Journal Information
Journal ID (publisher-id): jgi
ISSN: 1910-7595
Publisher: Centre for Addiction and Mental Health
Article Information
© 1999-2005 The Centre for Addiction and Mental Health
Publication date: December 2005
Publisher Id: jgi.2005.15.15
DOI: 10.4309/jgi.2005.15.15

The categorization of pathological gambling and the Impulse-Control Disorders Not Elsewhere Classified
Richard Rosenthal Role: Presenter
Proceedings of the 19th annual conference on prevention, research, and treatment of problem gambling. June 23–25, 2005, in New Orleans, Louisiana. National Council on Problem Gambling, Washington, DC.
Session IV: Towards the DSM-V

For correspondence:

(Introduction.)Jon Grant: It is my great pleasure to introduce Richard Rosenthal. Richard has been a pioneer in understanding pathological gambling. In fact, he was instrumental in drafting the DSM-IV criteria for pathological gambling. He continues his research at UCLA and treats individuals with gambling addiction in his private practice.

Richard Rosenthal: Thank you, Jon. It's my impression that pathological gambling is not a difficult diagnosis to make. The criteria are reasonably straightforward; they work well. They even provide a simple measure of severity. There is a problem, however, although not with the diagnosis. It's with the classification. In other words, we can say whether someone is a pathological gambler, but not what that means. What is it? Often the question is worded, “Is it an addiction or an impulse-control disorder?”

I'm going to be discussing the classification of pathological gambling, and the category of the impulse-control disorders, by attempting to answer that question and two other questions that are commonly asked: Isn't the category of Impulse-Control Disorders Not Elsewhere Classified a wastebasket category? And why is the categorization of the impulse disorders so confusing?

Is pathological gambling an addiction or an impulse-control disorder?

That question has a very simple answer. It is both. Addictions are impulse disorders. First, consider the name of the category in which pathological gambling appears: the Impulse-Control Disorders Not Elsewhere Classified (IDNEC). This is a residual diagnostic category for disorders of impulse control that are not classified elsewhere in DSM-IV. DSM-III and DSM-III-R called attention to this in the introduction to their respective IDNEC chapters, and both offered examples of some of the other impulse disorders. However, the IDNEC committee for DSM-IV was aware that this was often overlooked, and that there was still confusion about the residual nature of the category, so they tried to make the introduction clearer and to give a more complete listing of the other impulse disorders. They include substance-related disorders, paraphilias, antisocial personality disorder, conduct disorder, schizophrenia, and some mood disorders. Thus, substance dependence—which had been similarly listed in DSM-III and III-R—is understood to be an impulse disorder, albeit one classified elsewhere.

Addiction, it should be noted, is not a word, or concept, which appears anywhere in DSM-IV. It was considered a layperson's term: too difficult to pin down or define. Instead, the preferred terms for the substance-related disorders are abuse and dependence. The most obvious comparison of substance dependence with pathological gambling occurred in DSM-III-R, where the criteria for the latter were taken directly from the former. This was most obvious in an earlier published draft of DSM-III-R, where one can see that the criteria for pathological gambling were taken directly from the criteria for substance dependence, with only the substitution of the word “gambling” for “intoxication” or “use.” In a 1992 paper in Psychiatric Annals, I placed the two sets of criteria side by side. One can appreciate that they're almost identical, curiously differing only in the number of criteria needed for diagnosis. Historically, if alcohol and substance dependence were thought to be “addictions,” so too was pathological gambling.

And, finally, the original definition of pathological gambling that the IDNEC committee for DSM-IV unanimously agreed upon was “a continuous or periodic loss of control over gambling; a progression, in frequency and in amounts wagered, and in the preoccupation with gambling and with obtaining money with which to gamble; irrational thinking; and a continuation of the behavior despite adverse consequences.”

This has been repeated in the literature and on Web sites and in educational materials and appears to have received some acceptance as an official definition of the disorder. It clearly is the definition of an addiction. Unfortunately, the senior editors for DSM-IV substituted another definition so as to put it into conformity with the other disorders. What currently appears in DSM-IV is “persistent and recurrent maladaptive gambling behavior that disrupts personal, family, or vocational pursuits.” This is unfortunate in that it doesn't add to our understanding of the disorder. What is “maladaptive gambling behavior”? It appears that the preferred definition of pathological gambling is that of an addiction.

