Journal of Marital and Family Therapy 43(2): 245–263

doi: 10.1111/jmft.12204
© 2016 American Association for Marriage and Family Therapy

“FAMILY MATTERS”: A SYSTEMATIC REVIEW OF THE

EVIDENCE FOR FAMILY PSYCHOEDUCATION FOR
MAJOR DEPRESSIVE DISORDER
Pamela Brady and Maria Kangas
Macquarie University

Katherine McGill
Hunter New England Mental Health Service

The first aim of this systematic review was to evaluate the evidence for family psychoeduca-
tion (FPE) interventions for major depressive disorder (MDD). A second aim was to com-
pare the efficacy of different modes of delivering face-to-face FPE interventions. Ten studies
(based on nine distinct samples) were identified comprising four single-family studies, four
multifamily studies, one single versus multifamily comparative study, and one peer-led,
mixed-diagnosis study. Seven studies measured patient functioning and six reported positive
outcomes. Six studies measured carer’s well-being and four reported positive outcomes.
Results provide preliminary evidence that FPE leads to improved outcomes for patient func-
tioning and family–carer’s well-being for persons with depression. The implications for
future development and delivery of FPE interventions for MDD are discussed.

There is considerable evidence demonstrating that family interventions for mental disorders
lead to improved outcomes for both patient and carer (Henken, Huibers, Churchill, Restifo, &
Roelefs, 2007; Jewell, Downing, & McFarlane, 2009; Lucksted, McFarlane, Downing, & Dixon,
2012; McFarlane, Dixon, Lukens, & Lucksted, 2003). In particular, Family Psychoeducation
(FPE) is an evidence-based practice that is recognized as part of the optimal treatment for psy-
chotic disorders by the U.S.’s Department of Health and Human Services Substance Abuse and
Mental Health Services Administration (Substance Abuse and Mental Health Services Adminis-
tration (SAMHSA), 2009) and by the U.K.’s National Institute for Health and Care Excellence
(NICE, 2014). Based on models originally developed by Andersen, McFarlane, and Falloon
(Anderson, Hogarty, & Reiss, 1980; Falloon, Boyd, & McGill, 1984; McFarlane, Lukens, &
Link, 1995), FPE is a method of working with families and carers who are supporting a person
with mental illness with the aim of enhancing treatment outcomes by enabling those who are
closest to the person to manage the issues that can arise or which may exacerbate the illness.
The majority of care for people with depression is provided at home and in most cases the pri-
mary caregiver is a relative. The terms carer, relative, and family are therefore used interchange-
able throughout this article.
Within the mental health service context, FPE refers to interventions that are educationally
oriented. However, FPE is not simply family education per se; that is, FPE is more than just infor-
mation provision and ensuring that people have an understanding of the illness. It also importantly
focuses on the development of problem solving, communication and coping skills, and enhance-
ment of social supports to manage the depression. These interventions differ from traditional fam-
ily therapy in that they do not assume dysfunction in the family (Dixon et al., 2011). For example,
systemic family therapy views dysfunctional family relationships as causing or reinforcing

Pamela Brady, MClinPsych, Centre for Emotional Health, Department of Psychology, Macquarie University,
Sydney, Australia. Maria Kangas, MClinPsych, PhD, Centre for Emotional Health, Department of Psychology,
Macquarie University, Sydney, Australia. Katherine McGill, DPsy(Clinical), Mental Health Research, Evaluation,
Analysis and Dissemination (MH-READ) Unit, Hunter New England Mental Health Service.
Address correspondence to Associate Professor Maria Kangas, Macquarie University, Centre for Emotional
Health, Department of Psychology, Sydney, NSW 2109, Australia; E-mail: [email protected]

April 2017 JOURNAL OF MARITAL AND FAMILY THERAPY 245

symptoms, and interventions are aimed at restructuring these maladaptive patterns of family inter-
actions. In FPE, rather than being part of the problem and requiring psychotherapy, families are
considered crucial to the patient’s recovery and they are provided with knowledge and skills to sup-
port this (Jewell et al., 2009). FPE interventions are based on the premise that if you increase fam-
ily members’ understanding of the illness, this will reduce distress and improve their ability to
cope, which will result in improved outcomes for the patient (Anderson et al., 1986). FPE has been
shown to predict patient recovery and functioning, relapse and rehospitalization rates, and adher-
ence to medication (McFarlane et al., 1995; Pitschel-Walz, Leucht, B€ auml, Kissling, & Engel,
2001). Although the primary goals of FPE programs have traditionally been improved outcomes
for the patient, there is now an increasing focus on also improving carer well-being, recognizing
that the two are often interdependent (Lucksted et al., 2012).
Although the majority of research has been conducted with families of patients with psy-
chotic disorders, a number of studies have shown that family functioning is also important in
determining the course of major depressive disorder (MDD) (Bolkan et al., 2013; Keitner et al.,
1995). While involving the family in treatment for depression is important to a patient’s recovery
and ongoing functioning, the impact of the care-giving role on the well-being of the person pro-
viding the care is substantial. Studies show that while there are both similar and different chal-
lenges in supporting a relative with depression compared to psychotic disorders, the overall
carer burden is just as significant (Anderson et al., 1986; Angermeyer, Kilian, Wilms, & Witt-
mund, 2006; van Wijngaarden et al., 2009). Often referred to as “family burden” or “carer bur-
den”, the problems most frequently reported by people supporting a person with depression are
financial difficulties, feelings of isolation, reduction in social activities, relationship distress, feel-
ing confused, and overwhelmed by depressive symptoms, worry about stigma, worry about the
future, and being able to access treatment (Ahlstr€ om, Sk€ ars€
ater, & Danielson, 2009; Highet,
McNair, Davenport, & Hickie, 2004; Jeglic et al., 2005; Lemmens, Eisler, Buysse, Heene, &
Demyttenaere, 2009). Between 40% and 72% of carers experience clinical levels of depression
and anxiety (Coyne et al., 1987; Heru, Ryan, & Madrid, 2005; Jeglic et al., 2005), although for
the majority of carers, clinical symptoms tend to alleviate when the depressed relative recovers
(Coyne et al., 1987; Jeglic et al., 2005).
In 2012, Luciano and his colleagues published a nonsystematic review evaluating FPE inter-
ventions for persons with depression and their carers. On the basis of only four identified studies,
they concluded that although more evidence was needed, the emerging research indicated that
FPE leads to improved outcomes for patients and family members of people with MDD. A recent
study also demonstrated the cost effectiveness of FPE for MDD (Shimodera et al., 2012), and
other studies have shown that when FPE is offered to relatives, patients report higher satisfaction
with the care they receive (Bolkan et al., 2013).
Despite the apparent efficacy, interestingly, FPE is still not widely available for people with
depressive disorders and their families (Frank, Rummel-Kluge, Berger, Bitzer, & H€ olzel, 2014;
Lucksted et al., 2012). Moreover, some studies have shown that even when offered, FPE pro-
grams for depression are underutilized (Kronm€ uller et al., 2006; Lucksted et al., 2012; Shimazu
et al., 2011). Others have suggested that it is organizational factors such as lack of funding and
resources that are the major barriers to including families in treatment for MDD (Wirrell,
McGill, Kelly, & Bowman, 2014). Also, when carers do not attend programs, this is usually due
to practical difficulties of carers in being able to commit to attending programs (Kronm€ uller
et al., 2006; Shimazu et al., 2011). Timing of FPE programs has also been cited as a barrier to
effective implementation (e.g., Mental Health Council of Australia, 2012). Some thirty years
ago, Anderson et al. (1986) drew attention to the fact that families have different needs at differ-
ent stages of a depressive illness and that psychoeducation was most critical at the acute stage.
In short, to have an impact on mental health outcomes for either patient or carer, FPE pro-
grams need to be offered in a cost-effective, accessible, and timely manner. There are a number
of different modes of delivering FPE, and although the relative efficacy of these modes has been
extensively explored with schizophrenia, this is not the case for MDD (Luciano et al., 2012;
Lucksted et al., 2012).
One of the key differences in the way FPE is implemented is whether programs are delivered
to single-family or multifamily groups. In a climate of limited resources, multifamily programs

