Elsevier

Psychiatry Research

Volume 210, Issue 2, 15 December 2013, Pages 641-646
Psychiatry Research

French version of the Family Attitude Scale: Psychometric properties and relation of attitudes to the respondent's psychiatric status

https://doi.org/10.1016/j.psychres.2013.07.008Get rights and content

Abstract

The Family Attitude Scale (FAS) is a self-report measure of critical or hostile attitudes and behaviors towards another family member, and demonstrates an ability to predict relapse in psychoses. Data are not currently available on a French version of the scale. The present study developed a French version of the FAS, using a large general population sample to test its internal structure, criterion validity and relationships with the respondents' symptoms and psychiatric diagnoses, and examined the reciprocity of FAS ratings by respondents and their partners. A total of 2072 adults from an urban population undertook a diagnostic interview and completed self-report measures, including an FAS about their partner. A subset of participants had partners who also completed the FAS. Confirmatory factor analyses revealed an excellent fit by a single-factor model, and the FAS demonstrated a strong association with dyadic adjustment. FAS scores of respondents were affected by their anxiety levels and mood, alcohol and anxiety diagnoses, and moderate reciprocity of attitudes and behaviors between the partners was seen. The French version of the FAS has similarly strong psychometric properties to the original English version. Future research should assess the ability of the French FAS to predict relapse of psychiatric disorders.

Introduction

“Expressed emotion” (EE) refers to communication of criticism, hostility or rejection about someone with a psychiatric illness, or reports of emotional over-involvement with them. It originated in the 1950s, following observation of interactions between inpatients with schizophrenia and their families (Brown et al., 1958, Brown et al., 1962). This theory is conceptually akin to other contemporaneous theories on family communication (Bateson et al., 1956, Bateson and Ruesch, 1951), but unlike those theories, EE is focused on family factors that increase risks of relapse in psychiatric disorders, rather than advancing a general theory of human communication (Favez, 2010). Research on EE has shown that risks of relapse in schizophrenia are substantially higher in families with high EE, especially when there is a substantial amount of contact between the patient and their family (Brown et al., 1958, Brown et al., 1962, Butzlaff and Hooley, 1998, Kavanagh, 1992). EE was argued to act as a major stressor, which triggers intense physiological activation, which in turn increases the risk of psychotic symptoms, social withdrawal and ultimately, relapse (Brown et al., 1972, Rosenfarb et al., 1995, Rosenfarb et al., 2006). Links between EE and relapse are also seen in patients with mood (Hooley et al., 1986, Hooley and Teasdale, 1989), anxiety (Chambless et al., 2001) and alcohol use disorders (O'Farrell et al., 1998).

Several instruments have been developed to measure the presence of EE in the familial environment. The “Camberwell Family Interview” (CFI; Brown et al., 1972), updated and shortened by Vaughn and Leff (1976), is the gold standard instrument, but necessitates initial training for interviewers and several hours for each individual interview and its subsequent rating. A shorter alternative is the “Five Minute Speech Sample” (FMSS; Gottschalk and Gleser, 1969, Magana et al., 1986) during which a family member speaks about his or her perception of the patient and their relationship for 5 min without interruption. Responses are coded using the CFI scoring procedures. However, the FMSS still requires training, and it misses some instances of high EE that are seen in a full CFI. As a result, its predictive validity for relapse is less well established than for the CFI (Hooley and Parker, 2006).

An alternative is to use a self-report instrument, which can be delivered without incurring the costs of training and of the administration and scoring of interviews. The Family Attitude Scale (FAS; Kavanagh et al., 1997) is one such instrument with strong psychometric characteristics. In an initial study, the FAS was administered to undergraduate students and their parents, and to relatives of people with schizophrenia spectrum disorders (Kavanagh et al., 1997). Strong evidence for a single-factor solution was obtained, and the internal consistency of the scale was high for parental FAS scores in both student and clinical samples (Cronbach alpha≥0.95). In the student sample, parental FAS scores significantly correlated with State and Trait Anger and Anxiety on the State-Trait Personality Inventory (Spielberger et al., 1983) (Median r across the four scales=0.28, p<0.001 for mothers, 0.44, p<0.001 for fathers) and with anger expression (r=0.35, p<0.001 for mothers, 0.49, p<0.001 for fathers) on the Anger Expression Inventory (Spielberger et al., 1985). In parents of people with schizophrenia spectrum disorders, parental Hostility and Criticism on the CFI were significantly associated with more negative parental FAS scores, especially in the case of maternal Criticism on the CFI and maternal FAS (r=0.66, p<0.001). Subsequent studies have confirmed the validity of both the English (Kavanagh et al., 2008) and Japanese versions of the FAS (Fujita et al., 2002) against the CFI. The FAS has also shown predictive validity for illness relapse in two samples of patients with psychosis (Kavanagh et al., 2008), although the relationship was weaker than when the CFI was used.

