5 4

4

What Is (and Is Not) a Mental Disorder

One of the problems in DSM-5 is that it extends the concept of mental disorder and can be used to diagnose those who have only subclinical symptoms or problems. This is a danger because it could lead to the creation of new categories as well as broader definitions of existing ones.

We need to decide what we mean by “mental disorder” and to differentiate it from life’s vicissitudes—what Freud (1896/1957) once referred to as “normal human unhappiness.” This definition is crucial for determining the scope of psychiatry (Kagan, 2012; McNally, 2011). The ultimate question is whether DSM-5 describes a set of illnesses or problems associated with living.

Disease and Disorder

Medicine describes pathological states with terms such as disease or illness. Disease refers to physical abnormalities (e.g., anatomi- cal lesions and physiological or biochemical changes) that cause discomfort or dysfunction. Illness is often used as a synonym for disease, but it may also be used to describe the subjective feeling of “being ill” (Eisenberg, 1977).

In psychiatry, the use of the term mental disorder reflects a prob- lem in defining true diseases of the mind. A disease process is based on a known and specific etiology and pathogenesis. But there are no consistent biological markers in psychiatry reflecting the pathologi- cal mechanisms behind illness. This was so 40  years ago (Kendell,

Co py ri gh t @ 20 15 . Ox fo rd U ni ve rs it y Pr es s.

Al l ri gh ts r es er ve d. M ay n ot b e re pr od uc ed i n an y fo rm w it ho ut p er mi ss io n fr om t he p ub li sh er , ex ce pt f ai r us es p er mi tt ed u nd er U .S . or a pp li ca bl e co py ri gh t la w.

EBSCO : eBook Collection (EBSCOhost) – printed on 11/28/2018 9:05 PM via WALDEN UNIV AN: 939818 ; Paris, Joel.; The Intelligent Clinician’s Guide to the DSM-5® Account: s6527200.main.eds

4 W h a t I s ( a n d I s n o t ) a M e n t a l D i s o rd e r | 5 5

1975) and remains so today (Paris, 2008a). Thus, clinicians have to rely on signs and symptoms that cause distress or disability. That is why we use the term “disorder,” but psychiatrists may forget that disorder is not disease.

Finally, although the use of the term “mental disorder” is less potentially stigmatic than “mental illness,” a few clinicians and patients still avoid it in favor of misleading and vague concepts such as “mental health condition.” But whatever you call them, mental disorders are frightening and threatening to personal autonomy. For this reason, stigma can be reduced but not eliminated.

Defining Mental Disorder

DSM-5 offers a complex definition of mental disorder. Patients must have a behavioral or psychological syndrome or pattern that reflects an underlying psychobiological dysfunction, the consequences of which are clinically significant distress (e.g., a painful symptom) or disability (i.e., impairment in one or more important areas of func- tioning) that must not be merely an expectable response to com- mon stressors and losses (e.g., the loss of a loved one), a culturally sanctioned response to a particular event (e.g., trance states in reli- gious rituals), or a result of social deviance or conflicts with society. A disorder should have diagnostic validity based on a set of external validators (prognostic significance, psychobiological disruption, or response to treatment), and it should also have clinical utility (con- tributing to better conceptualization or to better assessment and treatment). Finally, diagnostic validators and clinical utility should help differentiate the disorder from its “near neighbors.”

As in all editions since DSM-III, the definition of mental disor- der includes a set of caveats. Thus, symptoms must not appear as a part of normal development or reflect cultural variations alone. They must not be developmental quirks (e.g., the moodiness of nor- mal adolescents) or cultural patterns (e.g., the possession states cul- tivated by some religions).

Co py ri gh t @ 20 15 . Ox fo rd U ni ve rs it y Pr es s.

Al l ri gh ts r es er ve d. M ay n ot b e re pr od uc ed i n an y fo rm w it ho ut p er mi ss io n fr om t he p ub li sh er , ex ce pt f ai r us es p er mi tt ed u nd er U .S . or a pp li ca bl e co py ri gh t la w.

EBSCO : eBook Collection (EBSCOhost) – printed on 11/28/2018 9:05 PM via WALDEN UNIV AN: 939818 ; Paris, Joel.; The Intelligent Clinician’s Guide to the DSM-5® Account: s6527200.main.eds

5 6 | Pa Rt I   D I a g n o s t I C P R I n C I P l E s

Each category in the manual needs to meet this overall defini- tion. But because pathology and normality can sometimes lie on a spectrum, it is often unclear whether symptoms meet these overall criteria. That is a reason why psychiatry needs to be conservative about decisions to change criteria for any disorder or to add or delete any category. Given that even minor changes in wording can vastly increase the prevalence of a diagnosis, a risk–benefit analysis needs to be applied to assess the impact of any changes from DSM-IV to DSM-5. We should be sure that benefits follow from changes. Yet over the years, the DSM system has been more notable for adding than for subtracting, even when additions carry an unknown risk.

