9.7 Public mental health
Oxford Textbook of Public Health
Public mental health
Peter Tyrer and Freya Tyrer
Historical review of mental health service developments in relationship to public health
Reasons why public health has historically ignored mental health
Classification of mental illness
Epidemiology of mental disorders
Neurotic, stress-related, and somatoform disorders
Behavioural syndromes associated with physiological disturbances and physical factors
Psychiatric services and public mental health
Schizophrenia and substance abuse
Mood and adjustment disorders
Personality and mental state disorders
Significance of diagnostic distribution of psychiatric disorders in public health
Evidence of public health efficacy of mental health services
Recent shift from mental health of individuals to whole communities with shift of public health into purchasing field
Targets for mental health services
Improving mental health of populations
Public mental health is a concept that has grown in recent years. In the past much of mental health has been in the private domain. We use the word ‘private’ in all its different meanings, as for many years mental health treatment (as opposed to custodial care) was only available to those with private means, the care of people with mental health problems has generally been a private secretive exercise, and because of the damaging influence of stigma there has been a tendency to keep much of mental illness shrouded in euphemism and understatement so that official statistics are of much less value than in other medical disciplines. However, with the growth of public mental health, a discipline which acknowledges mental health pathology in all settings from the special hospitals for forensic psychiatric patients through to large populations in the community, we are now beginning to obtain a comprehensive picture of mental illness which is immeasurably better than it was 10 years ago. The main disadvantages of private mental health are illustrated by the classical work of the psychoanalyst. The patient (or population with a specific psychiatric disorder) presented to the analyst is assessed and offered flexible, sometimes unlimited, treatment with no agreed measures of outcome (as the intention is for these to be developed by patient and therapist as treatment proceeds), and the outcome, even if positive, is usually of very little help in generalizing to other patients with psychiatric disorders. It adds to the simile to note that the psychoanalyst carries out most of the treatment sitting behind the patient who lies on a couch with little to look at apart from the ceiling. Although this is said to aid treatment, it does not immediately give the impression of improving communication.
However, in the last 10 years, and in particular since the last edition of this textbook, mental health services are being audited and examined in a way which is quite newand which has led to many difficulties. In the United Kingdom, targets in each of the key areas of health were identified at the beginning of this decade, but unfortunately it has to be admitted that there is no satisfactory outcome measure of public mental health that can be easily recorded and which could serve as a yardstick (Anonymous 1992). However, one measure has been selected—prevention of suicide—and although many in the mental health services were unhappy about this being regarded as an index of effectiveness of mental health interventions, it has been of considerable value. The British government set a national target of reducing the death rate from suicide and ‘undetermined injury by at least a further sixth (17 per cent) by 2010, from a baseline at 1996’ (DoH 1999), and in this regard they appear to have been bolstered by the likely success of their first target—to reduce suicide rates by 15 per cent between 1990 and 2000.
In this chapter we aim to introduce the reader to the wealth and breadth of psychiatric disturbance and to open some of the secret doors that have helped to make the discipline a private one but hindered its public health role. In illustrating the disorders, an epidemiological perspective will be adopted before describing the various interventions that are available and which can generally be described as evidence-based. In the final section we discuss the important question of need in mental health services, both met and unmet, and its implication for the public health physician.
Historical review of mental health service developments in relationship to public health
It is difficult to know to what extent historical developments in mental health are important in understanding the present and future. However, study of the public mental health aspects demonstrate that throughout the centuries there has been an ebb and flow of official attitudes towards mental illness that have affected its practice enormously. At one extreme was total integration of the mentally ill into the rest of the community with no special provision for their care and the belief that such people constitute a common responsibility for the community. This is perhaps best exemplified by mental health care in Ireland a thousand years ago (Robins 1986). At the other extreme is the isolation of the mentally ill from the rest of the community, partly because of stigma and partly because of fear, exemplified by the massive growth of mental hospitals in the United Kingdom in the nineteenth century (Scull 1959).
Although it might be expected that, with increasing knowledge, there would be a steady move towards a more enlightened view of mental illness, the fluctuations of the past do not inspire confidence, and in recent years there has been a strong reaction against the growth of what has become called ‘community psychiatry’ and the policies of incarceration, or the equivalents in less emotive language, are becoming increasingly attractive again (Coid 1994).
What has altered over the years is the interpretation of mental illness in its sufferers. For centuries there has been a distinction between those mental illnesses which are associated with an apparent loss of reason, the psychoses, from those in which distress is present but reason is maintained and the so-called ‘reality-based disorders’. Most of the latter were formally classified under the general heading of ‘neurosis’, a term which physicians are now taught to ignore as an outdated relic of psychoanalysis (Bayer and Spitzer 1985), but which still has its adherents (Tyrer 1985).
The psychoses have led to fear and loathing in equal measure. The most obvious interpretation of the totally unreasonable behaviour of psychotic disorders is that the person has been possessed, and this has been a common explanation throughout the ages. Supernatural possession is the most common attribution, and whereas innocent possession by alien influences is sometimes inferred on most occasions it is assumed that the devil or other malevolent influences are primarily responsible. There has also been a suspicion that the mad are often clever and devious. This is perhaps expressed most graphically in Shakespeare’s comment about Hamlet’s madness ‘If this be madness, yet there is method in’t’. There has also been the suspicion that mad people conspire together as a consequence of supernatural influences, and it is this that has provoked the most fear and consequent retribution; the execution of the witches of Salem in 1692 is a prime example.
Over the course of the past 200 years all forms of mental illness have gradually been brought together as varieties of sickness. Although little treatment was available, such people were cared for as though they were unwell, usually in almshouses or similar places of refuge, and the word ‘asylum’ developed in this context. However, as the mentally ill became increasingly isolated from other members of the community, the stigma of mental illness increased and integration back into the community became more difficult. Despite this, the period in which mental hospitals grew and isolation of the mentally ill increased was still the most formative in our understanding than any previous period. In the United Kingdom, merely 1046 people with mental illness were treated in hospital settings in 1827, but this had risen to 74 000 by the turn of the century and to a peak of 155 700 by 1959.
The growth of mental hospitals was greatest in countries influenced most by the Industrial Revolution and the consequent growth of large urban conurbations. The famous treatise of Faris and Dunham (1939) showing that schizophrenia was much more prevalent in urban than rural areas was adumbrated nearly a century earlier by Dorothea Dix who correctly inferred that ‘there are, in proportion to numbers, more insane in cities than in large towns, and more insane in villages than among the same number of inhabitants dwelling in scattered settlements’ (quoted in Porter 1991).
As mental hospitals increased in size and number it was realized that they sometimes created many more problems than they solved and the phenomenon of institutionalization was recognized (Barton 1959). Public opinion, a fickle but extremely important agent of change in psychiatry, steadily turned against mental hospitals which were heavily criticized from both a sociological viewpoint (Goffman 1961) and by the leaders of what became the antipsychiatry movement (Szasz 1960; Laing and Cooper 1964; Laing and Esterson 1970; Foucault 1973). Although the entry psychiatry phase of mental health was only brief, and had much more influence on literature than on psychiatry, it has recurred at various times throughout the history of the subject. In simple form it states that madness is a form of escape from intolerable external pressures—pressures induced by society and the family and not internally (endogenously) produced. It is an attractive hypothesis (hence its appeal in literature) but it has rarely, if ever, shown any value, except indirectly as offering a more humane and understanding approach to people with schizophrenia. Despite this, it continues to attract strong advocates who do not like data to get in the way of a good story, and there continues to be a minority of individuals who believe that schizophrenia and other major mental illness is a consequence of our abnormal society rather than genuine mental illness.
Reasons why public health has historically ignored mental health
Public health medicine has generally had few links with mental health until recently. However, with the growth of the discipline in the nineteenth century it was acknowledged that amidst the epidemics of diseases such as cholera and typhoid, and the need for preventive measures to avoid these (Chadwick 1842), there also needed to be adequate care for the mentally ill. To some extent the growth of mental hospitals followed from a public health policy: people with mental illness were vulnerable and had to be cared for in safe accommodation. The light and airy environment of the country asylums was a much healthier one than the squalor of the inner city.
It was argued that improving the productivity of the population, and spending money for the relief of the poor, led to an improvement in moral and mental behaviour. Early epidemiological studies were confined to the asylums. In 1828, the first statistics were recorded in the United Kingdom of those resident in the asylums and madhouses of the cities by the Lunacy Commissioners. At the General Register Office William Farr carried out the pioneering work in public mental health, and one can regard the birth of public mental health as probably beginning when he published his report on the mortality of the insane (Farr 1864).
In the early twentieth century, public health was preoccupied with preventive medicine and public hygiene and later became involved with surveillance of communicable diseases and the administration of health facilities in different parts of the country. It was not until the late 1950s that an epidemiological approach to public health became dominant again. The data derived from epidemiology informed the local provision and use of resources and this allowed mental illness to be examined in the same way as physical disease. In the early 1950s, most mental health treatment was carried out in asylum settings The peak occupancy was reached in the middle of this decade, but since then has fallen steadily to only a third of its original level. However, this change initially had little impact on public health in general, even though the implications were obvious after the seminal paper of Tooth and Brooke (1961) which predicted a dramatic fall in the resident population of psychiatric institutions.
One of the reasons why mental health did not figure highly on public health agendas was the major difficulty in measuring mental health. However, in the last 25 years there have been major advances in the measurement of mental illness. A large number of questionnaires and rating scales (e.g. the General Health Questionnaire (Goldberg 1972) and the Hospital Anxiety and Depression Scale (Zigmond and Snaith 1983)) have been developed which, despite some limitations, have allowed the symptoms of psychopathology to be measured reliably and accurately in public health settings. Better diagnostic criteria have increased confidence in the two main classification systems—the International Classification of Diseases (ICD) and the Diagnostic and Statistical Manual of Mental Disorders (DSM). The latter underwent a major reform in 1980 (American Psychiatric Association 1980) and is mainly responsible for the current groupings. Although the rigidity of these new systems has often been criticized, their introduction was a major advance in the classification of mental illness.
These developments, together with the policies of governments of all political persuasions to transfer the provision of services to the community, has forced mental health onto the agenda of public health. Mental health has developed a public voice with the inception of mental health advocacy organizations such as, in the United Kingdom, MIND, Sane, and the National Schizophrenia Fellowship. However, it is only in the last few years that public health has taken over an overarching role in advising and monitoring those involved in purchasing and providing mental health services. The view that ‘epidemiology should derive public health and that public health medicine should be responsible for assessing the health of the population, setting local objectives for health, specifying effective and efficient use of resources and monitoring the achievement of targets’, is now public policy (Acheson 1988). Since mental health services account for at least a tenth of the health budgets of virtually all countries, it was not surprising that public health medicine should take a much closer interest in mental health than it had ever done before.
Classification of mental illness
The system of classification of mental disorders has long been a subject of heated debate, disagreement, and criticism, both from within the medical profession and without. The current European classification has evolved over the years through collaboration, research, and debate, involving field trials in over 194 different centres in 55 different countries. The ICD was instituted in Paris in 1900 and has been revised regularly to the current 10th revision (ICD-10), where Chapter V(F) relates to mental and behavioural disorders. The equivalent American classification system (DSM-IV) is roughly similar in its categories to ICD-10 but there are some important discrepancies that hinder interpretation of worldwide epidemiological data. Poor concordance is shown for stress disorders, substance misuse, and some of the anxiety disorders (mainly involving panic) (Andrews et al. 1999), and so comparative figures for these diagnoses are questionable where ICD and DSM systems are both included.
