Trauma and Psychosis

New Directions for Theory and Therapy

Warren Larkin, Anthony P. Morrison, Warren Larkin, Anthony P. Morrison

• Publisher: Routledge
• Year: 2007
• Print ISBN: 9781583918203
• eBook ISBN: 9781135447458
• Edition: 1
• Topic: Medicine
• Subtopic: Psychiatry & Mental Health

Trauma and Psychosis provides a valuable contribution to the current understanding of the possible relationships between the experience of trauma and the range of phenomena currently referred to as psychosis.

Warren Larkin and Anthony P. Morrison bring together contributions from leading clinicians and researchers in a range of fields including clinical psychology, mental health nursing and psychiatry. The book is divided into three parts, providing comprehensive coverage of the relevant research and clinical applications. Part I: Research and Theoretical Perspectives provides the reader with a broad understanding of current and developing theoretical perspectives. Part II: Specific Populations examines the relationship between trauma and psychotic experiences in specific populations. Part III: From Theory to Therapy draws together current knowledge and investigates how it might be used to benefit individuals experiencing psychosis.

This book will be invaluable for clinicians and researchers interested in gaining a greater insight into the interaction between trauma and psychosis.

Part I
Research and theoretical perspectives

---

Chapter 1
The environment and psychosis

Rethinking the evidence

Richard P. Bentall

---

Many researchers studying the psychoses believe that the field is currently undergoing a paradigm shift. Long-held assumptions about the nature and origins of conditions such as ‘schizophrenia’ and ‘bipolar disorder’ are being questioned and new research strategies are being employed to examine old questions in new ways. One assumption that is long overdue for re-examination is the idea that severe psychiatric disorders are hardly at all influenced by the physical and psychological environment, for example, by where one is born, how one is raised or whether or not one has the misfortune to experience various kinds of adverse events.

This idea has a long history. During the late nineteenth and first half of the twentieth centuries, a period which the historian Edward Shorter (1997) has described as the first era of biological psychiatry, the psychoses were widely regarded as neurodegenerative diseases, the result of cerebral pathology that was almost certainly inherited. It is true that, during the 1950s and 1960s, this idea became briefly unfashionable when American psychiatry, under the influence of émigré psychoanalysts, attributed ‘schizophrenia’ to ‘doublebind’ patterns of interactions between parents and children (Bateson et al. 1956) or malevolent, ‘schizophrenogenic’ mothers (Fromm-Reichmann 1950). However, in the second era of biological psychiatry, ushered in by the publication of the third edition of the American Psychiatric Association’s Diagnostic and Statistical Manual (DSM-III: American Psychiatric Association 1980), environmental theories were once more rejected, and psychosis was again seen as exclusively a product of some kind of endogenous pathological process. From that time until the present day, it has usually been assumed that the search for environmental determinants, briefly stimulated by psychoanalytic theorising, has led only to negative results, and that further investigations of environmental influences are therefore unwarranted.

In this chapter I will argue that this prejudice has been ill founded, and that the role of environmental influences in psychosis has almost certainly been severely underestimated by psychiatric researchers in recent years. My argument will proceed in two stages. First, I will attempt to identify some impediments to thinking clearly about this issue, which may have contributed to the historical bias against environmental hypotheses, and second, I will briefly review some recent evidence that addresses the issue empirically, and which challenges traditional assumptions.

Some reasons why environmental influences on psychosis have been underestimated

It is possible to identify a number of serious conceptual, methodological and ideological impediments to understanding how the environment can affect the risk of severe mental health problems.

‘Schizophrenia’ as an explanandum

Undoubtedly, one serious problem has been researchers’ reliance on psychiatric diagnoses that have questionable scientific validity, particularly the diagnosis of ‘schizophrenia’. I have discussed the limitations of this concept at length elsewhere (Bentall 1990b, 1998, 2003) and therefore will provide only a brief overview of the relevant issues here.

In order to be useful, diagnoses have to meet two main criteria (Blashfield 1984; Kendell 1975). First of all, they have to be reliable, which is to say that different clinicians must use the diagnoses consistently (different clinicians should diagnose the same patients as ‘schizophrenic’). Second, they have to be valid, which is to say that they have to perform the functions for which they are designed, such as grouping together patients with similar symptoms and whose problems have similar aetiologies, or predicting outcome (what will happen to patients in the future) and response to treatment (for example, which kinds of drugs will work).