The Theoretical Agenda of DSM-5

Traditionally, medicine has defined disease in a way that separates pathology from normality. We all have illnesses from time to time but otherwise consider ourselves as normal. Psychiatry took the same view for most of its history, and it remains reasonable to separate disease-like disorders such as schizophrenia, bipolar dis- order, and melancholic depression from reaction patterns such as mild depression or anxiety disorders. The neo-Kraepelinian model of mental disorder was in accord with these principles. But practi- tioners wanted a system that covers all conditions they are asked to treat, and some clinicians see people who are more unhappy than ill. This is the main reason for the overinclusiveness of the DSM system.

Psychiatry is not alone in this regard. Medical theory and prac- tice has been gradually expanding its scope, “medicalizing” subclini- cal symptoms as well as life’s ups and downs. For example, people can go to doctors to adjust their cholesterol level, in the absence of any symptoms of disease. It has been suggested that this trend suits pharmaceutical companies, which engage in “disease-mongering” to increase profits (Moynihan et al., 2002).

DSM-5 sought to overturn the neo-Kraepelinian model and replace it with one in which illness is not separate from normality

Co py ri gh t @ 20 15 . Ox fo rd U ni ve rs it y Pr es s.

Al l ri gh ts r es er ve d. M ay n ot b e re pr od uc ed i n an y fo rm w it ho ut p er mi ss io n fr om t he p ub li sh er , ex ce pt f ai r us es p er mi tt ed u nd er U .S . or a pp li ca bl e co py ri gh t la w.

EBSCO : eBook Collection (EBSCOhost) – printed on 11/28/2018 9:05 PM via WALDEN UNIV AN: 939818 ; Paris, Joel.; The Intelligent Clinician’s Guide to the DSM-5® Account: s6527200.main.eds

4 W h a t I s ( a n d I s n o t ) a M e n t a l D i s o rd e r | 5 7

but, rather, defined by a cutoff point on a continuum. Kupfer and Regier (2011) claimed that diagnostic spectra are supported by neuroscience research. That implies that even if people feel normal, everyone may have a bit of illness. It has long been known that men- tal disorders lack a sharp separation from normal functioning—or from each other (Kendell, 1975). But if you identify mental disorder in everyone, the concept loses meaning, and the scope of psychiatry becomes broad to the point of absurdity.

The Boundary Between Illness and Life

An old witticism states that life is a disease for which psychiatry is the cure. Behind the joke lies a reality:  It is not obvious what dis- tinguishes mental disorder from unhappiness. Psychiatry must dis- tinguish between sadness and depression, between moodiness and bipolarity, and between eccentricity and psychosis. That is what has traditionally defined the very concept of psychopathology.

The DSM manuals suffer from what military historians call mission creep—the gradual but inevitable expansion of a mission beyond its original goals. The distinction between severe mental disorders and milder disorders that reflect distress in the face of circumstance has often been ignored (Horwitz, 2002). Many cat- egories are included that do not meet overall criteria for a mental disorder in that they present symptoms that produce distress but are reactive to circumstance. But DSM has been written to include every sort of problem, whether or not it constitutes a disorder. This problem undermines the validity of the system.

Because no one can say what is or is not a mental disorder, all editions of DSM have suffered from overinclusiveness. Moreover, “medicalization” reformulates the human condition as a set of illnesses—that is, problems that lie beyond one’s personal control (Conrad, 2007). Medicalization often comes not from physicians but, rather, from patient groups seeking to destigmatize problems. Thus, Alcoholics Anonymous promoted a medical model of problem drinking long before physicians accepted it. Similarly, consumer

Co py ri gh t @ 20 15 . Ox fo rd U ni ve rs it y Pr es s.

Al l ri gh ts r es er ve d. M ay n ot b e re pr od uc ed i n an y fo rm w it ho ut p er mi ss io n fr om t he p ub li sh er , ex ce pt f ai r us es p er mi tt ed u nd er U .S . or a pp li ca bl e co py ri gh t la w.

EBSCO : eBook Collection (EBSCOhost) – printed on 11/28/2018 9:05 PM via WALDEN UNIV AN: 939818 ; Paris, Joel.; The Intelligent Clinician’s Guide to the DSM-5® Account: s6527200.main.eds

5 8 | Pa Rt I   D I a g n o s t I C P R I n C I P l E s

groups have actively promoted diagnoses such as attention-deficit hyperactivity disorder (ADHD) and posttraumatic stress disorder.

Almost everything that creates trouble in human life can be found in the DSM manuals. Badly misbehaving children can be diagnosed with conduct disorder (Wakefield et  al., 2002). Adults who are painfully shy can be diagnosed with a social anxiety disor- der (Horwitz & Wakefield, 2012). Low mood after losses may justify a diagnosis of depression (Horwitz & Wakefield, 2007). Recurrent episodes of rage can be diagnosed as intermittent explosive disor- der. It does not matter how common the problem is—even tobacco addiction is listed as a mental disorder.

Given this level of inclusiveness, it should be not surprising that epidemiological studies, such as the National Comorbidity Survey (NCS-R), that examine the community prevalence of DSM-defined disorders have found mental disorders to be very common. Approximately 20% of the population will meet criteria for at least one disorder in any given year, and at least half will do so in a life- time (Kessler et al., 2005a). Some have argued that these numbers are still too low. Reporting on a prospective community study of a sample followed from childhood to age 32 years, Moffitt et al. (2009) found that prevalence of disorders measured at the time they actu- ally appear was nearly double than what people remembered and reported in retrospective studies.

Evidently, mental disorder is ubiquitous. If the lifetime preva- lence of physical illness is 100%, perhaps a 50% rate for mental dis- orders is an encouragingly low number. However, there are other explanations for these epidemiological findings. When prevalence is very high, you have to ask whether measurements are accurate. All these numbers assume the validity of the categories listed in the DSM manual. That is a very big assumption. In the first large-scale survey, the Epidemiological Catchment Area Study (Robins & Regier, 1991), the estimates were much more cautious. Since then, diagnostic inflation, based on expansion of many DSM categories, led to much higher prevalence. It is also possible that psychiatric epidemiology made a fundamental error by agreeing to measure DSM categories rather than the symptoms on which they are based.

Co py ri gh t @ 20 15 . Ox fo rd U ni ve rs it y Pr es s.

Al l ri gh ts r es er ve d. M ay n ot b e re pr od uc ed i n an y fo rm w it ho ut p er mi ss io n fr om t he p ub li sh er , ex ce pt f ai r us es p er mi tt ed u nd er U .S . or a pp li ca bl e co py ri gh t la w.

EBSCO : eBook Collection (EBSCOhost) – printed on 11/28/2018 9:05 PM via WALDEN UNIV AN: 939818 ; Paris, Joel.; The Intelligent Clinician’s Guide to the DSM-5® Account: s6527200.main.eds

4 W h a t I s ( a n d I s n o t ) a M e n t a l D i s o rd e r | 5 9

One also needs to be sure that a disorder is disabling. This prin- ciple led DSM-IV to require all diagnoses to be based on symptoms that are clinically significant. The problem is that this concept requires a serious judgment call. In major depression, Wakefield et al. (2010) noted that because symptoms already measure subjective dis- tress, adding such a requirement does not distinguish cases from non-cases. The real question concerns severity. What is the cutoff point at which distress and disability qualify as mental illness?

Many problems that merit a diagnosis under the current system are painful but not disabling. For example, mass screening methods for depression are more likely to uncover transient episodes than clinical conditions that could benefit from treatment (Patten, 2008; Thombs et  al., 2008). Thus, even if most people who meet criteria for psychiatric diagnoses are never treated (Kessler et  al., 2005b), that need not be a matter of great concern—as long as the sickest patients find a pathway to care.

Psychiatry is a branch of medicine, but one does not expect the majority of the population to have either clinical or subclini- cal disorders of the heart, kidney, or liver. This is what makes the findings of epidemiological research based on DSM categories hard to swallow. Some might say that a lifetime prevalence of 50% reflects a reality we just have to accept. The leaders of the National Comorbidity Study, a large-scale epidemiological survey based on DSM-IV (Kessler et al., 2003), took the view that psychiatry, like the rest of medicine, must make room for mild and subclinical disorders in its classification system. Much as general physicians treat com- mon colds as well as pneumonia, mental health clinicians need not actively discourage people with less severe problems from coming for help. Kessler et al. also argued that mild disorders could be pre- cursors of more severe disorders at some later point—in which case, early treatment might be preventive. However, they did not provide data on how often that actually happens or whether prevention is a practical option.

Admitting subclinical phenomena into a diagnostic classifica- tion is a very slippery slope. The lifetime prevalence of mental dis- orders could easily come to approach 100%. The boundary between

Co py ri gh t @ 20 15 . Ox fo rd U ni ve rs it y Pr es s.