Initially, the classifications provided brief thumbnail sketches of clinical concepts. Over the years the definitions of the various syndromes and disorders have become tighter and more rigorous, and both ICD-10 and DSM-IV now provide stricter diagnostic criteria to aid both clinicians and researchers to achieve greater diagnostic agreement. Although it is impossible to do justice to the main psychiatric syndromes in a short chapter, it is important for public health physicians to know the main features of each. Increasingly those with mental illness are being subdivided by diagnosis, and it is important to know something about the relative accuracy and nature of the psychiatric diagnostic process.
The classification of the main disorders is presented in Table 1. In the past, because of the heterogeneity of diagnostic criteria, the apparent incidence and prevalence of certain disorders appeared to vary considerably in different countries. Thus one study, the United States–United Kingdom Diagnostic Project, was able to show that the apparently high levels of schizophrenia in the United States compared with most European countries did not reflect a true difference in prevalence, but rather that different diagnostic criteria were being used for the same illnesses (Cooper et al. 1972). These differences are much less now; the International Pilot Study of Schizophrenia demonstrated, using the same criteria, that there were similar rates of schizophrenia in most of the countries in the study (Jablensky et al. 1992). This is of particular note since the countries included in the study were so varied (United Kingdom, United States, Taiwan, Switzerland, Russia, Nigeria, Colombia, Czechoslovakia, and Denmark).
Table 1 Classification of psychiatric disorders according to ICD-10
Further research into depression showed similar results, namely that the cross-cultural similarities were greater than anticipated. The World Health Organization (WHO) collaborative study assessment of depression looked at five centres—Switzerland, Canada, Iran, and Japan (two centres). This study showed that all countries with the exception of Iran (Tehran) had similar rates of depression, and that, despite a degree of variation in presentation, patients from all countries shared a core symptomatology of depressive symptoms (Sartorius et al. 1980).
Epidemiology of mental disorders
In an earlier edition of this book, it was emphasized that the backbone of public health strategy is the development and maintenance of an ‘accurate, reliable health information system upon which actions can be based’ (Detels and Breslow 1986). It has taken some time to develop an appropriate information base for mental disorders, but we have become much closer to it in the last 10 years. The epidemiology of the major mental disorders described in Table 1 is shown in Table 2 and will be described further according to the main groups in ICD-10. However, learning disability (mental retardation) and child disorders are not included in detail here.
Table 2 Annual community prevalence of the main groups of psychiatric disorders
The figures taken overall show that mental disorders are surprisingly common, with approximately one in three of the population having had a disorder which is sufficiently severe to warrant a diagnostic label, at least to a trained lay interviewer. However, five out of every seven patients identified in national surveys do not seek or receive any treatment for their disorders (Regier et al. 1984, 1998). Nonetheless, the figures from epidemiological surveys are likely to be underestimates, so that what Henderson (2000) calls the ‘exaggerated estimate’ problem may be incorrect. Only around 80 per cent of people complete epidemiological surveys, and the remaining 20 per cent who are never interviewed are likely to have significantly greater pathology than those who are interviewed.
This is one of the least satisfactory groups in the classification system, largely because the diagnosis of an organic disorder is based on identification of an organic cause and this is difficult to determine for many conditions until after death. It also includes a very disparate group of conditions, ranging from delirium at one extreme—short-lived episodes associated with confusion and transient psychotic symptoms—through to organic causes of anxiety, depression, personality change, and schizophrenia, and finally to the progressive dementias, which are much more important in public health terms because of their scale. Good epidemiological data are difficult to obtain for most of these, but the dementias are better understood.
Dementia is an acquired global impairment of intellect, memory, and personality occurring in a setting where there is no impairment of consciousness. It is detected first by evidence of increased forgetfulness and personality change. Loss of recent memory is the simplest way of identifying dementia by simple psychological tests; long-term memory is preserved until later in the condition. Two conditions—Alzheimer’s disease and vascular dementia—account for most of the dementing illnesses, with normal pressure hydrocephalus, Pick’s disease (affecting the frontal lobes initially), Huntington’s chorea, AIDS-related dementia, and Creutzfeldt–Jakob disease (with its links to a strain of bovine spongiform encephalopathy) occurring much less frequently. Alcoholic dementia is included among the substance misuse disorders. The incidence of Alzheimer’s disease is twice as great as vascular dementia, representing 2.7 per 1000 person-years at risk and vascular dementia 1.2 per 1000 in people over the age of 75 years (Brayne et al. 1995b). The risk of dementia doubles with each successive 5-year period over the age of 75 years (Paykel et al. 1994), and the data indicate that cognitive decline represents a continuum of which dementia is at the extreme (Brayne et al. 1995a). Vascular dementia is much less age sensitive than Alzheimer’s disease (Brayne et al. 1995b).
In both ICD-10 and DSM-IV substance misuse is classified according to the main substance involved (e.g. cocaine, amphetamine, alcohol dependence, nicotine, and caffeine). The nature of the misuse (e.g. intoxication, dependence, withdrawal, or psychosis) is then classified separately. Substance misuse has now become recognized as a subspeciality of psychiatry and in some areas there are special drug dependency centres or units. Those with dual diagnosis tend to be seen by general psychiatrists, but this is likely to change (Weaver et al. 1999).
Dependence on a drug includes several elements (not all of which need to be present to satisfy the diagnosis): a persistent wish to obtain the drug (drug-seeking behaviour), the development of tolerance with increased use, and the exhibition of a withdrawal syndrome after sudden cessation of the drug. Long-term complications of drug misuse include petty criminality (a large proportion of theft is carried out by addicts), social and occupational decline, and homelessness. There are also the risks of physical disease including hepatitis, oesophageal varices, hepatic cirrhosis, and dementia in clients abusing alcohol, and hepatitis B and C and HIV/AIDS in intravenous drug abusers. Substance misuse problems are growing in most countries of the world and, to date, no public or mental health intervention has had a significant preventive impact.
Schizophrenia is a syndromal diagnosis which, despite much interest in recent years, remains an enigma. It was originally described as dementia praecox and although this was later ridiculed when the condition became a ‘functional psychosis’, it has now been found that there is loss of brain tissue in many sufferers, with the whole brain, amygdala, and hippocampus being smaller than in controls (Lawrie et al. 1997). There is still a limited understanding of the aetiology and causal mechanisms involved, which explains why one of our foremost authorities on schizophrenia bleakly concludes that ‘as far as defining the clinical boundaries of schizophrenia/dementia praecox are concerned we have not moved very far in the past 100 years’ (Johnstone 1999).
Bleuler first coined the term schizophrenia in 1911 to refer to a group of psychoses which were characterized by splitting of mental functions, with delusions and hallucinations secondary to this splitting. Since then the definition of what constitutes schizophrenia has been revised and amended. Affect (the formal psychiatric word for mood), thinking, perception, and behaviour are all impaired in schizophrenia.
The characteristic affect is described as incongruent and inappropriate, with individuals laughing in a silly manner when discussing a serious matter. Alternatively, the affect may be blunted and the patient appears somewhat unresponsive, demonstrating a limited repertoire of apparent emotions. The patient may be formally thought disordered with incoherent illogical thought and lack of understandable associations between thoughts (derailment). This may be mild, making the patient difficult to follow at interview, or severe in which case it may not be possible to understand the patient at all. Poverty of content of thought may also be a characteristic feature. The nature of the content of thought may be abnormal, and patients often have delusions. Delusions can be varied and include persecutory, grandiose, and referential delusions, as well as more bizarre delusions such as control by external forces, mind reading and insertion, withdrawal, and broadcasting of thoughts. Perception is impaired and auditory hallucinations are common, particularly of voices discussing the patient in the third person, giving a running commentary about what the patient is doing, or giving commands. Other perceptual abnormalities include tactile (or somatic) hallucinations, visual hallucinations, and misperceptions not amounting to hallucinations. The patient’s behaviour may exhibit markedly abnormal behaviour because of responses to delusional beliefs or hallucinations, whereas social isolation and self-neglect seem to be more integral features of the underlying disease process.
Patients with schizophrenia are a broad group and have many ways of presenting to psychiatric services. At one end of the spectrum, patients present with predominantly negative symptoms associated with absence of motivation (social withdrawal, apathy, neglect of self-care). At the other end of the spectrum patients show predominantly positive symptoms—hallucinations, delusions, and abnormal behaviours, such as agitation, aggression (which is a serious public and community health issue), and posturing or stereotyped behaviour.
The classification of mood (or affective) disorders is probably the most complicated and controversial area of classification in psychiatry. These difficulties have largely been removed by recent classifications (ICD-10 and DSM-IV), which use a simpler atheoretical system. Now all mood disorders are classified together under the same rubric of affective disorders and largely classified in terms of severity of symptoms and associated disability. The previous separation of psychotic and neurotic depression and the similar separation of reactive and endogenous depression have been abandoned because they have failed to separate homogeneous diagnostic groups. Similarly, individuals with a chronic state of mild depression (dysthymia) are now classified within the affective disorders rather than with patients with personality disorders. Classification is based on symptomatology rather than aetiology.
An affective illness is primarily characterized by the quality of the affect or mood. The distinction is made between unipolar (depression only) and bipolar illnesses (depression and mania at different times). If a patient has an elated mood, this will suggest a diagnosis of hypomania or in more severe cases mania. A lowered mood with predominant feelings of worthlessness, guilt, self-reproach, and pessimism are suggestive of depression. The severity of the condition can further be defined as mild, moderate, or severe, further qualified as bipolar disorder if there has been a previous episode of mania or hypomania, and as a recurrent episode if there have been previous episodes of mania or depression. Dysthymia is a low-grade chronic depressive condition which is relatively common in the population (Table 2) and which shows strong comorbidity with other depressive conditions (sometimes called double depression).
Symptoms of depression include the following:
insomnia, particularly early morning wakening
diurnal variation of mood (feeling worse in the morning)
loss of appetite with weight loss
loss of libido
objective evidence of psychomotor retardation.
In addition the mood disorder can be yet further defined by the presence or absence of psychotic symptoms—hallucinations and delusions—which in contrast to schizophrenia are usually mood congruent (entirely consistent with the prevailing mood such as delusions of poverty in severe depression).
In severe depression there may be total inactivity (stupor), or the patient expresses nihilistic delusions that they do not exist or that parts of their body have died or stopped working. This is also sometimes described as Cotard’s syndrome, and is more prevalent in the elderly (Luque and Berrios 1994).
Neurotic, stress-related, and somatoform disorders
The disorders discussed under this heading fall into a number of different categories but have been grouped together for convenience. This is probably a more heterogeneous group than others, although all these conditions are ‘reality-based’ in that the symptoms are understandable and based on real experiences, feelings, and events.
These conditions are the most prevalent psychiatric disorders. They include anxiety disorders (generalized anxiety disorder), phobic disorders (agoraphobia, social phobia, and simple phobia), panic disorder, obsessive–compulsive disorder, reactions to acute stress and more prolonged stressors (adjustment disorder), reactions to very severe stress (post-traumatic stress disorder), dissociative (conversion) disorder (formerly called hysteria), and the somatoform disorders, in which the major symptoms are bodily ones. Mixed anxiety–depressive disorder is also included in this group; there are arguments whether this should be a subsyndromal rather than a full mental disorder. The neurotic and stress-related disorders are generally more difficult to classify than other psychiatric disorders, which is illustrated by the high levels of comorbidity shown in Table 2, and the dividing line between normal variation and pathological condition is often very difficult to draw.
The prevalence of any neurotic disorder in the previous week was recorded in a recent United Kingdom study (OPCS 1995) of 10 000 adults in private households (Fig. 1). One in seven had a psychiatric disorder, with mixed anxiety–depressive disorder (7 per cent) and generalized anxiety disorder (3 per cent) being the most common. These disorders were almost twice as common in women (male-to-female ratio of 6 to 11) and had similar prevalence in all age groups between 20 and 60 years of age (Fig. 1).
Fig. 1 Prevalence of any neurotic disorder: mixed anxiety and depressive disorder, generalized anxiety disorder, depressive episode, all phobias, and obsessive–compulsive disorder. (Data from OPCS (1995).)
Behavioural syndromes associated with physiological disturbances and physical factors
This is a smaller group comprising eating disorders (mainly anorexia nervosa and bulimia nervosa), sleep disorders, and sexual dysfunction. These are linked to neurotic disorders but have specific physiological consequences (e.g. amenorrhoea in anorexia nervosa) that do not apply to other disorders. Occasionally there are apparent ‘epidemics’ of eating disorders, particularly in girls’ schools.
It is only in the last 20 years that this concept has been refined sufficiently to stand with other psychiatric diagnoses and the categories of personality disorders have been more rigidly defined. There are numerous definitions of what constitutes a personality disorder but the main characteristics are a persistent abnormality in personal relationships, attitudes, and behaviour that develops early in life and leads to impaired social functioning.
Current psychiatric classifications adopt a categorical system although there is evidence that a dimensional approach is more valid. The main groups are listed in Table 3 in which the ICD-10 and DSM-IV classifications are compared.
Table 3 Classification of personality disorder in the international (ICD-10) and American (DSM-IV) systems of classification
The categorical system is not satisfactory because of the degree of overlap (comorbidity) between different diagnoses. Increasingly it is being recognized that three or four clusters—flamboyant (dissocial, histrionic, narcissistic, borderline, and impulsive), odd or eccentric (paranoid, schizoid, and schizotypal), anxious or fearful (anxious, dependent), and obsessive–compulsive (anankastic)—comprise a somewhat better classification system (Tyrer et al. 1991; Mulder and Joyce 1997). There is also a dimensional classification which covers the range between normal and severe personality disorder (Tyrer and Johnson 1996). Measuring personality disorder in this way has increased in recent years and has been accentuated by the recognition that some people with severe personality disorder are a threat to the public mainly because of their propensity for unprovoked violence. In the United Kingdom legislation is proposed that will allow the indefinite detention of those with dangerous severe personality disorder (as yet not formally defined). This has important public health connotations.
Psychiatric services and public mental health
In examining the public health issues in mental illness it is useful to adopt the filter model of Goldberg and Huxley (1980) (Table 4). This helps to explain the importance of the differences between the primary care and secondary care psychiatric services. The five levels are as follows:
the population as a whole (level 1)
the total mental morbidity (i.e. those who attend primary care with mental ill health, whether or not it is detected) (level 2)
the proportion of those identified as mentally ill by doctors, now know as conspicuous psychiatric morbidity (level 3)
the population being seen by the mental illness services (level 4)
the relatively small numbers who are treated as inpatients in psychiatric hospitals (level 5).
Goldberg and Huxley separated these five levels by four filters (Table 4) through which patients normally have to pass through before reaching a higher level. However, it is perfectly possible to go directly from level 1 to level 5 (e.g. by presenting as a psychiatric emergency at an accident and emergency department), and in some countries with less co-ordinated health services this form of direct transfer across levels is very common. This is not just a function of whether or not the country is well developed in terms of affluence and productivity; Goldberg and Huxley (1980, pp. 53–5) describe the ‘American bypass’—a tendency in the United States for patients to go directly from community to psychiatrist.
Table 4 Five levels and four filters, with estimates of annual period prevalence rates at each level
The annual period prevalence rates are shown (Table 4) and in epidemiological terms may seem large. However, it is important to understand that many of these cases comprise a low level of illness, and much activity in psychiatric epidemiology is concerned with deciding on the threshold for ‘caseness’—a threshold that may be decided more by the procrustean requirements of the classification system than by the need for intervention (Regier et al. 1998). It is also important to appreciate that diagnoses within the field of mental illness are much less specific that with physical disorders. Clearly, the diagnosis of diabetes is unequivocal in almost every case, it is associated with disabilities that are clearly secondary to the main disorder (e.g. retinopathy), and it has a fairly predictable time course and outcome. In mental illness we have the difficulties of case identification, problems of classification into homogeneous groups, overlap between disorders with difficulty in deciding which is primary and which is secondary, and unpredictable outcome.
There are some important differences between the annual prevalence of different diagnostic groups in different settings (Fig. 2). It will be noted that by far the largest proportion of patients with psychiatric disorder in the total population (two-thirds of the total) consists of neurotic and stress disorders or personality disorders. Alcohol and drug problems are more common in the community than in those presenting in primary care but, as expected, all other problems are more frequent in primary care within the population. Neurotic and stress disorders become less prevalent as one moves along the different levels of psychiatric care but all other disorders, except affective disorders, become more pronounced as one ascends the levels to inpatient status. The figures for affective disorder include a wide range of conditions from mild depression through to bipolar disorder and in psychiatric settings bipolar disorder becomes more frequent.
Fig. 2 Differences between the diagnostic distribution of mental disorders in (a) the general population, (b) primary care (conspicuous morbidity) (level 3), (c) psychiatric services (level 4), and (d) psychiatric patients (level 5). Neurotic, neurotic and stress-related disorders; personality, personality disorders; alcohol, alcohol and drug abuse; affective, depressive and manic disorders; organic, organic mental disorders (chiefly dementia); schizophrenic, the group of schizophrenic disorders. (Data from Regier et al. (1984), Goldberg and Huxley (1992), and de Girolamo and Reich (1993).)
Therefore it can be seen that the overall numbers for rates of mental illness in various populations only reveal part of the story. It is important to realize that these numbers are not as reliable as those for other disorders because there is great difficulty in separating psychiatric disorders from each other and deciding which is primary. The numbers in Table 2 illustrate one of the problems: the sum of those for individual disorders comes to much more than the total prevalence of mental disorder. This is explained by comorbidity. Thus in the National Comorbidity Survey in the United States, more than half of all lifetime psychiatric disorders in the population occurred in the 14 per cent of the sample who had a history of three or more comorbid disorders (Kessler et al. 1994). This suggests that these conditions are not truly comorbid in the original sense of the word—independent conditions occurring at the same time in the same person (Feinstein 1970)—and at least some are intimately associated and therefore consanguid (Tyrer 1996). The problem is that it is impossible to determine the relationship between these disorders through cross-sectional studies. The arguments over whether the conditions should be conjoined or separated is not ultimately a useful issue in clinical practice, but if one of the conditions is a predisposing vulnerability factor or a complication of the other it is important and should be investigated. However, it is only through longitudinal studies that such associations can be determined.
Some of the common examples of comorbidity and the problems it creates refer to four areas: schizophrenia and substance misuse, depression, neurotic and adjustment disorders, and mixtures of mental state and personality disorders. Many of these are referred to as ‘dual diagnoses’, but there are now so many of these this is becoming a valueless label.
Schizophrenia and substance abuse
This is perhaps the original dual diagnosis (Weaver et al. 1999). Abuse of many drugs has been associated with psychotic symptoms that to some extent simulate schizophrenia. The classical example is amphetamine, which can, in acute dosage and also during withdrawal, lead to a paranoid psychotic state that is indistinguishable from paranoid schizophrenia (Connell 1958). Other drugs, notably LSD, have also frequently been implicated in the aetiology of schizophrenia-like psychoses and, more recently, there have also been claims that cannabis can also create a psychotic state although not normally to the extent that it could be regarded as a definitive psychosis (Thomas 1993). Benzodiazepine dependence may also be associated with psychotic disorders after withdrawal (Roberts and Vass 1986). Opioid drugs and cocaine do not create or trigger psychotic disorders.
Cannabis, benzodiazepines, and amphetamines do not create psychotic symptoms in most users, but in those who are vulnerable to developing schizophrenia these drugs may trigger episodes which would not have occurred without the drug exposure. This has important implications in the debate over legalization of cannabis, and we would predict an increase in such dual diagnoses if it were legalized. Diagnostic practice is very difficult when substance misuse is present with schizophrenic symptoms. Clinicians are taught to be parsimonious and to choose only one diagnosis; the outcome is that some diagnose patients who otherwise look identical as having a drug dependence problem whereas others would diagnose schizophrenia.
Mood and adjustment disorders
Mood disorders constitute a separate section in ICD-10, and in this context ‘mood’ includes the spectrum between depression and mania only. Depression is fairly easy to recognize as a symptom but notoriously difficult to diagnose into homogeneous groups, and in previous classifications the symptom was present in all parts of the diagnostic system. To try and resolve this, all depressive diagnoses are now incorporated in one section in ICD-10. Most of the cases consist of mild and moderate depression which is not associated with any psychotic disorder. However, in most mild mental disorder there are mixed symptoms of anxiety and depression and this apparent ‘comorbidity’ is difficult to resolve. Anxiety is now officially a ‘neurotic’ symptom whereas depression is a ‘mood’ symptom, and so separation has to be achieved unless one adopts the mixed (subsyndromal) diagnosis of mixed anxiety and depression allowed in ICD-10 (WHO 1992). There are some who argue for the diagnosis of mixed syndromes such as the ‘general neurotic syndrome’ to cover this group of disorders (Tyrer 1985; Andrews et al. 1990; Tyrer et al. 1992; Sullivan and Kendler 1998), but this continues to be a bone of contention among psychiatrists, as both the general neurotic single model and the two-factor depression–anxiety model can marshal equally convincing arguments in their favour (Jacob et al. 1998).
Another category, termed ‘adjustment disorders’, is also included under the stress disorders in ICD-10 and DSM-IV. In such disorders the patient may be anxious or depressed, but this is clearly related to an obvious psychosocial stressor which is regarded as the cause of the symptoms and which, when removed, will lead to resolution of the clinical problem in most cases. However, while the stressor is still present, the patient will have significant anxiety and depressive symptoms and in many cases this will be recorded through the appropriate diagnostic groupings. However, these are not mental illnesses in any formal sense, because all of us can suffer from adjustment disorders when under conditions of sufficient stress. In symptomatic terms, they are indistinguishable from other forms of anxiety and depression and so are normally included amongst them.
Personality and mental state disorders
Personality disorder has been difficult to diagnose in the past and is regarded in many quarters as a pejorative label. There has also been a tendency to diagnose patients as either having a personality disorder or a mental illness whereas in fact both can exist together as they do not measure the same domain of psychiatric function (Tyrer et al. 1991). In old statistics personality disorder usually accounts for only between 2 and 5 per cent of psychiatric disorders but when formal assessments of personality status are made in the populations concerned the proportions increase several-fold (Tyrer and Ferguson 2000). The results in Fig. 3 are derived from a variety of studies in which such formal assessment of personality status has been made (de Girolamo and Reich 1993; Casey 2000). They show that personality disorder is remarkably common, particularly in psychiatric inpatients.