Isn't the category of Impulse-Control Disorders Not Elsewhere Classified a wastebasket category?

Various authors have argued that the five disorders categorized as Impulse-Control Disorders Not Elsewhere Classified have little in common, and were placed together only due to a lack of agreement as to where else to put them. The IDNEC category has been described as something of a wastebasket.

Only kleptomania and pyromania seem to belong together, although there has been essentially nothing published about pyromania for the last 40 years, not even case reports. Trichotillomania may better fit under obsessive compulsive disorder, although some would prefer it to be classified under disorders of childhood or as a stereotypy/habit disorder. The IDNEC committee, like its III-R predecessor, questioned whether intermittent explosive disorder (IED) really existed, and decided to include it in order to encourage research that might provide an answer. In its present form, IED is a disorder of exclusion, to be made only after a number of other disorders have been ruled out. As already discussed, similarities between pathological gambling and substance dependence were obvious to the committee. Pathological gambling could have been classified with the substance-related disorders. That it wasn't may have been at least partly for turf-related or political reasons.

Thus one can argue that the Impulse-Control Disorders Not Elsewhere Classified is a category that exists only by default, and that the five disorders represented in it don't really belong together other than by accident or as part of some politically minded compromise.

I would like to take issue with this, and suggest that these five disorders have much in common, and that one can easily understand why they are grouped together. First of all, they are all old disorders. Kleptomania was named and described by Mathey in 1816, pyromania by Marc in 1833. Gambling mania may have been around a lot longer, but was the subject of a famous painting by Gericault in 1822. Impulsive homicidal mania was described during this period. Only trichotillomania, which Hallopeau didn't introduce until 1889, came later. Thus, the IDNEC disorders came into existence or were first described within a short time of one another.

A second trait shared by the group is that they are all deviant disorders, in that they either describe criminal behaviors (kleptomania, pyromania, perhaps intermittent explosive disorder), behaviors that frequently lead to criminal behavior (pathological gambling), or behaviors viewed with shame and disgust (trichotillomania). And, finally, a third trait is that the behavior may occur in a seemingly normal or otherwise normal individual.

Most importantly, what holds the category together are its historical roots, dating back to Esquirol's 1810 description of the monomanias. The defining characteristic of these disorders was the idée fixe, a single pathological preoccupation in an otherwise sound mind. What was revolutionary in Esquirol's new classification was this notion of partial insanity; that a person could otherwise be normal or appear normal when you talked to them, and unless you asked them the right question or somehow brought out this preoccupation of theirs—some driven kind of activity or delusional identity—they would appear normal. Esquirol also described the “irresistible impulse”: these people were driven to set fires, or hurt people, or steal, drink, or gamble.

Monomania became an extremely popular concept for about 60, 70 years. It not only dominated French psychiatry, it spread to other countries, and was taken up by the intellectuals, the artists and writers, and by the general public. One of the most important and lasting effects of the concept was in its use as an insanity defense. In 1825 one of Esquirol's protégés, Georget, introduced monomania into the courts. Prior to that the best witnesses for somebody accused of a crime, the so-called experts, were his neighbors—people who knew him when he was growing up. “Well, he was very quiet, and he always was nice to the children.” Now a new idea appeared, that it required an expert who would know what questions to ask. That the person could appear normal except in this one area of their behavior. This was the beginning of forensic psychiatry. The notion of the irresistible impulse remained in the court system for quite a while, although it's now in disfavor.

Why is the categorization of the impulse disorders so confusing?

First of all, the irresistible impulse: it's a wonderful phrase, but it's kind of like a ghost that has remained hanging over the category and has followed it from DSM-III to III-R to IV. People still think that the category talks about the inability to resist. In DSM-III, the pathological gambling section does mention being “unable to resist impulses to gamble.” But starting in DSM-III and dominating III-R and then IV, it doesn't say “unable to resist.” It says “failure to resist,” and broadens this further by saying not only “failure to resist an impulse,” but “failure to resist an impulse, drive, or temptation.”