Al l ri gh ts r es er ve d. M ay n ot b e re pr od uc ed i n an y fo rm w it ho ut p er mi ss io n fr om t he p ub li sh er , ex ce pt f ai r us es p er mi tt ed u nd er U .S . or a pp li ca bl e co py ri gh t la w.

EBSCO : eBook Collection (EBSCOhost) – printed on 11/28/2018 9:05 PM via WALDEN UNIV AN: 939818 ; Paris, Joel.; The Intelligent Clinician’s Guide to the DSM-5® Account: s6527200.main.eds

6 0 | Pa Rt I   D I a g n o s t I C P R I n C I P l E s

normality and pathology would then be completely lost. Unless disorders are defined in a way that requires severe dysfunction, almost every bump on the road of life will be considered pathologi- cal. These problems also follow from the view that psychopathology of all kinds is dimensional and lies on a spectrum with normality (Pierre, 2010). Everyone has a mental disorder, the only question being how severe. This paradigm threatens to trivialize psychiatry. To be taken seriously, the specialty has to define disorder in a way that recognizes a difference between problems of living and mental illness.

Harmful Dysfunction

Jerome Wakefield, a professor of social work at New York University, is a seminal figure in the debate about the boundaries between nor- mality and pathology. He has proposed defining mental disorder in terms of a construct he calls harmful dysfunction (Wakefield, 1992).

These are two words, each of which requires a precise definition. For Wakefield, dysfunction refers to an inability to carry out life tasks specified by evolutionary mechanisms. Thus, conditions such as psychosis, melancholic depression, or severe substance abuse prevent people from looking after themselves or from living in families and raising children. In severe mental illness, dysfunction is obvious because it leads to striking disability. The problem lies with boundary cases. At what point is reduced function considered dysfunction?

The word “harmful” adds a component of values. It means that symptoms hurt those who suffer them and/or other people with whom they are involved. But nearly every symptom patients experi- ence is harmful in some way.

The usefulness of Wakefield’s definition is that to define disor- der, both harm and dysfunction are required. Thus, behavior that is only harmful (e.g., laziness and rudeness) would not justify a medical diagnosis. Nor would behavior that is only dysfunctional (e.g., drunkenness). A  hybrid definition, combining harm and

Co py ri gh t @ 20 15 . Ox fo rd U ni ve rs it y Pr es s.

Al l ri gh ts r es er ve d. M ay n ot b e re pr od uc ed i n an y fo rm w it ho ut p er mi ss io n fr om t he p ub li sh er , ex ce pt f ai r us es p er mi tt ed u nd er U .S . or a pp li ca bl e co py ri gh t la w.

EBSCO : eBook Collection (EBSCOhost) – printed on 11/28/2018 9:05 PM via WALDEN UNIV AN: 939818 ; Paris, Joel.; The Intelligent Clinician’s Guide to the DSM-5® Account: s6527200.main.eds

4 W h a t I s ( a n d I s n o t ) a M e n t a l D i s o rd e r | 6 1

dysfunction, aims to cut this Gordian knot. Even so, determining whether each of these criteria is present requires judgment calls that may not be strictly objective, and there is also an overlap between harm and dysfunction. The definition of mental disorder in DSM-5 is not very different from the concept of harmful dysfunction, but the devil lies in the details.

The Scope of DSM

Mission creep has steadily expanded the boundaries of mental ill- ness. If a survey examining the presence of mental disorder iden- tifies people who consider themselves normal but who actually meet criteria for a diagnosis, that constitutes a false negative. But if the same survey identifies people as having a disorder when criteria are not met, that constitutes a false positive. The concept of mental disorder used by the DSM system is most likely to lead to false positives. This problem bedevils DSM-5. It has no way to separate clinical from subclinical phenomena. And it is up to the clinician to decide what is “significant.” In the absence of a precise definition, the concept of “clinical significance” can only be imprecise.

Since the third edition, DSM has included an increasingly long list of diagnoses. Every edition since has grown larger in scope, and the size of the manual has also grown. Again, it seems that mis- sion creep rules. Observing this trend, Zorumski (2009, p.  xxvi) commented wryly, “One might conclude that either the field has advanced greatly or we have now generated a system that codifies many poorly studied and poorly validated descriptors.”

Robert Spitzer once told me he wrote DSM-III with the aim of being “inclusive”—he thought it best to include more categories and sort out their validity later. That was a mistake. What Spitzer did not take into account is that once a category is listed in the manual, it is very difficult to remove. Too many people have a stake in main- taining it. When it came time to publish DSM-IV, only a few diagno- ses were taken out, while quite a few others were added.

Co py ri gh t @ 20 15 . Ox fo rd U ni ve rs it y Pr es s.

