Data from two studies, one naturalistic and the other a controlled trial, were analysed to clarify the relationships between independent life events, Expressed Emotion of a key relative, maintenance neuroleptics and the relapse of schizophrenia. It was found that patients in the community who are unprotected by medication are vulnerable either to acute stress in the form of life events or to chronic stress in the form of living with a high Expressed Emotion relative. Patients on regular medication are protected against one or other stress, but are very likely to relapse if the two forms of stress occur together. A model of schizophrenic susceptibility to environmental stress is constructed to incorporate these observations.

The relapse and readmission rates of schizophrenic patients who participated in a controlled trial of a nine-month behavioural family intervention trial based on the EE status of their relatives are presented at two years. The patients who received the behavioural family intervention had lower rates of relapse and readmission than patients from high-EE homes who had received a short educational programme or routine treatment. The relapse rate of the behavioural family intervention group (33%) was the same as that of the low-EE group (33%), and significantly lower than that of the non-intervention high-EE group (59%).

Schizophrenic patients living in high contact with relatives having high expressed emotion (EE) were recruited for a trial of social interventions. The patients were maintained on neuroleptic medication, while their families were randomly assigned to education plus family therapy or education plus a relatives group. Eleven out of 12 families accepted family therapy in the home, whereas only six out of 11 families were compliant with the relatives group. Non-compliance was associated with a poorer outcome for the patients in terms of the relapse rate. The relapse rate over nine months in the family therapy stream was 8%, while that in compliant families in the relatives group stream was 17%. Patients' social functioning showed small, non-significant, gains. The data from the current trial were compared with data from a previous trial. The lowering of the relapse rate in schizophrenia appears to be mediated by reductions in relatives' EE and/or face-to-face contact, and is not explained by better compliance with medication. Reduction in EE and/or contact was associated with a minuscule relapse rate (5%). Very little change occurred in families who were non-compliant with the relatives group. On the basis of these findings, we recommend that the most cost-effective procedure is to establish relatives groups in conjunction with family education and one or more initial family therapy sessions in the home. It is particularly important to offer home visits to families who are unable to or refuse to attend the relatives groups.

Schizophrenic patients were recruited into a trial of a prophylactic behavioural intervention with families. Families with at least one high Expressed Emotion (EE) relative were randomly allocated to one of four intervention groups: Behavioural Intervention Enactive; Behavioural Intervention Symbolic; Education Only; Routine Treatment. Patients from low-EE families were randomly allocated to two groups: Education Only or Routine Treatment. Relapse rates over nine months after discharge were significantly lower for patients in the two Behavioural Intervention, compared with Education Only and Routine Treatment groups. There was little difference between the two low-EE groups. Patients returning to high-EE relatives showed significantly higher relapse rates than those returning to low-EE relatives, in groups not receiving active intervention. Changes from high to low EE occurred in the Behavioural Intervention groups, and similar although less extensive changes occurred in the Education Only and Routine Treatment groups. Changes in criticism and marked emotional over-involvement (EOI) occurred generally in high-EE groups but were larger in magnitude in the Enactive and Symbolic groups. Reduction of hostility only occurred in the Behavioural Intervention groups. These results give partial support for the causal role of EE in relapse. There were no significant differences between the groups with respect to contact with the psychiatric services or medication.

A two-year follow-up study was conducted, mainly from case notes, of patients who had originally taken part in a study of the influence of relatives' expressed emotion on schizophrenic relapse. It was found that in the two years following discharge from hospital patients from high expressed emotion homes had a significantly greater relapse rate than those from low expressed emotion homes.

The prophylactic effect of maintenance drugs was no longer evident for patients from high expressed emotion homes at the two-year follow-up. By contrast, a significant protective effect of maintenance therapy emerged for patients from low expressed emotion homes.

The interpretation of these findings is discussed.

A history of life events in the three months before onset of illness was taken in a group of schizophrenic and a group of depressed neurotic in-patients. The Expressed Emotion (EE) of the patients’ key relatives was measured. There was no difference between the schizophrenic and depressed patients in the rate of events in the three month period before onset or in the proportion of undesirable events. However, the two groups exhibited significantly different patterns of interaction between life events and relatives’ EE. Schizophrenic patients living with high EE relatives had a low rate of life events compared with those living with low EE relatives, whereas depressives living with high criticism relatives had a high rate of events compared with those living with low criticism relatives. The implications of these findings are discussed in relation to schizophrenic and depressive patients’ differential vulnerability to environmental stress.

This study is a replication and extension of past work carried out by Brown, Birley and Wing (1972) concerning the influence of family life on the course of schizophrenia. In the original research the index of emotion expressed by a key relative about the patient at the time of key admission proved to be the best single predictor of symptomatic relapse in the nine months after discharge from hospital. In the present study this main finding of Brown et al has been replicated for two clinically different groups of psychiatric patients. The expressed emotion of the relative again seems to be associated with relapse independently of all other social and clinical factors investigated. In addition, important additive effects between various social influences and pharmacological treatments have been revealed which make it possible to predict relapse patterns in schizophrenia with considerable precision. The patterns of these relationships with relapse are different for the two clinical groups studied, patients with schizophrenic psychosis and with depressive neurosis.