So the essential characteristic of all of these disorders is failure to resist temptation. In other words it's a purely voluntary thing. If I fail to stop at a stop sign it's not because I can't. It may be because I don't want to. I look around, don't see a policeman, there are no cars. Maybe I don't believe in stop signs. Anyhow, I make a decision not to stop. There's the notion of volition. And the idea of loss of control has not been pinned down satisfactorily in the category.

In fact, there are no definitions offered either in the IDNEC chapter or in the glossary to DSM-III, III-R, or IV for what an impulse is, what impulsive means, or impulsivity. I think this has hindered research in this area, and, of course, there is difficulty with the construct of impulsivity. Just because somebody engages in self-destructive behavior, we say they're impulsive. They gamble. They set fires. They steal things.

An important distinction one needs to make is between specific and generalized types of impulsivity. For example, somebody can be a pathological gambler and act impulsively only in the area of their gambling, and perhaps in the rest of their life they are not impulsive. And the behavior, whether it's gambling or stealing or setting fires, may be purposive and defensive. They are engaging in this seemingly impulsive behavior for a reason. There is meaning to it. It has a defensive purpose. They are self-medicating.

One of the things I learned early in my training is that people will subject themselves to incredible amounts of pain in order to avoid pain. Therefore this purposive, meaningful kind of impulsivity as opposed to the more generalized, purposeless, random kind of impulsivity. Examples of the latter might be people who are more organic, who have no control over their behavior, and are more chaotic.

So that's one problem with the way impulsivity is used in relation to these disorders. There's another problem. When you talk about pathological gambling, there are about five kinds of impulsivity that are involved. You can be impulsive before you start gambling or you can be impulsive as a consequence of the gambling. The impulsivity before gambling can be because of some innate genetic predisposition or it can be secondary to a comorbid disorder, such as attention deficit hyperactivity disorder.

The impulsivity that's a consequence of the gambling may be because the individual is chasing. They have to get money because of some immediate debt or because somebody important to them is going to find out. They're afraid of losing their job, home, or marriage. The desperate behavior they're engaged in may look impulsive but is actually specific and goal directed. As gambling progresses, there may be an increase in shame, guilt, and depression. As a result, losing becomes more intolerable. Chasing increases in a desperate attempt to get even to undo the guilt and other painful effects.

Still another possibility is that the gambling progression leads to increasing disorganization, greater difficulty with self-regulation, and a general breakdown of executive functions, cognitive abilities, and cognitive skills. The increase in impulsivity would be part of this general deterioration; therefore it's not defensive or purposive. It's more of a spilling over or spilling out. This may be gradual or occur late in the disorder. So there are different kinds of impulsivity, and of course they may be present in combination. Impulsivity can lead to pathological gambling, which can lead to greater impulsivity.

Tension reduction is also a muddy conceptual problem. In DSM-IV, a central feature of the IDNEC disorders is an increasing tension or sense of tension prior to the act, relief or release with the commission of the act, and feelings of guilt or regret afterwards. Four of the five disorders list increasing tension as a central feature. Three of the five list increasing tension followed by relief as necessary for a diagnosis. This is carried over from DSM-III, where an increasing sense of tension followed by relief constituted two of the three essential characteristics, the other being the failure to resist.

Yet there is no definition of tension either in the chapter or in the glossary. And tension has multiple meanings. Tension can mean stress. It can mean dysphoria. It can mean what's going on in the environment that causes one to be upset. For example, we speak of tension in the workplace or tension at home. But the term “tension” also refers to a whole bunch of physical meanings. Tension headaches are the most common kind of headache—probably everybody here has had one at one time or another—but when we talk about tension headaches, it's not clear what the word tension means. Half of the literature on tension headaches talks about them as if the tension means stress. They even call them stress headaches, and there are a number of other synonyms relating to emotional stress.

However, there are just as many authors who think that the tension in a tension headache refers to muscle tension, and that it's the muscular band around the head or the muscular tension at the base of the neck and in the occipital region that gives it its name. There are also a whole bunch of uses of tension relating back to 19th-century physics and the energy models used by Freud. Tension there is defined by excess energy, which the body and mind seeks to reduce. Discharge of psychic energy leads to relaxation, while any increase in energy causes dysphoria or tension.

