Mental Illness in the Community

The pathway to psychiatric care

Prof David Goldberg, David Goldberg, Dr Peter Huxley, Peter Huxley

• Publisher: Taylor & Francis
• Year: 2012
• Print ISBN: 9780415264518
• eBook ISBN: 9781135644758
• Edition: 1
• Topic: Medicine
• Subtopic: Health Care Delivery

Tavistock Press was established as a co-operative venture between the Tavistock Institute and Routledge & Kegan Paul (RKP) in the 1950s to produce a series of major contributions across the social sciences.
This volume is part of a 2001 reissue of a selection of those important works which have since gone out of print, or are difficult to locate. Published by Routledge, 112 volumes in total are being brought together under the name The International Behavioural and Social Sciences Library: Classics from the Tavistock Press.
Reproduced here in facsimile, this volume was originally published in 1980 and is available individually. The collection is also available in a number of themed mini-sets of between 5 and 13 volumes, or as a complete collection.

---

Chapter 1

Introduction

---

Knowledge about mental illness and its social correlates has until recently largely been derived by studying those treated by the psychiatric services. This is reasonable for major disorders which are relatively rare and which are likely to reach the psychiatric services, but it is unreasonable for common conditions which often do not reach them. For example, a study that is based only on those under treatment by psychiatrists cannot possibly demonstrate the importance of a possible social correlate which may itself be associated with a reduced chance of receiving treatment.

Although there have been notable attempts to define psychiatric illness on theoretical grounds,¹ there is a more prosaic sense in which psychiatric illnesses are those disorders that occur among the clients of psychiatrists. These illnesses are enumerated in the World Health Organisation’s Glossary of Mental Disorders (World Health Organisation 1974). Since people consult psychiatrists for a variety of reasons, it is hardly surprising that the classification offered by the Glossary is strikingly inclusive; so that the clinician is even able to use ratings keys for classifying those with long-standing traits, such as lesbians, eccentrics, and stutterers. If only psychiatrists were in the habit of saying to their patients, ‘You don’t have a psychiatric illness, go away’, they might be said to have some part in the definition of mental illness: but such behaviour is very rare. Until quite recently psychiatrists did not define psychiatric illness, they described it. The descriptions which have resulted from their efforts are systematic and intelligent, but they are based on a study of a small subset of patients who present themselves – or who are presented by others – for psychiatric care.

What a given society understands by psychiatric illness is effectively defined by the characteristics of the referral pathway to the psychiatrist’s office. Paradoxically, psychiatrists have very little to do with the decisions which must be taken before a patient comes to see him, although naturally psychiatrists collectively contribute to the climate of ideas which will influence non-psychiatrists in their decisions concerning referral. Once a patient arrives in his office, the psychiatrist will typically concur with lay judgement and assign a diagnostic label, since his client has defined himself as psychiatrically ill by occupying the formal patient role.

Countries where there are large numbers of psychiatrists, and where members of the public can refer themselves directly to psychiatrists without the necessity of using a primary care physician as an intermediary, are therefore likely to have patients referred to psychiatrists with relatively minor disorders and life problems and to use rather over-inclusive criteria for deciding what constitutes a psychiatric illness. There is a real sense in which such psychiatrists are functioning as specialized primary care physicians, and it is hardly surprising that their colleagues in countries where patients are typically referred by primary care physicians have developed more conservative notions of what constitutes a psychiatric case.

In recent years an important development has occurred which allows psychiatrists to play a decisive role in the definition of what constitutes a psychiatric illness. The arrival of standardized psychiatric interviews and psychiatric screening questionnaires has allowed researchers to study systematically the distribution of symptoms among patients receiving psychiatric care.² It has become clear that, with certain interesting exceptions, most psychiatric patients have a common core of symptoms which relate to mood disorders – notably anxiety, depression, fatigue, irritability, and sleep disturbance.³ The exceptions fall into two groups: on the one hand, major disorders such as hypomania, certain forms of schizophrenia, and some organic states which can readily be diagnosed by the possession of other florid patterns of psychopathology; and on the other hand, various kinds of abnormal personality which may occur without the critical symptoms of mood disorder, and which therefore fail to meet the criteria for a psychiatric illness.