The diagnostic systems employed by modern psychiatry were based on clinical concepts developed by early German psychiatrists, particularly Emil Kraepelin (1899/1990), who believed he had identified a small number of discrete disorders, notably dementia praecox (later known as ‘schizophrenia’) and manic depression (now often known as ‘bipolar disorder’) (see Bentall 2003, for a historical review). Kraepelin believed that each of these disorders produced a particular cluster of symptoms (hallucinations, delusions and cognitive deterioration in the case of schizophrenia, extreme mood states in the case of manic depression) and had a particular outcome (poor versus good). Later researchers came to believe that schizophrenia responded well to dopamine receptor agonist drugs (the ‘neuroleptics’ or ‘antipsychotics’) whereas manic depression responded to mood stabilizers, particularly lithium carbonate.

In fact, none of these assumptions have stood the test of time. The reliability of psychiatric diagnoses, even using modern ‘operational’ definitions, remains unacceptably low (Kirk and Kutchins 1992) and moreover, as several different definitions have been proposed for each of the major diagnoses, different patients tend to be given each diagnosis depending on which criteria are employed (Brockington 1992). Patients with a mixture of schizophrenic and manic depressive symptoms are common (Kendell and Gourlay 1970), and the symptoms of patients with either diagnosis (however defined) fall into a number of separate clusters (positive symptoms, symptoms of cognitive disorganisation, negative symptoms), which do not observe traditional diagnostic borders (Liddle 1987; Toomey et al. 1998). Both diagnoses are associated with a wide range of outcomes (Ciompi 1984; Tsuang et al. 1979) and, more alarmingly for clinicians, appear to be poor predictors of response to different types of psychotropic medication (Johnstone et al. 1988). Overall, it is difficult to avoid the impression that the categorical diagnoses employed in modern psychiatry have much in common with astrological star signs, a diagnostic system that is also widely held to say something about the characteristics of individuals and what will happen to them in the future.

The poor validity of psychiatric diagnoses has profound scientific and clinical implications that are well beyond the scope of this chapter (see Bentall 2003). In the present context it is sufficient to note that the widespread use of meaningless diagnoses in research has impeded the identification of aetiological factors that might be responsible for severe psychiatric problems. How can it be possible to determine whether any particular factor (biological or environmental) influences the development of psychiatric conditions when the diagnoses employed in the research group together people with widely differing problems, undoubtedly with widely differing aetiologies?

Misunderstanding the genetic evidence

A second impediment has been over-reliance on genetic evidence that seemed to suggest that the psychotic disorders are under strong genetic control. According to this line of reasoning, disorders such as ‘schizophrenia’ are highly heritable, high heritability excludes the possibility of important environmental influences, and therefore the environment cannot play a crucial role in the development of psychosis. The conclusion of this argument is undermined by the fact that both of its premises are false.

In recent years it has become clear that early genetic studies exaggerated the heritability of psychiatric disorders, partly because researchers treated the genetic determination of psychosis as an axiom rather than a hypothesis to be rigorously tested (Marshall 1990). Numerous methodological problems beset widely disseminated studies carried out by investigators who believed that genetic influences could be demonstrated without giving consideration to confounding factors. For example, in the famous series of Danish-American adoption studies carried out by Kety et al. (1975) and Rosenthal et al. (1974), significant results in favour of the genetic hypothesis could be found only by adopting an impossibly broad definition of schizophrenia, by carrying out the wrong statistical analyses (Joseph 2003), and by including ‘interview’ data from participants who sceptical observers later discovered had been dead at the time of the alleged interview (Rose et al. 1985). Although the data from these studies is almost scientifically worthless, they continue to be widely cited. Moreover, some of these bad practices have persisted in the new era of molecular genetics. For example, the first investigators to report a linkage between a particular gene and ‘schizophrenia’ were only able to obtain significant results by including as cases of the disorder persons suffering from phobias and alcoholism (Sherington et al. 1988). Small wonder that, until recently, replicable associations between particular genes and schizophrenia have proven elusive (Crow 1997; Moldin 1997). Even when at last, such associations have been found, the genes identified appear to be ones of very small effect (for example, the gene neuregulin-1 is found in approximately 15 per cent of patients with a diagnosis of schizophrenia and about 7 per cent of unaffected members of the population (Sigurdsson 2004) which, because the patients are a very small proportion of the population, means that it is present in many more non-patients than patients) and not specifically associated with a particular diagnosis (Bramon and Sham 2004; Craddock et al. 2005).

A more subtle problem with the genetic objection to environmental influences concerns common misunderstandings about the implications of heritability estimates. These give values (between 0 and 1) indicating the proportion of the variance between individuals that can be attributed to genes. However, these values depend on the extent to which there is variability in the environment. If individuals all experience the same environment, variation in the occurrence of a disorder can only be accounted for by genetic differences, so important environmental determinants will not be recognised. For example, if everyone smoked, whether or not a particular person succumbs to lung cancer would presumably depend on genetic vulnerability alone, and researchers who paid too much attention to heritability estimates would suffer the illusion that an important environmental influence – exposure to tobacco smoke – plays no causal role whatsoever (Joseph 2003).