Up to now, information on the FAS scores of family members of patients with psychiatric disorders other than psychoses is sparse. Moreover, measures of EE have seldom been used to document attitudes and behaviors in relation to a marital partner. Nor are there many studies on potential effects of a respondent's own symptoms or diagnoses on EE or FAS scores. In parallel to a patient's symptoms influencing family members' adjustment to a given disorder (Albert et al., 2010), EE or attitudes towards another family member may well be affected by the respondent's own symptoms (Barrowclough and Parle, 1997). One study on 17 couples with depressed partners (Florin et al., 1992) showed that high EE of both the respondent and their depressed partner were significantly more common when the partner had a higher score on the Beck Depression Inventory (Beck et al., 1961). The study using the Japanese version of the FAS found the FAS ratings of 57 family members of 41 schizophrenic patients to be higher when they had more physical complaints themselves (Fujita et al., 2002). However, the FAS ratings of the family members were not significantly higher when they were more anxious or depressed, or had more social dysfunction (Fujita et al., 2002). A recent Polish study (Pankiewicz et al., 2012) showed no differences between mean FAS scores in 85 couples, where one or both partners suffered from Panic or Generalized Anxiety Disorders, than where neither had these disorders. Further research on this issue is needed. Moreover, studies on associations between EE and subthreshold mood disorders, which have gained increasing interest in contemporary psychiatry, are entirely lacking.

Given these gaps in existing literature regarding the sensitivity of the FAS to EE in non-psychotic disorders, the aims of the present study were to use a large general population sample: (a) to provide a short, internally coherent measure of EE in French-speaking cultures, by creating a French translation of the FAS, and testing its internal structure; (b) to establish the criterion validity of the translated FAS with other measures of relationship functioning (in particular, the Dyadic Adjustment Scale of Spanier (1976)); (c) to provide further data on the validity of the FAS, by examining relationships between FAS scores, and the respondent's own anxiety symptoms and anxiety, affective, psychotic and substance use disorders; and (d) to examine the extent of reciprocity between the FAS of respondents and their partners. An examination of associations between the respondent's own symptoms or diagnoses and the level of EE would extend the initial concepts, which primarily focused on the partner's psychopathology, to the role of the respondent's psychopathology in the development of emotional communication within the family system.

Section snippets

Participants

The present sample was derived from the CoLaus study, which included information on 6738 adults aged 35–75 years who were randomly selected from a list of residents of Lausanne, Switzerland, in 2003. That project assessed cardiovascular risk factors and collected DNA and plasma samples for the study of genetic variants and biomarkers (Firmann et al., 2008). The PsyCoLaus study (Preisig et al., 2009), which was based on a subsample of CoLaus, constituted its psychiatric arm. It included a

Sample characteristics

A total of 2072 (55.7%) of PsyColaus participants (49.9% female, mean age: 51.3 years, S.D.: 8.7 years) had completed the FAS after exclusion of questionnaires with more than 10% of missing data. In this subsample, 70.6% were married, 75% were of Swiss origin and 25.6% held professional specialty positions. Participants who completed all items of the FAS differed from those who did not in their gender (respectively, 50% vs. 57% female; χ2=18.0, d.f.=1, p<0.0001), age (51.3 vs. 50.5 years; F=

Discussion

Our first goal was to test the internal structure of the French FAS. Results showed satisfactory internal consistency, and the confirmatory factor analysis revealed an excellent fit of the single-factor solution. Our second goal was to establish the criterion validity of the French FAS. A lower FAS score, reflecting more positive attitudes and behaviors towards the partner, was strongly associated with a higher score on the Dyadic Adjustment Scale (predicting 69% of the variance). While more

Acknowledgments

This research was supported by three grants from the Swiss National Science Foundation (Grants nos. #32003b-105993, #32003b-118308 and #33csc0-122661) and two grants from GlaxoSmithKline clinical genetics to G. Waeber and M. Preisig.