Al l ri gh ts r es er ve d. M ay n ot b e re pr od uc ed i n an y fo rm w it ho ut p er mi ss io n fr om t he p ub li sh er , ex ce pt f ai r us es p er mi tt ed u nd er U .S . or a pp li ca bl e co py ri gh t la w.

EBSCO : eBook Collection (EBSCOhost) – printed on 11/28/2018 9:05 PM via WALDEN UNIV AN: 939818 ; Paris, Joel.; The Intelligent Clinician’s Guide to the DSM-5® Account: s6527200.main.eds

6 2 | Pa Rt I   D I a g n o s t I C P R I n C I P l E s

DSM-5 has also failed to remove invalid diagnoses. But, to its credit, it did not accept every proposal for new categories. One example is the category of “relationship disorder” (First et  al., 2002). Problems with other people, without overt symptoms, are ubiquitous and do not constitute a mental illness. Many reason- ably normal people get divorced, never marry, or fail to get  along with their children. In this case, mission creep is driven by the fact that insurance companies expect DSM categories to be coded for reimbursement. However, people who are unhappy with their rela- tionships (sometimes called “the worried well”) may seek psycho- therapy, but they do not deserve a medical diagnosis. If DSM-5 had agreed to include relationship disorders, psychiatry could have con- gratulated itself on finally succeeding in raising the prevalence of mental disorder to 100%.

DSM is already sufficiently elastic that people with very mild symptoms can be diagnosed with something. As shown in a survey of patients in psychoanalysis (Doidge et al., 2002), many patients who are functional enough to afford that expensive treatment meet criteria for common mental disorders (anxiety and mood disorders). Some also meet overall criteria for a personality dis- order, although there is a difference between lifelong dysfunc- tion with a wide range of problems and having trouble relating to an intimate partner. Quite a few people who seek therapy hold steady jobs and have relationships (even if they are not quite sat- isfactory). These people are troubled but not ill. DSM allows for the possibility that a patient can have no mental disorder but still have problems that are a focus of treatment. These problems can receive “V codes” (meaning that the patient has life problems but not a mental disorder). But then insurance would not pay for treatment.

To avoid mission creep, DSM could have confined itself to prob- lems that almost anyone would call a mental illness. This could be accomplished by reducing the number of categories and/or by mak- ing severity criteria more stringent. Then we might truly have a manual of mental disorders—not of life.

Co py ri gh t @ 20 15 . Ox fo rd U ni ve rs it y Pr es s.

Al l ri gh ts r es er ve d. M ay n ot b e re pr od uc ed i n an y fo rm w it ho ut p er mi ss io n fr om t he p ub li sh er , ex ce pt f ai r us es p er mi tt ed u nd er U .S . or a pp li ca bl e co py ri gh t la w.

EBSCO : eBook Collection (EBSCOhost) – printed on 11/28/2018 9:05 PM via WALDEN UNIV AN: 939818 ; Paris, Joel.; The Intelligent Clinician’s Guide to the DSM-5® Account: s6527200.main.eds

4 W h a t I s ( a n d I s n o t ) a M e n t a l D i s o rd e r | 6 3

Sensitivity and Specificity

Every psychiatric diagnosis is a trade-off. Sensitivity measures the proportion of positive cases correctly identified. Specificity measures the proportion of negative cases correctly identified. Every time one widens the criteria for any diagnosis, cases that might have been missed will be found, but one runs the risk of diagnosing people who are not cases.

From the very beginning, DSM has lacked specificity, resulting in multiple diagnoses (“comorbidity”). Moreover, there been more diagnoses in each edition, and criteria have tended to soften over time, with more unhappy people seen as depressed, more moody people considered to be bipolar, and more inattentive people con- sidered to have ADHD. DSM-5 moves even further in the same direction, loosening up the criteria for many disorders. This result is what one should expect from workgroups filled with academic mavens. Experts usually believe that the disorders that most inter- est them are more prevalent than any one realizes, even if they mas- querade as other problems. The inevitable result is that increasingly more people are defined as ill.

Mental Illness and Stigma

Much more than any physical illness, mental illness is associated with stigma (Corrigan, 2005). Despite all the progress psychiatry has made, the situation has not changed much. Stigma reflects neg- ative social judgments about people who suffer from psychological problems of any kind. Perhaps these attitudes derive from the fear we all have of being out of control of our own minds, leading to a critical view of mental illness and the mentally ill.

Stigma reflects the way we think about ourselves. Life is rarely easy, even for those who think of themselves as mentally healthy. When upsetting things happen, such as the loss of a job or a rela- tionship, it is normal to experience psychological symptoms. It is

Co py ri gh t @ 20 15 . Ox fo rd U ni ve rs it y Pr es s.

