Treatment and Management of Schizophrenia

Treatment and management of schizophrenia.

Topics covered: 

  • Introduction
  • Evidence for efficacy
    • Drug treatment
    • ‘Positive' features
    • ‘Negative' features
    • Cognitive performance
    • Non-specific symptomatology
    • Behaviour
    • Maintenance
    • Treatment resistance
    • Psychosocial interventions
    • Psychodynamic psychotherapy
    • Social skills training and illness self-management programmes
    • Family interventions
    • Cognitive-behavioural therapy
    • Alternatives to hospitalization
  • Principles of treatment and management
    • Preliminaries Identifying goals and defining structure
    • The acute phase
    • Low-potency regimens
    • High-potency regimens
    • The post-acute phase
    • The maintenance phase
    • Psychiatric emergencies
    • Poor response and treatment resistance
    • Antipsychotics
    • First-line adjunctive medications
    • Second-line adjunctive treatments
    • Psychosocial interventions in management
  • Concluding remarks 
  • References

Introduction

Historically, therapeutic interventions in patients with psychotic disorders were largely palliative and predicated on essentially psychosocial principles, (1) although the administration of medicines and other physical techniques has been recommended since antiquity (2) and actively employed in Britain for some 300 years. (3)

Over the last 50 to 60 years, a number of physical treatments of schizophrenia have aimed at promoting more specific benefits. Insulin coma, electroconvulsive therapy (ECT), and prefrontal leucotomy (4) became widely applied on the basis of enthusiasm rather than scientific study, but since the development of safe and effective medications they have largely fallen into disuse. Since chlorpromazine was introduced in 1952 and shown to have specific antipsychotic properties, (5) numerous antipsychotic agents have come into use, which, along with a range of psychosocial interventions, have been favourably evaluated.

There remains an obvious discrepancy between the generally favourable results of, especially drug, trials in this field and an on-going concern about the difficulties posed by schizophrenia. Some of this can be explained by the difference between responses evident in clinical trials (efficacy) and those that pertain in ordinary circumstances (effectiveness). The unique conditions of the controlled trial provide essential information on which to base rational therapeutic decisions, but not necessarily information that translates to all clinical contexts.

Evidence for efficacy

Drug treatment

Within 15 years of the introduction of chlorpromazine, the efficacy of antipsychotics in schizophrenia became one of the most comprehensively researched areas in psychiatry. (6) In addressing drug efficacy in schizophrenia, however, several different issues need to be considered. (7) These include efficacy in relation to the following:

  • ‘positive' features
  • ‘negative' features
  • cognitive performance
  • non-specific (affective) symptomatology
  • behaviour
  • maintenance
  • treatment resistance.

‘Positive' features

That standard antipsychotic drugs promote resolution of positive symptomatology characterizing acute schizophrenic episodes cannot be doubted. In a review covering the first two decades of antipsychotic use, Davis and Garver (8) found that 86 per cent of controlled studies showed chlorpromazine to be superior to placebo. They also noted that all 26 trials that had used more than 500 mg/day reported chlorpromazine to have definitely greater efficacy than placebo, while in this dose range no trials found only marginal or no benefits.

The question of ‘acute' efficacy is now settled to a sufficient degree that further attempts at replication would probably be unethical. Furthermore, this effect, or magnitude of effect, is not shared by other psychotropic agents and can be considered a class effect and a valid basis for classification.

This is not to say that antipsychotics are entirely satisfactory. While these drugs can be dramatically effective in individual cases, overall their efficacy in acute states is only partial. The major first-generation efficacy study, (9) which looked at treatment responses to a phenothiazine of each chemical type (i.e. chlorpromazine, thioridazine, and fluphenazine) in schizophrenic inpatients, showed three things: comparable efficacy for the three active drugs, considerable residual group symptomatology, and a small, but definite, response in the placebo group. Overall, 61 per cent of those who received one of the active drugs were considered ‘improved' to ‘very much improved'. As a result, it has since been taken that standard antipsychotics produce a satisfactory clinical response in around 60 to 70 per cent of patients with acute schizophrenia, while approximately 6 to 8 per cent do not respond at all. (10) This interpretation is based on exclusion of the placebo response rate which, when taken into account, suggests that significant benefits could be more realistically expected in 40 to 50 per cent of patients. There thus remains substantial room for improvement, even on this primary measure of efficacy. 

With one exception, all antipsychotic drugs (standard or new generation) are of equal efficacy on positive schizophrenic symptomatology. The exception is clozapine in treatment-resistant cases, whose advantages may relate only to this specific clinical context. (7)

Despite their limitations, antipsychotics are unquestionably the most satisfactory treatment modality for acute episodes. In a unique study, May (11) compared response rates and outcomes in a large group of schizophrenic patients randomly assigned to one of five treatment regimens: individual psychotherapy alone, antipsychotics alone, individual psychotherapy plus antipsychotics, ECT, and ‘milieu' therapy (i.e. a ward environment). Patients who received physical treatments clearly did better in terms of increased rates of discharge, reduced lengths of stay, and decreased need for additional treatments. Two years after discharge, twice as many of those treated with antipsychotics as with psychotherapy were in employment, (12) while in the 3 years post-discharge, the drug-treated patients spent less time back in hospital. (13)

Data from this trial suggested that those not subjected to antipsychotic treatment early on fared less well in terms of long-term outcome than those who had received it in the initial phase of the investigation. This topic has assumed considerable prominence in recent years and the impression is that early intervention is associated with a better outcome than a policy of ‘wait and see'. (14) The implication is that antipsychotic drugs achieve their best results when administered as soon after the onset of illness as possible. How long is ‘soon' remains unclear, although there is a suggestion that those with first episodes in whom exposure is delayed for longer than 12 months may have a worse long-term outcome than those coming to treatment earlier. (15) This has raised concerns that even brief periods of abstinence, as might be associated with participation in placebo trials, may be detrimental, but available evidence does not support this. (16)

In recent years, a new generation of antipsychotics has been developed, with clozapine the instrument of change. Pharmacologists first applied the term ‘atypical' to the benzamides following the widespread launch of sulpiride in the late 1970s, but it was unclear whether the clinical profiles of these drugs were sufficiently different to merit the designation, especially with dose-equivalent usage. (7) Clozapine is unquestionably different and was the first drug that could be legitimately called ‘atypical' on clinical grounds. However, a general definition of ‘atypicality' remains elusive. Criteria proposed by Lieberman (17) include the following:

  1. enhanced efficacy (i.e. in treatment-resistant cases);
  2. no, or only slight and ill-sustained, elevations of prolactin levels;
  3. most importantly of all, strikingly reduced liability to promote extrapyramidal dysfunction.

The development of new-generation antipsychotics has resulted in a revival of basic and comparative efficacy studies. Results from studies of risperidone, (18) olanzapine, (19) quetiapine, (20) amisulpride, (21) and zotepine (22) (at the time of writing, the compounds that have been commercially launched) have shown each of these drugs to possess antipsychotic efficacy which, on present evidence, appears comparable to each other and a standard comparitor. (Sertindole, an antipsychotic of proven efficacy, was withdrawn in December 1998 amid continuing concerns about its cardiac actions.)

Where the new-generation drugs appear to have advantage is in neurological tolerability, (5) although it is as yet hard to quantify the extent of this. New-generation drugs seem to have a reduced liability to promote acute dystonias, and are associated with lower levels of parkinsonism and akathisia. While it would seem reasonable to conclude that these benefits will translate into reduced rates of tardive motor disorders also, for which some supportive data have been presented, (23,24) firm evidence for this remains to be accrued.

‘Negative' features

The issue of whether antipsychotics exert therapeutic efficacy on negative schizophrenic features became of importance largely because of the implications of Crow's type I/type II hypothesis. (25) The inference of this is that drugs exerting beneficial effects on positive features, i.e. antipsychotics, would be unlikely to exert similar therapeutic effects on negative features, owing to the differing pathological substrates proposed for the two states—a structural one for negative states, and a functional (i.e. dopaminergic) disruption for positive symptomatology. Although this hypothesis was originally formulated when only standard drugs were available, the issues it raises have become even more pertinent with the introduction of new-generation compounds.

One of the strongest challenges to Crow's hypothesis came from Goldberg. (26) Reviewing five of the early placebo-controlled American efficacy studies, he clearly demonstrated that what he defined as negative features did respond to antipsychotic treatment. Other investigators, using more modern rating instruments, have also demonstrated similar beneficial effects. However, such findings are not unanimous and, importantly, in some studies, ratings of negative features have been found to covary with positive ratings, (27) raising the question of whether negative features are in fact state dependent. (28)

The key factor is the lack of consensus about what comprises negativity, which makes it difficult to evaluate drug efficacy. Historically, the problem of ‘negativity' in schizophrenia was defined longitudinally and primarily in terms of psychosocial functioning. It is the cross-sectional requirements of psychopharmacological research that have raised the conceptual difficulties. Carpenter and colleagues have emphasized the varied clinical states that can present similarly and the need to bear in mind a differential diagnosis (29). If one's concept of negativity encompasses the withdrawal and reserve of those preoccupied with hallucinations or delusions, then negative features are likely to resolve as part of the symptomatic response to antipsychotics—and standardized ratings of each domain will covary. While this is clearly a therapeutic action, it does not address the fundamental issue.

A major confound in assessing trial data in this field is the bradykinesia of drug-related parkinsonism. After nearly half a century, the boundaries of the parkinsonian syndrome associated with antipsychotic drug use remain undefined, particularly in relation to subjective symptomatology. (5) Improvements in negativity on changing from one drug to another may say more about differing liabilities to promote extrapyramidal disorder than about actions on primary negative features.

