SEARCH

SEARCH BY CITATION

Keywords:

  • expressed emotion;
  • family burden and distress;
  • family intervention;
  • schizophrenia;
  • State-Trait Anxiety Inventory

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES

Abstract  This study examined the direct effects of short-term psychoeducation on relatives of inpatients with schizophrenia, with the goal of introducing this type of support program into standard care. The subjects were 46 relatives of inpatients with schizophrenia who attended three or four sessions of psychoeducation. Levels of anxiety and subjective burden and distress were measured before and after sessions using self-administered rating scales. In addition, levels of expressed emotion were also measured. Results showed that both state and trait anxiety on the State-Trait Anxiety Inventory were significantly lower after psychoeducational intervention than before intervention. In addition, subjective burden and distress reported by the family significantly decreased on the subscales for family confusion resulting from a lack of knowledge of the illness and anxiety about the future, subjective burden and depression resulting from the patient’s illness, and difficulties in the relatives’ relationships with the patient. Comparison of high and low expressed emotion families showed that the intervention was almost equally effective for the two groups. However, its effectiveness with regard to the subjective burden and depression experienced by the families was significantly greater among high expressed emotion families. The present study confirmed that family psychoeducation during hospitalization, even for a short period, is effective for all families, whether high or low expressed emotion. Moreover, the results suggested that the intervention may have a greater effect on emotional factors in high expressed emotion families than in low expressed emotion families.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES

The excellent effectiveness of psychoeducational intervention for relatives of patients with schizophrenia in the treatment of this illness, particularly for preventing relapse, was first established in studies in the 1970s and has since been confirmed in countries throughout the world, including Japan.1 It is now considered necessary to integrate psychoeducation into standard therapy as an adjunct to medical treatment. However, psychoeducation is not widely used in clinical settings, as evidenced by the fact that behavioral family management (BFM), which is considered a standard form of psychoeducational intervention, is almost never used in Japan. One reason for this is that many such programs are costly and time-consuming. BFM, for example, involves 21 sessions over a period of 9 months. Consequently, providing such a program for all inpatients places a heavy burden on both those implementing the program and the participants. Therefore, in order to incorporate psychoeducation into standard therapy, an investigation is needed to examine whether a method of short-term education that is easier to implement can be used.

Linszen et al. reported that individual psychoeducational therapy for patients with recent onset schizophrenia and two sessions of psychoeducation for their relatives during hospitalization resulted in a relapse rate equal to that seen when psychoeducation for the relatives was continued during the subsequent outpatient period.2 Despite such reports, emphasis has been placed on the disadvantage of short-term programs in that they are less effective than long-term programs in preventing relapse. Little consideration has been given to the fact that short-term programs have been shown to be significantly more effective than standard therapy without psychoeducation.1 The reasons for this poor assessment of short-term programs include the fact that few studies have examined the direct effect of such programs on the relatives of patients with schizophrenia and the fact that the advantages of these programs, as compared with ordinary programs, have not been verified. The present study investigated the direct effects of psychoeducation on the relatives of patients with schizophrenia by assessing whether short-term psychoeducation results in changes in the levels of anxiety and burden and distress felt by the family members. The Japanese-language version of the State-Trait Anxiety Inventory (STAI) and an independently developed scale for assessing the levels of burden and distress in the patient’s family were used in the investigation. In addition, differing effects on relapse have been reported depending on whether the family is a high or low expressed emotion (EE) family.3,4 It is, therefore, desirable for psychoeducation to be at least as effective in high EE families as in low EE families. Therefore, the present study also examined differences in the effect of psychoeducation on high and low EE families, using the Five Minute Speech Sample (FMSS) to assess EE.

METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES

Subjects

Subjects were relatives of inpatients diagnosed with schizophrenia according to the Diagnostic and Statistical Manual of Mental Disorders, Third Edition, Revised5 criteria, who were hospitalized in the neuropsychiatry ward of the Nihon University Itabashi Hospital, Tokyo, Japan, between October 1997 and November 1999 or in the psychiatric ward of Ome Municipal General Hospital, Tokyo, in 2001. The subjects were recruited by the physician in charge of the patients at each hospital. Next, the individuals in charge of psychoeducational therapy explained the intervention sessions, research protocol, and privacy protection measures to the inpatients’ families. Informed written consent was obtained from all subjects. The 37 patients averaged 25.1 years of age and had been hospitalized an average of 1.6 times. The mean time from initial examination to hospitalization was 1066 days. The duration of morbidity was 1 year or less for 23 of the patients (62%). Because most of the patients at both hospitals were inpatients who had been hospitalized early during the acute phase, many were relatively young and a short time had elapsed since the onset of their illness. All patients were receiving standard drug therapy and individual psychotherapy from the attending physician.

Among the 50 relatives who consented to participate, 46 relatives who attended at least three of the four psychoeducational sessions were included as subjects. The average age of the 46 relatives was 52.7 years. A total of 32 were the mother of the patient, nine were the father, one was a mother-in-law, two were siblings, and two were the patient’s spouse. Therefore, most were the mothers of the patients, and their mean age was not overly high.

Psychoeducational intervention

Multi-family sessions for small groups of between two and six people were performed for a total of 13 groups between May 1998 and December 1999 at Nihon University Itabashi Hospital and between June 2001 and December 2001 at Ome Municipal General Hospital. As a rule, each group underwent four sessions in 2 months. Each session lasted approximately 2 h.

The first half of each session consisted of a lecture with some interaction with attendees. The second half consisted of a discussion. Each session was attended by two or three doctors and a nurse, who provided information and instruction to the attendees. A social worker was responsible for explaining the social support programs. Each lecture consisted of the following contents: orientation, information about schizophrenia (outline, signs and symptoms, prognosis and relapse, cause of the illness, and treatment), rehabilitation, social support programs, and management of patient behaviors. The discussions were held for as long as possible in order to take advantage of the small size of the groups. While discussions about patient behavior management after discharge and problem solving were included in the discussion sessions, communication skills training was not undertaken due to time constraints.

Evaluation of the intervention

Before and after psychoeducation, the relatives of the patients were interviewed and asked to complete a self-administered rating scale. Rather than objectively measuring variables such as knowledge of the illness, this investigation examined changes in subjective problems that affected the relatives, including anxiety, confusion, depression, and burden. Anxiety was assessed using the Japanese-language version of the STAI and an independently developed scale for assessing factors such as the levels of distress and burden of the illness on the relatives. The STAI measures anxiety and distinguishes between trait anxiety (A-trait), which indicates the extent to which that individual is predisposed to anxiety, and state anxiety (A-state), which indicates the level of anxiety experienced by the individual at a given time.6 The reliability and validity of the Japanese-language version of the STAI, prepared by Nakazato and Mizuguchi, has been verified.7 The Family Burden and Distress Scale was a newly developed tool for assessing various problems resulting from the patient’s illness, including anxiety, depression, confusion, and the burden on the family. This scale consists of 24 questions, and a factor analysis has shown that the scale has three subscales. The first factor, which consists of 10 items, is subjective burden and depression resulting from the patient’s illness. The second factor, which consists of nine items, is confusion resulting from a lack of knowledge of the illness and anxiety about the future. The third factor, which consists of five items, is difficulties in the relationship with the patient. The reliability and validity of each subscale have been verified (Cronbach’s alpha coefficients were 0.79–0.91, Pearson’s correlation coefficients were 0.78–0.91 as assessed by the test–retest method).8 Each item was assessed using a 4-step Likert scale similar to the STAI, with high scores indicating a high level of family distress. To examine whether the effectiveness of the intervention varied according to a high or low EE level, patients’ relatives were interviewed before the session using the FMSS. The FMSS is a method of measuring EE developed at University of California, Los Angels, and proposed by Magana et al.9 The individuals being assessed were asked to speak freely for 5 min about what kind of person the patient was and how the family and patient got along together. The family was then assessed as high EE, borderline high EE or, by elimination, low EE. Two Japanese-speaking assessors certified by UCLA observed and assessed the family members using FMSS criteria. If the two assessments differed, the assessors conferred to arrive at a final assessment. Changes observed in the assessments obtained before and after intervention were tested for significance using the Wilcoxon signed-ranks test, and differences according to EE type were tested using the Mann–Whitney U-test.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES

Expressed emotion assessment

The EE assessment of the 46 participating relatives, using the FMSS, classified 10 relatives as high EE, five as borderline-high EE, and the remaining 31 as low EE.