Al l ri gh ts r es er ve d. M ay n ot b e re pr od uc ed i n an y fo rm w it ho ut p er mi ss io n fr om t he p ub li sh er , ex ce pt f ai r us es p er mi tt ed u nd er U .S . or a pp li ca bl e co py ri gh t la w.

EBSCO : eBook Collection (EBSCOhost) – printed on 11/28/2018 9:05 PM via WALDEN UNIV AN: 939818 ; Paris, Joel.; The Intelligent Clinician’s Guide to the DSM-5® Account: s6527200.main.eds

6 4 | Pa Rt I   D I a g n o s t I C P R I n C I P l E s

not helpful to label these reactions with a diagnosis. Actually, people may benefit from normalizing difficult times in life. Periods of low or anxious mood can be seen as a “rough patch” rather than an illness. And whereas nobody feels stigmatized by a common cold, receiving a diagnosis of depression can have negative effects. Similarly, what is the benefit of diagnostic labeling for other common problems ranging from social awkwardness to the benign loss of memory that people experience with age?

DSM-5 could have maintained a boundary between true illness and life’s bumpy road. But influenced by the principle of dimension- ality, it chose not to do so. Ironically, the view that we are all just a little ill was held by Sigmund Freud and was long a principle of psychoanalysis—the theory overthrown by DSM-III. Neurobiological models of mental disorder have brought us full circle.

Diagnosis in Childhood

DSM-5 is designed for patients of all ages and makes a point of not separating adult and child psychiatry. This is a good idea because so many disorders begin in childhood and continue into adulthood. But children, by and large, do not always come to clinical attention unless parents and teachers are worried about them. Moreover, most patients seen in child psychiatry are boys, in contrast to the female predominance in adult psychiatry. The reason is that boys are more likely to have externalizing disorders that create trouble for others, which is what usually motivates a referral.

Some mental disorders are dormant in childhood and only emerge in adulthood so that many symptoms begin in adolescence (Copeland et  al., 2009). We often do not know whether diagnoses made in childhood are early forms of an adult disorder, separate disorders, or a bump on a developmental pathway. To answer this question, we need long-term follow-up research. Psychiatry has few studies of this kind largely because prospective research is so expen- sive. The patients seen in child psychiatry do not always come back as adults, and many adults seen by psychiatrists were never patients

Co py ri gh t @ 20 15 . Ox fo rd U ni ve rs it y Pr es s.

Al l ri gh ts r es er ve d. M ay n ot b e re pr od uc ed i n an y fo rm w it ho ut p er mi ss io n fr om t he p ub li sh er , ex ce pt f ai r us es p er mi tt ed u nd er U .S . or a pp li ca bl e co py ri gh t la w.

EBSCO : eBook Collection (EBSCOhost) – printed on 11/28/2018 9:05 PM via WALDEN UNIV AN: 939818 ; Paris, Joel.; The Intelligent Clinician’s Guide to the DSM-5® Account: s6527200.main.eds

4 W h a t I s ( a n d I s n o t ) a M e n t a l D i s o rd e r | 6 5

as children. Only a few categories (severe ADHD and conduct disor- der) are known to show developmental continuity. A  complex pic- ture has emerged in which some forms of pathology improve with age, whereas others are precursors for serious problems later. We are only at the beginning of the research that could address such questions.

It is difficult to determine the community prevalence of mental disorders in children and adolescents (Roberts et  al., 1998). Much information has to be gleaned from interviewing parents. When researchers try to determine how many children meet criteria for any DSM category, results tend to be inconsistent. The British are generally more conservative about diagnosis, and one survey in the United Kingdom (Ford et al., 2003) found an overall prevalence of diagnosable disorders of 9.5% in a community population of chil- dren. In contrast, a community survey of children ages 9–13 years in an American rural area (Costello et al., 2003) found that 31% of girls and 42% of boys met criteria for at least one DSM-IV disorder. These numbers depend on the validity of information, the choice of cutoff points for dysfunction, and the vagaries of DSM definitions. Using a more stringent cutoff for severity, Costello et  al. (2005) found that approximately one-fourth of these children met diagnos- tic criteria in a year, more similar to what one sees in adults (Kessler et al., 2005a). But that is still a very high number. These difficulties raise questions about the scope of diagnosis in child psychiatry. The underlying problem is similar to what we have seen in adults: What is a disorder, what is a time-limited problem, and what is a normal variant?