246 JOURNAL OF MARITAL AND FAMILY THERAPY April 2017

would appear to be more cost efficient if found to be as effective. Evolving from its family ther-
apy routes, FPE programs have traditionally been single-family based and often conducted in
the family home when the patient was discharged from hospital (Anderson et al., 1980; Falloon
et al., 1984). The primary goal of these original programs was to promote patient recovery and
prevent relapse by reducing family “Expressed Emotion” (or unsupportive, critical interactions).
Starting with an intensive engagement period with the family, the programs emphasized educa-
tion and problem solving based on the specific needs and goals of the family and patient (e.g.,
Anderson et al., 1986; Falloon & Shanahan, 1992). In contrast, the multifamily format pio-
neered by McFarlane, Dixon, and others emphasizes the importance of the mutual support fami-
lies can derive from joining with others in similar situations (Jewell et al., 2009; McFarlane
et al., 2003). Multi-Family Psychoeducation (MFPE) offers opportunities for sharing experiences
and joint problem solving, but also encourages families or partners to form a social support net-
work that extends beyond the formal stages of the group. While multifamily groups might seem
an obvious benefit in terms of both social networking and cost, there can be a lower treatment
compliance rate in multifamily therapy where family members are required to attend a group
session, compared to a therapist home visit (Leff et al., 1990). On balance, however, evidence
suggests that MFPE can be at least as effective as single-family psychoeducation (SFPE) for psy-
chotic disorders, unless the family are unusually emotionally resilient, have already developed
good coping skills, or are from minority groups (Jewell et al., 2009; Lucksted et al., 2012;
McFarlane et al., 2003). To date there has been no review of multi- versus single-family inter-
ventions for MDD.
A second difference in the implementation of FPE programs is whether FPE programs are
peer led or clinician led. In clinician-led programs, patient outcomes such as relapse prevention are
often the primary goal, particularly if it is conducted in an in-patient setting and the participants
are both patient and relatives (Luciano et al., 2012; Lucksted et al., 2012). Peer-led FPE programs
are generally conducted in a community setting and the primary focus is the carer’s well-being
(Dixon et al., 2011; Stephens, Farhall, Farnan, & Ratcliff, 2011). The patient is not required to
attend and does not have to be receiving treatment; hence, the attitude of the relative with the ill-
ness is not a barrier. Peer-led FPE has been shown to be effective in schizophrenia with proponents
of this model highlighting both its cost effectiveness as well as the benefits of a group leader who
can personally relate to the participants experience (Chien & Chan, 2004; Dixon et al., 2011).
However, no study has reviewed the evidence for peer-led FPE for MDD.
A third factor for consideration concerns the homogeneity of multifamily groups. In par-
ticular, whether FPE needs to be designed and delivered specifically for people supporting a
person with MDD or whether FPE for families caring for a person with any type of mental
illness can be equally effective. If heterogeneous groups can be designed to be equally effective,
this has significant implications for the provision of FPE. The requirement to recruit a mini-
mum number of participants to conduct a community-based group means programs cannot
always be provided at the time which would be of most benefit to the family member, gener-
ally at the time the illness is first diagnosed. The more inclusive the participation criteria, the
more likely a group will be scheduled in timely manner. To date, no study has evaluated the
evidence regarding the effectiveness of mixed-diagnosis groups for people specifically caring for
a person with MDD.
As noted, although Luciano et al. (2012) published a recent review of studies evaluating FPE,
they only included four FPE intervention studies. A notable limitation was that two of the four
identified treatment studies comprised mixed groups of samples diagnosed with various clinical
disorders, not limited to depression per se. Hence, the direct effects of FPE for persons with
depression and their carers could not be differentiated. On the basis of increasing studies testing
the efficacy of FPEs for MDD, it is timely to re-examine this body of literature. Accordingly, the
objective of this study was to conduct a systematic review of FPE interventions for MDD and
extend it by specifically comparing the efficacy of different modes of FPE interventions. The speci-
fic aim was to evaluate the efficacy of face-to-face FPE for persons with depression and their car-
ers; and to evaluate whether the effectiveness was dependent upon delivery mode, including
comparison of outcomes for (a) multi-family and single-family psychoeducation, (b) peer-led and

April 2017 JOURNAL OF MARITAL AND FAMILY THERAPY 247

clinician-led groups, and (c) FPE groups for MDD only compared to FPE for any mental illness
(mixed-diagnosis groups).

METHOD

Search Methods
The literature search was conducted using the PRISMA framework (Moher et al., 2015).
Search terms were selected based on past reviews. Cochrane, Embase, Medline, PubMed, and Psy-
cInfo databases were searched using combinations of the following terms as key words, titles, and
subject headings: “family”, “carer”, “caregiver”, “relative”, “spouse”, “partner”, “psychoeduca-
tion”, “intervention”, “therapy, “treatment”, “program”, “depression”, “major depression”,
“unipolar depression”, “depressive disorder”, “affective disorder”, “mood disorder”. Bibliogra-
phies of identified articles were scanned for further relevant studies and “find similar” or “find cit-
ing” functions were used for seminal articles within the PubMed and PsycInfo databases. The
search period covered studies published from 1985 to January 2016.

Eligibility Criteria
Studies were eligible for inclusion if they met the following conditions:
Study design. Randomized controlled trials (RCTs) and within-subject studies without a con-
trol or comparative group were included if they also met the other inclusion criteria (outlined
below).
Participants. Family psychoeducation intervention studies for family/relatives with MDD
aged 14–85 years were included. Studies of interventions restricted to families of people with a
specific type of depression (e.g., bipolar disorder) and FPE programs for younger children, or for
the dependent children of depressed parents were excluded because the psychoeducational inter-
ventions for these groups are substantially different and the nature of the presenting issues are also
disparate. Moreover, in the case of interventions involving children, these issues have also been
explored in other studies including a recent review (Institute of Medicine National Research Coun-
cil, 2009). For similar reasons, studies of FPE interventions where depression was secondary to a
physical or neurological condition (e.g., dementia) were also excluded.
Intervention. The intervention had to be presented in-person format, and be primarily psy-
choeducational in nature, comprising both education/information components and skills building.
Studies that investigated the efficacy of specific therapies (e.g., cognitive-behavior therapy, marital
therapy) rather than psychoeducation were excluded. Self-help interventions (psychoeducation
relying on written information) were not included. Interventions had to be delivered weekly or
fortnightly over a period of at least 6 weeks, with or without the patient being present; this was
based on the recommended minimum number of sessions required for efficacy for FPE for psy-
chotic disorders (Substance Abuse and Mental Health Services Administration (SAMHSA),
2009). Interventions could be clinician or peer led and use open or closed group formats.
Outcome measures. Studies that assessed treatment outcomes with standardized or validated
measures for capturing changes in depressive symptoms or other psychological well-being symp-
toms for either the person with depression and/or their family–carers as a result of an FPE inter-
vention were included. Studies were excluded if participant satisfaction and/or engagement in the
FPE program were the only variables assessed, as these variables were not considered to be repre-
sentative of a well-being outcome. Studies of FPE interventions for carers of patients with a mix-
ture of mental health disorders, but which did not report separate outcomes for the MDD
subgroup were also excluded.
Other. Non-English language studies were included where possible and translated using on-
line software with further translation by a native speaker if required.

Data Collection and Analysis

Two of the authors (deidenitified) conducted the study selection review and coding of articles.
Abstracts of all identified studies were initially screened with regard to participant group (specific
or possible reference to MDD), intervention type, and study design (case studies excluded). The
full texts of remaining articles were reviewed against the eligibility criteria by both authors. For

248 JOURNAL OF MARITAL AND FAMILY THERAPY April 2017

each study, information was recorded on the study population, design, intervention, outcome mea-
sures, and results. Studies were categorized along the following dimensions: (a) single-family or
multifamily group intervention, (b) studies where FPE was aimed at families of people with MDD
only or studies which were aimed at families of people with MDD and other mental disorders, and
(c) studies where FPE was led by clinicians and studies where FPE was led by peers. The selected
studies were heterogeneous in quality, design, and in outcomes measured, which did not enable a
meta-analysis to be conducted.