Fig. 3 European standardized mortality rates of suicide and undetermined injury, by social class. Men, aged 20 to 64 years, England and Wales, 1991 to 1993. (Data from Drever and Bunting (1997).)
Significance of diagnostic distribution of psychiatric disorders in public health
The data show that the population with psychiatric disorder in the community consists almost entirely of people with some degree of personality disorder who have anxiety and depressive conditions, possibly in association with stressors. Other conditions, with the exception of alcohol abuse, account for very few cases of psychiatric disorder. By contrast, those seen by psychiatric services consist of significant proportions of patients with schizophrenia and mood disorder (including a higher proportion of patients with affective psychosis in which both mania and depression may be present at different time) and a relatively low proportion of those with neurotic and stress-related disorders.
This has to be borne in mind when considering interventions in mental illness. Most of the minor disorders are self-limiting and even if the general practitioner does not have a good detection rate for those who present with mental illness, the amount of distress caused is still relatively small (as what is lost is the more rapid treatment of a condition that will resolve in any case). However, in the psychiatric services there is a greater preponderance of the disorders that need active treatment or have persistent disorders that are extremely difficult to treat (e.g. personality disorder) and the disorder with the greatest disability, schizophrenia, is over 40 times more prevalent in inpatient settings than in the population as a whole.
Because these figures conceal a great deal of variation it is sometimes useful to give the rough limits for the prevalence of each condition within a population. These are shown in Table 5 for a population of 500 000 (including 70 000 over the age of 65 years).
Table 5 Prevalence of and needs for the main psychiatric disorders in a population of 500 000
Evidence of public health efficacy of mental health services
Recent shift from mental health of individuals to whole communities with shift of public health into purchasing field
Since the Acheson Report Public Health in England highlighted public health medicine’s role in reviewing the health of the population, there has been greater awareness of the need to examine mental health as an essential component of population health. As departments of public health in health authorities are now responsible for purchasing mental health services, they need much greater knowledge of the workings of their local mental health provision. The mandatory annual report of the health of the population in the local area also needs to address the mental health of this population. This is a difficult task. Knowledge of whether mental health services are working effectively are seldom possible through official statististics or so-called ‘quality indicators’ such as waiting times or proportion of attenders to non-attenders. Intimate knowledge of the services is necessary and this can often be boosted by data from patients’ organizations or patients themselves, as their levels of satisfaction are better indicators than most others of good functioning services (Shipley et al. 2000). Because patients with severe mental illness are often extremely mobile, there is a tendency for disproportionately large numbers to present to psychiatric services in some areas, of which inner cities are the best example. These people do not appear on census figures and therefore their numbers cannot be predicted by sociodeprivation indices such as Jarman scores (Jarman 1983). Some are regarded as ‘psychiatric tourists’ and may create imbalance in local services (Montgomery and Parshall 1998). These issues need to be considered by public health physicians when planning and funding local services if this is to be done sensitively and appropriately.
Inequalities in health provision are currently high on the political agenda. The recently published Report of the Independent Inquiry into Inequalities of Health (Anonymous 1998) focuses on social inequalities and their relationship to health and suicide. It aims to improve ‘socioeconomic and living conditions and social cohesions’ which will ‘have many benefits in addition to their contribution to the prevention of suicide’. In this report young people are targeted, particularly men and the mentally ill. The suicide rate in those in social class 5 is three times greater than in those in social class 1 (Drever and Bunting 1997) (Fig. 3), and so when areas of great social deprivation are found to have higher suicide rates it should not be assumed that they have poor local psychiatric services.
Measuring outcomes in mental health services represents a major challenge to public health as well as an important responsibility. Public health now has to grapple with questions that have bothered the mental health services for years: How is it possible to decide whether a service is providing good or poor care? What targets and outcome measures are available? The need for an easily completed outcome measure (which can also be used as a record of mental health status) has been met to some extent by the development of a Global Assessment of Functioning Scale (Endicott et al. 1976) which can be separated into scaled measuring clinical symptoms and social functioning. This is now widely used in the United States and is the fifth axis in DSM-IV (American Psychiatric Association 1994). In the United Kingdom, in response to the Health of the Nation initiative, Health of the Nation Outcome Scales have been developed to record outcome in the main areas of change and have commanded wide use (Bebbington et al. 1999). The major problem remains that any instrument that is carried out with an unco-operative and psychotic patient cannot be assessed by self-report and the reliability of instruments such as the Global Assessment of Functioning is not particularly good.
Targets for mental health services
In 1992, targets were set for mental illness for the first time, most notably in The Health of the Nation, a United Kingdom White Paper setting forward one of the clearest statements on public mental health ever. Like all good public health it emphasizes that the prevention of ill health and promotion of good health is as least as important as treatment of illness in its various forms. Unfortunately, when mental health is considered (and it is a key area) most of the targets evaporate because, to use the words of the document ‘there is at present no straightforward and objective way of describing, aggregating and monitoring outcomes of care nor any agreement on clear and reliable measures which could confidently be used as proxies for outcome measures’ (Anonymous 1992).
However, there is one unambiguous measure of outcome that is addressed prominently and has been adopted as a major target: suicide. In The Health of the Nation (Anonymous 1992) two targets were set: to reduce the overall suicide rate by at least 15 per cent from 11.1 per 100 000 in 1990 to no more than 9.4 per 100 000 population by the year 2000, and to reduce the suicide rate of severely mentally ill people by at least 33 per cent over the same period. Bearing in mind that the suicide rate in the United Kingdom is considerably lower than in most other countries in Europe (Denmark, Finland, Austria, Switzerland, France, and Belgium all have rates greater than 20 per 100 000 population (Moens 1990)), the target appeared to be ambitious but looks likely to be met.
There have been doubts as to whether these targets are achievable by improvements in mental health services (MacDonald 1993), but these are not the only ways they can be achieved. Indeed, the steadily falling suicide rate in older people compared with the rising rate in the young (Charlton et al. 1993) means that demographic factors may achieve part of the reduction without any other form of intervention. Suicide is also a good index of priority for mental health services. Almost all mental illness—schizophrenia, affective (manic–depressive) psychosis, substance misuse, learning disability, personality disorder, dementia, panic and anxiety disorders, neurotic, stress and adjustment disorders—have standardized mortality ratio suicide rates that are at least three times greater than expected, and that for psychiatric inpatients exceeds 1000 (Harris and Barraclough 1998).
Doubts still persist; severe mental illness is not defined in any formal classification, and suicide figures can include or exclude many forms of undetermined death that could have an important impact on future figures. Nevertheless, the danger that the figures can be massaged in various ways to achieve the targets unfairly is much less likely with the figures for suicide than any other measure of mental illness. Better measures of outcome in mental illness are urgently needed but the suicide rates have set down a useful yardstick for comparison.
Our Healthier Nation (DoH 1999) was developed as a sequel to The Health of the Nation (Anonymous 1992), but with broader targets concentrating on improving health and narrowing the health gap. It stresses the widening of health inequalities in the United Kingdom, and one of its main aims is ‘to improve the health of the worst off in society and to narrow the health gap’ (DoH 1999).
Improving mental health of populations
All public health physicians are supporters of prevention. If important factors responsible for poor mental health can be identified, these can then be addressed and altered, and if identified correctly, improvement in the mental health of the whole population follows.
Community intervention studies
One of the major problems with severe mental illness is that patients do not present voluntarily to the psychiatric services at an early stage. However, it has often been suggested that if care was available at an earlier stage than at the point of hospital presentation or at the accident and emergency department, better care could be given and unnecessary suffering avoided.
A series of programmes which collectively are often described as ‘assertive community treatment’, generating from the work of Stein and Test (1980), has demonstrated that such intervention can reduce admission to hospital. Much of this work has been carried out with the severely mentally ill, who account for most of the work of the psychiatric services (Table 3). It has been shown that Stein and Test’s model is replicated in the United Kingdom (Muijen et al. 1992) and that it is effective. It has also been shown that day hospital care can be a useful substitute for inpatient treatment (Creed et al. 1990) and that improving care outside hospital, both by home visiting (Burns et al. 1993) and early intervention (Merson et al. 1992) can reduce the number and duration of admissions significantly. In the last few years in countries with publicly funded mental health services there has been a growth of multidisciplinary community mental health teams, and these have been shown in a recent systematic review to not only reduce hospital admissions and improve patient satisfaction but also reduce suicide rates (Tyrer et al. 1999).
These studies have been carried out by special teams, often created for research purposes, and the findings have not been generalized to services as a whole. However, there is now evidence that those services which have a higher level of community resources do have lower admissions to the country as a whole (Jarman et al. 1992). What is much less clear is whether the assertive treatment model is appropriate for all countries and all settings. It has certainly proved to be effective in the United States but here the standard service for state patients is often poor (Lehman and Steinwachs 1998). Where standard services are reasonably competent in administering evidence-based psychiatry, and where there is an existing comprehensive network of services, the apparent gains of assertive community treatment and its rough equivalent, intensive case management, are lost (Holloway and Carson 1998; Thornicroft et al. 1998; UK700 Group 1999).
Care programming and care management
For the last 5 years, the care of psychiatrically ill patients has been formalized in a new configuration. The two central elements are care programming and care management. Care programming applies to all individuals seen by the psychiatric service, and extends from a single assessment by a member of the mental health team followed by discharge, to a multidisciplinary arrangement whereby several workers from mental health, social, and voluntary services are all integrated together in providing care. This is too large a range to have much meaning, and the phrase ‘discharged into the community under the care programme approach’ means the same if the last five words are removed. Care programming overlaps with care management, a term which describes the assessment of needs by social services, who hold the budget for the provision of such needs and therefore prioritize services. The intention is that patients with severe mental illness with complex needs will receive both care programming and care management and ensure that proper allocation of funds is made, whereas those who have less severe needs can be cared for using care programming alone.
There are many aspects of this system which are unsatisfactory and which are difficult to address by purchasers of services. The notion that ‘packages of care’ can be designed on the basis of needs assessments (see below) is reasonable in principle but impossible to achieve in practice for many people. Needs can change unexpectedly during care and are difficult to anticipate, and many psychiatric patients change their minds and alter their levels of co-operation with treatment..
Care programming and care management are British terms but they are similar to case management in the United States and contain some features that are found in managed care in that country. In October 1994, the supervision register was introduced, owing to concern about the risks created by those with the most severe mental illness and who need the closest supervision. Any patient with severe mental illness who has a significant risk of self-harm, self-neglect, or is dangerous to others, is placed on the register and requires regular reviews.
Unfortunately, all these systems have been introduced without any additional resources and, whilst they may be indicative of good practice and therefore should be followed in any case, the additional bureaucracy involved in their implementation necessitates some additional resources. It is a difficult decision for public health departments involved in commissioning services to decide what level of resources is required and whether or not the implementation of these new arrangements does improve the total mental health of the populations.