Michael Rutter (2011), a senior British child psychiatrist who has always been dubious about the validity of existing psychiatric diagnoses, commented in detail on the directions being taken by DSM-5 and ICD-11 in children, and he made a number of provoca- tive recommendations:

1. There are far too many diagnoses, leading to a high rate of sup- posed comorbidity. If the number were drastically reduced, so would the overlap between diagnoses.

Co py ri gh t @ 20 15 . Ox fo rd U ni ve rs it y Pr es s.

Al l ri gh ts r es er ve d. M ay n ot b e re pr od uc ed i n an y fo rm w it ho ut p er mi ss io n fr om t he p ub li sh er , ex ce pt f ai r us es p er mi tt ed u nd er U .S . or a pp li ca bl e co py ri gh t la w.

EBSCO : eBook Collection (EBSCOhost) – printed on 11/28/2018 9:05 PM via WALDEN UNIV AN: 939818 ; Paris, Joel.; The Intelligent Clinician’s Guide to the DSM-5® Account: s6527200.main.eds

6 6 | Pa Rt I   D I a g n o s t I C P R I n C I P l E s

2. There is no need for a separate grouping of disorders with an onset specific to childhood. Instead, the various specific disor- ders should be placed in appropriate groups in a classification that cuts across all developmental stages. (DSM-5 has adopted this change.)

3. A group should be formed of disorders known to occur but for which further testing for their validity is needed. This would allow new categories to be tested before being reified and cast in stone. (DSM-5 also adopts this stance.)

4. Categorical and dimensional approaches to diagnosis can be combined. However, dimensions should be introduced only where there is good evidence to support them.

5. The requirement of impairment should be removed from all diagnostic criteria, given problems in reliability and validity. Instead, functioning should be coded separately.

6. Research and clinical classifications should be kept separate. (That would have been one of my own recommendations for DSM-5.) Doing so would make clinical utility much easier to achieve. Similarly, there is a need to develop a primary care classification for both medical and nonmedical primary care.

Each of Rutter’s suggestions would lead to a more conservative, evidence-based approach to psychiatric diagnosis. They also have implications that go far beyond child psychiatry and address prob- lems that afflict the DSM system as a whole. I  can only state my approval—and my regret that only a few of these principles were adopted.

DSM-5 and the Role of the Specialist

DSM-5 has many purposes. If it were to concentrate on being a sci- entific categorization of mental illness, it would be less inclusive. When every human problem finds a place in a diagnostic manual, psychiatry’s mission to provide specialized medical care to severely ill people is undermined.

Co py ri gh t @ 20 15 . Ox fo rd U ni ve rs it y Pr es s.

Al l ri gh ts r es er ve d. M ay n ot b e re pr od uc ed i n an y fo rm w it ho ut p er mi ss io n fr om t he p ub li sh er , ex ce pt f ai r us es p er mi tt ed u nd er U .S . or a pp li ca bl e co py ri gh t la w.

EBSCO : eBook Collection (EBSCOhost) – printed on 11/28/2018 9:05 PM via WALDEN UNIV AN: 939818 ; Paris, Joel.; The Intelligent Clinician’s Guide to the DSM-5® Account: s6527200.main.eds

4 W h a t I s ( a n d I s n o t ) a M e n t a l D i s o rd e r | 6 7

When psychiatry moved back into the medical mainstream, it returned to its historical roots and to the treatment of severely ill patients (Paris, 2008a). The focus of practice has greatly changed. Psychiatrists now define themselves by their expertise in psy- chopharmacology, backed up by their diagnostic acumen. Most now spend little time on psychotherapy, which in the future may not be carried out by medical specialists. Psychotherapy for psychiatrists could become like physiotherapy for orthopedic surgeons—a procedure to be referred out to another clinician. The role of the psychiatrist now focuses on patients who need their unique skills, not on those who could be managed by other mental health professionals.

Psychiatrists play a crucial role as consultants. They are trained to evaluate pathology and to establish a diagnosis. DSM-5 should support that priority, not undermine it. This is why it needs to distinguish between patients who have illness and who need medical treatment and those who can see other profession- als for life problems. Although psychiatrists still have to know how to carry out psychotherapy, they should not offer it to nor- mal people, even if the DSM system has categories that could justify doing so. At the same time, psychiatrists are prescribing vast amounts of medication to relieve unhappiness and common human problems. As Norman Sartorius (2011), one of the prime movers behind ICD-11, has warned, psychiatry could lose public respect if its classification conflates mental illness with normal life experience.

Yet in all fairness, psychiatry is only doing what the rest of medicine has been doing for some time—diagnosing patients who are not ill but who have risk factors for illness and treating people who may not need treatment. One only has to look at the history of medicine’s approach to cholesterol levels or to hypertension to find examples. Moreover, early diagnosis has become such a priority that people without a disease are being treated as if they had one. The misuse of blood tests for prostate cancer (and of mammograms for breast cancer) shows that even biological markers provide no protection from overdiagnosis and overtreatment.