Study Quality and Validity Assessment

The methodological quality of the studies was evaluated according to the following eight
validity criteria that were adapted from the Consolidated Standards of Reporting Trials (CON-
SORT) guidelines (Altman, Schulz, & Moher, 2001) and Delphi criteria list (Verhagen et al.,
1998): randomization; allocation concealment; blinding of outcome assessments; comparability of
groups at baseline; withdrawals and handling of dropouts in analyses; and use of intention-to-treat
analysis (ITT); and multiple follow-up assessments. These criteria have been used in previous sys-
tematic and meta-analytic reviews to evaluate study quality (e.g., Kangas, Bovbjerg, & Mont-
gomery, 2008). The methodological quality for each of the studies was evaluated according to the
eight criterion summarized above, and a point was allocated for each criterion satisfied (range: 0–8
points; See Table S1). A study was deemed as having good internal validity/quality if they scored
>4 points. Understandably, as two of the eight criterion were based strictly on RCT methodology
(i.e., RCT method and concealment of allocation), the maximum points possible for non-RCTs,
pre–post design studies was 6.

RESULTS

Figure 1 presents a summary of the literature search in accord with PRISMA guidelines.
From the 5262 studies identified, the full texts of 102 studies were reviewed and a total of 10 studies
met inclusion criteria including 2 studies (Bernert et al., 2006; Schomerus et al., 2007) that
reported on different outcomes for the same sample. Hence, 10 studies based on nine distinct data-
sets/samples were included in this review. The most common reason for exclusion was participant
factors such as child focus (i.e., less than 14 years), depression was secondary to other issues, or
outcomes for depressed participants were not specifically reported.
The characteristics and results of the included studies are summarized in Table 1 (inclusive of
the method quality total score). The methodological quality details of these studies are presented
in the Table S1. Nine of the ten studies described FPE interventions for patients with MDD or
Mood Disorders only. Of these studies, four were single-family psychoeducational (SFPE) studies
(Clarkin et al., 1990; Fiorillo, Malangone, & Del Vecchio, 2011; Kumar & Gupta, 2015; Sanford
et al., 2006), four were multifamily psychoeducational (MFPE) studies (Bernert et al., 2006; Kat-
suki et al., 2011; Schomerus et al., 2007; Shimazu et al., 2011), (although two of these MFPE stud-
ies were based on the same sample; [Bernert et al., 2006; Schomerus et al., 2007]), and one was a
SFPE and MFPE comparative study (Lemmens et al., 2009). Only one MFPE study for people
with any mental disorder was identified, and which reported outcomes for nonpsychotic disorders
including depression separately (Stephens et al., 2011).

SFPE Studies
Consistent with the traditional model of SFPE (Falloon, 2003), the four identified SFPE inter-
ventions included the patient, of which three were based on an RCT design (Clarkin et al., 1990;
Fiorillo et al., 2011; Sanford et al., 2006); although two of the three RCTs did not specify the ran-
domization method used inclusive of concealment of allocation used. For the fourth study by
Kumar and Gupta (2015), although the design was referred to as an RCT, the description did not
strictly meet randomization criterion as participants were allocated to each condition via alternat-
ing sequence. For this study, patient outcome measures were only reported. The other three studies
reported measures of both patient and family functioning.
Three of the four studies reported positive outcomes for patient measures; two of these studies
were based in a community setting and adapted Falloon’s SFPE model of 12 sessions over

April 2017 JOURNAL OF MARITAL AND FAMILY THERAPY 249

 Identification
Records identified through Additional records identified
database searching through other sources
(n = 5262) (n = 5)

Duplicates removed
(n = 480)
Screening

Records screened Records excluded

(n = 4787) (n = 4683)

Full-text articles assessed Full-text articles excluded, with

for eligibility reasons
Eligibility

(n = 102) (n = 91)

Sampling issues= 35
(No MDD-caregiver specific
outcomes= 9; Child focus= 14; MDD
secondary to other illness/ issue= 12)

Intervention issues= 21
(Not psychoeducation= 15
Studies included in Intervention too short= 7)
Included

synthesis
Methodological issues= 13
(n = 10
(Poor design= 1; No standardised
outcomes= 12)

Other= 22
(Review=14; Repeated sample= 5;
Other= 3)

Figure 1. Search and review outcomes.

6 months in the patients home, involving as many family members who opted to be involved (Fior-
illo et al., 2011; Sanford et al., 2006). Specifically, the Fiorillo et al. (2011) study comprised adult
patients and their families. In contrast, the Sanford et al. (2006) study was based on adolescent
patients and their families, and positive effects for family social functioning were reported. More-
over, patient outcomes were found to be mediated by the quality of the relationship between
adolescent patient and parents; however, family burden was not assessed. The Sanford et al. study
was also found to have medium effect size for patient outcomes after both 3 and 6 months of
treatment.
The third study which found positive effects for patient outcomes was conducted in an inpa-
tient setting (Kumar & Gupta, 2015). The researchers utilized a short intervention consisting of
four sessions of SFPE conducted over 8 weeks. Kumar and Gupta (2015) reported significant dif-
ferences in depression symptoms in the intervention group after just one session, with differences
in significance peaking 4 weeks after the intervention concluded.
The only SFPE study which reported negative outcomes for patients with MDD (and also
included patients with bipolar) was also conducted in an inpatient setting using an RCT design. In
particular, Clarkin et al. (1990) program consisted of six sessions with no follow-up sessions after
discharge. Patients with MDD had improved at the time of discharge from hospital but by

250 JOURNAL OF MARITAL AND FAMILY THERAPY April 2017

 Table 1

April 2017
Studies Included in the Review

Type of study & Method

Author quality scorea Sample Intervention Measures and timeframe Results

Single-family interventions, all clinician led, and patient included

Clarkin RCT 50 hospitalized adult 6 9 1 hr sessions Patient Functioning: All MDD patients
et al., SFPE + TAU versus TAU patients with affective conducted in hospital Global Assessment improved symptoms
1990; only (medication plus disorders + relatives focusing on education, Scale, Psychiatric and functioning from
USA group therapy) MDD (n = 29), 17 family attitude, and Evaluation Form. admission to discharge.
*Quality Score = 5.5 rec’d intervention. coping strategies. Family Attitude: Family Male MDD patients
Bipol (n = 21), 12 Timing of sessions not Attitude Inventory. worse after 6 months,
rec’d intervention. specified. Admission, Discharge, 6 Male and female worse
*Female 60%. mean Part of larger study of and 18 months after 18 months
age 35.9 (14.5) FPE intervention for postdischarge. compared to TAU.
schizophrenia. No differences in family
measures for MDD
patients compared to
TAU.
No effect sizes
reported.
Fiorillo RCT 44 patients in 12 9 90 min home-based Patient Symptoms and Significant improvements

JOURNAL OF MARITAL AND FAMILY THERAPY

et al., SFPE + TAU versus community mental sessions over 6 months. Functioning: Brief on all measures for
2011; Control TAU + info health setting with Falloon model of FPE for Psychiatric Rating Scale, intervention group
Italy package MDD + families, 22 Schizophrenia with Disability Assess Scale. compared to TAU for
*Quality Score = 1.5 received intervention. content modified for Family Burden and both patients and
*Female 59%, mean MDD. Social Functioning: families.
age 48.6 (10.8) Family Problems Qu, No effect sizes
Social Network Qu. reported.
Pre- and posttreatment