This issue is of crucial importance in evaluating claims made for new-generation drugs, where improvements in negative features are likely to reflect tolerability rather than efficacy. Sophisticated statistical applications, such as path analysis, which can be used to substantiate a therapeutic action for some new antipsychotics (30,31) do not address the issue, as the problem lies at the clinical, and not the statistical, level. The evidence to date is that even the undoubted benefits of clozapine are restricted to secondary negative symptomatology and reflect that compound's unique neurological tolerability. (32)

Carpenter et al. (33) have also emphasized ‘durability' as a defining criterion of the ‘deficit' syndrome. By reintroducing longitudinal evaluation, this is much more akin to the historical concept of schizophrenic negativity. It is of interest that this group found the benefits of clozapine on negative features were confined to patients who did not conform to criteria for a ‘deficit' state. (34)

Crow's hypothesis has not yet been disproven, and it remains to be shown that antipsychotics exert any therapeutic benefits on primary negative schizophrenic symptomatology. This conclusion applies to new-generation compounds as much as to standard drugs.

Cognitive performance

Recently, attention has focused on the question of whether antipsychotic drugs in general possess, as one of their target actions, therapeutic efficacy on specific aspects of cognition and, increasingly, whether certain compounds exert preferential actions in this regard. (35)

Evaluation of the cognitive effects of antipsychotics faces major methodological problems and the contradictory nature of findings in this field have been highlighted in several reviews. (35,36 and 37) Standard antipsychotics have been reported to initially impair aspects of attention and motor behaviour which improve with continued exposure, while working memory and long-term recall do not appear to be fundamentally affected. (35) Beyond this, it is hard to extract a consensus— a conclusion that also applies to studies of new-generation drugs.

In terms of assessment measures, performance tests of executive function that are not timed tend to be insensitive to antipsychotic effects, while paced (i.e. timed) tests are affected.(37) Clearly, certain findings in this field may reflect neurological status.

Non-specific symptomatology

For many schizophrenic patients, affective symptomatology is a prominent part of the clinical picture (38) but the efficacy of antipsychotics on these features has been poorly researched. Their value in the treatment of anxiety experienced during acute episodes has not been systematically addressed and rests largely on clinical wisdom.

There is a long tradition, mainly in Continental Europe, of attributing antidepressant actions to antipsychotic drugs administered in low doses, at which they may exert preferential actions at presynaptic dopaminergic (autoreceptor) sites. (39) Trial evidence has focused on flupenthixol (40) and the L-enantiomer of sulpiride (41) and, more recently, amisulpride. It is unlikely that such an action would be relevant to long-term maintenance in schizophrenia, where the doses used are still considerably above those reportedly associated with antidepressant actions.

A greater concern is whether antipsychotics may contribute to depressive mood change. At one time they were implicated as being ‘depressogenic', particularly when administered in depot formulation. (42) This, again, raises the difficulty of distinguishing between similar presentations of pathophysiologically different states. Van Putten and May(43) described a state of dysphoric mood change resembling depression which they referred to as ‘akinetic depression'. As this state resolved following administration of an anticholinergic, it is hard to see it as other than a subjective manifestation of bradykinesia.

Depression is a common feature of schizophrenia prior to introduction of medication and it has been shown that depressed mood tends to resolve as positive psychotic symptomatology diminishes. (44) As with negative states, such findings probably reflect symptom covariation.

Behaviour

Efficacy studies of antipsychotics seldom consider behaviour as a separate dimension of disorder. It is clear, however, that particular, largely confrontational, behaviours such as hostility, belligerence, and resistiveness do improve with antipsychotic treatment. (6) This is usually taken as secondary to improvements in other domains, especially positive features.

However, behavioural disorganization in schizophrenia may not reflect a single dimension of disorder. In long-stay patients, behavioural disturbance appears to correlate with negative symptomatology and not positive.(45) Antipsychotics are thus unlikely to be efficacious in treating all behavioural disorders with which schizophrenic patients may present.

Maintenance

The long-term use of antipsychotic drugs in the treatment of schizophrenia is most frequently aimed at minimizing the risk of florid exacerbations of a disorder usually characterized by some persisting symptomatology. Hence, the term ‘maintenance' is preferable to ‘prophylaxis'.

The efficacy of antipsychotics in preventing relapse has also been extensively investigated and is beyond doubt. (46) Despite this, it remains difficult to quantify this effect, as the published literature presents figures that are widely discrepant. Nonetheless, reviewing the randomized, double-blind trials of antipsychotic versus placebo covering variable follow-up intervals, Janicak et al. concluded that on average 55 per cent of those on placebo relapsed compared with 21 per cent on active medication. (46) Statistically, these figures provide overwhelming support for the maintenance action of antipsychotics (1 × 10 –7).

Such comparisons are compelling, but studies addressing the question of long-term maintenance tend to be biased towards patients who have already shown a degree of response. There is reason to question whether the magnitude of the antipsychotic effect, at least over longer than 12 to 24 months, is as great in routine practice as trial-based analyses suggest. (47)

While most studies have concerned patients with established recurrent illness, there is additional evidence that long-term treatment is effective in preventing relapse after first episodes also. In the Northwick Park First Episodes Study, (15) significantly more patients in the placebo group than in the active treatment group relapsed over 2 years (62 per cent versus 46 per cent respectively: p < 0.002).

In understanding the benefits and principles of long-term antipsychotic administration, several issues need to be addressed:

  • the natural history of relapse
  • the history of relapse following successful maintenance/prophylaxis
  • practice choices
  • the pattern of intervention
  • adjunctive measures.
The natural history of relapse

Drawing from three large maintenance studies (48,49 and 50) an average constant rate of relapse can be calculated at 11.5 per cent per month. Thus, in any population of schizophrenic patients around 10 per cent will relapse each month after cessation of antipsychotic treatment.

Davis et al. (51) plotted relapse rates over a 2-year placebo follow-up period and showed that relapses tended to occur along an exponential function. This indicates a progressive and regular rate of relapse over time, and suggests that time to relapse may be a fixed aspect of the illness which is characteristic to each individual.

The history of relapse following successful maintenance/prophylaxis

A pertinent clinical question relates to how long medication should be continued in patients in whom stable maintenance has been achieved. Johnson (52) compared patients discontinuing antipsychotics with matched patients remaining on drugs, all of whom had been stable for one of three distinct periods: 1 to 2 years, 2 to 3 years, or 3 to 4 years. Relapse rates at 18 months were similar in all three groups who stopped, regardless of how long they had been stable beforehand (80 per cent, 90 per cent, and 70 per cent, respectively), and substantially higher than those who remained on treatment (35 per cent, 15 per cent, and 19 per cent, respectively). Others have come to the same conclusion using intervals of from 12 months to almost 8 years. (53,54 and 55) A similar picture of relapse has also been established in target populations selected on the specific basis of having responded particularly well to long-term medication and hence predicted as being of high likelihood to maintain well being. (56)

The implication seems to be that, no matter the duration or quality of well being on antipsychotic medication, relapse is inevitable following discontinuation in those with an established relapsing–remitting pattern of illness, with a time course that, in general, is consistent for each individual.

Practice choices

There is no evidence that any one drug is better than any other for long-term maintenance if one's ‘endpoint' is relapse, although there is persuasive, if largely anecdotal, evidence that in terms of quality-of-life parameters, clozapine offers superior benefits. (57)

There is powerful evidence that the choice of a depot formulation is associated with substantial improvements in relapse prevention. Support for the value of depot over oral administration comes largely form ‘mirror-image' studies, which compare a period on depot with a corresponding period prior to starting depot, in terms of, primarily, time spent in hospital. Six such studies were unanimous in showing a substantial reduction of time spent in hospital after the switch to depot (range 55–90 per cent; average 77.8 per cent). (58) There is no evidence that depots are themselves inherently more efficacious in maintenance. Their advantage seems to spring entirely from the enhanced compliance that ensues from their long-term use.

The other treatment practice variable is dosage. There is substantial evidence that long-term, low-dose antipsychotic regimens are associated with an increased risk of relapse. This has been investigated mainly in relation to depots, particularly fluphenazine decanoate. Kane et al.(59) found 1.25 to 5 mg every two weeks to be significantly less effective than 25 mg, a finding supported by other authors and with other preparations. (60,61,62 and 63)

Interpretation of such data is, nonetheless, not clear, and applying a dose–response model, Baldessarini et al. (47) have argued that in long-term treatment the half-maximal effective doses (ED50) may be as low as one-fifth to one-tenth those normally employed. Even the statistical findings of the studies above indicate that some individual patients may be successfully maintained on low-dose depot regimens. Kane and colleagues showed that while relapse rates were higher in the low-dose group, neurological adverse effects and psychosocial/quality-of-life parameters were superior. (59) Similarly, the Northwick Park First Episodes Study found that at endpoint, significantly fewer of those treated with active drug achieved some specific advance or attainment in their life than those on placebo. (64) This raises concern that maintenance antipsychotic treatment may have hidden costs.