Anxiety and burden and distress levels in families before intervention

State anxiety and trait anxiety scores as measured by the STAI and the Family Burden and Distress Scale before and after the intervention are shown in Table 1. Before intervention, the level of state anxiety was very high at 54.2, and the trait anxiety level was also high at 48.7. On the Family Burden and Distress Scale, high scores were obtained for confusion resulting from a lack of knowledge of the illness and anxiety about the future. The proportion of relatives for whom a score of 2.5 points or higher was obtained was 84.8%.

Table 1.  Scores on the State-Trait Anxiety Inventory and Family Burden and Distress Scale before and after family intervention and change in scores
  STAIFamily Burden and Distress Scale
S-anxietyT-anxietyBurdenConfusionRelationship§
  •  Subjective burden and depression resulting from the patient’s illness.

  •  Confusion resulting from a lack of knowledge of the illness and anxiety about the future.

  • §

     Difficulties in the relationship with the patient.

  • STAI, State-Trait Anxiety Inventory.

Total n = 46Pre 54.2 48.7  2.19  3.07  2.02
Post 47.2 43.9  1.85  1.92  1.87
Pre-post  7.0  4.8  0.34  1.15  0.15
P<0.0001<0.001<0.001<0.0001<0.05

Changes in anxiety and burden and distress levels after intervention

Compared with the STAI scores obtained before intervention, scores after intervention demonstrated a significant decrease of 7.0 points for state anxiety and a significant decrease of 4.8 points for trait anxiety. On the Family Burden and Distress Scale, scores for subjective burden and depression resulting from the patient’s illness, confusion resulting from a lack of knowledge of the illness and anxiety about the future, and difficulties in the relationship with the patient showed a significant decrease. Therefore, significant differences were seen in the scores for all items.

Differences in effectiveness according to expressed emotion level

The effectiveness of psychoeducation according to EE level was then examined. Based on previous studies indicating that borderline high EE ought to be considered high EE,10,11 because the FMSS produces a high number of false-negatives in Japan, the analysis for the present study divided relatives into two groups: the high EE group combined the high EE and borderline high EE relatives, and the low EE group comprised the remaining relatives. The changes in the scores for the high and low EE groups, before and after intervention, are shown in Fig. 1. The significance of the assessments pre and post intervention and the decrease for each item were assessed using the Mann–Whitney U-test (Table 2).

image

Figure 1. Change in scores on the State-Trait Anxiety Inventory and Family Burden and Distress Scale after family intervention according to expressed emotion level. (+) P < 0.1 * P < 0.05 ** P < 0.01 *** P < 0.001.

Download figure to PowerPoint

Table 2.  High expressed emotion versus low expressed emotion (Mann–Whitney U-test)
  Average rankP
High EE (n = 15)Low EE (n = 31)
  • *

     P < 0.05,

  • **

     P < 0.1.

  •  Subjective burden and depression resulting from the patient’s illness.

  •  Confusion resulting from a lack of knowledge of the illness and anxiety about the future.

  • §

     Difficulties in the relationship with the patient.

  • EE, expressed emotion.

PreState-anxiety26.422.10.308
Trait-anxiety26.522.10.297
Burden28.221.20.096**
Confusion24.822.90.648
Relationship§28.621.00.069**
PostState-anxiety23.823.40.916
Trait-anxiety25.022.80.590
Burden23.023.80.851
Confusion22.424.00.700
Relationship§27.321.70.183
Pre-postState-anxiety19.425.50.153
Trait-anxiety19.525.40.163
Burden16.526.90.014*
Confusion21.324.60.430
Relationship§22.324.10.660

For all items, scores for the indices before intervention were higher in the high EE group (high anxiety and distress level) than in the low EE group. With regard to subjective burden and depression resulting from the patient’s illness and difficulties in the relationship with the patient, the high EE group showed a trend toward higher levels than the low EE group (P < 0.1, Table 2).