251
 252
Table 1
Continued

Type of study & Method

Author quality scorea Sample Intervention Measures and timeframe Results

Sanford RCT 31 Adolescents 12 9 90 min SFPE home- Patient Symptoms: Improvements in patient
et al. SFPE + TAU versus TAU diagnosed with MDD based sessions over Reynolds Adolescent symptoms and social
(2006) (medication + supportive using K-SADS-P. 16 6 months with one Depression Scale. functioning at both time
Canada counseling). received intervention. booster session at Patient Social points compared to
*Quality Score = 6 * Female: 64%; mean 9 months. Functioning: Struct. TAU mediated by
age 15.6 (1.1) Based on Falloon model Social Adj. Interview parent/adolescent
of FPE. Family functioning, relationship. Medium
Family Assessment effect size on both
Device. measures
Baseline, 2 weeks, 3, 6 & No significant difference
9 months in family functioning
compared to TAU.
Kumar RCT 80 adults diagnosed 4 sessions delivered Patient Symptoms: Sig diff in mean HDRS
and SFPE + TAU versus TAU with MDD + family– fortnightly. Hamilton Dep Rating for intervention group
Gupta (medication + supportive carer. 38 received Scale (HDRS), GAF, compared to TAU from

JOURNAL OF MARITAL AND FAMILY THERAPY

(2015) counseling). intervention Psychological General 4 weeks with difference
India *Quality Score = 4.5 *Female 40%; mean Well-being Index peaking at 12 weeks. Sig
age 33 (11.63) (PGWBI) diff in mean GAF and
Baseline, 2, 4, 8, 12 weeks PGWB in intervention
group at 12 weeks. No
effect sizes reported

April 2017
 Table 1

April 2017
Continued

Type of study & Method

Author quality scorea Sample Intervention Measures and timeframe Results

Multifamily interventions, all clinician led, and patient excluded

Bernert Within-subjects MFPE, MFPE for spouses of 12 sessions over 6 months Carer Burnout: Maslach No sig. effect on burnout
et al. with control group MDD outpat. covering information, Burnout Inventory; pre–postprogram or
(2006) *Quality Score = 2 (n = 66). Interv. plus coping strategies, and Spouse psychological compared to control.
Schomerus control group of comm. skills. One distress: BAI, BDI. No significant effect on
et al. nonattendees booster session Baseline, postprog., distress at any time
(2007) (n = 50). Participant 3 months postprogram. 3 months, 6 month f/u. point compared to
Germany gender and age not control. 96% positive
provided. program evaluation
rating.
Katsuki Within-subjects MFPE, no 32 relatives of adult 4 9 2 hr sessions over Carer psychological Carer distress, care
et al. control group inpatients and 6 weeks. Based on distress: K6. Caregiver burden, and families
(2011), *Quality Score = 2 outpatients diagnosed McFarlane model but burden: Zarit Burden expressed emotion all
Japan with MDD without patient. 30 mins Interview, Dysfunction showed statistically
**Female 72%, male psychoeducation and in Ordinary Family Life significant
28%. Mean age 61.7 90 min group sharing Scale; Family Expressed improvements.
(12.5) and problem solving. Emotion: Family Number of relatives

JOURNAL OF MARITAL AND FAMILY THERAPY

Attitude Scale. indicating clinical levels
Pre- and postintervention of depression and
measures. anxiety reduced from 16
(50%) to 3 (9.3%) after
intervention.
Medium effect size on all
measures.

253
 254
Table 1
Continued

Type of study & Method

Author quality scorea Sample Intervention Measures and timeframe Results

Shimazu RCT Adult relatives of 4 bi-weekly sessions over Patient symptoms/ Patient Relapse over
et al. MFPE + TAU versus outpatients with 6 weeks, 30 mins remission: Hamilton 9 months: significant
(2011) TAU MDD living with psychoed and 90 min Rating Scale, BDI; improvement v TAU.
Japan (medication + supportive patient. MFPE for group sharing and Family Expressed No signif. change in
psycho-therapy carers, n = 54. 24 problem solving. Max 5 Emotion: Family family expressed
fortnightly) received intervention. participants per group Attitude Scale, Five- emotion and no
*Quality Score = 6.5 *44% female, mean age Minute Speech Samples, interaction affect with
59.2 (14.6). **56% Baseline and 9 months patient symptoms.
female, mean age 59 posttreatment No effect sizes
(11.4) reported.
Single-family versus multifamily interventions, clinician led, and patient included
Lemmens RCT. 83 hospitalized patients SFPE: 6 bi-weekly 90 min Patient depression: BDI, Both intervention groups
et al., MFPE + TAU versus diagnosed with MDD sessions: education and Hamilton Rating Scale; showed signif better
2009; SFPE + TAU versus and partners. skills within systemic Patient medication use; outcomes versus TAU
Belgium TAU (2–3 months in- *MFPE (n = 35), framework. Children Rehospitalization rates; on all measures after
patient treatment female 80%, mean attend two sessions. Subjective emotional 15 months. MFPE

JOURNAL OF MARITAL AND FAMILY THERAPY

including indiv/group age 43.9 (8.3). 3 months follow up health of patient/ group slightly better
therapy and medication. *SFPE (n = 25), female session. partner, rating by each than SFPE but
*Quality Score = 6 64%, mean age 40.2 MFPE: Groups of 4–7 other: Dyadic differences not
(9.1) patients and partner. Adjustment Scale significant (small
*TAU (n = 23), female More emphasis on Measures after 3 months sample). No effect sizes
69.9%, mean age 43.2 group problem solving. (discharge) and reported.
(8.4) 15 months

April 2017
 April 2017
Table 1
Continued

Type of study & Method

Author quality scorea Sample Intervention Measures and timeframe Results

Mixed-diagnosis multifamily groups, peer led, and patient excluded

Stephens Within subjects study 459 Relatives of “Well Ways” peer-led Negative consequences of Positive outcomes on all
et al., without control group patients with any Family to Family Care giving: (worrying, measures, small effect
2011; *Quality Score = 1 mental illness. MFPE, 8 weekly 3 hr tension, supervision, size. Mostly continued
Australia Patient diagnosis sessions. urging); Involvement to improve over
subgroup (n = 159), Participants encouraged Evaluation 6 months.
psychotic to develop support Questionnaire – 27 core Psychotic benefited more
disorder = 66%, groups postprogram. items than nonpsychotic on
mood disorder 39%, Caregiver psychological worrying. Females
anxiety disorder 21%. distress: GHQ-12 benefited more than
**Female: 77.9%. Preprogram, male on tension.
mean age 55.19 postprogram, 3/
(10.14) 6 months follow-up

JOURNAL OF MARITAL AND FAMILY THERAPY

Notes. NB, *patient demographics, **carer demographics;
a
Methodological Quality Score – for full details of criterion scores – see also Table S1.

255
6 months, in the case of men, and by 18 months for women, patient functioning was actually
worse in the intervention group than the nonintervention (treatment as usual: TAU) group. Mea-
sures of family burden and attitudes showed similar patterns. This was not the case with the bipo-
lar patients; these patients and their families who received the intervention continued to improve in
their functioning over time compared to the control group.
For these four SFPE studies, three scored at least 4 on the quality ratings (Clarkin et al., 1990;
Kumar & Gupta, 2015; Sanford et al., 2006) indicating sound methodology, with the strongest
quality demonstrated by the Sanford et al. (2006) adolescent trial. The fourth study by Fiorillo
et al. (2011) was evaluated to have very low methodological quality. This was the only study iden-
tified which was not published in manuscript format. Rather, only a conference abstract was avail-
able in the referenced journal. Hence, only scant methodological information in accord with
CONSORT guidelines was available to evaluate the methodological quality.

