30th October 2023

Expressed Emotion

What is expressed emotion?

Expressed emotion means the level of spontaneous comment about and reaction to a patient from their family in relation to their mental or physical health condition. First developed by George Brown and his colleagues in the second half of the 20th century, the concept is based on five principles (Myung and Furrow, 2019):

Levels of expressed emotion are culturally dependent. As an example, what is seen as emotional overinvolvement in one culture may be seen as completely normal in another culture. To have less involvement in that culture may actually be seen as a lack of care and concern. Therefore, cultural norms must be taken into account when assessing the level of expressed emotion (Bhugra and McKenzie, 2003).

Measures of expressed emotion

There are many different ways of determining the level of expressed emotion within a family. We have discussed a few of them below. An important thing to remember is that most of the methods require deep knowledge and understanding of how to interpret interviews and are complex in how they come to their conclusions. These summaries give a flavour of the methods used, but they are not in-depth representations of them.

Camberwell family interview (CFI)

This is the most commonly used tool for measuring expressed emotion. It was first developed as a way of assessing ‘emotional involvement’ when Brown and colleagues were studying the effect of a patient’s family environment on the likelihood of relapse in schizophrenia patients. The interview schedule focuses on the circumstances of the family home in the three months before admission for a relapse. Information is gathered on how the current relapse has affected family life, including arguments, contributions around the house and interactions between family members and the patient. It also looks at the patient’s feelings towards their family by assessing the emotional quality of speech and the content of the comments made.

When the CFI was developed in 1972, interviews were undertaken with 80 psychiatric patients and their spouses. Each patient and spouse were given separate interviews, followed by a short interview together. 30 additional families were then interviewed to assess the reliability and validity of the interviews. Emotions were given ratings on two scales. These were observed emotions, and several remarks, positive or negative, were made about a family member by the interviewee. The development of this tool allowed an index of expressed emotion to be created.

This was felt to be a reliable measure of expressed emotion. However, interviews could last up to five hours and would need to be recorded for coding later. This could be challenging for patients and their families, as well as requiring many researcher hours. As a result, in 1976, Vaughn and Leff set about to reduce the time needed to conduct the interview by around half without compromising on reliability or validity. They found that by prioritising certain topics, all areas of criticism and emotion that the interviewee had were covered without needing the full interview.

In practice, the CFI is still a lengthy process and requires significant interviewer training. Therefore, alternative tools for assessing expressed emotion have been developed (Vaughn and Leff, 1976).

Five minute speech sample (FMSS)

As the name suggests, this method of assessing expressed emotion is much quicker. It was developed in 1969 by Gottschalk and Gleser. An important family member is asked to talk about the patient for five minutes. The only direction they are given is to talk about what they think about the patient and how they get along with them. The interviewer does not interrupt in this time.

The interpreter of the interview then uses the following scales to assess the level of expressed emotion:

1. Quality of the initial statement.
2. Quality of the relationship.
3. Criticism.
4. Emotional overinvolvement.

Definitions of high expressed emotion statements are then split into critical or emotional overinvolvement. The definition of critical high expressed emotion is making one or more critical comments in the statement or if the first sentence of the statement is negative. Emotionally overinvolved high expressed emotion is noted if there is self-sacrificing behaviour or emotions are obviously displayed during the interview. Emotional overinvolvement can also be detected if there are two of the following characteristics: excessive past details, positive remarks totalling more than five, or a statement of feelings about the patient.

The FMSS gives similar results to the CFI when it detects highly expressed emotion. Families with high expressed emotion in the FMSS are also highly likely to have high expressed emotion on the CFI. However, more borderline cases of expressed emotion are under-detected on FMSS when compared with CFI (van Humbeeck et al., 2002).