When the scores for pre and post intervention were compared, the high EE group showed a greater overall improvement than the low EE group. On both the STAI and the Family Burden and Distress Scale, confusion resulting from a lack of knowledge about the illness and anxiety about the future decreased in a comparable manner. However, in terms of the other items, the postintervention scores for the high EE group improved to a point approximately equal to those for the low EE group (Fig. 1). The changes for pre to post intervention in the STAI score and burden/distress levels were examined in the low EE group and the high EE group using the Wilcoxon signed-ranks test. The high EE group showed significant decreases in all indices. In contrast, non-significant decreases were seen for subjective burden and depression resulting from the patient’s illness and difficulties in the relationship with the patient, even though the low EE group had significant decreases in state and trait anxiety, confusion resulting from a lack of knowledge of the illness and anxiety about the future. Therefore, differences between the high and low EE groups were observed. In particular, the change in the subjective burden and depression resulting from the patient’s illness were significantly greater in the high EE group than in the low EE group, as indicated by the Mann–Whitney U-test (Table 2). These findings indicated that there were larger changes in emotion-related factors, such as burden and depression, in the high EE group than in the low EE group.

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES

Effectiveness of short-term psychoeducation

Pitschel-Walz et al. compared the effect of short-term and long-term intervention on the relapse rate and hospitalization rate and found long-term intervention to be superior based on the mean effect size. However, they also found short-term intervention to be significantly more effective than usual treatment in preventing relapse.1 As noted above, Linszen et al. reported similar relapse rates with individual psychoeducational therapy for patients with schizophrenia of recent onset during hospitalization and two sessions of family psychoeducation performed according to the method of Anderson, as compared with the case when family psychoeducation was subsequently continued while the patient was treated as an outpatient.2 Because short-term intensive psychoeducation does not place a heavy burden on the participants or treatment staff, it makes it relatively easy to introduce psychoeducation for family members of patients with schizophrenia on a widespread basis in the clinical setting. As was indicated by Linszen et al. it can be expected to provide efficacy over the short-term, particularly when the patient is hospitalized soon after the onset of the illness. The study by Linszen et al. did not examine the direct effects on family members other than on the relapse rate. The following discussion compares the results of studies of the other individual effects of short-term psychoeducation with the results of the present study.

There have been numerous studies concerning increasing knowledge of schizophrenia, and the results have generally been positive. For example, Barrowclough et al. reported that two educational sessions for patients in the acute inpatient ward significantly increased scores on the Knowledge About Schizophrenia Interview.12 Because the primary purpose of the present study was to examine changes in the subjective difficulties experienced by family members, such as anxiety and distress, changes in objective measures of the knowledge level were not examined. However, the scores for confusion subjectively experienced by the family as a result of a lack of knowledge of the illness and anxiety about the future showed the largest reductions after psychoeducational intervention, decreasing significantly from 3.07 points to 1.92 points (Table 1). This finding is consistent with previous studies and confirmed the effectiveness of this approach. Moreover, there was no significant difference in the effectiveness of this point between the high and low EE groups, with the effectiveness being equal in these groups.

In addition, several studies have examined the effects of short-term psychoeducation with respect to factors other than knowledge. Cozolino et al. reported a significantly increased sense of support from the treatment team and a nearly significant tendency toward a decrease in self-blame.13 Solomon et al. reported an increase in self-efficacy regarding mentally ill relatives who had never participated in a support or advocacy group.14 Smith and Birchwood, and Birchwood et al. reported an increase in optimism concerning the family’s role in treatment.15,16