A further complication is that heritability estimates, even when environmental influences are understood and measured, pay no heed to gene– environment correlations (the tendency of individuals with particular genetic constitutions to experience particular environments). In fact, correlations of this kind are the norm in nature because individuals (even plants!) tend to create their own environments (Lewontin 1993). As a consequence, the occurrence of a trait such as intelligence, which is under very strong genetic control, may still depend on environmental influences such as exposure to an emotionally supportive and stimulating environment in the first few years of life. In a mathematical treatment of this issue, Dickins and Flynn (2001) use sporting prowess as a readily understandable hypothetical example illustrating the effect. As certain physical characteristics (for example, height) predispose individuals to excellence in sport, whether one becomes a world champion undoubtedly depends on the possession of particular genes. However, actual success also crucially depends on receiving the right coaching, an environmental influence. This apparent paradox is resolved when it is recognised that professional coaches seek out individuals with particular physical characteristics, and hence the relevant genes.

Overall, these arguments show that genetic research has very few implications for the possible contribution of the environment to psychosis. The only way of establishing whether particular environmental factors play a role is by studying them.

Misunderstanding the neurobiological evidence

Similar misunderstandings have beset the interpretation of neurobiological data. As in the case of the genetic evidence, much of the evidence is less clear cut than biological psychiatrists have been prepared to admit. For example, while it is true that a proportion of ‘schizophrenia’ patients have enlarged cerebral ventricles, the magnitude of this effect is much less than initially thought, and the phenomenon has also been found in patients with bipolar disorder and psychotic depression (Raz and Raz 1990). Hypofrontality (reduced activity in the frontal lobes), once touted as a clear biological concomitant of schizophrenia, is not found in the resting state and observed only when patients are asked to perform tasks that engage the frontal lobes (Liddle 1996), tasks which are highly sensitive to motivation (Green et al. 1992). Most remarkably of all, the dopamine hypothesis (which proposed that ‘schizophrenia’ is caused by some kind of overactivity of pathways employing the neurotransmitter dopamine), once the front-running biological theory of schizophrenia, has been abandoned by many researchers because of numerous failures to find evidence of dopamine abnormalities in the brains of patients (Carlsson 1995; McKenna 1994). (Two studies which have implicated the dopamine system suggest that it may be implicated only during acute psychotic crises (Laruelle and Abi-Dargham 1999) or in psychotic responses to trauma (Hamner and Gold 1998).)

However, it is not the weakness of the biological findings that concerns us here (given that minds are ultimately brains it would be nothing short of amazing if no neurobiological differences could be detected between patients and other people: Rose 1984) but, once again, the inferences that have been drawn from them. Numerous studies have demonstrated that experience quite literally shapes the structure of the brain. For example, animals separated from their parents early in life show neurotransmitter abnormalities in adulthood (Suomi 1997) and human females who have been sexually abused in childhood show reduced hypocampal volume when adults (Stein et al. 1997) whereas the size of the hypocampus increases the longer London taxi drivers remain in their profession (Maguire et al. 2000). Structural and neurotransmitter differences between patients and others might therefore be taken as evidence of environmental causation rather than evidence against it. This is precisely the argument developed by Read et al. (2001), who have noted similarities between the brain abnormalities reported in ‘schizophrenia’ patients and those observed in victims of childhood sexual abuse.

Politics

A final, more difficult to document set of explanations for the relative neglect of environmental determinants of psychosis can be loosely labelled ‘political’, and to some extent subsumes the sources of misunderstanding that we have already considered. Under this rubric we can include various ideological and economic forces that have conspired, over recent decades, to encourage simplistic reductionistic, biological explanations of psychosis, to discourage investigation of social influences such as the family environment and sexual abuse, and to foster those misconceptions about the biological evidence that we have already noted.

The historical development of these forces is complex, but they came together in the early 1970s, when the battle between biological psychiatry and the antipsychiatry movement was widely regarded as having being won by the former group (Shorter 1997). At this time, organised psychiatry in North America and Britain was seeking to reinforce its identity as a branch of medicine, partly as a reaction to the ideas expressed by psychiatric dissidents such as Laing (1967) and Szasz (1960), which threatened to undermine any rationale for the profession’s existence. DSM-III (American Psychiatric Association 1980) was one product of this renewed enthusiasm for biological approaches, and was authored by psychiatrists who saw themselves as returning to their Kraepelinian roots (Klerman 1978). Two further influences that reinforced the dom...

Table of contents

• Cover Page
• Title Page
• Copyright Page
• Figures and tables
• Foreword
• Acknowledgement
• Introduction
• Part I: Research and theoretical perspectives
• Part II: Specific populations
• Part III: From theory to therapy