Pattern of intervention

A further question is whether it is better to support long-term remission via a continuous regimen of administration or whether equal benefits can accrue from interventions targeted on early relapse, identified in prodromal symptomatology. The latter has strong intuitive appeal, especially to patients, but the trial evidence is, alas, not encouraging. None of the controlled studies to investigate the issue found advantage in targeted intervention, and in a meta-analysis of published data, Davis et al. (58) calculated that while relapse rates in those continuously treated were 25 per cent, the corresponding figure for those on targeted intervention was 50 per cent (p < 1 × 10–20). While this approach may have merit in carefully selected individuals, it cannot be recommended as a routine strategy. A further potential caution to the use of intermittent treatment regimens is the increasing concern that this pattern of drug utilization may increase the risk of tardive dyskinesia. (5)

Adjunctive measures

While antipsychotics can clearly protect against relapse, it has also been strongly advocated that additional benefits can be achieved when medication is combined with other forms of management, especially social/family interventions. This principle has considerable support in the literature, (58) although difficulties persist with most studies.

Treatment resistance

Treatment resistance came to prominence following publication of the United States' multicentre clozapine study, (65) one of the most influential studies of recent years. Within an operationally defined concept of treatment resistance, Kane and colleagues found that while 30 per cent of those on clozapine improved, only 4 per cent of those on the comparison regimen of chlorpromazine and benztropine improved ( p < 0.001). This was the first time that superior efficacy could be attributed to an antipsychotic drug, although the population to which this conclusion applied was circumscribed and the criteria for ‘improvement' were modest.

Although enhanced efficacy has been claimed ‘by association' for other new-generation antipsychotics (where comparable efficacy has been shown with clozapine as comparitor), specific tests of their advantages in operationally defined treatment resistance have not always shown these drugs to be superior to standard treatments. (66)

Psychosocial interventions

Controlled studies of psychosocial interventions are less extensive than those concerning drug treatments, and assessments in many of the studies have not been blind, but it is possible for the psychosocial therapy of schizophrenia to be evidence based.

The major types of intervention are as follows:

  • psychodynamic psychotherapy
  • social skills training and illness self-management programmes 
  • family interventions
  • cognitive-behavioural therapy.

A further group, while strictly relating more to administrative measures, may nonetheless be conveniently considered here:

  • alternatives to hospitalization.

Psychodynamic psychotherapy

Analytical psychotherapy was widely used in the first half of this century and continues occasionally to be utilized, although usually in combination with antipsychotic medication. The studies of May and colleagues, mentioned above, provide no support for the view that psychotherapy is an appropriate treatment for first-episode schizophrenia.

Gunderson et al. (67) reported a randomized trial comparison of two forms of psychotherapy in schizophrenia: reality adaptive–supportive, which focused on practical problems, and exploratory insight-orientated psychotherapy. The latter was stated to be the more effective in terms of ‘ego-functioning' but was clearly less effective than reality adaptive–supportive in terms of measures of rehospitalization and vocational and social adjustment.

Social skills training and illness self-management programmes

Social skills training refers to a structured learning-orientated approach towards the acquisition of skills relevant to the individual and the demands of his or her environment. (68) The studies of Claghorn et al. (69) in which emphasis was put upon tasks of daily living, and of Malm, (70) which stressed communication skills, both demonstrated effectiveness in terms of enhanced insight and socialization. Spencer et al. (71) compared social skills treatment with group discussion in a sample of chronic schizophrenic patients and found that conversational skills were improved in the social skills treatment group but not in the others, with improvements maintained at 2-month follow-up. Hogarty et al. (72) studied family psychoeducation, social skills training, and maintenance chemotherapy in the aftercare of schizophrenic patients, considering the effects after 1 and 2 years. They found an additive beneficial effect of social skills and family therapy after 1 year but this was less after the second year. Dobson et al. (73) randomized 33 schizophrenic patients to social skills training or milieu therapy and found that at 3 months there was a reduction in negative symptoms in the social skills as compared with the milieu therapy group.

Benton and Schroeder (74) conducted a meta-analysis of 27 studies of social skills training and found benefits in terms of assertiveness, general social skills, and speed of discharge. There was a possible reduction in relapse rates and limited evidence for generalization of skills and of their maintenance. Generalization of skills is an important issue for social skills training. Skills are generally taught in a role-play situation, and although some studies have indicated that behaviour learned in this way will generalize to unprompted situations, (75) some reviews have concluded that it is not clear that changes in treatment settings generalize to community settings.(76)

Individuals with substantial cognitive deficits and/or high levels of conceptual disorganization are poor candidates for most social skills training programmes. Illness self-management is an element of a number of these, but this particular area of patients' functioning has been more specifically addressed in separate plans of psychosocial management. Eckman et al. (77) studied the effects of an intensive, twice-weekly group programme directed to illness management which included video modelling, role playing and problem-solving activities. Patients received focused instructions, active coaching, and homework. This package was compared to a similarly intensive, supportive psychotherapy group. The package resulted in improved self-management in medication but little change in symptom levels. Linszen et al. (78) reported that a programme of illness self-management was alone as effective in reducing relapses as it was when combined with a family intervention.

Therefore available trials offer support for various types of social training in relevant skills. In general they have not been assessed on a blind basis. However, it has been shown(66,79) that this can be done successfully using edited video recordings.

Family interventions

Forms of management for schizophrenia focusing upon the patient's family arose from the observation that criticism and hostility on the part of close relatives was an important determinant of relapse.(80) Such criticism and hostility was referred to as high ‘expressed emotion', and forms of management designed to reduce this were devised. They combined attempts to improve family interactions with education about the disorder and direct advice about dealing with crises etc. Such packages have been extensively evaluated and their efficacy demonstrated, but which of the diverse elements is necessary for that efficacy is still unclear. (81,82)

Leff and colleagues have published a series of studies comparing education and relatives' groups with family therapy designed to reduce high expressed emotion. They found that relapse rates were reduced in those patients who remained on medication and whose relatives participated in family therapy. (83,84) It was concluded that reduction of expressed emotion and of time in face-to-face contact were important elements in determining the efficacy of this approach, although the authors acknowledge the difficulties of persuading relatives to participate in this type of intervention. Tarrier and colleagues have found essentially similar results. (85,86)

Family education and social skills training were also examined by Hogarty et al. (87) in 103 patients with schizophrenia and schizoaffective psychosis from high expressed emotion households. Relapse rates over 1 year were reduced by about 20 per cent by both treatments. It was concluded that some of the benefit was attributable to high expressed emotion reduction but that on-going antipsychotic medication was essential for schizophrenic patients in high expressed emotion households and that the effect of family therapy was in terms of delaying, rather than preventing, relapse. Glick et al. (88) reported that family education improved symptom ratings over 6 months, while McFarlane et al., (89) comparing psychoeducation in single family and multiple family treatment styles, showed advantages for the multiple family groups, and suggested that family support may have been the important element.

A meta-analysis of family intervention studies in the Cochrane database found that these reduce relapse rates by about 50 per cent for up to 2 years, and that medication compliance is similarly improved. (90) It was calculated that about six families need to be treated to prevent one relapse. It is therefore clear that family interventions are effective. The extent to which this effectiveness relates to improved compliance with medication, encouragement of realistic expectations, increased family involvement in clinical decision-making, improved family interaction, or indeed reduction of criticism and hostility towards the patient, is unclear.

These issues are underlined by the findings of a large multicentre study in which patients were allocated to one of three antipsychotic drug regimens and either supportive or intensive family treatment strategies. (61) There were clear advantages for standard as opposed to low or intermittent antipsychotic regimens, but no differences could be demonstrated between the two family interventions.

Cognitive-behavioural therapy

In recent years, controlled trials of cognitive-behavioural therapy in schizophrenia have been published. Tarrier et al. (91) randomly allocated 27 patients with residual symptoms on antipsychotic medication to coping strategy enhancement or problem solving, and compared them with 22 patients allocated to a waiting list condition. Both treatments reduced positive symptoms, but not negative symptoms or social functioning. Drury et al. (92) undertook a trial of cognitive-behavioural therapy in drug-treated patients with ‘acute non-affective psychosis'. Subjects were randomized to cognitive therapy or equivalent hours of therapeutic input. Of 117 patients, 69 satisfied inclusion criteria and 62 were randomly allocated to the two treatments, but 22 were withdrawn, essentially for non-cooperation. Both groups showed a decline in symptoms, but this was more marked (p < 0.001) in the cognitive therapy group. The authors emphasize that the study was not blind and observe that blindness is not possible in studies of psychosocial interventions, though as noted above, this is not now the case. (66,79)

The London–East Anglia randomized controlled trial of cognitive-behavioural therapy for psychosis (93,94) allocated 60 patients with at least one positive schizophrenic symptom resistant to medication to cognitive-behavioural therapy + standard care, or standard care. Fifty per cent of the cognitive-behavioural therapy group, as compared with 31 per cent of the control group, improved over 9 months. A major determinant of improvement with cognitive-behavioural therapy was ‘cognitive flexibility concerning delusions'. (94)

Cognitive remediation has recently been applied to the care of schizophrenic patients, where the desired endpoint is not symptom reduction but amelioration of cognitive deficits. Benedict et al. (95) compared 16 patients who received computerized vigilance task training with 17 who did not. The training improved task performance. Vollema et al.(96) found that performance on the Wisconsin Card Sorting Test was improved by training but not financial reward, while Corrigan et al. (97) comparing the effects of a 1-h session of vigilance with and without memory training on social cue recognition in 40 patients randomized to these two conditions, found that the combined training had greater benefits.