Co py ri gh t @ 20 15 . Ox fo rd U ni ve rs it y Pr es s.

Al l ri gh ts r es er ve d. M ay n ot b e re pr od uc ed i n an y fo rm w it ho ut p er mi ss io n fr om t he p ub li sh er , ex ce pt f ai r us es p er mi tt ed u nd er U .S . or a pp li ca bl e co py ri gh t la w.

EBSCO : eBook Collection (EBSCOhost) – printed on 11/28/2018 9:05 PM via WALDEN UNIV AN: 939818 ; Paris, Joel.; The Intelligent Clinician’s Guide to the DSM-5® Account: s6527200.main.eds

6 8 | Pa Rt I   D I a g n o s t I C P R I n C I P l E s

Diagnostic Inflation and Diagnostic Epidemics

Failure to draw boundaries between pathology and normality leads to diagnostic inflation (Frances, 2009c). There have been several examples in recent years, with increasing identification of condi- tions such as bipolar disorder, ADHD, autism, and generalized anxi- ety disorder (GAD). Each of these diagnoses has a fuzzy boundary with normality:  Bipolarity could just be moodiness, ADHD could just be impulsivity or inattentiveness, autism could just be social ineptness, and GAD could just be a tendency to worry too much. But the DSM system has encouraged physicians to identify all these conditions as mental disorders. This has led to an enormous number of false positives.

In some cases, the increase in identification has been so dramatic than one can speak of a diagnostic epidemic. Conditions that once seemed rare have now become common. No one seems to be able to escape the possibility that being different will be labeled as being disordered. The problem of false positives in diagnosis is encouraged by a system that is more concerned about “missing” something than about giving incorrect labels and unnecessary treatment to normal people. The consequences can sometimes be severe. If moody people are called bipolar, they will be treated with drugs that can have dan- gerous side effects. If impulsive or inattentive people are routinely diagnosed with ADHD, they may be put on stimulants for years. If socially awkward people are seen as falling within the autistic spec- trum, they will suffer unneeded stigma. If worried people are diag- nosed with GAD, they will be put on antidepressants that may or may not be helpful. Similarly, the interest in identifying mental dis- orders even before they start has created another set of problems. If one is too quick to diagnose early psychosis, or neurocognitive disorder, patients who will never develop a serious mental disorder will be stigmatized and treated unnecessarily.

The diagnostic categories in DSM, which are at best provisional, are unavoidably fuzzy, blending into normality at the edges. But

Co py ri gh t @ 20 15 . Ox fo rd U ni ve rs it y Pr es s.

Al l ri gh ts r es er ve d. M ay n ot b e re pr od uc ed i n an y fo rm w it ho ut p er mi ss io n fr om t he p ub li sh er , ex ce pt f ai r us es p er mi tt ed u nd er U .S . or a pp li ca bl e co py ri gh t la w.

EBSCO : eBook Collection (EBSCOhost) – printed on 11/28/2018 9:05 PM via WALDEN UNIV AN: 939818 ; Paris, Joel.; The Intelligent Clinician’s Guide to the DSM-5® Account: s6527200.main.eds

4 W h a t I s ( a n d I s n o t ) a M e n t a l D i s o rd e r | 6 9

through constant usage, clinicians have come to think of them as valid, in the same way as pneumonia. Thus, diagnoses in psychiatry easily become “reified”—hypothetical constructs treated as if they were real. We forget that categories are only a way of communicat- ing. We can only await the day when we truly understand mental illness, but in the meantime, we should be careful about making diagnoses too easily. It takes time to know what patients are really like; some diagnoses are made “on the fly” by practitioners who are too busy to take the necessary time. Although psychiatrists need a classification system to talk to each other and to patients and fami- lies, in a shared language, they should not use diagnosis to describe life itself.

Co py ri gh t @ 20 15 . Ox fo rd U ni ve rs it y Pr es s.

Al l ri gh ts r es er ve d. M ay n ot b e re pr od uc ed i n an y fo rm w it ho ut p er mi ss io n fr om t he p ub li sh er , ex ce pt f ai r us es p er mi tt ed u nd er U .S . or a pp li ca bl e co py ri gh t la w.

EBSCO : eBook Collection (EBSCOhost) – printed on 11/28/2018 9:05 PM via WALDEN UNIV AN: 939818 ; Paris, Joel.; The Intelligent Clinician’s Guide to the DSM-5® Account: s6527200.main.eds