiles: An integrated system of multi-

informant assessment. Burlington, VT: ASEBA.

Ackerson, J., Scogin, F., McKendree-Smith, N., & Lyman, R. D.

(1998). Cognitive bibliotherapy for mild and moderate adolescent

depressive symptomatology. Journal of Consulting and Clinical

Psychology, 66, 685–690.

American Academy of Child and Adolescent Psychiatry. (1998). Prac-

tice parameters for the assessment and treatment of children and

adolescents with depressive disorders. Journal of the American

Academy of Child and Adolescent Psychiatry, 37(Suppl.), 63S–83S.

American Academy of Child and Adolescent Psychiatry. (2004). Prac-

tice parameter for use of electroconvulsive therapy with adolescents.

Journal of the American Academy of Child and Adolescent Psy-

chiatry, 43, 1521–1539.

American Psychiatric Association. (1987). Diagnostic and statistical

manual of mental disorders–third edition–revised. Washington, DC:

American Psychiatric Press.

American Psychiatric Association. (1994). Diagnostic and statistical

manual of mental disorders–fourth edition. Washington, DC: Author.

American Psychological Association. (2003). Guidelines on multicul-

tural education, training, research, practice, and organizational

change for psychologists. American Psychologist, 58, 377–402.

Asarnow, J. R., Jaycox, L. H., Duan, N., LaBorde, A. P., Rea, M. M.,

Murray, P., et al. (2005). Effectiveness of a quality improvement

intervention for adolescent depression in primary care clinics: A

randomized controlled trial. Journal of the American Medical

Association, 293, 311–319.

Asarnow, J. R., Scott, C. V., & Mintz, J. (2002). A combined

cognitive–behavioral family education intervention for depression

in children: A treatment development study. Cognitive Therapy

and Research, 26, 221–229.

Barbe, R. P., Bridge, J., Birmaher, B., Kolko, D., & Brent, D. A. (2004).

Suicidality and its relationship to treatment outcome in depressed

adolescents. Suicide and Life Threatening Behavior, 34, 44–55.

Baron, R. M., & Kenny, D. A. (1986). The moderator–mediator vari-

able distinction in social psychological research: Conceptual, stra-

tegic, and statistical considerations. Journal of Personality and

Social Psychology, 51, 1173–1182.

Barrett, P. M., Farrell, L. J., Ollendick, T. H., & Dadds, M. R. (2006).

Long-term outcomes of an Australian universal prevention trial of

anxiety and depression symptoms in children and youth: An evalu-

ation of the FRIENDS program. Journal of Clinical Child and Ado-

lescent Psychology, 35, 403–411.

Barrett, P. M., & Turner, C. (2001). Prevention of anxiety symptoms

in primary school children: Preliminary results from a universal

school-based trial. British Journal of Clinical Psychology, 40,

399–410.

Beardslee, W. R., Gladstone, T. R. G., Wright, E. J., & Cooper, A. B.

(2003). A family-based approach to the prevention of depressive

symptoms in children at risk: Evidence of parental and child change.

Pediatrics, 112, 119–131.

Beck, A. T. (1970). Depression: Causes and treatment. Philadelphia:

University of Pennsylvania Press.

Beck, A. T., Steer, R. A., & Garbin, M. G. (1988). Psychometric

properties of the Beck Depression Inventory: Twenty-five years of

evaluation. Clinical Psychology Review, 8, 77–100.

Beck, A. T., Ward, C. H., Mendelson, M., Mock, J., & Erbaugh, J.

(1961). An inventory for measuring depression. Archives of General

Psychiatry, 4, 561–571.

Bernal, G. & Scharron-del-Rio, M. R. (2001). Are empirically

supported treatments valid for ethnic minorities? Toward an alterna-

tive approach for treatment research. Cultural Diversity and Ethnic

Minority Psychology, 7, 328–342.

Birmaher, B., Brent, D. A., Kolko, D., Baugher, M., Bridge, J.,

Holder, D., et al. (2000). Clinical outcome after short-term

psychotherapy for adolescents with major depressive disorder.

Archives of General Psychiatry, 57, 29–36.

Birmaher, B., Ryan, N. D., & Williamson, D. E. (1996). Depression in

children and adolescents: Clinical features and pathogenesis. In K. I.

TREATMENTS FOR YOUTH DEPRESSION 99

Shulman, M. Tohen, & S. P. Kutcher (Eds.), Mood disorders across

the life span (pp. 51–81). New York: Wiley.

Birmaher, B., Ryan, N. D., Williamson, D. E., Brent, D. A., Kaufman,

J., Dahl, R. E., et al. (1996). Childhood and adolescent depression:

A review of the past 10 years: Part I. Journal of the American

Academy of Child and Adolescent Psychiatry, 35, 1427–1439.

Boris, B., Brent, D. A., Kolko, D., Baugher, M., Bridge, J., Holder, D.,

et al. (2000). Clinical outcome after short-term psychotherapy for

adolescents with major depressive disorder. Archives of General Psy-

chiatry, 57, 29–36.

Bowlby, J. (1980). Attachment and loss: Volume 3. Sadness and

depression. Harmondsworth, UK: Penguin.

Breland-Noble, A. M., Bell, C. C., & Nicolas, G. (2006). Family first:

The development of an evidence-based family intervention for increas-

ing participation in psychiatric clinical care and research in depressed

African American adolescents. Family Process, 45, 153–169.

Brent, D. A., Holder, D., Kolko, D., Birmaher, B., Baugher, M., Roth,

C., et al. (1997). A clinical psychotherapy trial for adolescent

depression comparing cognitive, family, and supportive therapy.

Archives of General Psychiatry, 54, 877–885.

Brent, D. A., Kolko, D. J., Birmaher, B., Baugher, M., Bridge, J., Roth,

C., et al. (1998). Predictors of treatment efficacy in a clinical trial of

three psychosocial treatments for adolescent depression. Journal of

the American Academy of Child and Adolescent Psychiatry, 37, 906–914.