Alternatives to hospitalization

Several randomized controlled trials have demonstrated that alternatives to hospitalization can be successful. Assertive (or intensive) case treatment involves multidisciplinary teams with high staff/patient ratios, working assertively with patients on an ‘as-required' basis to avoid hospitalization. Case management involves a designated key worker writing out a care plan for each patient and reviewing progress in meeting their needs via a number of agencies. In randomized controlled studies of mentally ill people not specified as having schizophrenia, Rosenheck et al. (98) and Quinlivan et al. (99) reported that, as compared with standard care, assertive case treatment reduced inpatient service use and reduced overall care costs. Aberg-Wistedt et al. (100) studying schizophrenic patients, reported that assertive case treatment (in comparison with standard care) reduced the burden upon relatives and increased the patients' social networks. On the other hand, Telles et al. (101) found that case management and a family management programme appeared to be associated with increased symptomatology in schizophrenic patients.

Case management has been the subject of a meta-analysis in the Cochrane database. Marshall et al. (102) in a review of nine studies, found that case management does lead to increase in contact with patients but doubles the rates of hospital admission and is not associated with any definite improvements in symptomatology or social functioning.

Principles of treatment and management

Preliminaries

Medically, ‘treatment' and ‘management' are often used synonymously. In dealing with the complexities of schizophrenia, however, ‘treatment', with its narrower, patient-orientated, and medical connotations, is best restricted to primarily psychopharmacological issues targeted on dimensions of efficacy while the latter is more appropriate to broader care issues. The views expressed here are the authors' own and relate to work within the United Kingdom context. They will not be shared by all others, and the strategies we suggest will not always be appropriate in the very different circumstances of some other countries.

The doctor confronted with a patient believed to be acutely psychotic must address three preliminary questions.

  1. Can the clinical situation be dealt with informally or are compulsory powers required?
  2. Does the patient require admission or can they be managed as an outpatient?
  3. Must medication be started immediately or is it legitimate to observe the evolution of the illness over time?

The answer to the first question will depend on a thorough risk assessment. The latter has already been addressed. Research and clinical experience indicate that not all acutely ill patients will require admission, although in certain scenarios admission should be a priority.

Admission provides the psychiatrist with a number of advantages, including the following:

  • adequate observation to delineate the extent of symptomatology
  • comprehensive assessment of risk
  • establishing a solid therapeutic alliance
  • stabilization of medication in a supervised environment in which rapid changes of plan can be implemented swiftly and safely
  • ready opportunities for avoiding or treating psychiatric emergencies.

However, there is an important additional advantage from the patient's, and often the carer's, point of view: inpatient facilities can provide that security and support which define the cardinal elements of ‘asylum'.

Identifying goals and defining structure

The key to avoiding confusion and ‘decision paralysis' in dealing with the complex clinical situations schizophrenia presents is to delineate the structure within which it is hoped to achieve a series of treatment/management goals.

The acute phase

Goals

The acute phase is characterized by the perception of significant change in well being. The goals of treatment are as follows:

  1. controlling intrusive non-specific symptomatology, such as anxiety, agitation, and, especially, insomnia;
  2. ensuring the safety and well being of the patient and others by containing chaotic and socially damaging behaviour; 
  3. engaging the patient in therapeutic recommendations, and gaining consent for treatment plans (mental state permitting);
  4. implementing an appropriate foundation drug regimen;
  5. stabilizing productive symptomatology;
  6. preventing psychiatric emergencies where possible, and treating them appropriately where unavoidable.

Stabilization of ‘positive' symptomatology is the ultimate, but not the first, goal of this phase. It is important not to overlook other goals that can be achieved quickly while awaiting specific antipsychotic efficacy.

There is no single first-line drug for the treatment of acute schizophrenic episodes. Choices traditionally reflect individual preferences. Nowadays, one might assume that high-potency compounds are the gold standard, an impression encouraged by the fact that haloperidol has become the first-line choice for comparitor in assessing the efficacy of new agents. Such an assumption would be erroneous. The apparent pre-eminence of high-potency compounds largely springs from the influence of American practice, where these drugs enjoy a prominence less evident elsewhere, although surveys suggest that this represents a relatively recent shift. (47) The multinational phase III efficacy study of risperidone, conducted in 15 countries across the world, (18) suggests that, internationally, low-potency phenothiazines remain the most commonly prescribed.

There are real problems with the use of high-potency drugs in acute-phase treatment. Although relatively safe, the risks of neurological adverse effects are substantially greater than with low-potency compounds, particularly in drug-naive patients. While potency is itself an issue in this, a further factor is that high-potency drugs tend, in relative terms, to be used in higher doses than low-potency compounds. (103) This probably follows from the fact that doses of low-potency drugs are to some extent self-limited by the anti-autonomic and sedative side-effects intrinsic to the group.

There is, however, persistent evidence that, in general, psychiatrists still tend to use higher doses of antipsychotics to treat schizophrenia than are necessary. No studies have been able to demonstrate any therapeutic advantage from high-dose regimens, (46) while neurological side-effects are predictably more prevalent. (5) The complex kinetics of antipsychotics and the variability of both the natural history and response patterns of the illness make it impossible to provide reliable advice on dosage from dose–response data. However, flexible dosing studies suggest that the majority of responses will be achieved in the range of 500 to 600 mg/day of chlorpromazine or its equivalent, (47) with some suggestion that first-episode patients may respond at lower doses.

It has been repeatedly pointed out that there is no evidence to support the view that low-potency drugs are better in the treatment of excited psychotic states. It may be that this is one of those areas in which data from the somewhat idealized circumstances of clinical trials do not travel well to the day-to-day context. In one study of manic patients, which showed no difference at 2 weeks between those treated with chlorpromazine and those on the high-potency thiothixene, subjects were monitored at intervals of 2 h, (104) a level of care which would seem far removed from most routine standards currently.

Controversy persists as to whether antiparkinson medication should, as a matter of routine, be prescribed when standard antipsychotic treatment is started. Coincidental antiparkinson medication does reduce the risk of early extrapyramidal side-effects, (5) although these data relate mainly to high-potency drugs. However, even with these compounds, only a minority will develop significant extrapyramidal disorders, while the most commonly utilized antiparkinson drugs (i.e. anticholinergics) may interfere with antipsychotic actions (105) and have their own profile of unwanted side-effects. (5) Current recommendations are that antiparkinsonian medication should only be prescribed if neurological adverse effects become evident. (5,106)

Perhaps the most taxing question now is the place of new generation drugs. Should they be considered as routine first-line treatments or reserved for specific indications? The core issue is cost and whether the perceived benefits justify the cost implications. With clozapine, the position is clear. This drug is licensed for use in treatment resistance and neurological intolerance only, which removes it from first-line treatment. As for the others currently available, this remains a matter of individual choice and local prescribing policy. However, intolerance to standard antipsychotic regimens relates not only to individual susceptibility, which resides with the patient, but to treatment and management decisions, which reside with the doctor. (5) It is possible to utilize standard antipsychotics with minimal risk of adverse effects in the majority. In the study by Janicak et al. (104) noted above, patients were successfully treated with relatively low doses of medication and none experienced intolerable adverse effects.

In deciding treatment regimens in those with past treatment histories—be advised by the patient! Medications in which they have confidence should always be one's first-line approach in treating acute relapses.

The boundary of the acute phase of treatment is not rigidly defined, nor is it necessarily set by major symptomatic reversals. The important goal is ‘stabilization', as evidenced by the psychotic process becoming less ‘active'. Should side-effects emerge, it is part of the doctor's on-going efforts at engagement and maintaining consent to explain the issues to the patient, and in the process, educate them about the nature, and possible treatment, of adverse experiences. Psychiatric emergencies are discussed below.

Formulating treatment plans
Initial planning

Reasonable, effective treatment plans can be devised around the first-line use of standard antipsychotics and at present, these should be the initial choice in the majority of cases. However, caution is urged with the routine use of high-potency compounds, especially in first-episode patients and those in relapse who have not been receiving medication for some time.

Low-potency regimens

Foundation recommendations

With disturbance in both non-specific and behavioural domains, a low-potency drug is a reasonable first choice for the initial treatment plan:

  • in the drug-naive or frail, chlorpromazine 50 mg twice daily and 100 mg at night
  • in those with prior exposure, chlorpromazine 100 mg three times daily or 100 mg twice daily and 150 mg at night.
Added flexibility

‘As-required' prescriptions allow nursing colleagues to intervene at their discretion, in the face of increasing disturbance. These should be carefully prescribed with indications, maximum frequencies, dosages, and modes of administration written up separately and unambiguously. A major advantage of ‘as-required' prescriptions is to inform the judgement about how far short initial treatment recommendations fall from practical requirements, and information on the use of this component of the plan should be incorporated into each revision. In order to avoid polypharmacy, choices of drug for ‘as-required' prescription should, as far as possible, be limited to additional doses of the initial compound.

Adjunctive medications

If the initial plan achieves less than was anticipated, it is important to consider adding adjunctive treatments, in order to avoid over-rapid increments in antipsychotic doses. Benzodiazepines are best in this situation.

High-potency regimens

Foundation recommendations

Initial plans with these drugs might comprise the following:

  • in the drug naive, haloperidol 1.5 to 2 mg twice daily
  • in those previously exposed, haloperidol 2.5 to 3 mg twice daily.
Added flexibility

It is important with high-potency regimens to be circumspect about the prescription of ‘as-required' doses. Additional doses of one's first-choice drug or single doses of another high-potency compound will inevitably lead to rapid dose escalation and the likelihood of neurological adverse effects. A benzodiazepine should again be considered as an alternative. Fifty per cent of acute dystonias present within 3 days of the start of treatment or dose increments, and features of akathisia can be evident even after single, especially parenteral, doses of high-potency compounds. (5) Prescribing an ‘as-required' anticholinergic does allow experienced nursing staff to intervene immediately should such features emerge, although this should only be implemented after medical confirmation of the diagnosis.