MFPE Studies
From the four identified MFPE studies, as aforementioned, two were based on the same data-
set/sample (Bernert et al., 2006; Schomerus et al., 2007), and the researchers utilized a within-sub-
ject designs with a nonrandom control group of nonattendees. These two studies scored low on
methodological quality as minimal information was reported on design and analytical issues. The
other two studies comprised an RCT design (Shimazu et al., 2011) which was assessed to have
strong methodological quality, and a pre–post study design (Katsuki et al., 2011), which was eval-
uated to have low quality due to minimal information being reported pertaining to CONSORT
criteria.
None of the MFPE interventions included the patient and only the RCT by Shimazu et al.
(2011) measured patient outcomes in addition to family functioning. These researchers found a sig-
nificant improvement in patient time to relapse over a 9-month follow-up period compared to the
control condition. All four MFPE studies measured family functioning; one reported positive
effects (Katsuki et al., 2011) and three (including the two related studies) did not (Bernert et al.,
2006; Schomerus et al., 2007; Shimazu et al., 2011). In particular, Katsuki et al. (2011) utilized the
same MFPE program as Shimazu et al. (2011) with a different sample and found significant
improvements in carer psychological distress and carer burden, using a pre–post design, with med-
ium effect size; although no longer-term follow-up measures were reported. However, while Shi-
mazu et al. found positive outcomes for patients, family expressed emotion did not improve. The
two related studies using an MFPE intervention designed for spouses of people with MDD
reported no significant difference in carer outcomes (including carer burnout, depression, and anxi-
ety symptoms) compared to a nonattendee control group postprogram and up to 6-month follow-
up (Bernert et al., 2006; Schomerus et al., 2007). Interestingly, the intervention was longer than
the one reported by Katsuki et al. and Shimazu et al. (see Table 1).

MFPE versus SFPE Program

Only one published study was identified that directly compared outcomes between a MFPE
and SFPE intervention utilizing an RCT design (Lemmens et al., 2009) with strong methodologi-
cal quality. The intervention consisted of six sessions in an in-patient setting with a 3-month fol-
low-up session postdischarge. Given all groups included the patient, the authors primarily focused
on measures of patient functioning rather than family functioning. By 15 months postdischarge,
patient depressive symptoms, medication use, and rehospitalization rates were significantly better
for both the intervention groups compared to a TAU/control group. Patients and their partners
also rated each other’s subjective emotional health after 15 months, and both intervention groups
reported better outcomes compared to the TAU control; although no significant difference
emerged between the two FPE formats.

FPE Program
Only one non-RCT FPE study based on a mixed-diagnosis group was identified (Stephens
et al., 2011), and was evaluated to have low methodological quality as minimal information was
reported pertaining to key design factors. The FPE intervention consisted of eight 3-h weekly ses-
sions over 2 months and was also the only peer-led intervention identified. The researchers focused

256 JOURNAL OF MARITAL AND FAMILY THERAPY April 2017

on carer outcomes and found significant improvement in carer psychological distress, worrying,
and tension with small effect sizes. Participants with psychotic relatives improved significantly
more on “worrying” than the nonpsychotic group, but no other significant interactions with
patient diagnosis were found.

DISCUSSION

The overarching objective of this review was to systematically evaluate the published literature
regarding the efficacy of FPE interventions for persons caring for individuals with MDD. The find-
ings indicate tentative support for FPE for MDD based on a small number of international stud-
ies. Although the majority of trials (k = 8) were published in English, two were based on Japanese
samples, one was Italian, one was Belgium, and one was based on an Indian sample (see Table 1).
The two related studies were derived from a German sample and the work was translated from
German. Overall, the findings are consistent with the preliminary findings from Luciano et al.
(2012) nonsystematic review, and also with the evidence of FPE for other disorders, particularly
schizophrenia and bipolar disorder (Lucksted et al., 2012). The results of this systematic review
extend the existing evidence base and build on Luciano et al.’s narrative review, by identifying
eight additional studies (based on seven distinct samples) and separately exploring the outcomes of
FPE interventions for patients and family–carers, which has not been previously evaluated. Of the
seven studies measuring patient variables, only the SFPE study by Clarkin et al. (1990) failed to
find a positive effect for patient outcomes, and interestingly, found the intervention made MDD
patient functioning worse. The authors suggested this may be because the intervention was
designed primarily for those with psychotic disorders and it is possible that the three subsequent
SFPE studies found positive results because the intervention was tailored more specifically for
MDD (Fiorillo et al., 2011; Sanford et al., 2006). Taken together, these outcomes suggest that
FPE needs to be specifically tailored to the needs of MDD patients and their families/carers.
When reviewing the evidence on family–carer outcomes from FPE interventions, it is impor-
tant to distinguish between the types of carer outcomes that were reported. These comprised two
main categories; measures of family attitude/expressed emotion and measures of carer well-being/
perceived burden. The former tended to be used in interventions where the focus of the interven-
tion was on improving patient functioning and where family attitude was hypothesized to be a
mediator for this change. The latter tended to be used for interventions aimed at improving well-
being of the carer. Of the four studies which measured family attitudes or expressed emotion, only
one of these studies which was a clinician-led MFPE was found to have a positive effect for this
measure based on a non-RCT design (Katsuki et al., 2011).
Six studies measured carer well-being or carer burden and four of these studies documented
positive effects which were based on variable designs (Fiorillo et al., 2011; Katsuki et al., 2011;
Lemmens et al., 2009; Stephens et al., 2011). Specifically, three of the four studies with positive
carer outcomes were evaluated to have low methodological quality (Fiorillo et al., 2011; Katsuki
et al., 2011; Stephens et al., 2011), whereas the fourth study by Lemmens et al. (2009) was evalu-
ated to have strong methodological quality. The remaining two studies which assessed carer burn-
out were based on the same dataset (Bernert et al., 2006; Schomerus et al., 2007). In this sample,
baseline levels of carer distress were not significantly different from the general population and the
authors of these studies acknowledged that the timing of the program could explain outcomes as
the intervention was not offered at the acute stage of the illness when distress is likely to be highest
(Anderson et al., 1980). These studies did not include any longer-term follow-up measures and, as
Bernert et al. (2006) propose, the benefits families derive from improved coping skills and building
greater support networks may be more apparent in the longer term rather than immediately
postintervention.
In summary, this pattern of results suggests that FPE interventions are beneficial to patient
functioning and also for family–carer well-being. Notably, the findings suggest that the FPE inter-
ventions for depression were not consistently associated with improvements in family attitudes/ex-
pressed emotion, which have been argued to be important variables that mediate patient recovery.
This further supports the view that family functioning is a solution rather than a cause of depres-
sion (Jewell et al., 2009), and that outcomes such as carer well-being and reduced carer burden

