Cognitive-behavioral therapy has been recommended for patients with persistent symptoms of schizophrenia by guidelines from the American Psychiatric Association (
1 ) and the Schizophrenia Patient Outcomes Research Team (
2 ). In the United Kingdom the National Institute of Clinical Excellence (
3 ) issued guidelines stating that cognitive-behavioral therapy should be "offered to any individual with schizophrenia." Meta-analyses of randomized controlled studies have repeatedly attested to its efficacy (
4,
5,
6 ), and another study has recently attested to its cost-effectiveness (
7 ).
Cognitive-behavioral therapy is an individualized intervention that encourages patients and caregivers to take an active role in treatment. Drop-out rates in studies on cognitive-behavioral therapy are low (12 to 15 percent) (
8 ). Cognitive-behavioral therapy uses "guided discovery," enabling the individual to gain insight into his or her symptoms, and systematically assists in developing more appropriate attributions and coping strategies for various symptoms, such as hearing voices and having paranoid beliefs. This in turn improves attitudes toward treatment (
9 ) and medication adherence (
10 ) and reduces relapse rates and rehospitalization (
7,
11 ). In trials so far, cognitive-behavioral therapy appears to be a safe treatment and has not been shown to bring about any rise in suicidal ideation, agitation, or violence (
12 ). Treatment manuals are available that fully describe its use (
13 ).
Availability of therapists skilled in cognitive-behavioral therapy is very limited. However, training courses with continuing supervision have made these therapists increasingly available in the United Kingdom, especially in centers where research on cognitive-behavioral therapy in psychosis has been most active. In a few areas, such as West Southampton (total population of 56,500) in southern England, this has meant that cognitive-behavioral therapy has been readily available for five to ten years for patients with schizophrenia who are referred. However, even where this has been the case, it has been apparent that some patients are still not being referred for cognitive-behavioral therapy. This audit aimed to establish, first, the proportion of people with a diagnosis of schizophrenia, schizoaffective disorder, or delusional disorder in West Southampton who had not been referred for cognitive-behavioral therapy and, second, the possible reasons for nonreferral.
Methods
This study is a retrospective analysis of reasons behind nonreferral for cognitive-behavioral therapy for psychosis by psychiatrists working in an urban area. All of the patients aged 18 to 64 years under the care of the community mental health team for the West Southampton area on a census date (November 25, 2004) with a previous inpatient diagnosis of schizophrenia, schizoaffective disorder, or delusional disorders were identified. Each of these patients had therefore had an admission at some time over the previous six years to Royal South Hants Hospital under one of the four consultant psychiatrists serving the area who would have been involved in the development of their care plan. The psychiatrists would usually have seen the patients in an outpatient clinic after the patients were discharged. Some patients with schizophrenia in the area would have been in contact with primary health care services but would not have been in contact with mental health services, so this audit underestimates the proportion of patients who could potentially have benefited from referral to cognitive-behavioral therapy. However, very few people with schizophrenia would have been receiving services from other, non-National Health Service providers of mental health services, because very few such providers exist within the U.K. health system. Therefore, the sample in our study is composed of people with schizophrenia who are more severely affected and who would benefit most from cognitive-behavioral therapy (
13 ).
This list was cross-referenced with psychology referral records since 1996 obtained from the Hampshire Partnership National Health Service Trust (the mental health service for that area). If a patient with one of the above diagnoses was referred to the hospital's Department of Psychology so he or she could see a psychologist, the department would have recommended cognitive-behavioral therapy for psychosis symptoms but might have also focused on depression or trauma work. Eight patients in our sample had also been included in a study of brief cognitive-behavioral therapy (
14 ) or had been seen by psychiatrists or nurses training or with an interest in cognitive-behavioral therapy for psychosis.
To determine the reasons for nonreferral, four adult psychiatric consultants (one of whom was also a rehabilitation specialist) were given a semistructured interview by the medical student researcher (HK) in the hospital or mental health center where the community team is based. They were asked to identify reasons for not referring particular patients for cognitive-behavioral therapy. This study was approved by the Clinical Audit Department of the Hampshire Partnership Trust.