It will be seen from the above account that needs assessment is an essential component of care management and, indeed, of care programming. Although this has a central role in public health strategies, it is important to be aware of its limitations. Needs assessment is ideal for conditions that are well researched, unequivocal in nature, and for which clear guidelines of therapy exist. Thus an elderly person who fractures the neck of a femur has a clear pathway set out for treatment and aftercare that can be easily monitored and costed. The pathway of, for example, a patient admitted with an episode of schizophrenia is very different. Some patients with schizophrenia improve dramatically after a few days and do not have further episodes, whereas other require care almost indefinitely, are unable to look after themselves, and need permanent support and accommodation. The cost of treating the episode for the first group of patients is unlikely to be more than around £1000, whereas treatment for the second group can cost well over £1000 000. Deciding on the relative needs of these individuals, particularly early in treatment, is extraordinarily difficult and often the outcome represents little more than a set of formalized value judgements (Slade 1994).
Stevens and Gabbay (1991), in criticizing much of the literature on needs assessment, emphasize that ‘needs’ for health care are quite separate from ‘wants’ and must include the population’s ability to benefit from health care and health service interventions. One approach to needs assessment is to consider the prevalence and incidence of mental disorders at both national and local levels and to use this in order to inform the commissioning of mental health services. Unfortunately, this can only be done well when the effectiveness and costs of mental health services and interventions are available. Because most of the information is insufficient, models of service care are often used in order to help purchasers decide which approach is most valuable. Although these can be of use, it is important to realize that they do involve a great deal of guess work. At present it is probably better to use ‘proxy measures’ such as the Jarman score of the area (Jarman 1984) to predict all the different elements of mental health service required, including the expensive item of bed numbers (Jarman et al. 1992).
Service use and effectiveness of intervention
It is useful to discuss interventions in psychiatry in each of the five levels defined by Goldberg and Huxley (1980) (Table 4). Although psychiatric patients can be involved at all of these levels and in different countries there may be much more intervention at one level than another because of the services available. There are still sufficient fundamental differences with the activities at each of the levels to keep them separate.
Level 5: inpatient care
This is the most expensive form of care and in the United Kingdom accounts for 85 per cent of the total costs of psychiatric services. The dramatic reduction in psychiatric inpatient beds in the past 30 years has led to a major shift in the diagnostic representation of inpatients. Organic states (particularly dementias), schizophrenia, and affective psychosis now constitute most of the diagnostic groups admitted to beds for mental illness and there have been significant falls in the proportions of patients admitted with neurotic and stress disorders, and depression (apart from that associated with bipolar affective disorder) (Tyrer et al. 1989). When there is greater use of beds than expected, it is usually because there is a lower threshold to the admission of affective disorders in general (Flannigan et al. 1994).
There is now a general notion in almost every country that psychiatric admission is not desirable and should be used as an option of last resort. This is because of the perceived perils of institutionalization (the gradation from independence to passive acceptance of a machine-like existence in an impersonal system), stigmatization of admission (the labelling of the victim as permanently ‘mad’ and alienated from society), and the relatively high cost of inpatient care (around £200 per patient per day), accounting for the disproportionate costs of hospital compared with community services (Merson et al. 1996).
In the past some of the dangers of hospital admission have been exaggerated, particularly in the 1960s in which all institutions of society were attacked (Goffman 1961), and they have been reduced by transferring inpatient care from large free-standing mental hospitals into psychiatric units in district general hospitals. It has also been appreciated that institutionalization is not just a consequence of institutions. Patients with psychiatric illness can lose all personal rights, be abused regularly, live in squalor under rigid control, and show all aspects of institutionalized behaviour while living at home with their families. It has also been appreciated that some symptoms that were formerly regarded as a consequence of institutional care, such as posturing, stereotyped behaviour, lack of motivation, and apathy, can be long-term consequences of psychiatric illness, notably schizophrenia, rather than created by the environment.
Nevertheless, public health policy in almost all countries, including developing countries which have never had a significant mental hospital base, has been to promote community care, the results of which are discussed above. However, if additional resources are available beyond those of the personnel in community teams, this can reduce the bed base further. Organizations such as the Richmond Fellowship and St Mungo’s Housing Association provide a network of hostels offering varying degrees of independence for up to several years. This can prove invaluable in stabilizing those with psychiatric problems and preventing the cycle of recurring admissions that is common in those who are discharged to substandard accommodation with all its attendant stresses. Other reasons include the reluctance of hospital-based staff to involve themselves in community work (Tyrer 1993) (a reluctance which is in itself another form of institutionalization), poor levels of integration between health and social services in many areas, and restrictions preventing adequate treatment of patients who are non-compliant. Patients who are at risk to themselves or others as a consequence of mental illness can be compulsorily treated, and how this should be done with proper respect for individual rights has long been a subject for debate.
A small number of psychiatric patients are recurrent criminal offenders or are particularly dangerous, and this has led to the development of forensic psychiatry as a separate discipline. Until quite recently forensic psychiatry has been mainly a tertiary referral service which has taken its patients from those who have already been in psychiatric care. Unfortunately, dangerousness and criminal behaviour are extremely difficult to predict. Many such patients have a combination of schizophrenia and personality disorder and may require long periods of detention to protect the public. For this reason special hospitals for the ‘criminally insane’ have long existed for such patients and, even in the era of community psychiatry, these necessarily have a high threshold for discharge. There is also increasing concern that, with the reduced numbers of psychiatric beds available, mentally ill patients may be diverted to the prison services where they will treated much less appropriately (Gunn et al. 1991). To avoid this, there has been a rapid growth of ‘court diversion’ schemes to assess the psychiatric status of those appearing in court (often for minor offences) so that those who are clearly mentally ill can be transferred to more appropriate care in psychiatric hospitals or other outpatient settings (Joseph 1994).
Forensic psychiatric patients occupy beds for a proportionally longer time, and if there is no appropriate district forensic service with its own bed base the numbers of those in hospital will be greater. Mental health legislation in most countries, largely because of concern for human rights, only has a limited role for compulsory treatments outside hospital, although in the United Kingdom reforms to the Mental Health Act to allow compulsory treatment in the community are likely to be made shortly. Many safeguards would have to be introduced to ensure that this power was not abused, but many countries are now reviewing their mental health legislation because so many more of the seriously mentally ill are being treated outside hospital.
In deciding on the number of beds available in a district, it is important to take into account the levels of social deprivation and particular characteristics of the area. In the United Kingdom Jarman (1983, 1984) has provided an extremely useful set of scores for each district based on their level of social deprivation (underprivileged area scores) which can be used to predict psychiatric admission rates (Jarman et al. 1992). However, even these predictions can be distorted by, for example, a small number of homeless mentally ill in the areas concerned. The homeless, because of their extreme levels of social deprivation, absence of contact with statutory services, and reluctance to engage in treatment, account for a disproportionate amount of psychiatric beds used. This is accentuated by the difficulties in finding accommodation once they have been treated and are ready for discharge on clinical grounds. The differences in levels of prevalence of severe mental illness between deprived inner-city areas and affluent rural areas are great and, as yet, inadequately quantified. They need to be borne in mind when extrapolating from data recorded in only one setting. For example, in the county of Buckinghamshire in England, a highly successful community psychiatric service, with a full range of crisis intervention and other services, has been able to reduce hospital admissions dramatically (Falloon and Fadden 1993); such results would be much more difficult to achieve in the inner city.
Level 4: patients attending psychiatric services outside hospital
This group, which can be termed ‘extra-cubilar’ psychiatry as it does not use hospital beds (Tyrer and Malone 1991), mainly involves outpatient and day care. In the last 20 years there has also been a shift of psychiatric resources towards primary care in the form of liaison psychiatry in general practice. There is now good evidence that a proportion of the severely mentally ill can be treated very effectively as day patients and this may obviate the need for psychiatric admission (Creed et al. 1990), but this may not be a realistic option in the less compliant patient. There is also reasonable evidence that the growth of liaison psychiatry in primary care has improved the skills of general practitioners and allowed more patients to continue their episodes of illness entirely in the primary care setting (Tyrer et al. 1990a). However, the advantage of such liaison still needs to be determined in controlled investigations and to date no impact has been found on psychiatric admission rates (Jackson et al. 1993) and the total cost of care is greater because more patients are referred (Goldberg et al. 1996).
Levels 2 and 3: hidden and conspicuous psychiatric morbidity in primary care
There has been considerable interest in the impact of educating primary care physicians so that they area able to detect important mental disorders because most psychiatric disorders are seen and assessed initially in primary care. If those disorders that were likely to lead to greater psychiatric morbidity (and progression to higher levels in service terms) and intervene early with effective treatments, this would be a very valuable form of secondary prevention. Evidence that screening psychiatric patients with instruments such as the General Health Questionnaire (Goldberg 1972) improve patient care is largely lacking. There is conflicting evidence from research studies (Johnstone and Goldberg 1976; Hoeper et al. 1984) and in ordinary practice this is rarely used. One of the problems is that most of the minor disorders seen in primary care have a favourable outcome which is determined much more by social factors than clinical ones (Huxley and Goldberg 1975) and clinical intervention may have relatively little impact. This may be another area of public health in which screening has failed to live up to its promise and might be replaced by greater input to the population specifically at risk (Holland et al. 1994).
More favourable results have been reported in depression, in which the risk of suicide is considerable. The most favourable results have been reported from education programmes aimed at improving the general practitioner’s ability to detect significant depression. The most important evidence of the public health value of this approach is a study from the island of Gotland. In Sweden in 1975, the International Committee for the Prevention and Treatment of Depression was founded, with the aim of improving both the understanding and the management of individuals with depressive illnesses. From this evolved the Swedish Committee for the Prevention and Treatment of Depression which was established in 1977 to examine the relationship between postgraduate medical education and depression in the community. In 1982 a study of the incidence of depression was set up in the island of Gotland, off the east coast of Sweden, to investigate the effects of an educational programme for general practitioners. The impact of this programme was evaluated in succeeding years.
The educational programme included lectures and discussions relating to patients with depression, and was aimed at improving the ability of general practitioners to detect and treat cases of depression. The outcome of the educational programme was most marked in the year after it was given. There was a significant decrease in the number of days spent in hospital because of depressive disorders, a reduction in days of work lost because of depression, and a reduction in the suicide rate by over 50 per cent (Rutz et al. 1989). A separate cost–benefit analysis showed that the costs of the educational programme were far outweighed by the benefits shown in terms of health-care savings (Rutz et al. 1992a). Unfortunately, in subsequent years the suicide rate crept upwards again (Rutz et al. 1992b) and it had been suggested that continuous education might be necessary to maintain this level of improvement.
There have been some criticisms about this study and the merits of generalizing from its findings. It is far from clear that taking data from this study and using it to make generalizations about medical practice elsewhere in different settings is appropriate as it is unclear to what extent the results of this study represent the unique island environment. There is also little evidence that the intervention has had a long-term effect on suicide rate. To this extent it has had less impact on suicide than other issues that are independent of mental health services. The most marked of these is the replacement of gas containing carbon monoxide (coal gas) by natural gas (containing no carbon monoxide) in the 1960s and this has had a much greater effect on the suicide rate that the Gotland data. Nevertheless the results of this study have encouraged those who feel that mental health services can have an impact on overall suicide rates.
Level 1: psychiatric morbidity in the community
Most people in developed countries with significant psychiatric morbidity present to either the primary care or psychiatric services (Fig. 3).