Adjunctive medications

Reliance on benzodiazepines to achieve early, acute-phase treatment goals may be greater if one's initial plan incorporates a high-potency antipsychotic, as the low introductory doses recommended here may be inadequate to improve sleep and ease anxiety/agitation as swiftly as is ideal.

Where neither non-specific symptomatology nor behavioural disturbance is prominent, advocacy for any particular drug type is more problematic. While the use of standard compounds as the first-line treatment of acute schizophrenic episodes is advocated, there are two scenarios in which new-generation drugs should be considered:

  1. those first-episode patients in whom it is felt particularly important to minimize the risk of adverse effects, especially sedation and the unpleasant subjective experience of neurological adverse effects;
  2. patients in relapse who have been poorly tolerant, especially of extrapyramidal symptomatology, in the past.
Revision

Monitoring response to initial treatment decisions involves bringing together information from nursing colleagues, prescription data, mental state examination, and assessment of non-target effects, including extrapyramidal status. Initial treatment plans should be reviewed frequently—a key element in fostering engagement and compliance as well as in refining treatment recommendations.

The markers of early progress in treatment are stabilization and, over several days, containment of non-specific affective symptomatology, improvement in nocturnal sleep without undue daytime drowsiness, and control of overactive, disorganized, and potentially confrontational behaviour.

There are two temptations to avoid in revising early treatment plans. The first is changing drug prematurely and the second is introducing polypharmacy. While intolerance or patient pressure may force an early change of antipsychotic, it is in general important to give each drug an adequate trial, which may take several weeks, while adding other antipsychotics to the foundation drug will increase the antipsychotic ‘load' and the neurological risks. Rather than go down either of these roads, it is better to rely on adjunctive medications to temporarily contain difficult situations.

The post-acute phase

Goals

The post-acute phase is characterized by returning stability. The goals now are as follows:

  1. consolidation of clinical improvements
  2. rationalization of the treatment regimen
  3. resocialization.

All schizophrenic symptomatology does not resolve at the same rate and the first signs of improvement should not be taken as evidence that the tide has conclusively turned. Global improvements evident over the broad range of domains must be shown to deepen into specific improvements in particular symptomatology. Furthermore, improvements evident in the ward environment need to be shown to generalize to the patient's routine situation. Although there is no evidence that rationalizing and simplifying the treatment plan contributes to enhanced compliance, (107) this is intuitively sound. This is also the period when resocialization, and if necessary rehabilitation, should be commenced, and secondly, the issue of longer-term treatment raised.

Formulating treatment/management plans

Initial planning In the post-acute phase, there is often great pressure to reduce medication doses radically, but overzealous reductions at this stage are likely to be followed by symptomatic exacerbation. The main task in the early part of this phase is to persuade the patient of the merits of holding the regimen stable while resocialization is being established. As symptomatology begins to settle, those on low-potency regimens are likely to become less tolerant of sedative actions which may compromise resocialization and compliance. A cautious reduction on this basis is certainly justified. With high-potency drugs the problem is more likely to be the apathy and impaired initiative of bradykinesia, which, if not responsive to antiparkinson medication, would also merit dose reduction.

Revision

Once the initial steps in resocialization have been achieved, medication can be rationalized. Where multiple drugs continue to be necessary, they should be prescribed at similar times, with as much of the regimen as possible given at night. In addition, the need for adjunctive medications should be re-evaluated. There is no consensus as to whether or not stopping antiparkinson medication at this stage will result in recrudescence of neurological signs. (5) As it is likely that some patients will not require these treatments, systematic reductions should be started. If a depot is agreed, this should also be started. Depots are not ideally suited to acute-phase treatment and because of their unusual pharmacokinetics should be commenced ‘in parallel' with existing oral medication, which can be discontinued slowly as the depot approaches steady state (i.e. approximately five times the half-life).

The maintenance phase

Goals

The boundary between post-acute and maintenance phases is the least defined but is reached when improvements in all the major domains of disorder are felt to be maximal. The major goals of maintenance are as follows:

  1. maximum well being with minimum adverse effects
  2. monitoring efficacy/effectiveness and tolerability
  3. continuation or completion of rehabilitation and social reintegration.

Attempts to reduce medication should still be cautious during maintenance, but over the longer term, more determined. The major medical task in this phase is monitoring, to include effectiveness, as well as the narrower efficacy. Any long-term therapeutic regimen can only be properly evaluated by balancing efficacy/effectiveness against tolerability. Exploration of neurological tolerability should involve both an enquiry into subjective effects as well as an examination to elicit signs, and should be undertaken regularly.

The final, and to some extent most intensive, aim of maintenance comprises those management techniques geared towards helping patients attain the highest possible level of psychosocial functioning and carers the greatest degree of understanding and skills for coping.

Formulating treatment/management plans

For the first few months of maintenance, any social and interpersonal difficulties should be addressed with minimal changes to antipsychotic regimens, unless adverse effects are impeding this process. The more determined period of medication reduction should be reserved for the time when a stable set of routines has been established and the patient is considered to be functioning relatively competently. Any reductions should be followed by a prolonged period of observation before further changes are contemplated; because of their long half-lives, this principle especially applies to depots.

Antiparkinsonism medication should be continually justified by attempting dose reductions throughout the maintenance phase, but these should be delayed following reduction of the antipsychotic. Decreases in the clinical features of extrapyramidal disorder will not occur for some weeks after decreases in antipsychotic doses and premature cessation of the antiparkinson agent may result in a false conclusion that it is required long-term.

Depression during the maintenance phase can have as many causes as depression in other contexts, but despite initial pessimism, there is now good evidence that such mood states do respond to traditional (tricyclic) antidepressants (108) and they should be vigorously treated.

Psychiatric emergencies

Those who deal professionally with acutely ill patients must always bear in mind the ‘unpredictability factor' and the potential for violent behaviour during acute symptomatic ‘shifts'. Risk assessment of dangerousness is an imprecise science, but an attempt should be incorporated into all routine clinical assessments during acute phase treatment. Caution should be exercised with patients who are profoundly suspicious, verbally aggressive, resistant to engagement, whose pattern of disorder does not allow for a comprehensive mental state examination, or whose clinical condition is complicated by substance abuse, and especially (for those in relapse) who have a past history of assaultive or threatening behaviour. Principles of wider management are crucial in avoiding emergency situations, bearing on the quality of the environment and the amount and quality of support available. However, even with high levels of vigilance, pre-emptive plans, and good quality management, emergency situations may still occur and must be dealt with swiftly.

Poor response and treatment resistance

Several intervention strategies have been recommended in those whose response to standard regimens is disappointing.

Antipsychotics

Where a patient has failed to respond satisfactorily to an adequate dose (600–800 mg chlorpromazine or its equivalent per day) of the first-choice antipsychotic, given for an adequate time (at least 6 weeks), they might be considered a ‘poor responder' but not yet treatment resistant. Alternative plans have been proposed at this stage and include, in order of implementation:

  1. increasing the first-choice drug to a high-dose schedule (i.e. up to 1000 mg chlorpromazine or its equivalent) for 6 to 8 weeks;
  2. changing to a drug of different class, in standard dose ranges, for 6 to 8 weeks, which nowadays might involve changing to (or from) a new-generation compound;
  3. increasing the second-choice drug to a high-dose schedule, for 6 to 8 weeks.

As far as the research literature is concerned, it is only when at least one, and preferably two, of these steps have been tried without notable improvement that the patient would be considered ‘treatment resistant'. For routine practice, however, several other interventions can be recommended.

Should it be that antipsychotic doses have escalated almost ‘by default', it is worth reducing to average or low average ranges, as sometimes pervasive extrapyramidal symptomatology may be the source of impaired progress. (5) A further strategy is the addition of a depot with a view to transferring the bulk of treatment to this over time. Failure of the initial oral drug(s) may relate to adverse pharmacokinetic parameters, such as poor absorption or enhanced metabolism, which can be bypassed by depot administration.

First-line adjunctive medications

One should, in addition, look to simplifying any complexities that may have entered into the regimen, such as eradicating polypharmacy, reducing or stopping anticholinergics where possible, and assessing the need for other drugs such as antidepressants.

Second-line adjunctive treatments
Lithium

Lithium salts have been particularly recommended for patients with schizoaffective disorders, although at one time were more widely advocated. Although there is indirect evidence to support this general use, direct trial evidence is less encouraging (109) and their utilization is best considered as empirical in those with prominent affective symptoms.

Carbamazepine

The use of carbamazepine in schizophrenia has not been systematically researched and rests on anecdotal evidence. It may be considered as an alternative mood stabilizer in those with schizoaffective disorder, although another adjunctive use in schizophrenia may lie with those patients who have intransigent symptomatology, with a perceptual emphasis, reminiscent of temporal lobe disorder, even when the routine EEG is normal. A further indication may be patients whose EEG abnormalities show a temporal localization without clinical fits. Carbamazepine induces the metabolism of some antipsychotics, which may necessitate raising doses of the latter to prevent a paradoxical deterioration.

High-dose benzodiazepines

High-dose benzodiazepine treatment has been reported as being specifically beneficial in the treatment of acute schizophrenic episodes. (110) It is likely, however, that these benefits, which appear slight and inconsistent, result from actions on affective symptomatology or elements of extrapyramidal disturbance, and their adjunctive use in high-dose regimens would not seem justified.

Electroconvulsive therapy

ECT has become less popular as a treatment for schizophrenia, although it can be effective (111) and may still hold a place in the therapeutic armamentarium, but as a third-line treatment.

