By Daniel Freeman and Jason Freeman

According to the UK Centre for Economic Performance, mental illness accounts for nearly half of all ill health in the under 65s. But this begs the question: what is mental illness? How can we judge whether our thoughts and feelings are healthy or harmful? What criteria should we use?

This month sees the publication of the latest version of the psychiatrist’s bible: the American Psychiatric Association’s Diagnostic and Statistical Manual of Mental Disorders (DSM). The DSM is arguably the definitive reference work on mental illness, used by health services worldwide (though the World Health Organisation’s International Classification of Diseases and Health Related Problems is widely used in the UK). Sales of the previous edition, DSM-IV, are estimated at about a million copies — not bad for a book that runs to almost 1000 densely packed pages and retails for around £80.

What’s changed in DSM-5 — apart from the move from Roman to Arabic numerals in the title? Well, terms have been revised (“mental retardation” has become “intellectual disability”, for example). New disorders have been introduced. For instance, “premenstrual dysphoric disorder” has been added to the list of depressive disorders. And, perhaps most controversially, some professionals have worried that the threshold for diagnosis of certain disorders appears to have been lowered — meaning that more people may be classified as mentally ill. Indeed there is organised opposition to the new edition, exemplified by the International DSM-5 Response Committee.

The DSM’s basic approach, on the other hand, has remained consistent for more than 30 years: a painstaking enumeration of symptoms, designed to make the clinician’s task of diagnosis easier and more consistent. This is an objective that it has undoubtedly achieved. But are those diagnoses scientifically valid?

Take clinical depression, for example. Nine possible symptoms are listed in DSM-IV, and you’d need to report at least five of them to warrant a diagnosis. These symptoms must be sufficiently intense to really interfere with a person’s life and they must have lasted for a while.

One effect of this approach is to emphasize the severe end of a spectrum that also includes relatively mild psychological problems. So the DSM criteria won’t capture everyday fluctuations in mental health. And they won’t pick up people with, say, four symptoms rather than five.

Implicit here is a debate about the nature of mental illness. The DSM uses a medical model of psychiatric illness. It thinks in terms of separate, discrete disorders, just like physical medicine. The approach is binary: either you meet the criteria for a particular condition, or you don’t.

Many would argue that this kind of all-or-nothing attitude, with hundreds of separate conditions, doesn’t fit well with people’s real-life experience of psychological problems. Better instead to think of psychological experience as being dimensional — that is, encompassing a wide variety of experiences, from the unproblematic to the severely distressing. The further along that dimension, the more symptoms a person is likely to have and the more upsetting and disruptive those symptoms will be.

This is the psychological model of mental illness. It argues that there’s no binary opposition between disorder and ‘normality’. Psychological disorders are simply the extreme manifestation of traits that we all possess to varying degrees. For example, almost everyone experiences occasional feelings of anxiety. People who develop what the DSM classes as an anxiety disorder aren’t experiencing something qualitatively different. They’re simply undergoing a more intense version of the same thing.

There is a third approach to understanding mental illness: the sociological model. Proponents argue that psychological disorders aren’t illnesses at all. They’re a label used to stigmatize and control behaviour society deems objectionable — such as homosexuality, which featured in the DSM until 1980.

Our view is that psychological problems aren’t illusory. They are real expressions of distress, for which most people — understandably — want help. However there is variability in the validity of individual diagnoses. Therefore it is often wisest not to focus on particular diagnoses. Better instead to adopt a dimensional approach, and to concentrate on the key problems and day-to-day symptoms that lead people to seek assistance. To help us understand these problems, we can look at epidemiological information to see which experiences occur together, and therefore may share common causes. Psychologists call this a data-driven approach.

We can also be guided by our knowledge of how the brain works. For example, basic emotions such as fear or unhappiness are powered by relatively distinct circuits in the brain. So we can understand certain psychological problems as what follow when these emotional circuits don’t function properly. We can match up the emotion and the problem: sadness and depression, fear and anxiety disorders, for example. This is what we might call a theory-driven approach, though given the complexity of brain activity it may – at least at present — be a little optimistic.

Importantly, even such a psychological, evidence-based approach doesn’t get around the need to classify problems. Mental health professionals must still make decisions about how to label the problems people describe to them. Without some kind of classificatory system, we can’t communicate, research, and evaluate treatments.

But the problems inherent in the current systems arguably constitute the greatest obstacle to that work. Given the extent of the burden on society and individuals alike, improving the scientific understanding of psychological disorders remains a priority. And that means DSM-5 certainly won’t be the last word on the classification of mental illness.

Daniel Freeman is a Professor of Clinical Psychology in the Psychiatry Department at the University of Oxford. Jason Freeman is a writer and editor. Their latest book is The Stressed Sex: Uncovering the Truth about Men, Women, and Mental Health (Oxford University Press).

The OUPblog is running a series of articles on the DSM-5 in anticipation of its launch on 18 May 2013. Stay tuned for views from Donald W. Black, Michael A. Taylor, and Joel Paris. Read yesterday’s post “DSM-5 will be the last” by Edward Shorter.

Subscribe to the OUPblog via email or RSS.
Subscribe to only psychology articles on the OUPblog via email or RSS.
Image credit: Thinker, created by Auguste Rodin at the end of the 18 century. San Francisco Legion of Honor. © Rafael Ramirez Lee via iStockphoto.

SHARE: