Gabor I. Keitner, MD

About the author

Tamás Kurimay, MD, PHD

About the author

The McMaster Model of Family Functioning (MMFF) and the Family Assessment Device (FAD) as a Multidimensional Measurement of Family Functioning – International and Hungarian Experiences

The McMaster Model of Family Functioning (MMFF) and the Family Assessment Device (FAD) was developed over the past 60 years in the U.S. The therapy model and the questionnaire was introduced in Hungary 35 years ago. International and Hungarian findings from the past 30–60 years and recent results from the previous 2 years will be discussed in the symposium. The symposium is organized in honor of Professor Gabor Keitner (U.S., Rhode Island), who has a Hungarian origin.


The History of the McMaster Model of Family Functioning (MMFF) and the Family Assessment Device (FAD)

Gabor I. Keitner

Rhode Island Hospital and Brown University, Providence RI, USA

The McMaster Model was developed over the past 60 years under the leadership of Nathan Epstein. It took shape initially at McGill University in Montreal, McMaster University in Hamilton and established at Brown University in Providence… It evolved from an individual psychodynamic to an interpersonal and finally to a systems model. Throughout its development guiding principles included clinical utility and empirical testability. The MMFF emphasizes clarity, openness, family strengths and the family’s responsibility for change. The focus is on teaching the family ways to identify, understand and solve problems. The MMFF is grounded on a solid multidimensional theory of family functioning, presents a manualized and tested therapy model (The problem Centered Systems Therapy of the Family (PCSTF), and offers well-validated objective and subjective measures of family functioning across its multiple domains.

The Family Assessment Device (FAD) is a self-report questionnaire designed to assess the six dimensions of family functioning of interest to the MMFF including: problem solving, communications, affective involvement, affective responsiveness, roles, behavior control. It has been translated into 25 languages. It is able to discriminate between clinical populations and controls and among patients with different illnesses. It has good test-retest and concurrent reliability and sensitivity to change. Poor family functioning as assessed by the FAD is associated with lower recover rates and adherence to treatment, longer recovery time, poorer quality of life and increased risk of relapse and drop-out. The FAD is a suitable instrument for the evaluation of family functioning both in clinical and research settings.

Introduction and usage of the McMaster Model and FAD in Hungary: From the 90s to the present

Tamás Kurimay

Department of Psychiatry and Psychiatric Rehabilitation of Saint John Hospital, Budapest, Hungary

The McMaster Model and the Family Assessment Device (FAD) was introduced in Hungary in 1984–85 under the leadership and contribution of Prof. Gábor Keitner. The first step was the translation and validation of the FAD. Colleagues participating in this work were Judit Fodor, clinical psychologist at University of Szeged and Tamás Kurimay. Following the validation phase cross-cultural researches started in which family functioning of patients with major depression has been assessed by the FAD, and then North American and Hungarian clinical and non-clinical (control) families were compared (Keitner, Fodor et al., l991). A further application of the FAD covered the examination of patients with alcohol use disorder and their families in Hungary (Kurimay, l994). From the mid-1990s the initial version of the FAD was updated and used in the longitudinal Budapest Family Study led by Judit Gervai (Hungarian Academy of Science) and her team (Danis et al., 2005, 2008). Later, the FAD was applied to other patients suffering eating disorders and their families.

Recently, the FAD is being used in some other research projects, as, for example, in the field of perinatal care and early development.

In 2018 November, a large sample (n=1000) representative online community survey was conducted by Török and Danis et al. at Semmelweis University, Institute of Mental Health, which can serve new results on validation and standardization of FAD in Hungary.

In addition to research, the McMaster Model of Family Functioning has been included in family therapy trainings (Kardos &Kurimay, 1988). The model appears in a distance learning family therapy textbook, as well (Kurimay (ed.), 2004). A further experiment was the attempt for introducing the McSiff semi-structured interview in Hungary, however, this has not been used in Hungarian language yet.

The presentation shows also other applications and uses of the McMaster model in clinical settings and research.

Using FAD in the longitudinal Budapest Family Study (BFS)

Judit Gervai1, Krisztina Lakatos1, Ildikó Tóth1, Krisztina Ney1, Ildikó Danis2

1Institute of Cognitive Neuroscience and Psychology, Natural Science Research Center, Hungarian Academy of Sciences, Budapest

2Institute of Mental Health, Semmelweis University, Budapest

The longitudinal BFS aimed at studying first-born children’s behavior and relationship development from birth. The BFS started with 114 expecting couples, consisted of 103 families at the infants’ age of 12 months, and 89 families participated at the children’s age of 6 years. The parents were highly educated, mostly of middle class background.

Among the multiple contextual measures characterizing the environment of the BFS infants’ development, FAD was one of the instruments assessing family functioning. Here, we show descriptive statistics of maternal and paternal FAD subscales in the BFS sample, as well as within-individual stability and longitudinal changes from pregnancy through 6 months to 6 years of child’s age.