Comments

While in terms of professional guidelines it would be considered good practice to go through these steps in patients responding poorly or frankly resistant, most are largely empirically derived, and such evidence as there is does not encourage optimism about their overall likelihood of success. As was noted, the one treatment of established efficacy in resistant schizophrenia is clozapine. (65) While in individual cases impressive benefits may become evident, especially in quality-of-life parameters, improvements attributable to clozapine over standard drugs are far from dramatic. Despite this, clozapine is probably underutilized and should be more readily considered in treatment of resistant cases.

(See antipsychotic and anticholinergic drugs for further discussion of new-generation antipsychotics.)

Psychosocial interventions in management

While antipsychotic drugs may be considered the cornerstone of the treatment of schizophrenia, the most competent prescribing of these does not provide a comprehensive programme of management. Even if no formal programme is followed, all patients and their relatives will require appropriate information about the nature of the illness and the therapeutic possibilities. They will need support in coming to terms with the limitations that the illness may impose upon the expectations they have had, and guidance about alternative plans. The requirements of individual patients for rehabilitation and resocialization vary greatly, but general rules can be provided in terms of the classification of psychosocial approaches overviewed above.

Antipsychotic drugs can (and indeed overwhelming evidence indicates that they should) be given to all patients in whom the diagnosis of schizophrenia is appropriately made. This is not the case with psychosocial interventions. In general, these require co-operation from patients and/or relatives, and some of them can only be applied to certain types of patient.

There is no evidence for the value of psychodynamic psychotherapy in schizophrenia, but some will nonetheless wish to have it—and may well find it helpful. However, it should only be recommended in combination with treatments of proven effectiveness.

Family interventions have been shown to be of value, although the essential elements of any package remain far from clear. Education and support from trained staff should be offered, although not all families will be willing to accept this in a group format. While social skills training programmes of various kinds have been demonstrated to be efficacious in achieving specific goals in controlled circumstances with selected groups of patients, it is difficult to draw general guidelines about this kind of management. Nonetheless, while the goals may be restricted, enhanced skills in conversation and self-presentation may make a great deal of difference to the success of resocialization. The place of cognitive remediation is less clear and the requirement for it in routine practice cannot yet be said to be established.

Cognitive-behavioural techniques also require the co-operation of the patient and probably a degree of cognitive flexibility in respect of delusions. This again restricts the applicability of such programmes, but the benefits demonstrated by available trials indicate that they should be available for appropriate individuals.

As far as alternatives to hospital care are concerned, assertive case treatment provides benefits in terms of inpatient service utilization costs, and sometimes burden of care, but case management has not been shown to be helpful.

Concluding remarks

Traditionally, psychiatry has not always given to schizophrenia the attention that its complexities and chronicity merit. Views about what really helps in such a disorder are not, at the end of the day, based solely on research evidence, but are informed by personal experience—and, no doubt, personal prejudice. The distinction between theoretical and pragmatic psychiatry is nowhere more evident than in psychiatrists' actions in regard to schizophrenia. As a result, an element of variability in treatment and management recommendations is inevitable. What is often perceived as lacking, however, is less a consistency in practice recommendations than a consistency of personnel. One element sufferers and their families appreciate most is accessible care by experienced professionals, who, by being involved over the long years of illness, know the disorder, the patient, and the families—and are able to attempt some understanding of all three. Combining such an essentially ‘primary care' function with rapidly expanding knowledge of pathophysiological mechanisms and a wide, and evolving, array of therapeutic tools, schizophrenia can increasingly be seen as a legitimate and rewarding area for specialist psychiatric practice in which, contrary to widespread belief, the skilled doctor can make a difference. 

References

1. Clouston, T. (1884). Clinical lectures on mental diseases. Henry Lea, Philadelphia, PA.

2. Adams, F. (1856). Aretaeus, the Cappadocian: the extant works (translated and edited by F. Adams). Mulford House, Boston, MA.

3. MacDonald, M. (1981). Mystical Bedlam: madness, anxiety and healing in seventeenth century England. Cambridge University Press.

4. Johnstone, E.C. (1998). Psychiatry, its history and boundaries. In Companion to psychiatric studies (ed. E.C. Johnstone, C.P.L. Freeman, and A.K. Zealley), pp. 1–10. Churchill Livingstone, Edinburgh.

5. Owens, D.G.C. (1999). A guide to the extrapyramidal side-effects of antipsychotic drugs. Cambridge University Press.

6. Klein, D. and Davis, J.M. (1969). Diagnosis and drug treatment of psychiatric disorders. Baltimore. Williams and Wilkins, Baltimore, MD.

7. Owens, D.G.C. (1996). Adverse effects of antipsychotic agents: do newer agents offer advantages? Drugs, 51, 895–930.

8. Davis, J.M. and Garver, D.L. (1978). Neuroleptics—clinical use in psychiatry. In Handbook of psychopharmacology (ed. L.L. Iversen, S. Iversen, and S. Snyder), pp. 129–64. Plenum Press, New York.

9. Cole, J.O. and the NIMH Psychopharmacology Service Center Collaborative Study Group (1964). Phenothiazine treatment in acute schizophrenia. Archives of General Psychiatry, 10, 246–61.

10. Tuma, A.H. and May, P.R.A. (1979). And if that doesn't work, what next? A study of treatment failures in schizophrenia. Journal of Nervous and Mental Disease, 167, 566–71.

11. May, P.R.A. (1968). Treatment of schizophrenia: a comparative study of five treatment methods. Science House, New York.

12. May, P.R.A., Tuma, A.H., Yale, C., Potepam, P., and Dixon, W.J. (1976). Schizophrenia: a follow-up study of results of treatment: II. Hospital stay over two to five years. Archives of General Psychiatry, 33, 481–6.

13. May, P.R.A., Tuma, A.H., Yale, C., and Dixon, W.J. (1981). Schizophrenia: a follow up study of the results of five forms of treatment. Archives of General Psychiatry, 38, 776–84.

14. Wyatt, R.J. (1991). Neuroleptics and the natural course of schizophrenia. Schizophrenia Bulletin, 17, 325–51.

15. Crow, T.J., Macmillan, J.F., Johnson, A.L., and Johnstone, E.C. (1986). The Northwick Park study of first episodes of schizophrenia: II A randomised controlled trial of prophylactic neuroleptic treatment. British Journal of Psychiatry, 148, 120–7.

16. Johnstone, E.C., Crow, T.J., Davis, J.M., and Owens, D.G.C. (1999). Does a four week delay in the introduction of medication alter the course of functional psychosis? Journal of Psychopharmacology, 13, 238–44.

17. Lieberman, J.A. (1993). Understanding the mechanism of action of atypical antipsychotic drugs: a review of compounds in use and development. British Journal of Psychiatry, 163 (Supplement 22), 7–18.

18. Peuskins, J., on behalf of the Risperidone Study Group (1995). Risperidone in the treatment of patients with chronic schizophrenia: a multi-national, multi-centre, double-blind, parallel-group study versus haloperidol. British Journal of Psychiatry, 166, 712–26.

19. Tollefson, G.D., Beasley, C.M., Tran, P.V., et al. (1997). Olanzapine versus haloperidol in the treatment of schizophrenia and schizoaffective and schizophreniform disorders: results of an international collaborative trial. American Journal of Psychiatry, 154, 457–65.

20. Arvanitis, L.A., Miller, B.G., and the Seroquel Trial 13 Study Group (1997). Multiple fixed doses of ‘Seroquel' (quetiapine) in patients with acute exacerbation of schizophrenia: a comparison with haloperidol and placebo. Biological Psychiatry, 42, 233–46.

21. Puech, A., Fleurot, O., Tein, W., and the Amisulpride Study Group (1998). Amisulpride, an atypical antipsychotic, in the treatment of acute episodes of schizophrenia: a dose-ranging study vs. haloperidol. Acta Psychiatrica Scandinavica, 98, 65–72.

22. Pettit, M., Raniwalla, J., Tweed, J., Leutenegger, E., Dollfus, S., and Kelley, F. (1996). A comparison of an atypical and typical antipsychotic: zotepine versus haloperidol in patients with acute exacerbations of schizophrenia: a parallel group, double-blind trial. Psychopharmacology Bulletin, 32, 81–7.

23. Chouinard, G. (1995). Effects of risperidone in tardive dyskinesia: an analysis of the Canadian Multicentre Risperidone Study. Journal of Clinical Psychopharmacology, 15 (Supplement 1), 36s–44s.

24. Tollefson, G.D., Beasley, C.M., Tamura, R.N., Tran, P.V., and Potvin, J.H. (1997). Blind, controlled, long-term study of the comparative incidence of treatment-emergent tardive dyskinesia with olanzapine and haloperidol. American Journal of Psychiatry, 154, 1248–54.

25. Crow, T.J. (1980). Molecular pathology of schizophrenia : more than one disease process? British Medical Journal, 280, 66–8.

26. Goldberg, S.C. (1985). Negative and deficit symptoms in schizophrenia do respond to neuroleptics. Schizophrenia Bulletin, 11, 453–6.

27. Tandon, R., Riveiro, S.C.M., DeQuardo, J.R., Goldman, R.S., Goodson, J., and Greden, J.F. (1993). Covariance of positive and negative symptoms during neuroleptic treatment in schizophrenia: a replication. Biological Psychiatry, 34, 495–7.

28. van Kammen, D.P., Hommer, D.W., and Malas, K.L. (1987). Effects of pimozide on positive and negative symptoms in schizophrenic patients: are negative symptoms state dependent? Neuropsychobiology, 18, 113–17.