The motivational psychologists adopted this mechanistic, hydraulic model of energy, which has survived in references to energy being blocked or released therapeutically when one expresses anger. Sports was believed to be a way to get rid of excess anger and aggression. One of the most common theories for alcoholism, just prior to the writing of DSM-III, in fact, was the tension-reduction hypothesis. People drank to release tension. While these theories have been disputed, they continue to form the basis for the tension-reduction model expressed in the IDNEC chapter of DSM-IV.

Another source of confusion is the elimination of any reference to ego-syntonic and ego-dystonic behavior. In the draft of the text for pathological gambling, this distinction was thought to be extremely important. In fact, most of these behaviors started out as pleasurable, but at some point they took on a life of their own. There's no mention of this in the chapter as it was published, which I think interferes with our attempts to understand the loss of control.

Possibilities for DSM-V

On my final slide I listed the various possibilities for DSM-V. The first option would be to keep the Impulse-Control Disorders Not Elsewhere Classified as it is. Since there is a bias in favor of not making changes in the manual unless there's strong data and compelling arguments supporting the need for change, this would be the leading contender.

I've listed two possible modifications. One would be to add more disorders to the category. Various groups have made a case for adding compulsive shopping, Internet addiction, sexual addiction, and pathological lying. Interestingly, we talked about Internet addiction in yesterday afternoon's session. The one paper I'm familiar with on it used the diagnostic criteria for pathological gambling and adapted them to Internet addiction. Sexual addiction—there's a large group of treatment providers making a strong argument for including it. And, again, they're undecided whether it's a sexual compulsion or an addiction. But these are the disorders that are most likely to come into DSM-V and into this category.

Another modification, within the existing classification, would be to clarify what the essential features are. This would be extremely important, as would defining the various terms, such as “impulse,” “impulsive,” and “impulsivity.”

Next I list the spectrum disorders: obsessive compulsive spectrum disorders as described by Hollander, and the affective spectrum described by Susan McElroy. Carlos is going to talk about them and he'll also talk about the possibility of categorizing these disorders as addictions, specifically behavioral addictions. And, finally, I just want to mention these last two ways of conceptualizing these disorders. Under primitive subgroups, I list disorders of acquisition and disorders of grooming. Judy Rappaport, among others, has suggested this way of thinking about these disorders. Disorders of acquisition would include pathological gambling, and also kleptomania, compulsive shopping, and hoarding. Disorders of grooming would include trichotillomania, compulsive nail biting, skin picking, and a number of disorders found in various species of animals.

For example, canine acral lick disorder—where dogs bite off their fur, mostly on their forepaws—can cause terrible dermatological conditions. Feather plucking in birds is another well-known disorder. Similarities to trichotillomania and compulsive nail biting and skin picking are obvious. And both of those animal disorders are treated with selective serotonin reuptake inhibitors, such as Prozac or Zoloft. We had a dog who was on Prozac for canine acral lick disorder, and he complained terribly of the sexual side effects. (Laughter.)

With these primitive subgroups—the disorders of acquisition, and disorders of grooming—in addition to there being animal models, which are extremely useful to researchers, there are parts of the brain that have been localized for these disorders. Again, this suggests possibilities for research.

And, finally, there are some authors who believe that all of the impulse-control disorders are just different manifestations of the same disorder. Webster and Jackson feel that these are people who suffer from feelings of worthlessness, who self-medicate in different ways (stealing, shopping, setting fires), but that they're all trying to deal with the same underlying problem. And [S. W.] Kim, who has done naltrexone studies with Jon Grant, has said that the primary problem is one of uncontrollable urges and cravings, and that how they manifest themselves is what determines the name of the disorder. In other words, it's the drive, not how it's expressed, that defines the underlying, unifying problem.

So that's my talk. I don't know how much time we have for questions…

Jon Grant: We'll just take two questions.

Renee Cunningham-Williams: Hi, Renee Cunningham-Williams from Washington University. Very nice overview. One of the things that I was thinking about as I was sitting there and thinking about additional possibilities for DSM-V: what are your thoughts on subtyping based on age of onset of certain symptoms, as well as clustering of symptoms within a specific time period?

