The family attitude scale: reliability and validity of a new scale for measuring the emotional climate of families

Abstract

Research on outcomes from psychiatric disorders has highlighted the importance of expressed emotion (EE), but its cost-effective measurement remains a challenge. This article describes development of the Family Attitude Scale (FAS), a 30-item instrument that can be completed by any informant. Its psychometric characteristics are reported in parents of undergraduate students and in 70 families with a schizophrenic member. The total FAS had high internal consistency in all samples, and reports of angry behaviour in FAS items showed acceptable inter-rater agreement. The FAS was associated with the reported anger, anger expression and anxiety of respondents. Substantial associations between the parents' FAS and the anger and anger expression of students was also observed. Parents of schizophrenic patients had higher FAS scores than parents of students, and the FAS was higher if disorder duration was longer or patient functioning was poorer. Hostility, high criticism and low warmth on the Camberwell Family Interview (CFI) were associated with a more negative FAS. The highest FAS in the family was a good predictor of a highly critical environment on the CFI. The FAS is a reliable and valid indicator of relationship stress and expressed anger that has wide applicability. ©1997 Elsevier Science Ireland Ltd.

Introduction

Over the last 30 years, a substantial number of studies have examined the role of life events and social environment on the course of schizophrenia (Bebbington and Kuipers, 1992). One indicator of stressors within the living environment, expressed emotion (EE), has demonstrated a strong association with the course of psychiatric disorders (Kavanagh, 1992). High EE is usually defined by the level of criticism, hostility or emotional over-involvement (EOI) that is displayed by one or more of the patient's relatives on the Camberwell Family Interview, CFI (Vaughn and Leff, 1976). An increased risk of relapse is most reliably demonstrated with criticism or hostility, although high levels of emotional over-involvement are also associated with poorer outcomes (Kavanagh, 1992).

The CFI has strong psychometric data supporting its use. After training in administration and scoring, users can demonstrate high inter-rater reliability (Vaughn and Leff, 1976), and CFI ratings correlate strongly with behavioural observations of family interactions (Valone et al., 1983; Miklowitz et al., 1984). However, the CFI requires 4–5 h person to administer and score, and a valid assessment of the home environment usually requires the rating of at least two family members (Vaughn and Leff, 1976). Access to training courses also presents problems, especially for practitioners outside the UK. A partial solution is to develop a briefer interview method such as the Five-Minute Speech Sample (FMSS) which is rated on similar criteria to the CFI (Magaña et al., 1986). The FMSS is a valid measure of EE, although it misses some instances of high EE that are identified by the CFI (Malla et al., 1991). Like the CFI, the FMSS requires that raters be trained and monitored for reliability.

An alternative and potentially cost-effective method of obtaining information on family relationships is to use questionnaires. When this study began, four main questionnaires had been used in this area: the Parental Bonding Instrument, PBI (Parker et al., 1982); the Level of Expressed Emotion Scale, LEE (Cole and Kazarian, 1988); the Patient Rejection Scale, PRS (Kreisman et al., 1979); and the Family Environment Scale, FES (Moos and Moos, 1981). While all of them appeared conceptually linked to EE, evidence on their association with the CFI was weak (e.g. Parker et al., 1989).

There was a need to develop a questionnaire that could be completed by either relatives or patients, was sensitive to mild forms of criticism or annoyance, and would show an association with CFI variables. The current article describes the development of the Family Attitude Scale (FAS) which attempted to fill this need. The first study used a sample of undergraduate students and their parents to examine the psychometric characteristics of the measure.