The most elaborate attempt to specify a psychiatric illness in operational terms is John Wing’s Index of Definition derived from the Present State Examination (see Wing et al. 1977; Wing 1980), the latter being a 140 item research interview developed at the Institute of Psychiatry and now used throughout the world by the World Health Organisation. An alternative, rather cruder attempt to define a psychiatric illness is represented by psychiatric screening questionnaires such as the General Health Questionnaire.⁴ In order to satisfy the Index of Definition, one must have more than a critical number, type, and severity of PSE symptoms, while in order to be considered a ‘probable case’ on the GHQ a respondent must endorse more than a critical number of symptoms from a checklist offered to him. It is worth noticing two rather arbitrary characteristics of both these ways of identifying a psychiatric illness. In the first place, the ‘psychiatric patients’ who were used to generate the calibration groups on which each system depends were produced by a particular health care delivery system: in both cases, the British National Health Service. From a British viewpoint this may seem very reasonable, but it might seem less reasonable viewed from Washington DC, or New York, where the copious availability of analytically trained psychiatrists taking direct referrals from the community may result in many patients failing to meet the Index of Definition. The second point is that either measuring instrument will produce distributions of patients without a clear division between ‘cases’ and ‘normals’; so that the decision as to where subclinical disturbance ends and being a psychiatric case begins is, in the last analysis, arbitrary. For example, the concept of a ‘case’ which was used in the validation studies of the GHQ had regard to a degree of psychological disorder which was ‘just clinically significant’ in relation to a patient’s visit to his general practitioner. Several studies have shown that this is equivalent to a ‘Borderline Case’ in Wing’s scheme; if one required the two instruments to produce similar rates one would need to raise the threshold score used by the GHQ.

The same arbitrary standards are used by the rival American scheme, the Research Diagnostic Criteria of Spitzer, Endicott, and Robins (1975). In order to be diagnosed as, for example, a ‘major depression’ a patient must possess certain key symptoms and then at least five out of a shopping list of eight associated symptoms. Naturally, some patients just fail to make it to the criterion. This sort of procedure is perfectly reasonable; but it is also completely arbitrary.

In the past few years these research instruments have been used to measure rates for psychiatric illness in the general population in order to arrive at estimates of prevalence independent of the illness behaviour of the patient or the ability of his medical attendants to detect and treat any disorder that may present. When this is done the concepts of psychiatric illness which have been derived from those patients seen by psychiatrists are being back-projected onto the general population in order to assess the numbers of those with similar patterns of symptoms who have not sought psychiatric care.

Despite the somewhat different theoretical underpinnings of the various methods of psychiatric case findings now in use, two conclusions are unmistakeable. First, there is far less variation between recent estimates for rates of illness in random samples of populations than there were in the studies reported up to the early 1970s;⁵ and second, it is quite clear that even in the developed countries of the world, most mentally disordered patients are not being treated by the psychiatric services.⁶

Aim of the Book

We will attempt a summary of research findings that have used the case-finding techniques which have recently become available, with particular emphasis on research which deals with the detection and management of psychiatric disorders by family doctors. As we shall show, the greatest share of the burden falls on their shoulders both in England and the United States. The book has three aims:

1. To describe the selection processes which operate on psychologically disordered individuals which determine which of them will seek care; having sought care, which will have their disturbances detected; having been detected, which will be treated in a primary care setting and which will be referred for psychiatric care. A schematic model will be used to illustrate these steps.

2. To describe the kinds of psychiatric disorder commonly encountered among patients at each stage of the help-seeking process, and to summarize what is known about social factors associated with psychological disorders at each level.

3. Since most psychologically disordered patients who seek care will continue to receive treatment in primary care settings, our third aim is to describe the forms of treatment which should be available in such settings, and the training which primary care physicians and other health professionals should receive to enable them to provide such care.

The Model to be Used

A simplified model will be presented with five levels, each level representing different populations of subjects. In order to pass from one level to another it is necessary to pass through a filter.

Level 1 represents the community: at this level, our knowledge is derived from surveys of psychiatric morbidity which have either screened entire populations or which have been based on random samples of a particular population.

Level 2 is represented by studies of psychiatric morbidity among patients attending primary care physicians, irrespective of whether or not the physician has detected the illness. The first filter is between the first and second levels. The factors which determine whether or not an individual passes through the first filter are often referred to as ‘illness behaviours’ of the patient.

Level 3 consists of those patients attending primary care physicians who are identified as ‘psychiatrically sick’ by their doctor. These patients collectively represent psychiatric morbidity as it is seen from the vantage point of the primary care physician, and they will be referred to as the ‘conspicuous psychiatric morbidity’ of general medical practice. The second filter is represented by their doctor’s ability to detect psychiatric disorders among patients in the second level. It will be shown that passage is through this filter is determined by characteristics of both doctor and patient.

Level 4 is represented by patients attending psychiatrists in outpatient clinics and private offices. In England, the primary care physician is critically placed to determine who will be referred for psychiatric outpatient care, and he will therefore be thought of as the third filter. In the United States it will be shown that in addition to patients being referred to psychiatrists by primary care physicians, there is a considerable ‘short circuit’ of the second and third filters, in that a substantial number of patients are self-referred and thus pass directly from the first filter to level 4. However, even in the United States, many patients enter psychiatric care by referral from primary care physicians and thus passage is through the second and third filters (detection and subsequent referral).