Brent, D. A. & Poling, K. (1997). Cognitive therapy treatment manual

for depressed and suicidal youth. Pittsburgh, PA: Star Center.

Burns, D. (1980). Feeling good. New York: Signet.

Butler, L., Miezitis, S., Friedman, R., & Cole, E. (1980). The effect of

two school-based intervention programs on depressive symptoms in

preadolescents. American Education Research Journal, 17, 111–119.

Cardemil, E. V., Reivich, K. J., Beevers, C. G., Seligman, M. E. P., &

James, J. (2007). The prevention of depressive symptoms in low-

income, minority children: Two-year follow-up. Behaviour Research

and Therapy, 45, 313–327.

Cardemil, E. V., Reivich, K. J., & Seligman, M. E. P. (2002). The pre-

vention of depressive symptoms in low-income minority middle

school students. Prevention and Treatment, 5.

Chambers, W. J., Puig-Antich, J., Hirsch, M., Paez, P., Ambrosini, P.,

Tabrizi, M. A., et al. (1985). The assessment of affective disorders in

children and adolescent by semistructured interview: Test–retest

reliability of the schedule for affective disorders and schizophrenia

for school-age children, present episode version. Archives of General

Psychiatry, 42, 696–702.

Chambless, D. L., Baker, M. J., Baucom, D. H., Beutler, L. E.,

Calhoun, K. S., Crits-Christoph, P., et al. (1998). Update on empiri-

cally validated therapies, II. The Clinical Psychologist, 51, 3–16.

Chambless, D. L., & Hollon, S. D. (1998). Defining empirically supported

therapies. Journal of Consulting and Clinical Psychology, 66, 7–18.

Chambless, D., Sanderson, W., Shoham, V., Johnson, S., Pope, K.,

Crits-Cristoph, P., et al. (1996). An update on empirically validated

therapies. The Clinical Psychologist, 49, 5–18.

Chaplin, T. M., Gillham, J. E., Reivich, K., Elkon, A. G. L., Samuels,

B., Freres, D. R., et al. (2006). Depression prevention for early ado-

lescent girls: A pilot study of all girls versus co-ed groups. Journal of

Early Adolescence, 26, 110–126.

Cheung, A. H., Emslie, G. J., & Mayes, T. L. (2005). Review of the

efficacy and safety of antidepressants in youth depression. Journal

of Child Psychology and Psychiatry, 46, 735–754.

Cicchetti, D., & Toth, S. L. (1998). The development of depression in

children and adolescents. American Psychologist, 53, 221–241.

Clarke, G. N., Debar, L., Lynch, F., Powell, J., Gale, J., O’Connor, E.,

et al. (2005). A randomized effectiveness trial of brief cognitive–

behavioral therapy for depressed adolescents receiving antidepres-

sant medication. Journal of the American Academy of Child and

Adolescent Psychiatry, 44, 888–898.

Clarke, G. N., DeBar, L. L., & Lewinsohn, P. M. (2003). Cognitive–

behavioral group treatment for adolescent depression. In A. E.

Kazdin & J. R. Weisz (Eds.), Evidence-based psychotherapies for chil-

dren and adolescents (pp. 120–147). New York: Guilford.

Clarke, G. N., Hawkins, W., Murphy, M., Sheeber, L., Lewinsohn,

P. M., & Seeley, J. (1995). Targeted prevention of unipolar depress-

ive disorder in an at risk sample of high school adolescents: A ran-

domized trial of a group cognitive intervention. American Academy

of Child and Adolescent Psychiatry, 34, 312–321.

Clarke, G. N., Hops, H., Lewinsohn, P. M., Andrews, J., Seeley, J. R.,

& Williams, J. (1992). Cognitive–behavioral group treatment of

adolescent depression: Prediction of outcome. Behavior Therapy,

23, 341–352.

Clarke, G. N., Hornbrook, M., Lynch, F., Polen, M., Gale, J.,

Beardslee, W. R., et al. (2001). A randomized trial of a group cog-

nitive intervention for preventing depression in adolescent offspring

of depressed parents. Archives of General Psychiatry, 58, 1127–1134.

Clarke, G. N., Hornbrook, M., Lynch, F., Polen, M., Gale, J.,

O’Connor, E., et al. (2002). Group cognitive–behavioral treatment

for depressed adolescent offspring of depressed parents in a health

maintenance organization. Journal of the American Academy of

Child and Adolescent Psychiatry, 41, 305–313.

Clarke, G. N., Lewinsohn, P. M., & Hops, H. (1990). Adolescent cop-

ing with depression course. Retrieved from http://www.kpchr.org/

acwd/acwd.html. Portland, OR: Kaiser Permanent.

Clarke, G. N., Rohde, P., Lewinsohn, P. M., Hops, H., & Seeley, J.

(1999). Cognitive–behavioral treatment of adolescent depression: Effi-

cacy of acute group treatment and booster sessions. Journal of the

American Academy of Child and Adolescent Psychiatry, 38, 272–279.

Coopersmith, S. (1967). The antecedents of self-esteem. San Francisco:

Freeman.

Coopersmith, S. (1975). Self-esteen inventory–form B. Lafayette, CA:

Self-Esteem Institute.

Cottrell, D. (2003). Outcome studies of family therapy in child and

adolescent depression. Journal of Family Therapy, 25, 406–416.

Cramer, B., & Palacio Espasa, F. (1993). La pratique des psychotherapies

meres-bebes (The practice of maternal-infant psychotherapy). Paris:

PUF.

Curry, J., Rohde, P., Simons, A., Silva, S., Vitiello, B., Kratochvil,

C. J., et al. (2006). Predictors and moderators of acute outcome in

the Treatment for Adolescents with Depression Study (TADS).

Journal of the American Academy of Child and Adolescent Psy-

chiatry, 45, 1427–1439.

Curry, J. F., Wells, K. C., Lochman, J. E., Craighead, W. E., & Nagy, P.

