Problem and pathological gambling represent growing public health concerns. One to two percent of the general adult population is thought to meet past-year criteria for pathological gambling, and another 2% to 4% experience less severe yet still interfering gambling problems. Problem and pathological gambling have been reported to be associated with significant adverse life events including unemployment, receipt of welfare, arrest, incarceration, bankruptcy, divorce, poor or fair physical health and psychiatric problems (including substance use disorders). Despite the growing prevalence rates among adults from the general population and the associated negative events, relatively little research has been performed to investigate the etiology and treatment of problem and pathological gambling. Given the high rates of co-morbidity between gambling and substance use disorders, it is important for addiction psychiatrists to have an understanding of current advances in the epidemiology, etiology and treatment of pathological gambling.

In this symposium, Dr. Shaffer discussed the epidemiology of gambling. Dr. Petry addressed cognitive-behavioral therapy for the treatment of pathological gambling, and Dr. Potenza discussed the neuropsychopharmacology of pathological gambling.

Symposium Chair: Marc N. Potenza, MD, PhD, Yale University, New Haven, CT

Gambling is risking something of value on the outcome of an event when the probability of that outcome is less than certain. Gamblers have variable awareness that they are putting something of value at risk, that the bet that they make is irreversible and that the outcome is in fact determined by chance. The issue of cognitive impairment of some sort is critical in the development of gambling-related disorders.

Gambling has grown enormously in the last decade of the 20th century. Legalized gambling currently grosses approximately $63 or $64 billion annually. Noncasino revenue is greater than casino revenue, but they are similar. The growth in legalized gambling has led people to examine the role of the availability of gambling opportunities in the development of gambling-related disorders.

For the group of individuals 12 years of age and older, the current rate of lifetime gambling participation slightly exceeds that for alcohol use. Differences emerge when examining the rates of gambling participation over the past 25 years. In 1975, 61% of the population reported past-year gambling and 68% lifetime gambling. Now, approximately 86% of Americans have gambled during their lifetime, but the past-year gambling rate has not increased much (63%).

Like with early drug and alcohol experiences, most gambling begins with the family and at home. Many people with gambling problems report changes over time in their motivations for gambling, with less importance over time given to winning and more importance to the process of gambling or being "in action."

Initial gambling experiences often involve betting on sports. Young gamblers frequently report an exaggerated sense of skill involved in gambling. That is, they thought that the most skill was involved in card playing and sports betting, and they thought they had the most skill for those two games. There is a statistically positive correlation between skill and having gambling problems: the more that one believes skill is important, the greater the risk for developing or having a gambling problem.

One of the early warning signs of problem gambling is an early onset of gambling: the younger a child begins gambling, the more likely he or she is to develop a gambling problem. Pathological gamblers generally report starting gambling between the ages of 8 and 11 years, whereas nonpathological gamblers tend to report beginning between the ages of 11 and 13. The more money that people bet the first time, the more likely they are to develop a gambling problem.

People who live in poverty generally have very little to lose when they gamble and everything to gain. People of wealth generally have everything to lose and very little to gain. The lottery is predominated by people of lesser means on a regular basis. As a lottery jackpot grows, educated people of means buy a greater proportion of lottery tickets sold.

Pathological gambling is considered a persistent and recurrent maladaptive gambling behavior that disrupts personal, family or vocational pursuits. Problem and pathological gamblers cannot be identified before they encounter problems.

Pathological gambling often starts with a preoccupation with gambling and a need to gamble increasing amounts. Pathological gamblers often experience restlessness or irritability when they cut down or stop gambling. Gambling is often used to escape from dysphoric states. After losing money, pathological gamblers frequently chase losses (i.e., go back to gamble on another day to win back lost money).

Pathological gamblers frequently lie to family members, therapists or others to conceal the extent of their gambling. Their gambling often jeopardizes relationships. They often rely on others to provide a financial "bailout," and they have unsuccessful repeated attempts to control or cut back their gambling.

Almost 30% of adults in treatment report gambling-related problems. If psychiatrists are not screening for gambling on a routine basis, they are probably missing people who have gambling-related problems among their treatment population. College students have also been found to have very high rates of gambling-related problems.

The prevalence of gambling has increased slightly in the past 25 years among general-population adults. However, it has been stable for young people, psychiatric patients and criminals because these groups appear less sensitive to the legalization of gambling. The general adult population, however, appears much more sensitive, so that when the availability of legalized gambling increased, their involvement increased.

An additional 3.8% of adults are problem gamblers. Together, about 5% of the population has some degree of a gambling problem at some point in their lives. There are higher rates of problem and pathological gambling in substance abuse treatment patients and prisoner populations, up to about 30%.

In a survey of 2,638 American adults, pathological gamblers had a 28% rate of alcohol dependence. Nonpathological gamblers had a 1.2% rate. There is a high co-morbidity between alcohol use disorders and pathological gambling.

Not many pathological gamblers seek treatment. About 1% to 4% of problem or pathological gamblers actually come in for treatment.

Gambling problems in substance abusers are associated with greater psychiatric distress, higher levels of impulsivity, increased HIV risk behaviors and childhood maltreatment histories.

Substance abuse and pathological gambling are highly co-morbid, and they are associated with poor psychosocial functioning.