The authors would like to express their gratitude to the Lausanne inhabitants who volunteered to participate in the PsyCoLaus study. We would also like to thank all the investigators of the CoLaus study, who made the psychiatric study possible, as well as many GSK

References (62)

  • J. Angst et al.

    The depressive spectrum: diagnostic classification and course

    Journal of Affective Disorders

    (1997)
  • J. Baillargeon et al.

    Traduction francaise de l'Echelle d'Adjustment Dyadique (French translation of the Dyadic Adjustment Scale)

    Canadian Journal of Behavioral Science

    (1986)
  • C. Barrowclough et al.

    Appraisal, psychological adjustment and expressed emotion in relatives of patients suffering from schizophrenia

    British Journal of Psychiatry

    (1997)
  • G. Bateson et al.

    Toward a theory of schizophrenia

    Behavioral Science

    (1956)
  • G. Bateson et al.

    Communication. The Social Matrix of Psychiatry

    (1951)
  • A.T. Beck et al.

    An inventory for measuring depression

    Archives of General Psychiatry

    (1961)
  • G. Brown et al.

    The influence of family life on the course of schizophrenic illness

    British Journal of Preventive Social Medicine

    (1962)
  • G.W. Brown et al.

    Influence of family life on course of schizophrenic disorders: replication

    British Journal of Psychiatry

    (1972)
  • R.L. Butzlaff et al.

    Expressed emotion and psychiatric relapse: a meta-analysis

    Archives of General Psychiatry

    (1998)
  • D.L. Chambless et al.

    Predicting expressed emotion: a study with families of obsessive–compulsive and agoraphobic outpatients

    Journal of Family Psychology

    (2001)
  • D.A. Cole

    Utility of confirmatory factor analysis in test validation research

    Journal of Consulting and Clinical Psychology

    (1987)
  • D.R. Crane et al.

    Assessing marital quality with distressed and nondistressed couples: a comparison and equivalency table for three frequently used measures

    Journal of Marriage and the Family

    (1990)
  • L.J. Cronbach

    Coefficient alpha and the internal structure of tests

    Psychometrika

    (1951)
  • H.J. Eysenck et al.

    Manual of the Eysenck Personality Questionnaire

    (1975)
  • H.J. Eysenck et al.

    La structure de la personnalité chez les français confrontée à celle des anglais, comparaison “cross-culturelle” (The personality structure of the French compared to the English people, “cross-cultural” comparison)

    Personnalité

    (1980)
  • N. Favez

    L'examen Clinique de la Famille. Modèles et Instruments d'Évaluation (Clinical Examination of the Family. Assessment Models and Instruments)

    (2010)
  • I. Florin et al.

    Expressed emotion in depressed patients and their partners

    Family Process

    (1992)
  • M. Firmann et al.

    The CoLaus study: a population-based study to investigate the epidemiology and genetic determinants of cardiovascular risk factors and metabolic syndrome

    BMC Cardiovascular Disorders

    (2008)
  • D.P. Goldberg et al.

    A User's Guide to the General Health Questionnaire

    (1988)
  • L. Gottschalk et al.

    The Measurement of Psychological States Through Analysis of Verbal Behavior

    (1969)
  • W.K. Halford et al.

    Individual psychopathology and marital distress. Analyzing the association and implications for therapy

    Behavior Modification

    (1999)
  • Cited by (8)

    • Measuring the expressed emotion in Chinese family caregivers of persons with dementia: Validation of a Chinese version of the Family Attitude Scale

      2016, International Journal of Nursing Studies
      Citation Excerpt :

      The brevity and self-reporting format of the FAS present a minimal administrative burden compared with the CFI (Hooley and Parker, 2006). Furthermore, a French version of the scale demonstrated significant association with dyadic adjustment, which is of particular importance in dementia caregiving (Vandeleur et al., 2013). By using a cut-off score of >60, the FAS has reliably been found to predict relapse of psychiatric disorder among patients with schizophrenia (Kavanagh, 1992).

    View all citing articles on Scopus
    View full text