Richard Rosenthal: Clustering meaning lifetime versus last year in the diagnostic criteria?

Renee Cunningham-Williams: No, clustering of certain symptoms like, say, having preoccupation, chasing, and something else in the last 30 days in addition to some additional symptoms.

Richard Rosenthal: I think Marianna [Toce Gerstein] is going to be talking about that, so I'll hold off on that. As far as age of onset, of course, one of the reasons for wanting to exclude trichotillomania from this category was that they thought it was a childhood disorder that should be categorized under the childhood onset disorders. Other people thought it should be under OCD, but there was strong support for including it as a childhood disorder. Does that answer your question or is there—

Renee Cunningham-Williams: I was specifically thinking of—there are some gamblers who we know start early, early in age, like age eight, in early childhood; and then there are others who are pretty much new to gambling and may start later in age, like some women starting like in their 30s and their 40s. Are these different types of folks? Are they different types of pathological gamblers, and, if so, would it be helpful in the criteria to have a typing like adolescent onset or childhood onset or adult onset similar to—a little bit of what we do with ASPD [antisocial personality disorder] in looking at conduct disorder, and having to meet certain criteria in childhood before you can say something about this same behavior being manifested slightly differently in adulthood?

Richard Rosenthal: We can ask the other people on the panel, but my sense is that there have been a number of attempts to subtype pathological gamblers, and that clinically I think, the one that works best is Henry Lesieur's distinction between the action seekers and the escape seekers.

However, one of the problems has been that Henry and I at one time tried to develop an instrument to distinguish the action seekers from the escape seekers. We thought we had the right questions to ask, but no matter how we played with it, we couldn't get it on paper. However, when we knew the gambler, or were interviewing someone face to face, it wasn't that difficult. I think it's a distinction that still holds the greatest utility, and going back to the original question, the action seekers typically start gambling early in life, and the escape seekers—and this is usually true for women gamblers—typically start later in life, after their adult identities have been formed.

One of the things I remember being discussed at an early conference was the difference between gambling patterns in boys and in girls. I don't know how many of you remember Sirgay Sanger, a psychiatrist who was one of the first presidents of the National Council. He had the experience of having treated a lot of children, and he made the comment, which I think is valid, that gambling was normal in young people, and that it started a lot earlier than people thought, and with various games, but that a difference between boys and girls was that boys were more competitive, and they played for keeps. Girls, on the other hand, even when they played similar games, at the end of the game would give back what they had won. (Laughter.) Girls, it seems, are aware early on of the need to maintain the social order and the sense of the community. One of the things I've wondered about when I hear about changing drinking patterns among college students, with girls now binge drinking and drinking like guys, was whether we were going to see a difference in how girls gambled? I don't know if people who work with young people have seen that yet, but we may start to see a change in gambling patterns and in the characterization of women gamblers as typically late onset, noncompetitive, more luck-based escape-seeking gamblers.

Before we finish that, anyone else want to respond to that question? No? Okay.

Jon Grant: Thank you, Richard. We'll keep this kind of thing flowing. Otherwise, I won't get my moderator award. And I do have my own comment, which I think I'll just toss out for later thoughts. We're talking a lot about categorization, and, obviously, you can look at this on two levels as Richard has alluded to. The actual individual criteria of pathological gambling, and whether there should be changes there, are also how you conceptualize it.

And one of the things that I was thinking about as he was talking: I would just have people ponder the forensic aspect of gambling. The one criterion that's always personally bothered me is committing illegal acts in furtherance of gambling. Also interesting, as Richard mentioned, is the irresistible impulse.

This is a very big issue for the court system. For those of you who have testified in court about this, if you have an irresistible impulse to gamble, do you also have an irresistible impulse to embezzle to gamble? In order to feed your gambling? It's an interesting criterion to have it one step removed from the behavior, which we don't have—I mean if you say to alcoholics, “Well, driving drunk and having an accident is part of the alcoholism.” That would be kind of a parallel, and I think we have a difficulty in understanding how to legally deal with behaviors that are in furtherance. But it's interesting that it's a criterion for the disorder.

[End of presentation.]

Article Categories:
  • session IV