D. (2003). Cognitive–behavioral intervention for depressed, substance-

abusing adolescents: Development and pilot testing. Journal of the

American Academy of Child and Adolescent Psychiatry, 42, 656–665.

Debar, L., Clarke, G. N., O’Connor, E., & Nichols, G. A. (2001).

Treated prevalence, incidence, and pharmacotherapy of child and

adolescent mood disorders in an HMO. Mental Health Services

Research, 3, 73–89.

De Cuyper, S., Timbremont, B., Braet, C., De Backer, V., & Wullaert,

T. (2004). Treating depressive symptoms in school children: A pilot

study. European Child and Adolescent Psychiatry, 13, 105–114.

Delate, T., Gelenberg, A. J., Simmons, V. A., & Motheral, B. R.

(2004). Trends in the use of antidepressants in a national sample

of commercially insured pediatric patients, 1998–2002. Psychiatric

Services, 55, 387–391.

Diamond, G. S., Reis, B. F., Diamond, G. M., Siqueland, L., & Isaacs,

L. (2002). Attachment-based family therapy for depressed adoles-

cents: A treatment development study. Journal of the American

Academy of Child and Adolescent Psychiatry, 41, 1190–1196.

Durlak, J. A., Fuhrman, T., & Lampman, C. (1991). Effectiveness

of cognitive–behavior therapy for maladapting children: A meta-

analysis. Psychological Bulletin, 110, 204–214.

100 DAVID-FERDON AND KASLOW

Emery, R. E., & O’Leary, K. D. (1982). Children’s perceptions of

marital discord and behavior problems of boys and girls. Journal

of Abnormal Child Psychology, 10, 11–24.

Emslie, G. J., Heiligenstein, J. H., Wagner, K. D., Hoog, S. L., Ernest,

D. E., Brown, E., et al. (2002). Fluoxetine for active treatment of

depression in children and adolescents: A placebo-controlled, rando-

mized clinical trial. Journal of the American Academy of Child and

Adolescent Psychiatry, 41, 1205–1215.

Emslie, G. J., Hughes, C. W., Crimson, M. L., Lopez, M., Pliszka,

S. R., Toprac, M. G., et al. (2004). A feasibility study of the child-

hood depression medication algorithm: The Texas Children’s Medi-

cation Algorithm Project (CMAP). Journal of the American

Academy of Child and Adolescent Psychiatry, 43, 519–527.

Emslie, G. J., Kratochvil, C. J., Vitiello, B., Silva, S., Mayes, T.,

McNulty, S., et al. (2006). Treatment for Adolescents with

Depression Study (TADS): Safety results. Journal of the American

Academy of Child and Adolescent Psychiatry, 45, 1440–1455.

Emslie, G. J. & Mayes, T. L. (2001). Mood disorders in children and

adolescents: Psychopharmacological treatment. Biological Psy-

chiatry, 49, 1082–1090.

Emslie, G. J., Rush, A. J., Weinberg, W. A., Kowatch, R. A., Hughes,

C. W., Carmody, T., et al. (1997). A double-blind, randomized,

placebo-controlled trial of fluoxetine in children and adolescents

with depression. Archives of General Psychiatry, 54, 1031–1037.

Endicott, J., Cohen, J., Nee, J., Fleiss, J. L., & Sarantakos, S. (1981).

Hamilton depression rating scale: Extracted from regular and

change versions of the schedule for affective disorders and

schizophrenia. Archives of General Psychiatry, 38, 98–103.

Fagan, A. A., & Mihalic, S. (2003). Strategies for enhancing adoption

of school-based prevention programs: Lessons learned from the

blueprints for violence prevention replications of life skills training

program. Journal of Community Psychology, 31, 235–253.

Farrell, L. J., & Barrett, P. M. (2007). Prevention of childhood emotional

disorders: Reducing the burden of suffering associated with anxiety

and depression. Child and Adolescent Mental Health, 12, 58–65.

Filene, J. H., Lutzker, J. R., Hecht, D., & Silovsky, J. (2005). Project

SafeCare–Issues in replicating an ecobehavioral model of child mal-

treatment prevention. In K. A. Kendall-Tackett & S. M. Giacomoni

(Eds.), Child victimization: Maltreatment, bullying and dating

violence, prevention and intervention (pp. 18-11—18-18). Kingston,

NJ: Civic Research Institute.

Fine, M. A., & Carlson, C. I. (Eds.). (1991). Handbook of family-school

intervention: A systems perspective. Boston: Allyn & Bacon.

Fine, S., Forth, A., Gilbert, M., & Haley, G. (1991). Group therapy for

adolescent depressive disorder: A comparison of social skills and

therapeutic support. Journal of the American Academy of Child

and Adolescent Psychiatry, 30, 79–85.

Fristad, M. A., Arnett, M. M., & Gavazzi, S. M. (1998). The impact of

psychoeducational workshops on families of mood-disordered

children. Family Therapy, 25, 151–159.

Fristad, M. A., Gavazzi, S. M., & Soldano, K. W. (1998). Multi-family

psychoeducation groups for childhood mood disorders: A program

description and preliminary efficacy data. Contemporary Family

Therapy, 20, 385–402.

Fristad, M. A., Goldberg-Arnold, J. S., & Gavazzi, S. M. (2003). Multi-

family psychoeducation groups in the treatment of children with

mood disorders. Journal of Marital and Family Therapy, 29, 491–504.

Galambos, N. L., Leadbeater, B. J., & Barker, E. T. (2004). Gender differ-

ences in and risk factors for depression inadolescence: A 4-year longitudi-

nal study. International Journal of Behavioral Development, 28, 16–25.

Garber, J. (2006). Depression in children and adolescents: Linking risk

research and prevention. American Journal of Preventive Medicine,

31, S104–S125.

Gillham, J. E., Hamilton, J., Freres, D. R., Patton, K., & Gallop, R.