Level of expressed emotion scale (LEE)

In 1988, Cole and Kazarian set out to create a questionnaire for the measurement of expressed emotion. This would serve the purpose of eliminating the need for an interviewer and someone to code the interview transcript. Using the CFI as a basis, they established four domains for the questionnaire items. These were:

1. Intrusiveness.
2. Emotional response.
3. Attitude towards illness.
4. Tolerance and expectations.

The questionnaire is given to the patient only. The final version of the scale contains 60 items, 15 in each domain, each with a true or false response. As this is a self-report questionnaire for the patient, family member perspective is not recorded. This may mean that factors such as the patient’s own expressed emotion levels and previous feelings towards family members before illness onset may skew the results. In addition, thoughts and feelings of the family members that patients may not have been consciously aware of may not come to light. These feelings are still important in assessing the level of expressed emotion (Cole and Kazarian, 1988).

Impact of highly expressed emotion

The concept of expressed emotion in families was first conceived in the context of schizophrenia after Brown and colleagues noticed that inpatients who were discharged back to the same family environment as before were more likely to relapse than if they went to a new home, such as to different family members. From there, research showed that highly expressed emotion was an important predictor for relapse in schizophrenia.

Since this pioneering research, a similar link has been proposed for other psychological conditions such as depression, eating disorders and post-traumatic stress disorder, although the link has not been quite so well established yet.

Work has been done to reduce the levels of expressed emotion within families to try and reduce rates of relapse. A bonus effect of this would be to help with mental health of caregivers, reducing carer burnout. Many studies have shown that self-help interventions, group therapy and educational strategies can all help to improve outcomes for patients and their families, reducing expressed emotion and reducing relapse (Amaresha and Venkatasubramanian, 2012).

Many of these studies are from the 80s and 90s and often use traditional family models in Western cultures to establish the possibility and impact of reducing expressed emotion. Further studies are needed to assess the impact on other types of families, as well as other cultures, although these are starting to happen.

Studies have also evaluated the impact of healthcare staff expressed emotion on patients. Whilst staff tend to have less expressed emotion than families, there is still some present. This tends to mostly take the form of critical comments, with hostility and emotional overinvolvement being seen less often. Two theories for a difference in expressed emotion between staff and families is that staff have less emotional connection to patients and often have professional coping strategies and training to reduce expressed emotion. Perhaps unsurprisingly, research seems to show that staff expressed emotion has less of an effect on patients than family expressed emotion. The presence or absence of a positive relationship between staff and patient seems to be more important than the level of expressed emotion (Berry, Barrowclough and Haddock, 2011).

References

• Amaresha, A.C. and Venkatasubramanian, G. (2012) ‘Expressed emotion in schizophrenia: an overview’, Indian journal of psychological medicine, 34(1), pp. 12–20.
• Berry, K., Barrowclough, C. and Haddock, G. (2011) ‘The role of expressed emotion in relationships between psychiatric staff and people with a diagnosis of psychosis: a review of the literature’, Schizophrenia bulletin. 37(5), pp. 958–972.
• Bhugra, D. and McKenzie, K. (2003) ‘Expressed emotion across cultures’, Advances in Psychiatric Treatment. 9(5), pp. 342–348.
• Cole, J.D. and Kazarian, S.S. (1988) ‘The level of expressed emotion scale: A new measure of expressed emotion’, Journal of Clinical Psychology, 44(3), pp. 392–397.
• van Humbeeck, G. et al. (2002) ‘Expressed emotion: A review of assessment instruments’, Clinical Psychology Review, 22(3), pp. 321–341.
• Myung, H.S. and Furrow, J.L. (2019) ‘Expressed Emotion in Families’, in J.L. Lebow, A.L. Chambers, and D.C. Breunlin (eds) Encyclopedia of Couple and Family Therapy. Cham: Springer International Publishing, pp. 980–985.
• Vaughn, C. and Leff, J. (1976) ‘The Measurement of Expressed Emotion in the Families of Psychiatric Patients’, British Journal of Social and Clinical Psychology, 15(2), pp. 157–165.