29. Carpenter, W.T., Heinrichs, D.W., and Alphs, L.D. (1985). Treatment of negative symptoms. Schizophrenia Bulletin, 11, 440–52.

30. Moller, H.-J., Muller, H., Borison, R.L., Schooler, N.R., and Chouinard, G. (1995). A path-analytical approach to differentiate between direct and indirect drug effects on negative symptoms in schizophrenic patients—a re-evaluation of the North American Risperidone Study. European Archives of Psychiatry and Clinical Neuroscience, 245, 45–9.

31. Tollefson, G.D. and Sanger, T.M. (1997). Negative symptoms: a path analytic approach to a double-blind, placebo- and haloperidol-controlled clinical trial with olanzapine. American Journal of Psychiatry, 154, 466–74.

32. Kane, J.M., Safferman, A.Z., Pollack, S., et al. (1994). Clozapine, negative symptoms and extrapyramidal side-effects. Journal of Clinical Psychiatry, 55 (Supplement B), 74–7.

33. Carpenter, W.T., Heinrichs, D.W., and Wagman, A.M.I. (1988). Deficit and non-deficit forms of schizophrenia: the concept. American Journal of Psychiatry, 145, 578–83.

34. Breier, A., Buchanan, R.W., Kirkpatrick, B., et al. (1994). Effects of clozapine on positive and negative symptoms in outpatients with schizophrenia. American Journal of Psychiatry, 151, 20–6.

35. Sharma, T. and Mockler, D. (1998). The cognitive efficacy of atypical antipsychotics in schizophrenia. Journal of Clinical Psychopharmacology, 18 (Supplement 1), 12s–19s.

36. King, D.J. (1990). The effect of neuroleptics on cognitive and psychomotor function. British Journal of Psychiatry, 157, 799–811.

37. King, D.J. (1994). Psychomotor impairment and cognitive disturbances induced by neuroleptics. Acta Psychiatrica Scandinavica, 89 (Supplement 380), 53–8.

38. Johnstone, E.C., Owens, D.G.C., Frith, C.D., and Leary, J. (1991). Disabilities and circumstances of schizophrenic patients: a follow-up study. III. Clinical findings—abnormalities of the mental state and movement disorder and their correlates. British Journal of Psychiatry, 159 (Supplement 13), 21–5.

39. Jenner, P. and Marsden, C.D. (1982). The mode of action of sulpiride as an atypical antidepressant agent. Advances in Biochemical Psychopharmacology, 32, 85–103.

40. Margakis, B.P. (1990). A double-blind comparison of oral amitriptyline and low-dose intramuscular flupenthixol decanoate in depressive illness. Current Medical Research and Opinion, 12, 51–7.

41. Salminen, J.K., Lehtonen, V., Allonen, H., et al. (1980). Sulpiride in depression: plasma levels and effects. Current Therapeutic Research, 27, 109–15.

42. de Alarcon, R. and Carney, M.W.P. (1969). Severe depressive mood changes following slow-release intramuscular fluphenazine injection. British Medical Journal, 3, 564–7.

43. Van Putten, T. and May, P.R.A. (1978). ‘Akinetic depression' in schizophrenia. Archives of General Psychiatry, 35, 1101–7.

44. Koreen, A.R., Siris, S., Chakos, M., Alvir, J., Mayerhoff, D., and Lieberman, J. (1993). Depression in first-episode schizophrenics. American Journal of Psychiatry, 150, 1643–8.

45. Owens, D.G.C. and Johnstone, E.C. (1980). The disabilities of chronic schizophrenia: their nature and factors contributing to their development. British Journal of Psychiatry, 136, 384–95.

46. Janicak, P.G., Davis, J.M., Preskorn, S., and Ayd, F.J. (1993). Principles and practice of psychopharmacotherapy. Williams and Wilkins, Baltimore, MD.

47. Baldessarini, R.J., Cohen, B.M., and Teicher, M.H. (1990). Pharmacological treatment. In Schizophrenia—treatment of acute episodes (ed. S.T. Levy and P.T. Ninan), pp. 61–118. American Psychiatric Press, Washington, DC.

48. Caffey, E.M., Diamond, L.S., Frank, T.V., et al. (1964). Discontinuation or reduction of chemotherapy in chronic schizophrenia. Journal of Chronic Diseases, 17, 347–58.

49. Hogarty, G.E., Goldberg, S.C., and Collaborative Study Group (1973). Drug and sociotherapy in the aftercare of schizophrenic patients. Archives of General Psychiatry, 28, 54–64.

50. Prien, R.F., Cole, J.O., and Belkin, N.F. (1969). Relapse in chronic schizophrenics following abrupt withdrawal of tranquilising medication. British Journal of Psychiatry, 115, 679–86.

51. Davis, J.M., Schaffer, C.B., Killan, G.A., Kinard, C., and Chan, C. (1980). Important issues in the drug treatment of schizophrenia. Schizophrenia Bulletin, 6, 70–87.

52. Johnson, D.A.W. (1979). Further observations on the duration of depot neuroleptic maintenance therapy in schizophrenia. British Journal of Psychiatry, 135, 524–30.

53. Hogarty, G.E. and Ulrich, R.F. (1977). Temporal effects of drug and placebo in delaying relapse in schizophrenic outpatients. Archives of General Psychiatry, 34, 297–301.

54. Capstick, N. (1980). Long-term fluphenazine decanoate maintenance dosage requirements of chronic schizophrenic patients. Acta Psychiatrica Scandinavica, 61, 256–62.

55. Cheung, H.K. (1981). Schizophrenics fully remitted on neuroleptics for 3–5 years. British Journal of Psychiatry, 138, 490–4.

56. Morgan, R. and Cheadle, J. (1974). Maintenance treatment of chronic schizophrenia with neuroleptic drugs. Acta Psychiatrica Scandinavica, 50, 78–85.

57. Meltzer, H.Y., Burnett, S., Bastani, B., and Ramirez, L.F. (1990). Effects of six months of clozapine treatment on the quality of life of chronic schizophrenic patients. Hospital and Community Psychiatry, 41, 892–987.

58. Davis, J.M., Metalon, L., Watanabe, M., and Blake, L. (1993). Depot antipsychotic drugs: place in therapy. Drugs, 47, 741–73.

59. Kane, J.M., Rifkin, A., Woerner, M., et al. (1983). Low dose neuroleptic treatment of outpatient schizophrenics. Archives of General Psychiatry, 40, 893–6.

60. Marder, S.R., van Putten, T., Mintz, J., Lebell, M., McKenzie, J., and May, P.R.A. (1987). Low- and conventional-dose maintenance therapy with fluphenazine decanoate: two year outcome. Archives of General Psychiatry, 44, 518–21.

61. Schooler, N.R., Keith, S., Severe, J., et al. (1997). Relapse and rehospitalization during maintenance treatment of schizophrenia: the effects of dose reduction and family treatment. Archives of General Psychiatry, 54, 453–63.

62. Johnson, D.A.W., Ludlow, J.M., Street, K., and Taylor, R.D.W. (1987). Double blind comparison of half dose and standard-dose flupenthixol decanoate in the maintenance treatment of stabilised outpatients with schizophrenia. British Journal of Psychiatry, 151, 634–8.

63. Davis, J.M., Kane, J.M., Marder, S., et al. (1993). Dose response of prophylactic antipsychotics. Journal of Clinical Psychiatry, 54 (Supplement 3), 24–30.

64. Johnstone, E.C., Macmillan, F.J., Frith, C.D., Benn, D.K., and Crow, T.J. (1990). Further investigations of the predictors of outcome following first schizophrenic episodes. British Journal of Psychiatry, 157, 182–9.

65. Kane, J.M., Honigfeld, G., Singer, J., Meltzer, H.Y., and the Clozaril Collaborative Study Group (1988). Clozapine for the treatment-resistant schizophrenic: a double-blind comparison with chlorpromazine. Archives of General Psychiatry, 45, 789–96.

66. Mercer, G., Finlayson, A., Johnstone, E.C., Murray, C., and Owens, D.G.C. (1997). A study of enhanced management in patients with treatment resistant schizophrenia. Journal of Psychopharmacology, 11, 349–56.

67. Gunderson, J.G., Frank, A.F., Katz, H.M., Vannicelli, M.L., Frosch, J.P., and Knapp, P.H. (1984). Effects of psychotherapy in schizophrenia. II. Comparative outcome of two forms of treatment. Archives of General Psychiatry, 10, 564–98.

68. Kopelowicz, A., Corrigan, P.W., Schade, M., and Liberman, R.P. (1998). Social skills training. In Handbook of social functioning in schizophrenia (ed. K.T. Mueser and N. Tarrier), pp. 307–26. Allyn and Bacon, Needham Heights, MA.

69. Claghorn, J.L., Johnstone, E.E., Cook, T.H., and Itschner, L. (1974). Group therapy and maintenance treatment of schizophrenics. Archives of General Psychiatry, 31, 361–5.

70. Malm, U. (1982). The influence of group therapy on schizophrenia. Acta Psychiatrica Scandinavica (Supplement 297), 1–65.

71. Spencer, P.G., Gillespie, C.R., and Ekisa, E.G. (1983). A controlled comparison of the effects of social skills training and remedial drama on the social conversational skills of chronic schizophrenic patients. British Journal of Psychiatry, 143, 165–72.

72. Hogarty, G.E., Anderson, C.M., Reiss, D.J., et al. (1991). Family psychoeducation, social skills training and maintenance chemotherapy in the aftercare treatment of schizophrenia. II. Two-year effects of a controlled study on relapse and adjustment. Archives of General Psychiatry, 48, 340–47.