(2006). Preventing depression among early adolescents in the

primary care setting: A randomized controlled study of the Penn

Resiliency Program. Journal of Abnormal Child Psychology, 34,

203–219.

Gillham, J. E., & Reivich, K. J. (1999). Prevention of depressive symp-

toms in school children: A research update. Psychological Science,

10, 461–462.

Gillham, J. E., Reivich, K. J., Freres, D. R., Chaplin, T. M., Shatte,

A. J., Samuels, B., et al. (2007). School-based prevention of depress-

ive symptoms: A randomized controlled study of the effectiveness

and specificity of the Penn Resiliency Program. Journal of Consult-

ing and Clinical Psychology, 75, 9–19.

Gillham, J. E., Reivich, K., Freres, D. R., Lascher, M., Litzinger, S.,

Shatte, A. J., et al. (2006). School-based prevention of depression

and anxiety symptoms in early adolescence: A pilot of a parent inter-

vention component. School Psychology Quarterly, 21, 323–348.

Gillham, J. E., Reivich, K., Jaycox, L., & Seligman, M. E. P. (1995).

Prevention of depressive symptoms in school children: Two year

follow-up. Psychological Science, 6, 343–351.

Goodyer, I. M., Dubicka, B., Wilkinson, P., Kelvin, R., Roberts, C.,

Byford, S., et al. (2007). Selective serotonin reuptake inhibitors

(SSRIs) and routine specialist care with and without cognitive

behaviour therapy in adolescents with major depression: Rando-

mised controlled trial. British Medical Journal. doi:10.1136=bmj.

39224.494340.494355

Griffith, J., Zucker, M., Bliss, M., Foster, J., & Kaslow, N. J. (2001).

Family interventions for depressed African American adolescent

females. Innovations in Clinical Practice, 19, 159–173.

Hammad, T. A., Laughren, T., & Racoosin, J. (2006). Suicidality in

pediatric patients treated with antidepressant drugs. Archives of

General Psychiatry, 63, 323–329.

Hannan, A. P., Rapee, R. M., & Hudson, J. L. (2000). The prevention

of depression in children: A pilot study. Behaviour Change, 17, 78–83.

Holmbeck, G. N. (1997). Toward terminological, conceptual, and

statistical clarity in the study of mediators and moderators: Exam-

ples from the child–clinical and pediatric psychology literatures.

Journal of Consulting and Clinical Psychology, 65, 599–610.

Holmbeck, G. N. (2002). Post-hoc probing of significant moderational

and mediational effects in studies of pediatric populations. Journal

of Pediatric Psychology, 27, 87–96.

Horowitz, J. L., & Garber, J. (2006). The prevention of depressive

symptoms in children and adolescents: A meta-analytic review.

Journal of Consulting and Clinical Psychology, 74, 401–415.

Hughes, C. W., Emslie, G. J., Crismon, M. L., Posner, K., Birmaher,

B., Ryan, N. D., et al. (2007). Texas children’s medication algorithm

project: Update from texas consensus conference panel on medi-

cation treatment of childhood major depressive disorder. Journal

of the American Academy of Child and Adolescent Psychiatry, 46,

667–686.

Hughes, C. W., Emslie, G. J., Crismon, M. L., Wagner, K. D.,

Birmaher, B., Geller, B., et al. (1999). Texas children’s medication

algorithm project: Report of the texas consensus conference panel

on medication treatment of childhood major depressive disorder.

Journal of the American Academy of Child and Adolescent

Psychiatry, 38, 1442–1454.

Jaycox, L., Reivich, K., Gillham, J. E., & Seligman, M. E. P. (1994).

Prevention of depressive symptoms in school children. Behavioral

Research and Therapy, 32, 801–816.

Jayson, D., Wood, A., Kroll, L., Fraser, J., & Harrington, R. (1998).

Which depressed patients respond to cognitive–behavioral treat-

ment. Journal of the American Academy of Child and Adolescent

Psychiatry, 37, 35–39.

John, K., Gammon, D., Prusoff, B., & Warner, V. (1987). The Social

Adjustment Inventory for Children and Adolescents (SAICA):

Testing a new semi-structured interview. Journal of the American

Academy of Child and Adolescent Psychiatry, 26, 898–911.

TREATMENTS FOR YOUTH DEPRESSION 101

Kahn, J., Kehle, T., Jenson, W., & Clark, E. (1990). Comparison of

cognitive behavioral, relaxation, and self-modeling interventions

for depression among middle-school students. School Psychology

Review, 19, 196–211.

Kaslow, N. J., Deering, C., & Ash, P. (1996). Relational diagnosis of

child and adolescent depression. In F. W. Kaslow (Ed.), Handbook

of relational diagnosis and dysfunctional family patterns (pp. 171–

185). New York: Wiley.

Kaslow, N. J., Deering, C., & Racusin, G. (1994). Depressed children

and their families. Clinical Psychology Review, 14, 39–59.

Kaslow, N. J., Mintzer, M., Meadows, L., & Grabill, C. (2000). A fam-

ily perspective on assessing and treating childhood depression. In

E. Bailey (Ed.), Working with children: Using the family as a resource

in children’s therapy (pp. 215–241). New York: Norton.

Kaslow, N. J., & Thompson, M. (1998). Applying the criteria for

empirically supported treatments to studies of psychosocial interven-

tions for child and adolescent depression. Journal of Clinical Child

Psychology, 27, 146–155.

Kaufman, N. K., Rohde, P., Seeley, J. R., Clarke, G., & Stice, E.

(2005). Potential mediators of cognitive–behavioral therapy for ado-

lescents with comorbid major depression and conduct disorder.

Journal of Consulting and Clinical Psychology, 73, 38–46.

Keller, M. B., Ryan, N. D., Strober, M., Klein, R. G., Kutcher, S.,

Birmaher, B., et al. (2001). Efficacy of paroxetine in the treatment

of adolescent major depression: A randomized, controlled trial.

Journal of the American Academy of Child and Adolescent

Psychiatry, 40, 762–772.