Dr. Petry examined brief interventions for a larger group of patients who are problem gamblers but who do not meet pathological gambling criteria. Study participants were found through direct screening and recruitment efforts at substance abuse treatment programs at medical and dental clinics in the inner cities and at some gambling facilities.

Patients included in the study did not meet DSM-IV criteria for pathological gambling. Instead, they reported minimal gambling problems, and they were randomly assigned to one of four conditions. The control condition is a no-treatment condition in which subjects receive first an evaluation and then follow-up over time. In condition 2, immediately following the initial evaluation, the therapist reviews a one-page handout that provides brief advice as to how to prevent progression of their gambling to pathological gambling.

In condition 3, subjects get one 60-minute session in which a therapist uses motivational enhancement therapy. In condition 4, subjects receive the initial session and are asked to return for three more sessions of cognitive-behavioral therapy targeting ways to reduce internal and external cues to gambling.

The 150 study participants were equally divided among the groups. Patients were interviewed at intake, at week 6 and at month 9.

At the beginning of treatment, subjects were actually gambling more frequently than pathological gamblers (about 16 days on average). About 6 weeks later, all groups demonstrated a reduction in the number of days gambled. The median dollars gambled was about $500 a month as opposed to the $2,000 in the pathological gamblers. The figure goes down slightly in the no-treatment control condition and a little bit more in the group receiving one session of MET. However, there is a steeper slope down to about $150 a month in the groups receiving the four-session intervention or 5 minutes of brief advice condition.

Even more striking is the finding that, at 9 months later, the days of gambling stayed about the same in the no-treatment condition, the one-session intervention and the four-session intervention, but those who got the 5 minutes of brief advice continued to have a reduction in the number of days gambled, even over the course of the next 7 or 8 months.

The median dollars gambled went back up slightly in the no-treatment and the one-session intervention. Surprisingly, in the four-session intervention, the amount gambled returned to baseline levels. In the 5-minute intervention, the amount gambled continued to diminish.

Relatively little is known regarding the neurobiology of pathological gambling or impulse control disorders in general. A growing number of studies have implicated the ventromedial prefrontal cortex and other prefrontal and frontal cortical brain regions in impulse control.

To date, no neuroimaging studies of pathological gambling have been published. These studies will be important both for improving the understanding of pathological gambling and placing it within the context of impulse control disorders and behaviors characterized by impaired impulse control.

Dr. Potenza conducted some functional magnetic resonance imaging (fMRI) studies of individuals with substance use disorders and individuals with pathological gambling. A videotaped cue elicitation paradigm was used to study urge states.

The basis of the study is blood oxygen level-dependent fMRI. Pictures of the entire brain can be obtained within a 1- to 2-second period so that brain responses can be followed over time.

Several types of videotapes were developed including gambling-related ones for the pathological gambling studies and cocaine-related ones for the cocaine-dependent studies. In addition, happy and sad scenarios were generated to serve as active control states. The scenarios involved an actor or an actress speaking directly to the camera and engaging the viewer in a simulated interaction. The videotapes were presented in a counterbalanced order. Each was about 4 minutes in length. Individuals indicated with a button press the onset of an emotional or motivational response. In between the viewing of the videotapes, they described both the quality and the magnitude of their emotional and motivational responses.

The first study investigated the neural activities underlying gambling urges in response to gambling cues. The researchers compared the neural correlates of gambling urges and cocaine cravings.

Pathological gamblers and control subjects were examined in the first study, and individuals with other major neuropsychiatric disorders, except for nicotine dependence, were excluded. People could not use any psychotropic medications or substances other than tobacco or caffeine prior to the scan. By and large, the responses to the sad and happy tapes were mild to moderate. They were comparable for both the pathological gambling subjects and control subjects.

The pathological gambling subjects and control subjects showed differential brain activities, mainly in response to gambling cues and mainly during the initial period of tape viewing of the gambling scenario prior to the reported onset of emotional or motivational response. The decreased ventral anterior cingulate activity during the final period of tape viewing corresponded to the presentation of the most provocative stimuli.

The brain activities of pathological gambling and cocaine-dependent subjects were compared during viewing of their respective addiction tapes. Pathological gambling as compared with cocaine-dependent subjects’ urges displayed relatively decreased activity in the ventral and dorsal regions of the anterior cingulated. The time period in which the most significant similarities were seen was in the initial period of tape viewing prior to the onset of emotional or motivational response.

The findings that a brain region frequently implicated in impulse control (ventral anterior cingulate) is relatively less active in pathological gambling as compared with cocaine-dependent subjects may reflect some of the therapeutic differences seen in drug treatment trials.

A number of neurotransmitter systems have been implicated in the pathophysiology of pathological gambling. The current state of the psychopharmacology of pathological gambling is very young. Until a few years ago, there were no placebo-controlled trials with more than one subject, and only within the past several years have some larger randomized trials been published.

Presently, no drugs are FDA-approved for the treatment of pathological gambling. Effective treatments appear to be emerging, although more work is needed in this area. The limitations of the drug treatment trials published to date are that they are short-term in nature; they frequently exclude individuals with other co-occurring disorders; a limited number of sites have participated; and the studies have generally included small to moderate sample sizes. A significant need exists for early identification of people who are pathological gamblers. Aversive strategies are not particularly helpful.