Only a small number of those at risk might be helped to a greater extent if their problems were detected earlier and appropriate intervention made. One group that could be helped is those patients with schizophrenia who present late in the course of illness because of gradual social withdrawal and severing of links with others in society including health services. There are suggestions from studies of outcome (Johnstone et al. 1990) that such late-presenting patients fare particularly badly in terms of outcome and that if they were detected earlier much morbidity could be avoided. However, how this could be done is difficult to contemplate at present.
Prevention in psychiatry is more elusive than in other branches of medicine because of the relatively vague and multifactorial aetiology of most mental disorders and this had led to the conclusion that most resources should be directed at those affected by these disorders rather than preventive strategies per se (Doll 1983). Its list of achievements is also disappointgly small (Holland and Fitzsimons 1990). However, there have been successes and, increasingly awareness of secondary prevention in particular has created a more optimistic environment. Pardes et al. (1989) have identified four areas that need refining before preventative strategies in psychiatry can be implemented more widely: improved psychiatric diagnosis, better epidemiological studies involving longitudinal data linked to health service use, advances in genetics generally and the wider use of genetic counselling in particular, and the potential advantages of biotechnology allowing non-invasive observation of brain function (e.g. positron emission tomography scanning). All these have been improved in the last 10 years and lead us to revise Doll’s pessimism.
By far the most impressive evidence of primary prevention is in the field of mental handicap (now euphemistically termed learning difficulties). Advances in clinical genetics now mean that nearly 40 per cent of the causes of severe mental handicap (Down syndrome, phenylketonuria, tuberose sclerosis, Hurler’s syndrome, Lesch–Nyhan syndrome, and Tay–Sachs disease) are caused by identifiable chromosome abnormalities that are potentially preventable (Weatherall 1991). At present, apart from a few special high-risk groups (e.g. Tay–Sachs disease in Ashkenazi Jews) screening before conception is not considered fruitful. The revelation that fragile X syndrome accounts for around 6 per cent of all individuals with mental handicap (Lavoxa et al. 1977) has tremendous scope for prevention through general population screening and detection of carriers. Other preventable causes of mental handicap include better obstetric care and supervision of babies with low birth weight (Illsley and Mitchell 1984), the prevention of rubella syndrome by the combined measles–mumps–rubella vaccine (although evidence for this is still awaited) and of neural tube defects by folic acid supplements in pregnancy (MRC Vitamin Study Group 1991). Genetic counselling for most of these disorders is still in its infancy and misleading information or maladroit presentation can create its own morbidity (McGuffin 1994). An important area of public health research is to examine the correct application of this preventative strategy.
Another important example of primary prevention is the introduction of needle-exchange schemes to reduce the transmission of HIV in drug users. There is little doubt that those who use such schemes significantly reduce the risk of developing HIV infection (Stimson et al. 1989) but those who avoid such schemes are often most risky in their behaviour. It is also now well established that ready access to illicit drugs and alcohol is probably the single most important precursor of addiction: and if the source of supply is reduced much psychiatric, as well as other morbidity, could be avoided (Royal College of Physicians 1991). Reduction of alcohol use would also be likely to reduce the suicide rate as around a quarter of all those who successfully complete suicide are dependent on alcohol (Barraclough et al. 1974). Other potential preventative strategies include the political one of raising the standards of living of all people in the country so that significant social deprivation is reduced to a minimum, which would have a beneficial effect on child psychiatric disorders, personality disorder (particularly of the antisocial type), drug and alcohol abuse, and post-traumatic stress disorders. It might also reduce the incidence of schizophrenia by improving perinatal care in particular (Lewis 1989).
There are more opportunities for psychiatric services in secondary prevention, mainly by early intervention to those at risk. For example, reduction of alcohol availability is a political and legal decision that affects alcohol consumption in the whole population, but making those who are at special risk a target for more intensive intervention is more cost-effective and in most cases preferred to general health promotion. In both general practice (Wallace et al. 1988) and general hospital admissions (Chick et al. 1985) intervention of an educational nature has helped to reduce subsequent alcohol consumption. However, as with most of these endeavours, the demonstration that reduction can be achieved is rarely followed by generalization from this finding. It is clear that such policies have to be pursued assiduously and persistently in order to be effective. There is considerable argument over whether intervention in children at risk of psychiatric disorder is beneficial or not. Because many such children come from disadvantaged backgrounds it is easy for any such intervention to be nullified by a return to the same cycle of disadvantage. However, Kolvin and his colleagues in particular have argued that children who show behaviour characteristic of maladjustment improve to a significantly greater degree with group therapy (from social workers), behaviour therapy, and what is described as ‘nurture work’ (support and enrichment from teacher-aides) to a significantly greater degree than a control group of children, and these gains are maintained 3 years after intervention (Kolvin et al. 1990).
Affective disorders are common in women of anxious premorbid personality after childbirth and Barnett and Parker (1985) have compared the value of professional psychiatric help to lay support and a control group. The results of this additional support show some evidence that postnatal affective symptoms, particularly anxiety, were reduced in those who received professional help but the findings do not at this stage suggest that such intervention is cost-effective. Like many of these populations it is only secondary intervention that takes place at a late stage (i.e. just before the expression of a psychiatric syndrome) that seems to be effective in preventive terms.
The prevention of disability and relapse in those who already have psychiatric disorder is now part of the standard practice of psychiatry. Unfortunately, it is rarely carried out to its full potential as many psychiatric services, through a combination of limited resources and habitual functioning, are essentially reactive rather than proactive in nature.
Schizophrenia covers a range of conditions that last from a few months to many years, but unfortunately the main effective treatment, antipsychotic drugs, has a high incidence of adverse drug reactions. The acute symptoms, pseudoparkinsonism and akathisia (bodily restlessness), are almost invariably present at some point in treatment, and akathisia in particular is almost impossible to relieve with other drugs such as procyclidine and propranolol which have only a limited effect. In the longer term, the syndrome of tardive dyskinesia is even more disturbing as this syndrome, once developed, is almost impossible to treat and in over a third of patients become permanent irrespective of further antipsychotic drug treatment. Nevertheless, maintenance treatment of schizophrenia by a drug regime is still the preferred option since there is increasing evidence that each relapse and readmission leads to some loss of function which is never regained (Stevens 1982).
However, social measures may also prevent relapse. Combination of maintenance drug treatment and reduction of ‘expressed emotion’ (interactions of strong emotional content of any type between the schizophrenic patient and others, of whom relatives are usually the most important) is one of the more effective ways of preventing relapse (Leff and Vaughn 1981).
Similarly, long-term antidepressant therapy, including tricyclic antidepressants and mood stabilizers such as lithium and sodium valproate, are all effective in preventing relapse in recurrent depressive and bipolar affective disorders. In general, lithium is preferred for bipolar disorders and antidepressants for persistent (unipolar) disorders. Despite evidence from many studies that such measures are successful, the evidence that their introduction has led to a reduction in relapse in these disorders for the psychiatric population with affective disorders as a whole is not yet available.
Suicide has been referred to early in the context of improved detection of depression. Any preventative measure that reduces the incidence of depression should also reduce suicide rates as approximately 1 in 6 of all depressed patients end their life by suicide (Gunnell and Frankel 1994). If it is possible to reduce the number of fatal means available to commit suicide this may also assist, particularly if the suicidal act is an impulsive one. People in occupations who have ready access to means of successful suicide (e.g. veterinary surgeons, farmers, anaesthetists) have suicide rates of more than twice the average of the general population (Charlton et al. 1993). Death by carbon monoxide poisoning through car exhausts is also increasing rapidly (Charlton et al. 1993) and has led to initiatives to fit most cars with catalytic converters that reduce the carbon monoxide content of car exhausts.
There is much discussion about the choice of antidepressant drugs in depressed patients. Although there is no doubt that the older antidepressants (mainly tricyclic compounds) are more dangerous in overdose than the newer agents, such as the selective serotonin-reuptake inhibitors (Cassidy and Henry 1987), it is difficult to know how much suicide could be prevented by wider prescription of the newer drugs, since only around 5 per cent of patients committing suicide do so by taking antidepressants alone. Unfortunately, to date there is no successful way of preventing further episodes of parasuicide (attempted suicide) in those seen after completing such an act. The only treatment that has been shown to be effective is a very intensive form of individual and group treatment related to cognitive therapy (Linehan et al. 1991), but this is too intensive to be introduced into the National Health Service. Since 1 per cent of patients after an episode of parasuicide commit suicide in the following year and around 20 per cent do so eventually, some way of preventing repetition would be a major gain to public mental health. Shorter forms of Linehan’s treatment linked to short treatment booklets have shown initial promise and may present the way forward (Evans et al. 1999).
The impact of services in tertiary prevention is also important. Treatments that are effective in preventing relapse are of no value if patients are not maintaining contact with the service. Formal introduction of the care programming approach to psychiatric services in England has not been formally tested apart from one study in which close supervision of vulnerable psychiatric patients by a key worker at regular intervals was more effective than control care in preventing drop out but led to significantly more psychiatric admissions (Tyrer et al. 1995). There is a contradiction between these findings and the specific intervention of community mental health teams which have been shown to reduce admissions to hospital and often provide more superior care. The key difference appears to be the presence of a suitably resourced multidisciplinary team that can provide the appropriate skills for patients outside hospitals rather than the common situation in care programming for a single key worker to take most of the clinical responsibility. Under such circumstances it is perhaps not surprising that when the key worker is faced with a deterioration in the patient’s condition, admission to hospital becomes more likely.
This chapter illustrates the extensive degree of overlap, and opportunities for collaboration, between the public health and mental health services. There are still large areas that need to be properly developed which, despite the optimism of Caplan (1964), have yielded very little in terms of improving public mental health. It is reasonable to argue that this is partly because so little has been shared between the two disciplines until the last few years.
Public health has a right to decide on how resources are distributed and to ask for targets such as the suicide targets of Our Healthier Nation and the satisfaction of the population’s needs. Therefore it is right and proper for a public health physician, for example, to question the duties of community psychiatric nurses to ensure that they are addressing these needs with appropriate priority. Community psychiatric nurses have a range of skills and are unable to deploy them all simultaneously (Tyrer et al. 1990b), and so other professionals, including those from public health, can provide advice based on their relative efficacy and cost, and the needs of the population served.
Simultaneously, public health physicians need to acknowledge the special difficulties of psychiatry. Our failure to achieve good outcome measures in psychiatry is not a consequence of incompetence. Direct measures of psychiatric ill health are now well established but are impractical to apply to whole populations. The search for appropriate proxy measures that could substitute for direct measurement of mental health status continues to preoccupy us greatly and we are making suitable efforts to obtain them. This should not mean that public health should rush into assuming that assessments such as those for needs are necessarily accurate and valid (Stevens and Gabbay 1991; Slade 1994); there is a great deal more to be done before they can be regarded in the same way as visual acuity can be used to measure the success of a treatment policy for cataracts.
Mutual understanding also needs to develop between the different perspectives of public health and mental health professionals. The public health physician has to deal with the health of large numbers and the uniqueness of the individual is relatively unimportant amongst these large numbers. Nevertheless, if a very small number of individuals are receiving care whilst large numbers of equally deserving numbers are not, it is only the person with the wider perspective who will blow the whistle and ask for something to be done. But the public health physician also has to be aware of the difficulties facing the psychiatric services; the idiosyncratic nature of diagnosis with frequent changes from one to the other, varying needs of the psychiatric patient at different times, and the dynamics of interaction between the psychiatric patients and others around about them, including their psychiatric attendants.