Kennard, B., Silva, S., Vitiello, B., Curry, J., Kratochvil, C. J.,

Simons, A., et al. (2006). Remission and residual symptoms after

short-term treatment in the Treatment of Adolescents with

Depression Study (TADS). Journal of the American Academy of

Child and Adolescent Psychiatry, 45, 1404–1411.

Kerfott, M., Harrington, R., Harrington, V., Rogers, J., & Verduyn,

C. (2004). A step too far? Randomized trial of cognitive–behaviour

therapy delivered by social workers to depressed adolescents.

European Journal of Child and Adolescent Psychiatry, 13, 92–99.

King, C. A., & Kirschenbaum, D. S. (1990). An experimental

evaluation of a school-based program for children at risk: Wisconsin

early-intervention. Journal of Community Psychology, 18,

167–177.

Kleinman, A., & Good, B. (1985). Culture and depression. Berkeley:

University of California Press.

Kolko, D., Brent, D. A., Baugher, M., Bridge, J., & Birmaher, B.

(2000). Cognitive and family therapies for adolescent depression:

Treatment specificity, mediation, and moderation. Journal of

Consulting and Clinical Psychology, 68, 603–614.

Kovacs, M. (1992). Children’s depression inventory. North Tonawanda,

NY: Multi-Health Systems.

Kovacs, M., Sherrill, J., George, C. J., Pollock, M., Tumuluru, R. V.,

& Ho, V. (2006). Contextual emotion–regulation therapy for child-

hood depression: Description and pilot testing of a new intervention.

Journal of the American Academy of Child and Adolescent

Psychiatry, 45, 892–903.

Kowalenko, N., Rapee, R. M., Simmons, J., Wignall, A., Hoge, R.,

Whitefield, K., et al. (2005). Short-term effectiveness of a school-

based entry intervention program for adolescent depression. Clinical

Child Psychology and Psychiatry, 10, 493–507.

Kraemer, H. C., Wilson, T., Fairburn, C. G., & Agras, W. S. (2002).

Mediators and moderators of treatment effects in randomized clini-

cal trials. Archives of General Psychiatry, 59, 877–883.

Kratochvil, C. J., Emslie, G. J., Silva, S., McNulty, S., Walkup, J.,

Curry, J., et al. (2006). Acute time to response in the Treatment

for Adolescents with Depression Study (TADS). Journal of

the American Academy of Child and Adolescent Psychiatry, 45,

1412–1418.

Lewinsohn, P. M., Clarke, G., Hops, H., & Andrews, J. (1990). Cog-

nitive–behavioral treatment for depressed adolescents. Behavior

Therapy, 21, 385–401.

Lewinsohn, P. M., Clarke, G., Rohde, P., Hops, H., & Seeley, J.

(1996). A course in coping: A cognitive–behavioral approach to

the treatment of adolescent depression. In E. D. Hibbs & P. S.

Jensen (Eds.), Psychosocial treatments for child and adolescent disor-

ders: Empirically based strategies for clinical practice (pp. 109–135).

Washington, DC: American Psychological Association.

Liddle, B., & Spence, S. (1990). Cognitive behavior therapy with

depressed primary school children: A cautionary note. Behavioural

Psychotherapy, 18, 85–102.

Lonigan, C. J., Elbert, J. C., & Johnson, S. B. (1998). Empirically sup-

ported psychosocial interventions for children: An overview. Journal

of Clinical Child Psychology, 27, 138–145.

March, J. S., Silva, S., Vitiello, B., & The TADS Team. (2006). The

Treatment for Adolescents with Depression Study (TADS): Meth-

ods and message at 12 weeks. Journal of the American Academy of

Child and Adolescent Psychiatry, 45, 1393–1403.

Marsella, A. J., & Kaplan, A. (2002). Cultural considerations for

understanding, assessing, and treating depressive experience and dis-

order. In M. A. Reinecke & M. R. Davison (Eds.), Comparative

treatments for depression (pp. 47–78). New York: Springer.

May, D. E., Kratochvil, C. J., Puumala, S. E., Silva, S. G., Rezac, A. J.,

Hallin, M. J., et al. (2007). A manual-based intervention to address

clinical crises and retain patients in the Treatment of Adolescents

with Depression Study (TADS). Journal of the American Academy

of Child and Adolescent Psychiatry, 46, 573–581.

McClure, E. B., Connell, A., Zucker, M., Griffith, J. R., & Kaslow,

N. J. (2005). The Adolescent Depression Empowerment Project

(ADEPT): A culturally sensitive family treatment for depressed,

African American girls. In E. D. Hibbs & P. S. Jensen (Eds.),

Psychosocial treatments for child and adolescent disorders: Empiri-

cally based strategies for clinical practice (2nd edition) (pp. 149–

164). Washington, DC: American Psychological Association.

McDowell, T., Ingoglia, L., Serizawa, T., Holland, C., Dashiell, J. W.,

& Stevens, C. (2005). Raising multiracial awareness in family ther-

apy through critical conversations. Journal of Marital and Family

Therapy, 31, 399–411.

Melvin, G. A., Tonge, B. J., King, N. J., Heyne, D., Gordon, M. S.,

& Klimkeit, E. (2006). A comparison of cognitive–behavioral ther-

apy, sertraline, and their combination for adolescent depression.

Journal of the American Academy of Child and Adolescent

Psychiatry, 45, 1151–1161.

Merry, S., McDowell, H., Wild, C. J., Bir, J., & Cunliffe, R. (2004). A

randomized placebo-controlled trial of a school-based depression

prevention program. Journal of the American Academy of Child

and Adolescent Psychiatry, 43, 538–547.

Michael, K. D., & Crowley, S. L. (2002). How effective are treatments

for child and adolescent depression? A meta-analytic review. Clinical

Psychology Review, 22, 247–269.

Moos, R. H., & Moos, B. S. (1981). Family environment scale manual.

Palo Alto, CA: Consulting Psychological Press.