At the child’s age of 6 months and 6 years, we found a consistent pattern of sex of parent by sex of child interaction for the FAD total problems, and also for a number of subscales. In each case, fathers of boys and mothers of girls reported more family problems.

We examined relations between family functioning as measured by the FAD and some key variables of the BFS. Child attachment classified in the Strange Situation at 12 months with the mother, and at 18 months with the father was not related to either the 6‑month or the 6‑year FAD measures; nor was the 6‑year attachment to the mother related to the 6‑year FAD scale scores.

Child behavior problems assessed by the CBCL and SDQ questionnaires at 6 years of age had different relations with maternal and paternal FAD total problem and subscale scores depending on the sex of the child. Also, parents’ mental health measures (anxiety and depression) were significantly related to concurrent FAD total scores.

The nature of the BFS sample was likely to result in restricted ranges of FAD scores, so relations with other measures might turn out differently in families living in adverse circumstances. Nevertheless, we expected relations with insecure attachment in both infancy and childhood, but these failed to appear. At the same time, problems of family functioning indicated by parents were reflected in mothers’ report of multiple behavioral problems differentially for sons and daughters.

Rapid Assessment of Family Functioning: The Brief Assessment of Family Functioning Scale

Gabor I. Keitner

Rhode Island Hospital and Brown University, Providence RI, USA

The purpose of the present study is to compare results from the 12 item General Functioning scale (GF-FAD) of the Family Assessment Device (FAD) to a three-item version, the Brief Assessment of Family Functioning Scale (BAFFS), designed to be used when brevity is especially important.  We used principal components analysis of the GF-FAD, followed by multiple sample confirmatory factor analyses to test the robustness of the BAFFS in different samples. The BAFFS correlated highly with the GF-FAD, and demonstrated good concurrent validity with another measure of global marital functioning, the Dyadic Adjustment Scale‑4 in a help-seeking sample.  Like the 12 item version, the BAFFS moderately correlated with an objective, interview based rating of family functioning, the McMaster Clinical Rating Scale.  The BAFFS appears to serve as a good proxy for the GF-FAD when an ultra-brief family assessment measure is needed.

Psychometric and validation results on FAD, FAD-GF and BAFFS in a large representative community sample in Hungary

Ildikó Danis1, Szabolcs Török1, Réka Koren1, Tamás Kurimay2, Judit Gervai3

1Insitute of Mental Health, Semmelweis University, Budapest, Hungary

2 Department of Psychiatry and Psychiatric Rehabilitation of Saint John Hospital, Budapest, Hungary

3 Institute of Cognitive Neuroscience and Psychology, Natural Science Research Center, Hungarian Academy of Sciences, Budapest, Hungary

The initial Hungarian version of FAD was introduced in the 1980s by Kurimay and Fodor. The updated version created by Gervai and colleagues was used in small-sample studies and clinical settings from the late 1990s.

As we know from the international literature, FAD is an appropriate measure for examining family functioning both in clinical and non-clinical populations, however, its usage in family therapy and clinical studies is more widespread, than using it as a screening method in community samples. Perhaps the original 60-item FAD with its six subscales (Problem solving, Communications, Affective involvement, Affective responsiveness, Roles, Behaviour control) is not short enough for screening purposes.  Previously, the 12-item General Functioning (GF) scale was often used as a short measurement, but recently in 2017, the new very-short, 3‑item Brief Assessment of Family Functioning Scale (BAFFS) was introduced and validated also by Keitner and colleagues.

In November 2018, a large (n=1000) representative online survey was conducted in the Hungarian adult (18+) community population. The study was financed by an EFOP 5.2.5 grant, planned and organized by Török, Danis and colleagues at the Institute of Mental Health, Semmelweis University.

One of our aims was to measure the 60-item FAD, checking its factor structure and examining the internal consistency of the original 6 dimensions. We wanted to assess the cross-correlations between versions of different lengths, and also the proportion of the total variance of the six FAD subscales and the 12-item GF scale that is explained by the very short 3‑item BAFFS scale.

In order to examine concurrent validity, we measured other constructs, such as adult attachment styles (Experiences in Closed Relationships-Revised, ECR‑R, Fraley et al, 2000; Hungarian adaptation: Gervai et al, 2008), well-being (WHO Well-being Questionnaire, Hungarian adaptation and short version: Rózsa et al, 2003), perceived stress (Perceived Stress Scale – 4, PSS‑4, Cohen et al, 1983; Hungarian adaptation: Stauder & Konkoly Thege, 2006) and depressive symptoms (Depression Scale Questionnaire 1, DS1K, Halmai et al, 2008).

As our sample was representative of sex, age and residency in the community, we can also present Hungarian standards for all three versions (FAD, GF, BAFFS). Using cut-off points at 1 or 2 standard deviations from the mean can be useful in screening family functioning in general and in clinical practice, and also in research.

Main psychometric characteristics, validation and standardization results will be shown in the context of other international findings.

Recommended training