Results
A total of 142 patients with schizophrenia, schizoaffective disorder, or delusional disorders were identified (88 males, or 62 percent, and 54 females, or 38 percent). A total of 69 patients (49 percent) had been referred for cognitive-behavioral therapy (46 males, or 52 percent, and 23 females, or 43 percent). Seventy-three (51 percent) did not receive cognitive-behavioral therapy from nurses, psychologists, or psychiatrists (42 males, or 48 percent, and 31 females, or 57 percent). The four psychiatrists provided information on all but four of the 73 patients who had not received cognitive-behavioral therapy; information on these four was not available because the cases were not familiar to the relevant consultant and notes on the cases could not be found. Patients who were not referred had a mean±SD age of 47±11 years, and those who were referred were 39±10 years (t=-3.98, df=140, p<.001). Referral rates for each of the consultants were very similar, ranging between 44 and 51 percent.
Table 1 shows the frequency with which each reason for nonreferral was reported (two reasons were given for 27 of the patients). Seven of 42 male patients (17 percent) who were not referred for cognitive-behavioral therapy were not referred because they refused the offer of therapy, whereas only three of 31 female patients (10 percent) were not referred for such a reason. The most commonly identified reasons for non-referral were that the psychiatrist believed that the patient would not engage in therapy (12 of 42 males, or 28 percent, and 13 of 31 females, or 42 percent) or the psychiatrist considered the patient to be doing well and cognitive-behavioral therapy was not required. However, in these circumstances the patient still remained under the supervision of mental health services because of continuing needs for supervision and support (14 of 42 males, or 33 percent, and nine of 31 females, or 29 percent).
Discussion
Overall, 49 percent of patients with schizophrenia were referred for cognitive-behavioral therapy; patients who were significantly younger tended to be referred. This is confirmatory evidence that in areas where cognitive-behavioral therapy is readily available, psychiatrists do see the therapy as relevant to patients' needs, accept the published evidence of efficacy, and refer many of their patients. (A similar survey in the two other areas of Southampton, Central and East, where cognitive-behavioral therapy was relatively less available, found referral rates of 35 and 17 percent, respectively.) The psychiatrists in our sample ranged from being enthusiasts to relative skeptics of cognitive-behavioral therapy, but examination of the data showed that in practice all referred patients in similar proportions.
Reasons for not referring included feeling as though patients were not suitable for therapy because of the belief that they would fail to engage. Such patients may deserve initial assessment for cognitive-behavioral therapy, even it is subsequently decided that a full course of therapy is not indicated at that time. Some were not referred for cognitive-behavioral therapy because they were believed to be doing well and not in need of therapy. Cognitive-behavioral therapy can assist with adherence issues, contributes to relapse prevention, and contributes to improved social recovery; thus this type of therapy may also be relevant for those who are believed to be doing well.
Of the remaining, less frequently reported, reasons for nonreferral, a small number refused an offer of therapy. Even so, cognitive-behavioral therapy is a future treatment option, if the patient agrees to therapy later, although for some this may never be an option they wish to take up or indeed need. Substance abuse was not commonly given as the reason for nonreferral, but the group of patients who were thought not likely to engage may have included some who abused substances. However, substance abuse is not a contraindication, because it has been shown that cognitive-behavioral therapy for psychosis with motivational interviewing can bring about significant improvements in illness and service use among patients with comorbid substance use disorders (
15 ).
A significant limitation of the study was that sampling the attitudes of only four care providers can provide only very restricted information about general attitudes among mental health workers about cognitive behavioral therapy for psychosis.
Conclusions
In this study, in an area where cognitive-behavioral therapy for people with schizophrenia was readily available from skilled therapists, half of the patients eligible for this therapy were referred by their psychiatrist. Only a small number of patients who were referred refused treatment. Among those who were not referred, a substantial number were considered too difficult to engage or, paradoxically, too well. Cognitive-behavioral therapy was seen as a valuable additional option for this group of patients, and as services become more widely available, flexible, and responsive, many more patients will be able to benefit from this evidence-based intervention.