Mufson, L. H., Dorta, K. P., Olfson, M., Weissman, M. M., &

Hoagwood, K. (2004). Effectiveness research: Transporting interper-

sonal psychotherapy for depressed adolescents (IPT–A) from the lab

to school-based health clinics. Clinical Child and Family Psychology

Review, 7, 251–261.

Mufson, L. H., Dorta, K. P., Wickramaratne, P., Nomura, Y., Olfson,

M., & Weissman, M. M. (2004). A randomized effectiveness trial of

interpersonal psychotherapy for depressed adolescents. Archives of

General Psychiatry, 61, 577–584.

Mufson, L. H., Weissman, M. M., Moreau, D., & Garfinkel, R.

(1999). Efficacy of interpersonal psychotherapy for depressed ado-

lescents. Archives of General Psychiatry, 56, 573–579.

102 DAVID-FERDON AND KASLOW

Muratori, F., Picchi, L., Bruni, G., Patarnello, M., & Romagnoli, G.

(2003). A two-year follow-up of psychodynamic psychotherapy for

internalizing disorders in children. Journal of the American Academy

of Child and Adolescent Psychiatry, 42, 331–339.

Nagayama-Hall, G., & Okazaki, S. (2002). Asian American psychology:

The science of lives in context. Washington, DC: American

Psychological Association.

Nathan, P. E., & Gorman, J. M. (Eds.). (2002). A guide to treatments

that work (2nd ed.). London: Oxford University Press.

National Institute for Health and Clinical Excellence. (2005).

Depression in children and young people: Identification and manage-

ment in primary, community, and secondary care. London: National

Institute for Health and Clinical Excellence.

Nelson, E. L., Barnard, M., & Cain, S. (2003). Treating childhood depression

over videoconferencing. Telemedicine Journal and e-Health, 9, 49–55.

Nolen-Hoeksema, S. (2002). Gender differences in depression. In I. H.

Gotlib & C. L. Hammen (Eds.), Handbook of depression (pp. 492–

509). New York: Guilford.

Nolen-Hoeksema, S., & Girgus, J. S. (1994). The emergence of gender

differences in depression during adolescence. Psychological Bulletin,

115, 424–443.

Nolen-Hoeksema, S., Larson, J., & Grayson, C. (1999). Explaining the

gender difference in depressive symptoms. Journal of Personality and

Social Psychology, 77, 1061–1072.

Olfson, M., Gameroff, M. J., Marcus, S. C., & Waslick, B. D. (2003).

Outpatient treatment of child and adolescent depression in the

United States. Archives of General Psychiatry, 60, 1236–1242.

Olfson, M., Marcus, S. C., & Shaffer, D. A. (2006). Antidepressant

drug therapy and suicide in severely depressed children and adults:

A case-control study. Journal of the American Academy of Child

and Adolescent Psychiatry, 63, 865–872.

Pattison, C., & Lynd-Stevenson, R. M. (2001). The prevention of

depressive symptoms in children: The immediate and long-term out-

comes of a school-based program. Behaviour Change, 18, 92–102.

Pfeffer, C., Jiang, H., Kakuma, T., Hwang, J., & Metsch, M. (2002).

Group intervention for children bereaved by the suicide of a relative.

Journal of the American Academy of Child and Adolescent

Psychiatry, 41, 505–513.

Posner, K., Oquendo, M. A., Gould, M., Stanley, B., & Davies, M.

(2007). Columbia Classification Algorithm of Suicide Assessment

(C–CASA): Classification of suicidal events in the FDA’s pediatric

suicidal risk analysis of antidepressants. American Journal of Psy-

chiatry, 164, 1035–1043.

Possel, P., Baldus, C., Horn, A. B., Groen, G., & Hautzinger, M.

(2005). Influence of general self-efficacy on the effects of a school-

based universal primary prevention program of depression symp-

toms in adolescents: A randomized and controlled follow-up study.

Journal of Child Psychology and Psychiatry, 46, 982–994.

Possel, P., Horn, A. B., Groen, G., & Hautzinger, M. (2004). School-

based prevention of depressive symptoms in adolescents: A six-

month follow-up. Journal of the American Academy of Child and

Adolescent Psychiatry, 43, 1003–1010.

Poznanski, E. O., Grossman, J. A., Buschsbaum, Y., Banges, M.,

Freeman, L., & Gibbons, R. (1984). Preliminary studies of the

reliability of the children’s depression rating scale. Journal of the

American Academy of Child Psychiatry, 23, 191–197.

Radloff, L. S. (1977). The CES–D Scale: A self-report depression scale

for research in the general population. Applied Psychological

Measurement, 1, 385–401.

Reynolds, W. M. (1989). Reynolds child depression scale: Professional

manual. Odessa, FL: Psychological Assessment Resources.

Reynolds, W. M., & Coats, K. (1986). A comparison of cognitive–

behavioral therapy and relaxation training for the treatment of

depression in adolescents. Journal of Consulting and Clinical Psy-

chology, 54, 653–660.

Roberts, C., Kane, R., Thomson, H., Bishop, B., & Hart, B. (2003).

The prevention of depressive symptoms in rural school children: A

randomized controlled trial. Journal of Consulting and Clinical Psy-

chology, 71, 622–628.

Roberts, R. E., Alegria, M., Roberts, C. R., & Chen, I. G. (2005). Con-

cordance of reports on mental health functioning by adolescents and

their caregivers: A comparison of European, African, and Latino

Americans. Journal of Nervous and Mental Disease, 193, 523–534.

Robinson, L. A., Berman, J. S., & Neimeyer, R. A. (1990). Psycho-

therapy for treatment of depression: A comprehensive review of con-

trolled outcome research. Psychological Bulletin, 108, 30–49.

Rohde, P., Clarke, G. N., Lewinsohn, P. M., Seeley, J. R., &

Kaufman, N. K. (2001). Impact of comorbidity on a cognitive–beha-

vioral group treatment for adolescent depression. Journal of the

American Academy of Child and Adolescent Psychiatry, 40, 795–802.