73. Dobson, D.J., McDougall, G., Busheikin, J., and Aldous, J. (1995). Effect of social skills training and social milieu treatment on symptoms of schizophrenia. Psychiatric Services, 4, 376–80.

74. Benton, M.K. and Schroeder, H.E. (1990). Social skills training with schizophrenics: a meta-analytic evaluation. Journal of Consulting and Clinical Psychology, 58, 741–7.

75. Wong, S.E., Martinez-Diaz, J.A., Massel, H.K., et al. (1993). Conversational skills training with schizophrenic inpatients: a study of generalisation across settings and conversants. Behaviour Therapy, 24, 285–304.

76. Halford, W.K. and Hayes R. (1991). Psychological rehabilitation of chronic schizophrenic patients: recent findings on social skills and family psycho-education. Clinical Psychology Review, 23, 23–44.

77. Eckman, T.A., Wirshing, W.C., Marder, S.R., et al. (1992). Technique for training schizophrenic patients in illness self-management: a controlled trial. American Journal of Psychiatry, 149, 1549–55.

78. Linszen, D., Dingemans, P., Van der Does, J.W., et al. (1996). Treatment, expressed emotion and relapse in recent onset schizophrenic disorders. Psychological Medicine, 26, 333–42.

79. Owens, D.G.C., Finlayson, A., Mercer, G., and Johnstone, E.C. (1997). The use of videotaped assessment in relation to a study of enhanced management in treatment resistant schizophrenia. Journal of Psychopharmacology, 11, 357–60.

80. Brown, G.W., Birley, J.L.T., and Wing, J.K. (1972). Influence of family life on the course of schizophrenic disorders. British Journal of Psychiatry, 121, 242–58.

81. Goldstein, M.J., Rodnick, E.H., Evans, J.R., May, P.R., and Steinberg, M.R. (1978). Drug and family therapy in the aftercare of acute schizophrenia. Archives of General Psychiatry, 35, 1169–77.

82. Falloon, I.R.H., Boyd, J.L., McGill, C., Razani, J., Moss, H.B., and Gilderman, A.M. (1982). Family management in the prevention of exacerbations of schizophrenia. New England Journal of Medicine, 306, 1437–40.

83. Leff, J., Kuipers, L., Berkowitz, R., Eberlein-Vries, R., and Sturgeon, D. (1982). A controlled trial of social intervention in the families of schizophrenic patients. British Journal of Psychiatry, 141, 121–34.

84. Leff, J., Berkowitz, R., Shavit, N., Strachan, A., Glass, I., and Vaughn, C. (1990). A trial of family therapy versus a relatives' group for schizophrenia. Two year follow-up. British Journal of Psychiatry, 157, 571–7.

85. Tarrier, N., Barrowclough, C., Vaughn, C., et al. (1989). Community management of schizophrenia. A two-year follow-up of a behavioural intervention with families. British Journal of Psychiatry, 154, 625–8.

86. Vaughan, K., Doyle, M., McConaghy, N., Blaszczynski, A., Fox, A., and Tarrier, N. (1992). The Sydney intervention trial: a controlled trial of relatives' counselling to reduce schizophrenic relapse. Social Psychiatry and Psychiatric Epidemiology, 27, 16–21.

87. Hogarty, G.E., Anderson, C.M., Reis, D.J., et al. (1986). Family psychoeducation, social skills training, and maintenance chemotherapy in the aftercare treatment of schizophrenia: I. one-year effects of a controlled study on relapse and expressed emotion. Archives of General Psychiatry, 43, 633–42.

88. Glick, I.D., Clarkin, V., Haas, G.L., and Spencer, J.H., Jr. (1993). Clinical significance of inpatient family intervention: conclusions from a clinical trial. Hospital and Community Psychiatry, 44, 869–73.

89. McFarlane, W.R., Lukens, E., Link, B., et al. (1995). Multiple-family groups and psychoeducation in the treatment of schizophrenia. Archives of General Psychiatry, 52, 679–87.

90. Mari, J.J. and Streiner, D. (1996). Family intervention for those with schizophrenia. In Schizophrenia module of the Cochrane database of systematic reviews (updated 2 December 1996) (ed. C. Adams, J. De Jesus Mari, and P. White). Available in The Cochrane Library (database on disk and CD-ROM). Cochrane Collaboration, Issue 2. Update Software, Oxford. (See also Psychological Medicine, 24, 565–78, 1994; Evidence-Based Medicine, 1, 121, 1996.)

91. Tarrier, N., Beckett, R., Harwood, S., Baker, A., Yusopoff, L., and Ugarteburu, I. (1993). A trial of two cognitive-behavioural methods of treating drug-resistant residual psychotic symptoms in schizophrenic patients. I. Outcome. British Journal of Psychiatry, 162, 524–32.

92. Drury, B., Birchwood, M., Cochrane, R., and Macmillan, J.F. (1996). Cognitive therapy and recovery from acute psychosis: a controlled trial. British Journal of Psychiatry, 169, 593–601.

93. Garety, P., Fowler, D., Kuipers, E., et al. (1997). London–East Anglia randomised controlled trial of cognitive-behavioural therapy for psychoses. II. Predictors of outcome. British Journal of Psychiatry, 171, 420–6.

94. Kuipers, E., Fowler, D., Garety, P., et al. (1998). London–East Anglia randomised controlled trial of cognitive-behavioural therapy for psychosis. III. Follow-up and economic evaluation at 18 months. British Journal of Psychiatry, 173, 61–8.

95. Benedict, R.H., Harris, A.E., Markow, T., McCormack, J.A., Nuechterlein, K.H., and Asarnow, R.F. (1994). Effects of attention training on information processing in schizophrenia. Schizophrenia Bulletin, 20, 537–46.

96. Vollema, M.G., Geurtsen, G.T., and van Voorst, A.J. (1995). Durable improvements in Wisconsin card sorting performance in schizophrenic patients. Schizophrenia Research, 16, 209–15.

97. Corrigan, P.W., Huschbeck, J.N., and Wilfe, M. (1995). Memory and vigilance training to improve social perception in schizophrenia. Schizophrenia Research, 17, 257–65.

98. Rosenheck, R., Neale, M., Leaf, P., Milstein, R., and Frisman, L. (1995). Multisite experimental cost study of intensive psychiatric community care. Schizophrenia Bulletin, 21, 129–40.

99. Quinlivan, R., Hough, R., Crowell, A., Beach, C., Hofstetter, R., and Kenworthy, K. (1995). Service utilization and costs of care for severely mentally ill clients in an intensive case management program. Psychiatric Services, 46, 365–71.

100. Aberg-Wistedt, A., Cressell, T., Lidberg, Y., Liljenberg, G., and Osby, U. (1995). Two-year outcome of team-based intensive case management for patients with schizophrenia. Psychiatric Services, 46, 1263–6.

101. Telles, C., Karno, M., Mintz, J., et al. (1995). Immigrant families coping with schizophrenia. Behavioural family intervention v. case management with a low-income Spanish-speaking population. British Journal of Psychiatry, 167, 473–9.

102. Marshall, M., Gray, A., Lockwood, A., and Green, R. (1997). Family intervention for those with schizophrenia. In Schizophrenia module of the Cochrane database of systematic reviews (updated 2 December 1996) (ed. C. Adams, J. De Jesus Mari, and P. White). Available in The Cochrane Library (database on disk and CD-ROM). Cochrane Collaboration, Issue 1. Update Software, Oxford. (See also Evidence-Based Medicine, 1, 30, 1995.)

103. Baldessarini, R.J., Katz, B., and Cotton, P. (1984). Dissimilar dosing with high-potency and low-potency neuroleptics. American Journal of Psychiatry, 141, 748–52.

104. Janicak, P.G., Bresnahan, D.B., Sharma, R., Davis, J.M., Comaty, J.E., and Malinick, C. (1988). A comparison of thiothixene with chlorpromazine in the treatment of mania. Journal of Clinical Psychopharmacology, 8, 33–7.

105. Johnstone, E.C., Crow, T.J., Ferrier, I.N., et al. (1983). Adverse effects of anticholinergic medication on positive schizophrenic symptoms. Psychological Medicine, 13, 513–27.

106. World Health Organization, Heads of Centres Collaborating in WHO Co-ordinated Studies on Biological Aspects of Mental Illness (1990). Prophylactic use of anticholinergics in patients on long-term neuroleptic treatment: a consensus statement. British Journal of Psychiatry, 156, 412.

107. Fenton, W.S., Blyler, C.R., and Heinssen, R.K. (1997). Determinants of medication compliance in schizophrenia: empirical and clinical findings. Schizophrenia Bulletin, 23, 637–51.

108. Siris, S.H., Morgan, V., Fagerstrom, R., Rifkin, A., and Cooper, T.B. (1987). Adjunctive imipramine in the treatment of post-psychotic depression: a controlled trial. Archives of General Psychiatry, 44, 533–9.

109. Johnstone, E.C., Crow, T.J., Frith, C.D., and Owens, D.G.C. (1988). The Northwick Park ‘functional' psychosis study: diagnosis and treatment response. Lancet, ii, 119–26.

110. Lingjaerde, O. (1991). Benzodiazepines and the treatment of schizophrenia: an updated survey. Acta Psychiatrica Scandinavica, 84, 453–9.

111. Brandon, S., Cowley, P., McDonald, C., Neville, P., Palmer, R.R., and Wellstood-Eason, S. (1985). The Leicester ECT trial: results in schizophrenia. British Journal of Psychiatry, 146, 177–83.