So much can be gained by better understanding of the roles of each discipline. On the one hand, public health cannot function as an adequate purchaser of mental health services without an understanding of clinical sensitivities, and on the other the psychiatrist and mental health worker need to have a wider perspective than that of patients under active care. Goldberg and Huxley (1992) sadly conclude at the end of their review of common mental disorders that ‘in most countries of the world services for the mentally ill survive on the crumbs left by the banquet of general health care’. Public health can ensure they all eat at the same table.
Acheson, D. (1988). Annual review of the public health. HMSO, London.
American Psychiatric Association (1980). Diagnostic and statistical manual of mental disorders (3rd edn). American Psychiatric Association, Washington, DC.
American Psychiatric Association (1994). Diagnostic and statistical manual of mental disorders (4th edn). American Psychiatric Association, Washington, DC.
Andrews, G., Stewart, G., Morris-Yates, A., Holt, P., and Henderson, S. (1990). Evidence for a general neurotic syndrome. British Journal of Psychiatry, 157, 6–12.
Andrews, G., Slade, T., and Peters, L. (1999). Classification in psychiatry: ICD-10 versus DSM-IV. British Journal of Psychiatry, 174, 3–5.
Anonymous (1992). The health of the nation: a summary of the strategy for health in England. HMSO, London.
Anonymous (1998). Report of the independent inquiry into inequalities of health. HMSO, London.
Barnett, B. and Parker, G. (1985). Professional and non-professional intervention for highly anxious primiparous mothers. British Journal of Psychiatry, 146, 287–93.
Barraclough, B., Bunch, J., Nelson, B., and Sainsbury, P. (1974). A hundred cases of suicide: clinical aspects. British Journal of Psychiatry, 125, 355–73.
Barton, R. (1959) Institutional neurosis. John Wright, Bristol.
Bayer, R. and Spitzer, R.L. (1985) Neurosis, psychodynamics and DSM-III: a history of the controversy. Archives of General Psychiatry, 42, 187–96.
Bebbington, P., Brugha, T., Hill, T., Marsden, L., and Window, S. (1999). Validation of the Health of the Nation Outcome Scales. British Journal of Psychiatry, 174, 389–94.
Bleuler, E. (1950). Dementia praecox or the group of schizophrenias. International Universities Press, New York.
Brayne, C., Gill, C., Paykel, E.S., Huppert, F., and O’Connor, D.W. (1995a). Cognitive decline in an elderly population—a two wave study of change. Psychological Medicine, 25, 673–83.
Brayne, C., Gill, C., Huppert, F.A., et al. (1995b). Incidence of clinically diagnosed subtypes of dementia in an elderly population. Cambridge Project for Later Life. British Journal of Psychiatry, 167, 255–62.
Burns, T., Beadsmoore, A., Bhat, A.V., Oliver, A., and Mathers, C. (1993). A controlled trial of home-based acute psychiatric services. I. Clinical and social outcome. British Journal of Psychiatry, 163, 49–54.
Caplan, G. (1964). Principles of preventative psychiatry. Basic Books, New York.
Casey, P. (2000). Epidemiology of personality disorder. In Personality disorders: diagnosis, management and course (2nd edn) (ed. P. Tyrer). Butterworth Heinemann, Oxford.
Cassidy, S.L. and Henry, J.A. (1987). Fatal toxicity of antidepressant drugs in overdose. British Medical Journal, 295, 1021–4.
Chadwick, E. (1842). Report on the sanitary condition of the labouring population of Great Britain. Edinburgh University Press.
Charlton, J., Kelly, S., Dunnell, K., Evans, B., and Jenkins, R. (1993). Suicide deaths in England and Wales: trends in factors associated with suicide deaths. Population Trends, 71, 34–42.
Chick, J., Lloyd, G., and Crombie, E. (1985). Counselling problem drinkers in medical wards: a controlled study. British Medical Journal, 290, 965–7.
Coid, J. (1994) Failure in community care: psychiatry’s dilemma. British Medical Journal, 308, 805–6.
Connell, P.H. (1958). Amphetamine psychosis. Oxford University Press.
Cooper, J.E., Kendell, R.E., Gurland, B.J., Sharpe, L., Copeland, J.R.M., and Simon, R. (1972). Psychiatric diagnosis in New York and London. Oxford University Press.
Creed, F., Black, D., Anthony, P., Osborn, M., Thomas, P., and Tormenson, B. (1990). Randomised controlled trial of day-patient vs in-patient psychiatric treatment. British Medical Journal, 300, 1033–7.
de Girolamo, G. and Reich, J.H. (1993). Personality disorders. WHO, Geneva.
Detels, R. and Breslow, L. (1986). Current scope. In Oxford textbook of public health: history, determinants, scope, and strategies (ed. W.W. Holland, R. Detels, and G. Knox), Volume 1. Oxford University Press.
DoH (Department of Health) (1990). Community care in the next decade and beyond. HMSO, London.
DoH (Department of Health) (1993). The Health of the Nation. Key area handbook: mental illness. HMSO, London.
DoH (Department of Health) (1999). Our healthier nation. A contract for health. HMSO, London.
Doll, R. (1983). Prospects for prevention. British Medical Journal, 286, 445–53.
Drever, F. and Bunting, J. (1997). Patterns and trends in male mortality. In Health inequalities: decennial supplement. DS Series no. 15. HMSO, London.
Endicott, J., Spitzer, R.L., Fleiss, J.L., and Cohen, J. (1976). The global assessment scale: a procedure for measuring the overall severity of psychiatric disturbance. Archives of General Psychiatry, 33, 766–71.
Evans, K., Tyrer, P., Catalan, J., et al. (1999) Manual-assisted cognitive–behaviour therapy (MACT): a randomised controlled trial of a brief intervention with bibliotherapy in the treatment of recurrent deliberate self-harm. Psychological Medicine, 29, 19–25.
Falloon, I.R.H. and Fadden, G. (1993). Integrated mental health care: a comprehensive community-based approach. Cambridge University Press.
Faris, R.E.L. and Dunham, H.W. (1939) Mental disorders in urban areas: an ecological study of schizophrenia and other psychoses. University of Chicago Press.
Farr, W. (1864). Letter to the Registrar General on the years 1851–1860. Supplement to the 25th report of the Registrar General. General Register Office, London.
Feinstein, A. (1970). The pre-therapeutic classification of comorbidity in chronic disease. Journal of Chronic Diseases, 23, 455–62.
Flannigan, C.B., Glover, G.R., Feeney, S.T., Wing, J.K., Bebbington, P. E., and Lewis, S.W. (1994). Inner London collaborative audit of admissions in two health districts. I: Introduction, methods and preliminary findings. British Journal of Psychiatry, 165, 734–42.
Foucault, M. (1973). Madness and civilisation. Vintage Books, New York.
Goffman, E. (1961). Asylums. Doubleday, New York.
Goldberg, D. (1972). The detection of psychiatric illness by questionnaire. Maudsley Monograph 21. Oxford University Press.
Goldberg, D. and Huxley, P. (1980). Mental illness in the community: the pathway to psychiatric care. Tavistock Publications, London.
Goldberg, D. and Huxley, P. (1992). Common mental disorders: a biosocial model. Tavistock/Routledge, London.
Goldberg, D., Jackson, G., Gater, R., Campbell, M., and Jennett, N. (1996) The treatment of common mental disorders by a community team based in primary care: a cost effectiveness study. Psychological Medicine, 26, 487–92.
Gunn, J., Maden, A., and Swinton, M. (1991). Mentally disordered prisoners. Home Office, London.
Gunnell, D. and Frankel, S. (1994). Prevention of suicide: aspirations and evidence. British Medical Journal, 308, 1227–33.
Harris, E.C. and Barraclough, B. (1998). Excess mortality of mental disorder. British Journal of Psychiatry, 173, 11–53.
Henderson, S. (2000). Conclusion: the central issues. In Unmet need in psychiatry: problems, reources, responses. Cambridge University Press.
Hoeper, E., Nicz, G., Kessler, L., Burke, J., and Pierce, W. (1984). The usefulness of screening for mental health. Lancet, i, 33–6.
Holland, W. and Fitzsimons, B. (1990). Public health concerns. How can social psychiatry help? In The public health impact of mental disorder (ed. D. Goldberg and D. Tantam), pp. 14–19. Hogrefe and Huber, Toronto.
Holland, W.W., Fitzsimons, B., and O’Brien, M. (1994). ‘Back to the future’—public health research into the next century. Journal of Public Health Medicine, 16, 4–10.
Holloway, F. and Carson, J. (1998) Intensive case management for the severely mentally ill: controlled trial. British Journal of Psychiatry, 172, 19–22.
Huxley, P. and Goldberg, D.P. (1975). Social versus clinical prediction in minor psychiatric disorders. Psychological Medicine, 5, 96–102.
Illsley, R. and Mitchell, R.J. (1984). Low birth weight. Wiley, Chichester.
Jablensky, A., Sartorius, N., Ernberg, G., et al. (1992). Schizophrenia: manifestations, incidence and course in different countries. A World Health Organization ten-country study. Psychological Medicine, Supplement 20, 1–97.
Jackson, G., Gater, R., Goldberg, D., Tantam, D., Loftus, L., and Taylor, H. (1993). A new community mental health team based in primary care: a description of the service and the effect on service use in the first year. British Journal of Psychiatry, 162, 375–84.
Jacob, K.S., Everitt, B., Patel, V., Weich, S., Araya, R., and Lewis, G.H. (1998). The comparison of latent variable models of non-psychotic psychiatric morbidity in four culturally diverse populations. Psychological Medicine, 28, 145–52.
Jarman, B. (1983). Identification of underprivileged areas. British Medical Journal, 286, 1705–9.
Jarman, B. (1984). Validation and distribution of scores. British Medical Journal, 289, 1587–92.
Jarman, B., Hirsch, S., White, P., and Driscoll, R. (1992). Predicting psychiatric admission rates. British Medical Journal, 304, 1146–51.
Johnstone, E.C. (1999). Diagnostic issues: concepts of the disorder. In Schizophrenia: concepts and clinical management (ed. E.C. Johnstone, M.S. Humphreys, F.H. Lang, S.M. Lawrie, and R. Sandler), p. 38. Cambridge University Press.
Johnstone, A. and Goldberg, D.P. (1976). Psychiatric screening in general practice: a controlled trial. Lancet, i, 605–8.
Johnstone, E.C., McMillan, J.F., Frith, C.D., Benn, D.K., and Crow, T. (1990). Further investigations of the predictors of outcome following a first schizophrenic episode. British Journal of Psychiatry, 157, 182–9.
Joseph, P.L.A. (1994). Psychiatric assessment at the magistrates’ courts. British Journal of Psychiatry, 164, 722–4.
Kessler, R.C., McGonagle, K.A., Zhao, S., et al. (1994). Lifetime and 12-month prevalence of DSM-III-R psychiatric disorders in the United States. Results from the National Comorbidity Survey. Archives of General Psychiatry, 51, 8–19.
Kolvin, I., Charles, R., Nicholson, R., Fleeting, M., and Fundudis, T. (1990). Factors in prevention in inner city deprivation. In Public health impact of mental disorders (ed. D. Goldberg and D. Tantam), pp. 115–23. Hogrefe and Huber, Toronto.