Rohde, P., Clarke, G., Mace, D. E., Jorgensen, J. S., & Seeley, J. R.

(2004). An efficacy=effectiveness study of cognitive–behavioral

treatment for adolescents with comorbid major depression and con-

duct disorder. Journal of the American Academy of Child and

Adolescent Psychiatry, 43, 660–668.

Rohde, P., Seeley, J. R., Kaufman, N. K., Clarke, G. N., & Stice, E.

(2006). Predicting time to recovery among depressed adolescents

treated in two psychosocial group interventions. Journal of Consult-

ing and Clinical Psychology, 74, 80–88.

Rossello, J., & Bernal, G. (1996). Adapting cognitive–behavioral and

interpersonal treatments for depressed puerto rican adolescents. In

E. D. Hibbs & P. Jensen (Eds.), Psychosocial treatments for child

and adolescent disorders (pp. 157–185). Washington, DC: American

Psychological Association.

Rossello, J., & Bernal, G. (1999). The efficacy of cognitive–behavioral

and interpersonal treatments for depression in Puerto Rican adoles-

cents. Journal of Consulting and Clinical Psychology, 67, 734–745.

Sanford, M., Boyle, M., McCleary, L., Miller, J. G., Steele, M.,

Duku, E., et al. (2006). A pilot study of adjunctive family psychoe-

ducation in adolescent major depression: Feasibility and treatment

effect. Journal of the American Academy of Child and Adolescent

Psychiatry, 45, 386–395.

Santor, D. A., & Kusumaker, V. (2001). Open trial of interpersonal

therapy in adolescents with moderate to severe major depression:

Effectiveness of novice IPT therapists. Journal of the American

Academy of Child and Adolescent Psychiatry, 40, 236–240.

Schoenwald, S. K., & Hoagwood, K. (2001). Effectiveness, transport-

ability, and dissemination of interventions: What matters when?

Psychiatric Services, 52, 1190–1197.

Seligman, M. E. P., Peterson, C., Kaslow, N. J., Tanenbaum, R. L., &

Abramson, L. Y. (1984). Attributional style and depressive symp-

toms among children. Journal of Abnormal Psychology, 93, 235–238.

Sethi, S., & Nolen-Hoeksema, S. (1997). Gender differences in internal

and external focusing among adolescents. Sex Roles, 37, 687–700.

Sheffield, J. K., Spence, S. H., Rapee, R. M., Kowalenko, N., Wignall,

A., Davis, A., et al. (2006). Evaluation of universal, indicated, and

combined cognitive–behavioral approaches to the prevention of

depression in adolescents. Journal of Consulting and Clinical Psy-

chology, 74, 66–79.

Shochet, I. M., & Ham, D. (2004). Universal school-based approaches

to preventing adolescent depression: Past findings and future direc-

tions of the Resourceful Adolescent Program. International Journal

of Mental Health Promotion, 6, 17–25.

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.

Spielmans, G. I., Pasek, L. F., & McFall, J. P. (2007). What are the

active ingredients in cognitive and behavioral psychotherapy for

anxious and depressed children? A meta-analytic. Clinical Psy-

chology Review, 27, 642–654.

TREATMENTS FOR YOUTH DEPRESSION 103

Stark, K. D., Hargrave, J., Hersh, B., Greenberg, M., Herren, J., &

Fisher, M. (in press). Treatment of childhood depression: The

ACTION treatment program. In J. R. Z. Abela & B. L. Hankin

(Eds.), Child and adolescent depression: Causes, treatment, and pre-

vention. New York: Guilford.

Stark, K. D., Hargrave, J., Sander, J., Schnoebelen, S., Simpson, J., &

Molnar, J. (2006). Treatment of childhood depression: The

ACTION treatment program. In P. C. Kendall (Ed.), Child and ado-

lescent therapy: Cognitive–behavioral procedures (3rd edition) (pp.

169–216). New York: Guilford.

Stark, K. D., Reynolds, W. M., & Kaslow, N. J. (1987). A comparison

of the relative efficacy of self-control therapy and behavior problem-

solving therapy for depression in children. Journal of Abnormal

Child Psychology, 15, 91–113.

Stark, K. D., Rouse, L., & Livingston, R. (1991). Treatment of

depression during childhood and adolescence: Cognitive behavioral

procedures for the individual and family. In P. Kendall (Ed.), Child

and adolescent therapy (pp. 165–206). New York: Guilford.

Szigethy, E., Whitton, S. W., Levy-Warrne, A., Demaso, D. R., Weisz,

J. R., & Beardslee, W. R. (2004). Cognitive–behavioral therapy for

depression in adolescents with inflammatory bowel disease: A pilot

study. Journal of the American Academy of Child and Adolescent

Psychiatry, 43, 1469–1477.

Tharp, R. G. (1991). Cultural diversity and treatment of children.

Journal of Consulting and Clinical Psychology, 59, 799–812.

The Treatment for Adolescents with Depression Study Team. (2003).

Treatment for Adolescents with Depression Study (TADS): Ration-

ale, design, and methods. Journal of the American Academy of Child

and Adolescent Psychiatry, 42, 531–542.

The Treatment for Adolescents with Depression Study Team. (2005).

The Treatment for Adolescents with Depression Study (TADS):

Demographic and clinical characteristics. Journal of the American

Academy of Child and Adolescent Psychiatry, 44, 28–40.

Treatment for Adolescents with Depression Study (TADS) Team.

(2004). Fluoxetine, cognitive–behavioral therapy, and their combi-

nation for adolescents with depression: Treatment for Adolescents

with Depression Study (TADS) randomized controlled trial. Journal

of the American Medical Association, 292, 807–820.

Trowell, J., Joffe, I., Campbell, J. C., Clemente, C., Almqvist, F.,

Soininen, M., et al. (2007). Childhood depression: A place for

psychotherapy. An outcome study comparing individual psychody-

namic psychotherapy and family therapy. European Journal of Child

and Adolescent Psychiatry, 16, 157–167.