April 2017 JOURNAL OF MARITAL AND FAMILY THERAPY 257

may themselves be the most significant variables underpinning improved patient outcomes rather
than via changes in expressed emotion or family attitudes (Shimazu et al., 2011). It is noteworthy
that even the studies that failed to show a positive effect for carer outcomes mostly reported high
family–carer satisfaction with the FPE program (Bernert et al., 2006; Sanford et al., 2006;
Schomerus et al., 2007; Shimazu et al., 2011).
The current findings also revealed that the length of the intervention did not appear to deter-
mine whether participants reported positive effects. Participants who attended the shortest inter-
ventions (four sessions over 6 weeks) reported positive longer-term effects (e.g., Shimazu et al.,
2011), which was similar to participants who had been offered longer intervention programs (e.g.,
Fiorillo et al., 2011). However, the successful shorter FPE interventions mostly have a common
focus on joint problem solving, peer support and networking, and previous authors have suggested
that if these elements are included, an extended intervention length is not critical (Katsuki et al.,
2011; Shimazu et al., 2011; Stephens et al., 2011). However, while this peer support element was
missing from the recent four session SFPE study by Kumar and Gupta (2015), which also reported
positive outcomes, this study did not report longer-term outcomes for either patients or their
carers.
In terms of evaluating the evidence for different modes of FPE interventions, the current find-
ings provide preliminary evidence that MFPE is at least as effective as SFPE in improving out-
comes for patients with MDD, and that MFPE may be more effective than SFPE in improving
outcomes for family–carers. This is consistent with studies that compare SFPE and MFPE groups
for schizophrenia, with these studies suggesting that the mutual support problem solving and the
extended social network offered by MFPE provides a significant advantage to relatives (Jewell
et al., 2009; McFarlane et al., 1995).
Another dimension of the MFPE versus SFPE comparison is the question of whether the per-
son with depression is included in the FPE intervention. In this review, consistent with the Falloon
et al. (1984) model on which most SFPE is based, all identified SFPE studies included the patient.
In the McFarlane et al. (1995) MFPE model for schizophrenia, clinically stable patients were also
included in the treatment alongside their relatives. However, in the four MFPE studies identified in
this review, consistent with their increased focus on family–carer outcomes, the person with
depression was not included in the FPE intervention. In contrast, in the Lemmens et al. (2009)
study which evaluated MFPE and SFPE, patients were included in both formats; and results
showed positive outcomes for both patient and carer with no significant differences between for-
mats. On the basis of the evidence to date, it is difficult to assess whether including or not including
the patient in FPE is preferable. Further comparative research is needed that examines both
patient and carer outcomes for both models as well as hybrid models where some sessions include
the person with depression and some sessions are conducted exclusively with the family–carers so
that there is an opportunity to specifically focus on their needs.
Only one non-RCT study of a peer-led group was identified with low methodological quality
(Stephens et al., 2011). The researchers found significant improvements in well-being for relatives
of patients with depressive or anxious disorders, providing some initial support for peer-led inter-
ventions for FPE in depression. This study used an intervention that was provided to family–carers
of people across mixed-diagnosis groups and no studies were identified that investigated the impact
of peer-led FPE groups designed specifically for families of MDD patients. In light of the notable
paucity of studies in this field, it is not yet possible to determine whether peer-led interventions are
equally effective as clinician-led programs. In a climate of limited resources, peer-led groups have
the potential to play a significant role in FPE for MDD and more research on this topic is clearly
warranted.
Three studies provided preliminary evidence indicating that FPE might be more effective for
both the person with depression and their family–carers when it is targeted specifically at MDD
compared to FPE that is aimed at any mental disorder. In the SFPE study by Clarkin et al. (1990),
although the families of patients with MDD were treated separately, the program was not tailored
to MDD and the authors found that neither the person with MDD nor their relatives reported sig-
nificant benefits. As aforementioned, the authors suggested that this may be because the treatment
had been primarily designed for people with schizophrenia. Stephens et al. (2011) also cited this as
a reason why the “Well-Ways” MFPE program for carers of people with any mental disorder

258 JOURNAL OF MARITAL AND FAMILY THERAPY April 2017

showed better outcomes for families of persons with psychotic disorders compared to those with
nonpsychotic disorders, including MDD.
The need for specifically targeted interventions is also suggested by the findings from a study
by van Wijngaarden et al. (2009) who identified a number of differences between family–carers of
people with schizophrenia and MDD with regard to what had the greatest impact on their well-
being. Specifically, they found that family–carers of persons with depression identified relationship
difficulties and ability to communicate with their relative as key strains, whereas those with
schizophrenia tended to report more strain as a result of the practical care that had to be provided
to their relative and worry about the person’s future functioning. This pattern of findings suggests
that FPE for psychotic disorders may need to be separated from FPE for nonpsychotic disorders
to effectively address the differing issues experienced by these carer groups.
Importantly, however, van Wijngaarden et al. (2009) suggest that these differences may arise
from the care-giving context rather than the illness. Because schizophrenia typically starts in young
adulthood, care is typically provided by parents, who are not necessarily living with the person full
time. Onset of MDD is typically later, and the primary carer is most often the spouse or partner
who is more likely to be spending many hours a day with the person and coping with a complete
change in the person with whom they first formed a relationship. Thus, the positive outcomes for
interventions specifically targeted at MDD may be because the interventions were able to better
reflect the needs of partner–carers while the mixed-group FPE interventions were trying to address
the needs of partner–carers, parent–carers, and others. Unfortunately, there were insufficient stud-
ies identified in this review that clearly distinguished between couple dyads versus other family
members/relatives; however, it is possible that the nature of the relationship (as partner, parent, or
other) will affect the degree to which FPE is effective. This highlights the clear need for further
research exploring the components of FPE that account for positive outcomes, the degree to which
outcomes are affected by relationship type, and further work regarding whether FPE should be
adapted for different relationship carer groups (e.g., parent, partner, etc.).
The findings from this review also need to be considered in context of limitations including
gaps in the available evidence base. First, in relation to the limitations pertaining to the method of
this review, the studies identified were based on trials published in peer review journals inclusive of
conference abstracts. It is therefore possible that other FPE-based trials for MDD may have been
missed, but which have yet to be published in peer review journals. Second, as noted, due to devel-
opmental considerations the current review was restricted to studies based on patient samples
within a minimum age of 14 years or older. Interestingly, only one adolescent study was identified
based on a SFPE intervention, and the findings were comparable to other SFPE studies based on
adult samples. Notwithstanding these two limitations, this is the first most comprehensive system-
atic review which has focused on evaluating the efficacy of FPE in MDD. In doing so, as aforemen-
tioned, we identified two related non-English trials thus enabling a more international evaluation
of this field.
Importantly, the findings from this review accentuated a number of limitations and gaps in
this field. Only 50% of trials identified (i.e., N = 5) were based on an RCT design. Second, no stud-
ies reported level of session attendance by participants; hence, it cannot be determined whether
results were influenced by treatment dosage effect. Third, only one study directly compared out-
comes for a SFPE versus an MFPE intervention, and only one study of a peer-led program
reported outcomes for families of people with nonpsychotic disorders. A fourth limitation in this
field is that only two studies measured follow-up outcomes beyond 6 months to determine treat-
ment retention effects. Fifth, there was little consistency in the measures used across studies; conse-
quently, this precluded a meta-analyses from being conducted.
In terms of methodological quality, given the criteria used, RCT designs had greater scope of
incurring a higher score. Indeed, five of the ten studies were evaluated to have adequate-to-strong
methodological quality. Four of these five trials comprised RCTs (Clarkin et al., 1990; Lemmens
et al., 2009; Sanford et al., 2006; Shimazu et al., 2011), whereas one was based on a “pseudo”
RCT design as patients were alternatively allocated to trial conditions (Kumar & Gupta, 2015). As
aforementioned, the fifth RCT study (Fiorillo et al., 2011) scored very low on methodological
quality as this was the only study identified which was not published in full manuscript format.
Moreover, of the five RCTs, three of the trials failed to report on whether assessors were blind,