Laing, R.D. and Cooper, D.G. (1964). Reason and violence. Tavistock Publications, London.
Laing, R.D. and Esterson, A. (1970). Sanity, madness and the family. Harmondsworth, Penguin.
Lavoxa, R., Ridler, M.A.C., and Bowen-Bravery, M. (1977). An etiological survey of the severely retarded Hertfordshire children who were born between January 1st 1965 and January 31st 1967. American Journal of Medical Genetics, 1, 75–86.
Lawrie, S.M., Abukmei, S., Santosh, C., Chiswick, A., Rimmington J.E., and Best, J.J.K. (1997). Qualitative morphological abnormalities in schizophrenia: an MRI study and systematic literature review. Schizophrenia Research, 25, 155–66.
Leff, J. and Vaughn, C.E. (1981). The role of maintenance therapy and relatives’ expressed emotion in relapse of schizophrenia: a two year follow up. British Journal of Psychiatry, 139, 102–4.
Lehman A.F. and Steinwachs, D.M. (1998). Patterns of usual care for schizophrenia: initial results from the Schizophrenia Patient Outcomes Research Team (PORT) Client Survey. Schizophrenia Bulletin, 24, 11–20.
Lewis, S.W. (1989). Congenital risk factors for schizophrenia. Psychological Medicine, 19, 5–13.
Linehan, M.M., Hubert, A.E., Suarez, A., Allmon, D., and Heard, H.L. (1991). Cognitive–behavioral treatment of chronically parasuicidal borderline patients. Archives of General Psychiatry, 48, 1060–4.
Luque, R. and Berrios, G.E. (1994). Cotard’s syndrome in the eldery: historical and clinical aspects. International Journal of Geriatric Psychiatry, 9, 957–64.
Macdonald, A. (1993). The myth of suicide prevention by general practitioners. British Journal of Psychiatry, 163, 260.
McGuffin, P. (1994). Genetics. In Prevention in psychiatry (ed. E.S. Paykel and R. Jenkins), pp. 32–9. Gaskell, London.
Magee, W.J., Eaton, W.W., Wittchen, H.U., McGonagle, K.A., and Kessler, R.C. (1996). Agoraphobia, simple phobia, and social phobia in the National Comorbidity Survey. Archives of General Psychiatry, 53, 159–68.
Meltzer, H., Gill, B., and Petticrew, M. (1994). OPCS Surveys of Psychiatric Morbidity in Great Britain. Bulletin 1: The prevalence of psychiatric morbidity among adults aged 16–64, living in private households, in Great Britain. OPCS, London.
Merson, S., Tyrer, P., Onyett, S., et al. (1992). Early intervention in psychiatric emergencies: a controlled clinical trial. Lancet, 339, 1311–14.
Merson, S., Tyrer, P., Carlen, D., and Johnson, A.L. (1996) The cost of treatment of psychiatric emergencies: a comparison of hospital and community services. Psychological Medicine, 26, 727–34.
Moens, G.F.G. (1990). Aspects of the epidemiology and prevention of suicide. Leuven University Press.
Montgomery, A.J. and Parshall, A.M. (1998). Implications of urban drift on health care resources in inner London. Psychiatric Bulletin, 22, 494–6.
MRC Vitamin Study Group (1991). Prevention of neural tube defects: results of the Medical Research Council Medical Study Group. Lancet, 338, 131–7.
Muijen, M., Marks, I.M., Connolly, J., and Audini, B. (1992). Home based care and standard hospital care for patients with severe mental illness: a randomised controlled trial. British Medical Journal, 304, 749–54.
Mulder, R.T. and Joyce, P.R. (1997) Temperament and the structure of personality disorder symptoms. Psychological Medicine, 27, 99–106.
OPCS (1995). OPCS Surveys of Psychiatric Morbidity in Great Britain. Bulletin 1: The prevalence of psychiatric morbidity among adults living in private households: Report 1. HMSO, London.
Pardes, H., Silverman, M.M., and West, A. (1989). Prevention and the field of mental health: a psychiatric perspective. Annual Review of Public Health, 10, 403–22.
Paykel, E.S., Brayne, C., Huppert, F.A., et al. (1994). Incidence of dementia in a population older than 75 years in the United Kingdom. Archives of General Psychiatry, 51, 325–32.
Porter, R. (1991). The Faber book of madness, p. 8. Faber and Faber, London.
Regier, D.A., Myers, J.K., Kramer, M., et al. (1984). The NIMH Epidemiologic Catchment Area (ECA) Program: historical context, major objectives and study population characteristics. Archives of General Psychiatry, 41, 934–41.
Regier, D.A.., Narrow, W.E., Rae, D.S., et al. (1993). The de facto US mental and addictive disorders service system: Epidemiologic Catchment Area prospective one-year prevalence rates of disorders and services. Archives of General Psychiatry, 50, 85–94.
Regier, D.A., Kaelber, C.T., Rae, D.S., et al. (1998). Limitations of diagnostic instruments for mental disorders: implications for research and policy. Archives of General Psychiatry, 55, 109–15.
Roberts, K. and Vass, N. (1986). Schneiderian first-rank symptoms caused by benzodiazepine withdrawal. British Journal of Psychiatry, 148, 593–4.
Robins, J. (1986). Fools and mad: a history of the insane in Ireland. Institute of Public Administration, Dublin.
Royal College of Physicians, Faculty of Public Health Medicine (1991). Alcohol and the public health. Macmillan, London.
Rutz, W., von Knorring, L., and Wålinder, J. (1989). Frequency of suicide on Gotland after systematic postgraduate education of general practitioners. Acta Psychiatrica Scandinavica, 80, 151–4.
Rutz, W., Carlsson, P., von Knorring, L., and Wålinder, J. (1992a). Cost-benefit analysis of an educational program for general practitioners by the Swedish Committee for the prevention and treatment of depression. Acta Psychiatrica Scandinavica, 85, 457–64.
Rutz, W., von Knorring, F.L., and Wålinder, J. (1992b). Long term effects of an educational program for general practitioners given by the Swedish Committee for the Prevention and Treatment of Depression. Acta Psychiatrica Scandinavica, 85, 83–8.
Sartorius, N., Jablensky, A., Gulbinat, W., and Ernberg, G. (1980). WHO collaborative study: assessment of affective disorders. Psychological Medicine, 10, 743–9.
Scull, A. (1959). Museums of madness: the social organization of insanity in 19th-century England. Allen Lane, London.
Shipley, K., Hilborn, B., Hansell, A., Tyrer, J., and Tyrer, P. (2000). Patient satisfaction: a valid index of quality of care in a psychiatric service? Acta Psychiatrica Scandinavica, 101, 330–3.
Slade, M. (1994). Needs assessment. British Journal of Psychiatry, 165, 293–6.
Stein, L.I. and Test, M.A. (1980). Alternative to mental hospital treatment. 1: Conceptual model, treatment program and clinical evaluation. Archives of General Psychiatry, 36, 1073–9.
Stevens, J.R. (1982). Neurology and neuropathology of schizophrenia. In Schizophrenia as a brain disease (ed. F.A. Henn and H.A. Nasrallah), pp. 112–47. Oxford University Press.
Stevens, A. and Gabbay, J. (1991). Needs assessment needs assessment. Health Trends, 23, 20–3.
Stimson, G.V., Dolan, K., Donoghue, M.C., et al. (1989). Syringe exchange schemes: a report and some commentaries. British Journal of Addiction, 84, 1283–90.
Sullivan, P.F. and Kendler, K.S. (1998). Typology of common psychiatric syndromes: an empirical study. British Journal of Psychiatry, 173, 312–19.
Szasz, T. (1960). The myth of mental illness. American Psychologist, 15, 113–18.
Thomas, H. (1993). Psychiatric symptoms in cannabis users. British Journal of Psychiatry, 163, 141–9.
Thornicroft, G., Wykes, T., Holloway, F., et al. (1998). From efficacy to effectiveness in community mental health services. PRiSM Psychosis Study 10. British Journal of Psychiatry, 173, 423–7.
Tooth, G.C. and Brooke, E.M. (1961). Trends in the mental hospital population and their effect on future planning. Lancet, i, 710–13.
Tyrer, P. (1985). The hive system. A model for a psychiatric service. British Journal of Psychiatry, 146, 571–5.
Tyrer, P. (1993). Who is failing the mentally ill? Lancet, 341, 1199–201.
Tyrer, P. (1996) Comorbidity or consanguinity. British Journal of Psychiatry, 168, 669–71.
Tyrer, P. and Ferguson, B. (2000). Classification of personality disorder. In Personality disorders: diagnosis, management and course (2nd edn) (ed. P. Tyrer), pp. 13–43. Edward Arnold, London.
Tyrer, P. and Johnson, T. (1996) Establishing the severity of personality disorder. American Journal of Psychiatry, 153, 1593–7.
Tyrer, P. and Malone, S. (1991). Psychiatry without hospital beds: a review of treatment strategies. In Recent advances in clinical psychiatry (ed. K. Granville-Grosman), pp. 105–18. Churchill Livingstone, Edinburgh.
Tyrer, P., Turner, R., and Johnson, A.L. (1989). Integrated hospital and community psychiatric services and use of inpatient beds. British Medical Journal, 299, 298–300.
Tyrer, P., Ferguson, B., and Wadsworth, J. (1990a). Liaison psychiatry in general practice: the comprehensive collaborative model. Acta Psychiatrica Scandinavica, 81, 359–63.
Tyrer, P., Hawksworth, J., Hobbs, R., and Jackson, D. (1990b). The role of the community psychiatric nurse. British Journal of Hospital Medicine, 43, 439–42.
Tyrer, P., Casey, P., and Ferguson, B. (1991). Personality disorder in perspective. British Journal of Psychiatry, 159, 463–71.
Tyrer, P., Seivewright, N., Ferguson, B., and Tyrer, J. (1992). The general neurotic syndrome: a coaxial diagnosis of anxiety, depression and personality disorder. Acta Psychiatrica Scandinavica, 85, 201–6.
Tyrer, P., Morgan, J., Van Horn, E., et al. (1995). Randomised controlled study of close monitoring of vulnerable psychiatric patients. Lancet, 345, 756–9.
Tyrer, P., Coid, J., Simmonds, S., Marriott, S., and Joseph, P. (1999) Community mental health team management for those with severe mental illness and disordered personality. In Schizophrenia module of the Cochrane database of systematic reviews (ed. C.E. Adams, L. Duggan, J. de Jesus Mari, and P. White). Cochrane Collaboration, Oxford.
UK700 Group (1999). Intensive versus standard case management for severe psychotic illness: a randomised trial. Lancet, 353, 2185–9.
Wallace, P., Cutler, S., and Haines, A. (1988). Randomised controlled trial of general practitioner intervention in patients with excess alcohol consumption. British Medical Journal, 297, 663–8.
Weaver, T., Renton, A., Stimson, G., and Tyrer, P. (1999) Severe mental illness and substance misuse comorbidity: research is needed to inform policy and service development. British Medical Journal, 318, 137–8.
Weatherall, D.J. (1991). The new genetics and clinical practice. Oxford University Press.
WHO (1992). International classification of diseases, 10th revision. WHO, Geneva.
Zigmond, A.S. and Snaith, R.P. (1983). The Hospital Anxiety and Depression Scale. Acta Psychiatrica Scandinavica, 57, 361–70.