Vitiello, B., Rohde, P., Silva, S., Wells, K. C., Casat, C., Waslick, B. D.,

et al. (2006). Functioning and quality of life in the Treatment for

Adolescents with Depression Study (TADS). Journal of

the American Academy of Child and Adolescent Psychiatry, 45,

1419–1426.

Vostanis, P., Feehan, C., Grattan, E., & Bickerton, W. (1996a). A ran-

domized controlled outpatient trial of cognitive behavioral treat-

ment for children and adolescents with depression: 9-month

follow-up. Journal of Affective Disorders, 40, 105–116.

Vostanis, P., Feehan, C., Grattan, E., & Bickerton, W. (1996b). Treat-

ment for children and adolescents with depression: Lessons from a

controlled trial. Clinical Child Psychology and Psychiatry, 1, 199–

212.

Wade, T. J., Cairney, J., & Pevalin, D. J. (2002). Emergence of gender

differences in depression during adolescence: National panel results

from three countries. Journal of the American Academy of Child and

Adolescent Psychiatry, 41, 190–198.

Wagner, K. D., & Ambrosini, P. (2001). Childhood depression:

Pharmacological therapy=treatment (Pharmacotherapy of child-

hood depression). Journal of Clinical Child Psychology, 30, 88–97.

Wagner, K. D., Ambrosini, P., Rynn, M., Wohlberg, C., Yang, R.,

Greenbaum, M. S., et al. (2003). Efficacy of sertraline in the treat-

ment of children and adolescents with major depressive disorder:

Two randomized controlled trials. Journal of the American Medical

Association, 290, 1033–1041.

Wagner, K. D., Jonas, J., Findling, R. L., Ventura, D., & Saikali, K.

(2006). A double-blind, randomized, placebo-controlled trial of esci-

talopram in the treatment of pediatric depression. Journal of the

American Academy of Child and Adolescent Psychiatry, 45, 280–288.

Wagner, K. D., Robb, A. S., Findling, R. L., Jin, J., Gutierrez, M. M.,

& Heydorn, W. E. (2004). A randomized, placebo-controlled trial of

citalopram for the treatment of major depression in children and

adolescents. American Journal of Psychiatry, 161, 1079–1083.

Watanabe, N., Hunot, V., Omori, I. M., Churchill, R., & Furukawa,

T. A. (2007). Psychotherapy for depression among children and adoles-

cents: A systematic review. Acta Psychiatrica Scandinavica, 116, 84–95.

Weersing, V. R., & Brent, D. A. (2003). Cognitive–behavioral therapy

for adolescent depression: Comparative efficacy, mediation, moder-

ation, and effectiveness. In A. E. Kazdin & J. R. Weisz (Eds.), Evi-

dence-based psychotherapies for children and adolescents (pp. 135–

147). New York: Guilford.

Weersing, V. R., & Weisz, J. R. (2002). Mechanisms of action in youth

psychotherapy. Journal of Child Psychology and Psychiatry, 43, 3–29.

Weisz, J. R., Jensen, A. L., & McLeod, B. D. (2005). Development and

dissemination of child and adolescent psychotherapies: Milestones,

methods, and a new deployment-focused model. In E. D. Hibbs &

P. S. Jensen (Eds.), Psychosocial treatments for child and adolescent

disorders: Empirically based strategies for clinical practice (2nd

edition) (pp. 9–39). Washington, DC: American Psychological

Association.

Weisz, J. R., McCarty, C. A., & Valeri, S. M. (2006). Effects of psycho-

therapy for depression in children and adolescents: A meta-analysis.

Psychological Bulletin, 132, 132–149.

Weisz, J. R., Southam-Gerow, M. A., Gordis, E. B., & Connor-Smith,

J. (2003). Primary and secondary control enhancement training for

youth depression: Applying the deployment-focused model of treat-

ment development and testing. In A. E. Kazdin & J. R. Weisz (Eds.),

Evidence-based psychotherapies for chidlren and adolescents (pp. 165–

183). New York: Guilford.

Weisz, J. R., Thurber, C., Sweeney, L., Proffitt, V., & LeGagnoux, G.

(1997). Brief treatment of mild to moderate child depression using

primary and secondary control enhancement training. Journal of

Consulting and Clinical Psychology, 65, 703–707.

Wells, K. B., Kataoka, S. H., & Asarnow, J. R. (2001). Affective dis-

orders in children and adolescents: Addressing unmet need in pri-

mary care settings. Biological Psychiatry, 49, 1111–1120.

Wintre, M. G., Hicks, R., McVey, G., & Fox, J. C. (1988). Age and sex

differences in choice of consultant for various types of problems.

Child Development, 59, 1046–1055.

Wood, A., Harrington, R., & Moore, A. (1996). Controlled trial of a

brief cognitive–behavioural intervention in adolescent patients with

depressive disorders. Journal of Child Psychology and Psychiatry, 37,

737–746.

Young, J. F., Mufson, L. H., & Davies, M. (2006a). Efficacy of inter-

personal psychotherapy-adolescent skills training: An indicated

prevention intervention for depression. Journal of Child Psychology

and Psychiatry, 47, 1254–1262.

Young, J. F., Mufson, L., & Davies, M. (2006b). Impact of comorbid

anxiety in an effectiveness study of interpersonal psychotherapy for

depressed adolescents. Journal of the American Academy of Child

and Adolescent Psychiatry, 45, 904–912.

Yu, D. L., & Seligman, M. E. P. (2002). Preventing depressive symp-

toms in Chinese children. Prevention and Treatment, 5.

Zeiss, A. M., Lewinsohn, P. M., & Munoz, R. F. (1979). Nonspecific

improvement effects in depression using interpersonal skills training,

pleasant activity schedules, or cognitive training. Journal of Consult-

ing and Clinical Psychology, 47, 427–439.

104 DAVID-FERDON AND KASLOW