April 2017 JOURNAL OF MARITAL AND FAMILY THERAPY 259

whereas for two of the trials there was no reporting of whether concealment of allocation occurred
during the randomization process. Overall, from the nine distinct study samples identified, the
majority (k = 7) of authors failed to report on handling of missing data, drop-outs, and/or whether
an ITT analyses was utilized. Given this information is vital for determining quality of analyses, as
well as facilitating in identifying discernible patterns of drop-out trends, researchers need to ensure
to document this information for future published trials in this field.
Collectively, these findings attest to the need for more rigorously designed RCTs using stan-
dardized measures that address both patient and carer outcomes over longer time periods. Indeed,
it is promising that recent published protocols suggest that these gaps will begin to be addressed
(e.g., Frank et al., 2015; Katsuki et al., 2014). Meta-analyses to identify strength of treatment
effects, comparative clinical trials where FPE is combined with effective individual treatments for
patients (e.g., cognitive-behavioral therapy), and a better understanding of what limits uptake of
FPE for MDD will further consolidate the evidence base in future years.
The finding that MFPE is at least as effective for the person with MDD and potentially
also for the family–carers has significant cost and resource implications for potential provi-
ders of FPE. Furthermore, the finding that FPE programs that are targeted at MDD (or at
least at nonpsychotic disorders), generally achieve better outcomes for MDD patients and
carers, is also relevant to the further evaluation and dissemination of FPE protocols. How-
ever, there is also a need for FPE at the acute stage of the illness (Anderson et al., 1986);
yet, MFPE programs are often provided at the less acute stage due to having to wait for suf-
ficient participant numbers to form groups. Models of FPE that can balance the need for
timeliness with the benefits of MFPE need to be developed and tested for efficacy. On-line
psychoeducation is now becoming common place for many disorders, particularly for provid-
ing information about the illness and treatment options (Stjernsw€ ard, 2012) and guided self-
help as a mode of delivery for psychoeducation is beginning to be investigated (e.g., McCann,
Songprakun, & Stephenson, 2015). In addition to the advantages pertaining to cost and par-
ticipant reach, providing at least some sessions of an FPE program on-line may provide
opportunity to tailor program content to different participant needs and to provide some ini-
tial psychoeducation at the time it is most needed. Nevertheless, the importance of connecting
with other family–carers in the same situation is a consistent theme arising from research into
what is important in FPE and this needs to be built into any on-line protocol (Stjernsw€ ard,
2012). It may be that a combination of on-line education sessions and face-to-face sessions,
primarily addressing skills and support components of FPE is an effective option. Such a
model may also offer opportunities for peer-led sessions which could complement the clini-
cian-led sessions. In summary, further research is needed into FPE intervention models that
are both flexible in delivering content for differing participant needs and sensitive to the orga-
nizational resource constraints.
The results from this systematic review highlight the promising outcomes that may be gained
by offering FPE for depression for both the patient and their family–carers. The findings also pro-
vide preliminary evidence that MFPE may be as effective as SFPE for patient outcomes and may
be potentially more effective for family–carers. If this is the case, it is important to consider how
multifamily interventions can be offered in both a timely and resource efficient manner. Given the
infancy of this field, further well-designed RCTs are clearly warranted to confirm the optimal deliv-
ery format and the types of outcomes that align with the needs for people living with depression
and their family members.

REFERENCES

Ahlstr€
om, B. H., Sk€ars€ater, I., & Danielson, E. (2009). Living with major depression: Experiences from families’ per-
spectives. Scandinavian Journal of Caring Sciences, 23(2), 309–316. doi: 10.1111/j.1471-6712.2008.00624.x.
Altman, D. G., Schulz, K. F., & Moher, D., . . ... CONSORT GROUP (Consolidated Standards of Reporting Trials).
(2001). The revised CONSORT statement for reporting randomized trials: Explanation and elaboration. Annals
of Internal Medicine, 17, 663–694.

260 JOURNAL OF MARITAL AND FAMILY THERAPY April 2017

Anderson, C. M., Griffin, S., Rossi, A., Pagonis, I., Holder, D. P., & Treiber, R. (1986). A comparative study of the
impact of education vs. process groups for families of patients with affective disorders. Family Process, 25(2),
185–205. doi: 10.1111/j.1545-5300.1986.00185.x.
Anderson, C. M., Hogarty, G. E., & Reiss, D. J. (1980). Family treatment of adult schizophrenic patients: A psycho-
educational approach. Schizophrenia Bulletin, 6(3), 490–505.
Angermeyer, M. C., Kilian, R., Wilms, H.-U., & Wittmund, B. (2006). Quality of life of spouses of mentally ill peo-
ple. International Journal of Social Psychiatry, 52(3), 278–285. doi: 10.1177/0020764006067186.
Bernert, S., Heider, D., Schomerus, G., Wilms, H. U., Wittmund, B., Bull, N., et al. (2006). Efficacy of an
intervention program on burnout symptoms of partners of depressed patients. Gesundheitswesen, 86(8–9),
545–550.
Bolkan, C. R., Bonner, L. M., Campbell, D. G., Lanto, A., Zivin, K., Chaney, E., et al. (2013). Family involvement,
medication adherence, and depression outcomes among patients in veterans affairs primary care. Psychiatric
Services, 64(5), 472–478. doi: 10.1176/appi.ps.201200160.
Chien, W. T., & Chan, S. W. C. (2004). One-year follow-up of a multiple-family-group intervention for Chinese fami-
lies of patients with schizophrenia. Psychiatric Services, 55(11), 1276–1284.
Clarkin, J. F., Glick, I. D., Haas, G. L., Spencer, J. H., Lewis, A. B., Peyser, J., et al. (1990). A randomized clinical
trial of inpatient family intervention V. results for affective disorders. Journal of Affective Disorders, 18(1), 17–
28. doi: 10.1016/0165-0327(90)90113-M.
Coyne, J. C., Kessler, R. C., Tal, M., Turnbull, J., Wortman, C. B., & Greden, J. F. (1987). Living with a depressed
person. Journal of Consulting & Clinical Psychology, 55(3), 347–352. doi: 10.1037/0022-006X.55.3.347.
Dixon, L. B., Lucksted, A., Medoff, D. R., Burland, H., Stewart, B., Lehman, A. F., et al. (2011). Outcomes of a ran-
domized study of a peer-taught family-to-family education program for mental illness. Psychiatric Services, 62
(6), 591–597. doi: 10.1176/appi.ps.62.6.591.
Falloon, I. R. (2003). Family interventions for mental disorders: Efficacy and effectiveness. World Psychiatry, 2(1),
20–28.
Falloon, I. R., Boyd, J., & McGill, C. (1984). Family care of Schizophrenia: A problem-solving approach to the treat-
ment of mental illness. New York, NY: Guilford Press.
Falloon, I. R., & Shanahan, W. (1992). Prevention of major depressive episodes: Early intervention. Journal of Men-
tal Health, 1(1), 53–53.
Fiorillo, A., Malangone, C., & Del Vecchio, V. (2011). The effect of family psychoeducational interventions on
patients with depression. European Psychiatry, 26(Supplement 1), 2209. doi: 10.1016/S0924-9338(11)73912-8.
Frank, F., Rummel-Kluge, C., Berger, M., Bitzer, E. M., & H€ olzel, L. P. (2014). Provision of group psychoeducation
for relatives of persons in inpatient depression treatment - a cross-sectional survey of acute care hospitals in Ger-
many. BMC Psychiatry, 14(1), 88–104. doi: 10.1186/1471-244X-14-143.
Frank, F., Wilk, J., Kriston, L., Meister, R., Shimodera, S., Hesse, K., et al. (2015). Effectiveness of a brief psychoed-
ucational group intervention for relatives on the course of disease in patients after inpatient depression treat-
ment compared with treatment as usual – study protocol of a multisite randomised controlled trial. BMC
Psychiatry, 15, 259. doi: 10.1186/s12888-015-0633-4.
Henken, T., Huibers, M. J., Churchill, R., Restifo, K. K., & Roelofs, J. J. (2007). Family therapy for depression.
Cochrane Database of Systematic Reviews, 3, CD006728.
Heru, A. M., Ryan, C. F., & Madrid, H. (2005). Psychoeducation for caregivers of patients with chronic mood disor-
ders. Bulletin of the Menninger Clinic, 69(4), 331–340. doi: 00259284 (ISSN).
Highet, N. J., McNair, B. G., Davenport, T. A., & Hickie, I. B. (2004). “How much more can we lose?”: Carer and
family perspectives on living with a person with depression. Medical Journal of Australia, 181(8), S6–S9. doi:
0025729X (ISSN).
Institute of Medicine National Research Council. (2009). Depression in parents, parenting, and children: Opportunities
to improve identification, treatment, and prevention. Washington, DC: The National Academies Press.
Jeglic, E. L., Pepper, C. M., Ryabchenko, K. A., Griffith, J. W., Miller, A. B., & Johnson, M. D. (2005). A caregiving
model of coping with a partner’s depression. Family Relations, 54(1), 37–45. doi: 10.1111/j.0197-
6664.2005.00004.x.
Jewell, T. C., Downing, D., & McFarlane, W. R. (2009). Partnering with families: Multiple family group psychoedu-
cation for schizophrenia. Journal of Clinical Psychology, 65(8), 868–878. doi: 10.1002/jclp.20610.
Kangas, M., Bovbjerg, D. H., & Montgomery, G. H. (2008). Cancer-related fatigue: A systematic and meta-analytic
review of non-pharmacological therapies for cancer patients. Psychological Bulletin, 134, 700–741. doi: 10.1037/
a0012825.
Katsuki, F., Takeuchi, H., Konishi, M., Sasaki, M., Murase, Y., Naito, A., et al. (2011). Pre-post changes in psy-
chosocial functioning among relatives of patients with depressive disorders after Brief Multifamily Psychoedu-
cation: A pilot study. BMC Psychiatry, 11(1), 56–62. doi: 10.1186/1471-244X-11-56.
Katsuki, F., Takeuchi, H., Watanabe, N., Shiraishi, N., Maeda, T., Kubota, Y., et al. (2014). Multifamily psychoed-
ucation for improvement of mental health among relatives of patients with major depressive disorder lasting