In contrast, few studies focusing on the distress associated with occurrence of the patient’s illness have examined the effectiveness of short-term psychoeducation on anxiety and distress in the family. Smith and Birchwood, and Birchwood et al. performed four educational sessions for the relatives of patients with schizophrenia, including some inpatients in the acute phase (same study as mentioned above) and found significant reductions in relatives’ reported stress symptoms and fear of the patient.15,16 Abramowitz and Coursey performed six sessions of psychoeducation at a community mental health center and reported significant reductions in anxiety and personal distress and significantly more active coping behaviors by the participants.17 The present study examined a variety of factors in addition to the anxiety experienced by relatives of patients during hospitalization. As a result, significant decreases were seen in factors such as anxiety resulting from the occurrence of the patient’s illness and in the distress and burden on the family. As families’ anxiety and confusion resulting from the onset of symptoms or relapse of a family member are at their highest levels during the hospitalization of a family member, it is necessary for all families to receive psychoeducation, not only to increase knowledge of the illness, but also to reduce the anxiety about the illness and the distress caused by the patient’s illness. It is particularly important for relatives of patients with recent onset to receive early psychoeducational interventions, which introduce long-term strategies for coping with patients. Family psychoeducation is important for reducing such anxiety and burden in both high and low EE families. In the present study, this direct effect was confirmed to occur even with short-term psychoeducation. This result can be considered evidence of the importance of introducing psychoeducation into standard therapy.

Difference in effectiveness of psychoeducation according to expressed emotion level

Smith et al. reported that although they found no significant difference in knowledge of schizophrenia between high EE and low EE families, the high EE families experienced a greater subjective burden and perceived themselves as coping less effectively.18 In the present study, no significant differences were observed between high and low EE families with the Mann–Whitney U-test. However, regarding subjective burden and depression resulting from the patient’s illness and difficulties in the relationship with the patient, the high EE group showed a trend toward higher levels than the low EE group. These findings were concordant with those previously reported. The absence of a significant difference may have been related to the small sample size, and further examination is, therefore, needed.

Few previous studies have examined whether the improvements in the anxiety and burden felt by the family as a result of short-term psychoeducation vary according to EE level. Cozolino et al. performed a single session for relatives of patients with schizophrenia that had initially occurred within the previous 2 years and found that although low EE families showed no significant change in attitude, high EE families displayed a significantly increased sense of understanding of the illness and increased feelings of support from the treatment team.13 In the present study, using the Wilcoxon rank-sum test, significant improvements in all factors were observed for the high EE group. In contrast, improvements were not significant in any factor for the low EE group. Moreover, the Mann–Whitney U-test showed significant differences between the high and low EE groups with respect to subjective burden and depression resulting from the patient’s illness. This resulted from the fact that, as is shown in Fig. 1, except for confusion resulting from a lack of knowledge of the disease, the STAI score and burden/distress levels showed an initial trend toward being higher in the high EE group than in the low EE group, and the score in the high EE group subsequently improved to roughly the same level as in the low EE group after psychoeducation. Specifically, the high EE group responded more sensitively to the emotional factors associated with psychoeducation, therefore, increasing the effectiveness of the intervention. The greater change seen in the high EE group with even short-term education appears to be one reason why psychoeducation produces large long-term effects, such as a reduced relapse rate, in high EE families. In providing psychoeducation, group work with this sort of focus on emotional factors is effective in reducing anxiety and distress in high EE families. This is an important point to keep in mind when implementing psychoeducation. Discussions among a small number of participants, in the group psychoeducation format used here, were effective in reducing anxiety and distress experienced by the family members.

Future topics

This study showed that the direct effects of psychoeducation may extend not only to improvements in the knowledge level of the family members but also to emotional factors such as anxiety and distress. Moreover, the results suggested that these effects, while present in the low EE group, are greater in the high EE group. There have been few reports of similar results, and it will be necessary to conduct numerous studies, including controlled studies, in order to verify these findings. In addition, the long-term effectiveness of such short-term interventions, including changes in EE levels, should be verified. As was mentioned earlier, a uniform effect was seen with short-term psychoeducation, but it cannot be concluded that this alone is sufficient support for the family. Birchwood et al. reported that improvements in knowledge and anxiety scores were maintained during six months of follow up, while the decrease in the fear and overall stress perceived by the family attenuated after the 6-month follow-up period.16 It will be necessary to examine continuous intervention that enhances short-term psychoeducational intervention. For example, for groups at a high risk for relapse, such as high EE and high-stress groups, continuous intervention will be examined, by introducing methods of improving the communication skills of family members with the aim of reducing the rate of relapse, after providing short-term psychoeducational intervention. It will also be necessary to further investigate what the approximate duration and frequency of this type of family intervention should be and the types of intervention that should be undertaken.

ACKNOWLEDGMENTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES

The authors would like to thank: Dr Hiromi Watanabe, Tokyo Metropolitan Umegaoka Hospital, the late Dr Kosei Kido, Nihon University School of Medicine, and the staff of Nihon University Itabashi Hospital and Ome Municipal General Hospital for their cooperation and encouragement.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES
  • 1
    Pitschel-Walz G, Leucht S, Bauml J, Kissling W, Engel RR. The effect of family interventions on relapse and rehospitalization in schizophrenia – a meta-analysis. Schizophr. Bull. 2001; 27: 7392.
  • 2
    Linszen D, Dingemans P, Van der Does JW et al. Treatment, expressed emotion and relapse in recent onset schizophrenic disorders. Psychol. Med. 1996; 26: 333342.
  • 3
    Vaughn CE, Leff JP. The measurement of expressed emotion in the families of psychiatric patients. Br. J. Soc. Clin. Psychol 1976; 15: 157165.
  • 4
    Mino Y, Inoue S, Tanaka S, Tsuda T. Expressed emotion among families and course of schizophrenia in Japan: A 2-year cohort study. Schizophr. Res. 1997; 24: 333339.
  • 5
    American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, 3rd edn revised. Cambridge University Press, Washington, DC, 1987.
  • 6
    Spielberger CD, Gorsuch RL, Lushene RE. Manual for the State-Trait Anxiety Inventory. Consulting Psychologists Press, Palo Alto, 1970.
  • 7
    Nakazato K, Mizuguchi T. Development and validation of Japanese version of State-Trait Anxiety Inventory. Shinshin Igaku 1982; 22: 107112 (in Japanese).
  • 8
    Yamaguchi H, Takahashi A, Shiraishi H, Takano A, Kojima T. Development of the Family Burden and Distress Scale and an investigation of its reliability and validity. Jpn. J. Clin. Psychiatry 2006; 35: 449456 (in Japanese).
  • 9
    Magana AB, Goldstein JM, Karno M, Miklowitz DJ, Jenkins J, Falloon IR. A brief method for assessing EE in relatives of psychiatric patients. Psychiatry Res. 1986; 17: 203212.
  • 10
    Uehara T, Yokoyama T, Goto M et al. Expressed emotion from the five-minute speech sample and relapse of out-patients with schizophrenia. Acta Psychiatr. Scand. 1997; 95: 454456.
  • 11
    Shimodera S, Mino Y, Inoue S, Izumoto Y, Kishi Y, Tanaka S. Validity of a Five-Minute Speech Sample in measuring expressed emotion in the families of patients with schizophrenia in Japan. Compr. Psychiatry 1999; 40: 372376.
  • 12
    Barrowclough C, Tarrier N, Watts S, Vaughn C, Bamrah JS, Freeman HL. Assessing the functional value of relatives’ knowledge schizophrenia: A preliminary report. Br. J. Psychiatry 1987; 151: 18.
  • 13
    Cozolino LJ, Goldstein MJ, Nuechterlein KH, West KL, Snyder KS. The impact of education about schizophrenia on relatives varying in expressed emotion. Schizophr. Bull. 1988; 14: 675687.
  • 14
    Solomon P, Draine J, Mannion E, Meisel M. Impact of brief family psychoeducation on self-efficacy. Schizophr. Bull. 1996; 22: 4150.
  • 15
    Smith J, Birchwood M. Specific and nonspecific educational intervention with families living with a schizophrenic relative. Br. J. Psychiatry 1987; 150: 645652.
  • 16
    Birchwood M, Smith J, Cochrane R. Specific and non-specific effects of educational intervention for families living with schizophrenia. A comparison of three methods. Br. J. Psychiatry 1992; 160: 806814.
  • 17
    Abramowitz I, Coursey R. Impact of an educational support group on family participants who take care of their schizophrenic relatives. J. Consult. Clin. Psychol. 1989; 57: 232236.
  • 18
    Smith J, Birchwood M, Cochrane R, George S. The needs of high and low expressed emotion families: A normative approach. Soc. Psychiatry Psychiatr. Epidemiol. 1993; 28: 1116.