Couple and Family Treatments: Study Quality and Level of Evidence

• Published date: 01 March 2015
• Date: 01 March 2015
• DOI: http://doi.org/10.1111/famp.12106
• Author: Joëlle Darwiche,Yves Roten

Couple and Family Treatments: Study Quality and

Level of Evidence

JO

€

ELLE DARWICHE*

YVES DE ROTEN

†

This paper examines the application of the guidelines for evidence-based treatments

in family therapy developed by Sexton and collaborators to a set of treatment models.

These guidelines classify the models using criteria that take into account the distinctive

features of couple and family treatments. A two-step approach was taken: (1) The qual-

ity of each of the studies supporting the treatment models was assessed according to a

list of ad hoc core criteria; (2) the level of evidence of each treatment model was deter-

mined using the guidelines. To reflect the stages of empirical validation present in the

literature, nine models were selected: three models each with high, moderate, and low

levels of empirical validation, determined by the number of randomized clinical trials

(RCTs). The quality ratings highlighted the strengths and limitations of each of the

studies that provided evidence backing the treatment models. The classification by level

of evidence indicated that four of the models were level III, “evidence-based” treat-

ments; one was a level II, “evidence-informed treatment with promising preliminary

evidence-based results”; and four were level I, “evidence-informed” treatme nts. Using

the guidelines helped identify treatments that are solid in terms of not only the number

of RCTs but also the quality of the evidence supporting the efficacy of a given treat-

ment. From a research perspective, this analysis highlighted areas to be addressed

before some models can move up to a higher level of evidence. From a clinical perspec-

tive, the guidelines can help identify the models whose studies have produced clinically

relevant results.

Keywords: Evidence-based Treatment; Randomized Clinical Trial; Guidelines

Fam Proc 54:138–159, 2015

Evidence supporting the efficacy and effectiveness of couple- and family-based treat-

ments has been provided in various reviews and meta-analyses (e.g., Carr, 2009a,b,

2014; Hazelrigg, Cooper, & Borduin, 1987; Sexton, Datchi, Evans, LaFollette, &

Wright, 2013; Shadish & Baldwin, 2003; Sprenkle, 2012; Stratton, 2010; Von Sydow,

Beher, Retzlaff, & Schweitzer, 2007; Von Sydow, Beher, Schweitzer, & Retzlaff, 2010;

*Institute of Psychology, Faculty of Social and Political Sciences, University of Lausanne, Lausanne,

Switzerland.

†

Institute of Psychotherapy, Department of Psychiatry, Lausanne University Hospital and University of

Lausanne, Lausanne, Switzerland.

Correspondence concerning this article should be addressed to Jo€

elle Darwiche, Institut de Psychologie,

Facult

e des Sciences Socialeset Politiques, G

eopolis, 1015 Lausanne, Switzerland.E-mail: Joelle.Darwiche@

unil.ch

The authors thank Sarah Zimmer for editing the English version of the manuscript, Nahema El Ghaziri

and Gloria Repond for the coding of the material, and the developers of the models for providing precious

information about the studies reviewed.

138

Family Process, Vol. 54, No. 1, 2015 ©2014 Family Process Institute

doi: 10.1111/famp.12106

Von Sydow, Retzlaff, Beher, Haun, & Schweitzer, 2013). For a treatment to be consid ered

evidence-based, its evidence has to be assessed and found to meet certain standards. A

number of classification systems have been developed to assess the quality of the studies

providing evidence and rank treatment models according to their level of evidence (e.g.,

Chambless & Hollon, 1998; Chambless & Ollendick, 2001). Under these classifications,

the outcome data, the methodology used, and the transportability of the results are gener-

ally considered important factors in whether a treatment is rated empirically supported

(Westen, Novotny, & Thompson-Brenner, 2004).

When applied to couple and family therapy, these classifications were of limited

interest. Recently, Sexton et al. (2011) developed specific guidelines for classifying

evidence-based couple and family treatments to include the multiple perspectives and

dimensions needed to assess family based intervention research (Sexton & Alexander,

2002). The guidelines are based on the recommendations of the Society for Family

Psychology of the American Psychological Association Task Force on Evidence-Base d

Treatments (APA, 2007). Sexton et al.’s (2011) guidelines recommend evaluatin g the

available research outcomes using criteria that take into account the specific features

of couple and family treatments. Such an evaluation would therefore assess whether

the treatments are effective for the problems for which couples and families

seek therapy (e.g., marital conflict, parent–child coalitions, family disengagement).

It would also test whether the observed changes take place at multiple levels

(e.g., improvement in individual symptomatology, parent–child relations, and family

functioning).

The guidelines are applied to well-defined and described clinical interventions or treat-

ment models and classify them from the least to the most evidence, using a levels-of-

evidence-based approach. Level I designates “evidence-informed treatments”; level II

designates “evidence-informed treatments with promising preliminary evidence-based

results”; and level III designates “evidence-based treatments.” The guidelines go beyond

merely sorting out which treatments work and which do not—they also assess the amount

of detail provided about the intervention, the clinical validity and transportability of the

outcomes, and the overall quality of the study.

The aim of this paper was to test Sexton et al.’s guidelines to understand the evidence

of a set of selected couple and family treatment models. The guidelines may be used with

models that are supported by various types of studies—not only randomized clinical trials

but also process-to-outcome studies. As a first step, we limited our focus in this paper to

models validated by randomized trials, to assess the quality of studies that adopted a simi-

lar methodological rigor.

According to Sexton et al.’s method, before treatment models can be classified by

levels of evidence, the quality of the studies providing the supporting evidence has

to be assessed. We therefore took a two-step approach: First (1) the quality of each of

the studies supporting the treatment models was assessed according to a list of ad hoc

core criteria selected from Sexton et al.’s papers (2011, 2013) and from psychotherapy

research (Chambless & Hollon, 1998; Chambless & Ollendick, 2001; Gerber et al.,

2011; Sprenkle, 2012); then (2) the level of evidence of each treatment model was

determined by conducting an overall assessment of the available research data and the

information provided on each model’s theoretical background and specific techniques.

Our goals were: (1) to assess both the quality of the randomized clinical trials

(RCTs) supporting different couple and family treatment models and the level of evi-

dence of each of the models to test the usefulness of Sexton’s guidelines; (2) to high-

light the strengths of the models and specify the areas that need to be addressed to

improve their level of evidence.

Fam. Proc., Vol. 54, March, 2015

DARWICHE & DE ROTEN

/

139