April 2017 JOURNAL OF MARITAL AND FAMILY THERAPY 261

 more than one year: Study protocol for a randomized controlled trial. Trials, 15, 320. doi: 10.1186/1745-6215-
15-320.
Keitner, G. I., Ryan, C. E., Miller, I. W., Kohn, R., Bishop, D. S., & Epstein, N. B. (1995). Role of the family
in recovery and major depression. American Journal of Psychiatry, 152(7), 1002–1008. doi: 10.1176/
ajp.152.7.1002.
Kronm€ uller, K. T., Kratz, B., Karr, M., Schenkenbach, C., Mundt, C., & Backenstraß, M. (2006). Treatment utilisa-
tion of a psychoeducational group for relatives of patients with affective disorders. Der Nervenarzt, 77(3), 318–
326. doi: 10.1007/s00115-005.
Kumar, K., & Gupta, M. (2015). Effectiveness of psycho-educational intervention in improving outcome of unipolar
depression: Results from a randomised clinical trial. East Asian Archives Of Psychiatry: Official Journal Of The
Hong Kong College Of Psychiatrists = Dong Ya Jing Shen Ke Xue Zhi: Xianggang Jing Shen Ke Yi Xue Yuan Qi
Kan, 25(1), 29–34.
Leff, J., Berkowitz, R., Shavit, N., Strachan, A., Glass, I., & Vaughn, C. (1990). A trial of family therapy versus a rel-
atives’ group for schizophrenia. Two-year follow-up. The British Journal of Psychiatry, 157(4), 571–577. doi:
10.1192/bjp.157.4.571.
Lemmens, G. M. D., Eisler, I., Buysse, A., Heene, E., & Demyttenaere, K. (2009). The effects on mood of
adjunctive single-family and multi-family group therapy in the treatment of hospitalized patients with
major depression: A 15-month follow-up study. Psychotherapy and Psychosomatics, 78(2), 98–105. doi:
10.1159/000201935.
Luciano, M., Del Vecchio, V., Giacco, D., De Rosa, C., Malangone, C., & Fiorillo, A. (2012). A “family affair”? The
impact of family psychoeducational interventions on depression. Expert Review of Neurotherapeutics, 12(1), 83–
92. doi: 10.1586/ern.11.131.
Lucksted, A., McFarlane, W. R., Downing, D., & Dixon, L. (2012). Recent developments in family psychoeducation
as an evidence-based practice. Journal of Marital and Family Therapy, 38(1), 101–121. doi: 10.1111/j.1752-
0606.2011.00256.x.
McCann, T. V., Songprakun, W., & Stephenson, J. (2015). Effectiveness of guided self-help in decreasing expressed
emotion in family caregivers of people diagnosed with depression in Thailand: A randomised controlled trial.
BMC Psychiatry, 15, 258. doi: 10.1186/s12888-015-0654-z.
McFarlane, W. R., Dixon, L., Lukens, E., & Lucksted, A. (2003). Family psychoeducation and schizophrenia: A
review of the literature. Journal of Marital and Family Therapy, 29(2), 223–245. doi: 10.1111/j.1752-
0606.2003.tb01202.x.
McFarlane, W. R., Lukens, E., & Link, B. (1995). Multiple-family groups and psychoeducation in the treatment of
schizophrenia. Archives of General Psychiatry, 52(8), 679–687. doi: 10.1001/archpsyc.1995.03950200069016.
Moher, D., Shamseer, L., Clarke, M., Ghersi, D., Liberati, A., Petticrew, M., et al. (2015). Preferred reporting items
for systematic review and meta-analysis protocols (PRISMA-P) 2015 statement. Systematic Reviews, 4(1), 1.
doi: 10.1186/2046-4053-4-1.
NICE. (2014). Psychosis and Schizophrenia in Adults: Treatment and Management. Retrieved from https://www.ni-
ce.org.uk/guidance/cg178/.
Pitschel-Walz, G., Leucht, S., B€auml, J., Kissling, W., & Engel, R. R. (2001). The Effect of Family Interven-
tions on Relapse and Rehospitalization in Schizophrenia—A Meta-analysis. Schizophrenia Bulletin, 27(1),
73–92.
Sanford, M., Boyle, M., McCleary, L., Miller, J., Steele, M., Duku, E., et al. (2006). A pilot study of adjunctive fam-
ily psychoeducation in adolescent major depression: Feasibility and treatment effect. Journal of the American
Academy of Child and Adolescent Psychiatry, 45(4), 386–395. doi: 10.1097/01.chi.0000198595.68820.10.
Schomerus, G., Heider, D., Wittmund, B., Wilms, H. U., Brieger, P., Nause, B., et al. (2007). A group intervention
for spouses of patients with depression; Is it effective? Psychiatrische Praxis, 34(5), 239–245.
Shimazu, K., Shimodera, S., Mino, Y., Nishida, A., Kamimura, N., Sawada, K., et al. (2011). Family psychoeduca-
tion for major depression: Randomised controlled trial. The British Journal of Psychiatry, 198(5), 385–390. doi:
10.1192/bjp.bp.110.078626.
Shimodera, S., Furukawa, T., Mino, Y., Shimazu, K., Nishida, A., & Inoue, S. (2012). Cost-effectiveness of family
psychoeducation to prevent relapse in major depression: Results from a randomized controlled trial. BMC Psy-
chiatry, 12(1), 40. doi: 10.1186/1471-244X-12-40.
Stephens, J. R., Farhall, J., Farnan, S., & Ratcliff, K. M. (2011). An evaluation of Well Ways, a family education pro-
gramme for carers of people with a mental illness. Australian & New Zealand Journal of Psychiatry, 45(1), 45–53.
doi: 10.3109/00048674.2010.
Stjernsw€ard, S. M. J. (2012). Online self-help tools for the relatives of persons with depression - a feasibility study.
Scandinavian Journal of Caring Sciences, 26(1), 70–80. doi: 10.1111/j.1471-6712.2011.00905.x.
Substance Abuse and Mental Health Services Administration (SAMHSA). (2009). Family psychoeducation: How to
use the evidence-based practices KITS. Rockville, MD: HHS.

262 JOURNAL OF MARITAL AND FAMILY THERAPY April 2017

Verhagen, A. P., de Vet, H. C., de Bie, R. A., Kessels, A. G., Boers, M., Bouter, L. M., et al. (1998). The Delphi list:
A criteria for qulaity assessmnet of randomized clinical trials for conducting systematic reviews developed by
Delphi consensus. Journal of Clinical Epidemiology, 51, 1235–1241.
van Wijngaarden, B., Koeter, M., Knapp, M., Tansella, M., Thornicroft, G., Vazquez-Barquero, J. L., et al. (2009).
Caring for people with depression or with schizophrenia: Are the consequences different? Psychiatry Research,
169(1), 62–69. doi: 10.1016/j.psychres.2008.06.013.
Wirrell, J., McGill, K. M., Kelly, P. J., & Bowman, J. (2014). Caring for someone with depression: Attitudes and clin-
ical practices of Australian mental health workers. Australian Psychologist, 49(6), 403–411. doi: 10.111/
ap.12058.

SUPPORTING INFORMATION

Additional Supporting Information may be found in a supplemental file next to the online version
of this article:

Table S1. Methodological Quality of Studies.

April 2017 JOURNAL OF MARITAL AND FAMILY THERAPY 263