Level 5 is represented by patients admitted to psychiatric hospitals and mental hospitals. They form the population most commonly referred to in national statistics of mental illness. The psychiatrist now appears for the first time, as the gate-keeper to in-patient beds. Even here, his powers are not absolute, since the number of patients he allows through the fourth filter depends on the number of beds made available to him by the health authorities. In all countries it is possible for an acutely psychotic patient to short-circuit the entire system and pass directly from level 1 to level 5, pausing only to be vetted by the psychiatrist acting as the fourth filter. However, these patients typically display major psychotic syndromes and once more the psychiatrist usually plays little part in deciding that a patient is referred to him in this way.

It will be seen that ‘psychiatric illness’ proper begins at level 4; yet psychiatrists do not define such illnesses, since they seldom send patients away undiagnosed. However we choose to define a psychiatric illness in theory, in practice it is defined by the process of passing through the first three filters. Each of the filters is selectively permeable, so that some individuals are more likely to pass through than others. And we can already see that the key people deciding who shall pass through are the patient and his family doctor.

Measures of Psychiatric Morbidity

If one wishes to study the distribution of a disorder in a human population, it is necessary to distinguish between the inception and prevalence. Inception refers to the rate at which new cases occur per unit time, and prevalence to the level of disorder, either at a point in time, or over a period of time. Generally speaking, surveys of illness in random samples of the general population will report point prevalence, while surveys in consulting populations will report period prevalence and sometimes inception rates. The definitions of each will be given so that the relationships between them may be more readily understood:

Annual Inception Rate (Synonym: Incidence Rate)

This refers to the number of individuals with a new episode of a given disorder each year, per 1,000 of the population at risk. If the disorder only affects a particular age group, it is permissible to adjust the population at risk to take this into account.⁷ The decision as to what is to be considered a ‘new’ episode is of course arbitrary; where psychiatric morbidity is concerned, it is usual to define it as one for which the patient has not previously consulted for at least one year.⁸

Point Prevalence

This refers to the number of people with a given disorder in a population at a point in time. It can obviously be expressed either as a percentage or as a rate per 1,000 at risk. If the age of the population surveyed is known, one can use point-prevalence data to calculate ‘morbid risks’ or disease expectancies in populations.⁹ In a large survey it is often impracticable to assess everyone on the same day unless the condition is very easy to count. Provided that each member of the population is only considered once, it is usual to allow such surveys to continue over a short time-period.

One-Year Period Prevalence (Synonym: Annual Patient Consulting Rate)¹⁰

This refers to the number of people who suffer from a disorder during the course of a calendar year on at least one occasion, per 1,000 population at risk. Individuals may be seen on numerous occasions during a year, and they will be counted as cases if they display the condition at any time during the survey year.

It is always possible to calculate one of these parameters if one knows the other two, since: One-year period prevalence = Point prevalence + Annual inception rate, and it is possible to calculate the mean duration of a disorder if one knows period prevalence and inception, since: Point prevalence = Annual inception × Mean duration of episode.

It will be noticed that both the prevalence and the inception rates are expressed ‘per 1,000 population at risk’ rather than as a percentage of all those attending doctors. In countries where people are free to shop around for medical care – going perhaps to one primary care physician when their child has a rash, but to another for a gynaecological complaint – such measures are almost impossible to calculate. In Britain, where as part of the National Health Service, every member of the population is registered with a single general practitioner, it is relatively straightforward matter, and estimates have been available for many years.¹¹ In the United States it has only recently become possible to calculate similar rates by studying populations registered for care with Neighborhood Health Centers and various forms of Health Insurance Plan; but even recent estimates suffer from the disadvantage that the populations receiving care from such schemes may not be fully representative of the general population.

We will not be concerned in this book with the numerous studies which report that ‘x per cent of a particular physician’s patients are emotionally disturbed’, since such estimates tell us nothing about the population at risk, and indeed the size of the estimate tells us more about the physician making the assessment than it does about the level of symptomatology among his patients.¹² However, if we have a large representative sample of primary care physicians, and we know what percentage of their patients are thought sick, then we can make a rough estimate of treated sickness by multiplying this percentage by the proportion of people in that population who seek care each year. This procedure wa...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright
• Contents
• Acknowledgements
• Foreword
• 1. Introduction
• 2. Level 1: Mental illness in the community
• 3. The first filter: The decision to consult
• 4. Level 2 and the second filter: Psychiatric disorders among primary care attenders
• 5. Level 3: Conspicuous psychiatric morbidity in primary care settings
• 6. The third filter: Referral to psychiatric services
• 7. Levels 4 and 5: Psychiatric morbidity treated by psychiatrists
• 8. Interview techniques in primary care settings
• 9. Non-medical models of care
• 10. Overview
• References
• Name index
• Subject index