Addiction as Disease
In J. Henningfield, W. Bickel, and P. Santora (Eds.), Addiction Treatment in the 21st Century: Science and Policy Issues. (Baltimore: Johns Hopkins, 2007) pp. 153-163.
Policy, Epidemiology, and Treatment Consequences of a Bad Idea
The effect of conceptualizing addiction as a disease on policy and treatment has been predictably disastrous. The disease conception that addiction can only become progressively worse and never selfameliorate is decisively wrong. Accepting this misconception leads to the belief that addictions can be remedied only through treatment when, in fact, natural recovery is typical. Moreover, treatment predicated on the notion of addiction as a disease treats those who are addicted as though they were unable to affect their own outcomes (represented by the twelvestep idea of powerlessness). The most effective treatments, by contrast, convey greater power and selfcontrol (i.e., self-efficacy) to addicts and understand that environmental conditions and skills at coping with them are crucial to remission. In the policy realm, the view that addiction inevitably requires treatment leads to support for coercion into treatment, usually twelvestep treatment. At the broader cultural level, the highest incidence of addiction occurs where there is widespread cultural belief in the disease model of addiction.
We can’t solve problems by using the same kind of thinking
we used when we created them.
— Albert Einstein
Policy amid the Consensus That Addiction Is a Disease
The idea that addiction is a disease has become almost received wisdom among conservatives and liberals alike. For example, people on both sides of the drug policy debate—those who favor continued repressive measures toward drug use and those who feel drug laws should be relaxed—agree in this view. Those who favor maintaining the status quo believe that any drug use causes irreversible compulsive drug use, whereas those who favor changing drugs’ legal status want to offer compulsive drug users treatment in place of criminal sanctions. At the same time, accepting addiction as a disease means ignoring important views held by those on both ends of the political spectrum. Conservatives who advocate the disease model neglect the ideas of responsibility and selfdetermination that are part of the conservative model of behavior, and liberals accept that people cannot control their drug use even though this view infantilizes drug users and places them under institutional and state control, contrary to the essential liberal view that humans can and should control their destinies.
These contradictions present fundamental dilemmas for conservatives and liberals. If conservatives accept that drug use can be an uncontrollable disease, they open up the possibility of addiction as a criminal defense—“Once I began using drugs, I could not choose to stop or control my behavior while addicted or intoxicated.” Liberals meanwhile, including drug policy reformers, sacrifice the idea that many people are able to control their drug use or that, even when addicted, people are human beings who should be allowed to make choices for themselves. They make this sacrifice to allow those caught using drugs—regardless of whether they are addicted—the option of being treated rather than incarcerated. In this framework, treatment is coercively foisted on people because it is perceived to be a humane alternative to other criminal penalties. The idea that people cannot help themselves out of addiction, however, undermines the very cultural beliefs that are associated with lower rates of addiction.
The Epidemiology of Addictive Drug Use
There are two worlds of addiction: the one glimpsed through clinical dealings with addicts and the one viewed through broad population, epidemiologic studies (Room, 1980). At times, these two worlds are almost unrecognizably different. In the clinical world, patients require constant attention and direction, hardly ever make progress, at least without tremendous therapeutic support and input, and must abstain. In the epidemiologic survey world, treatment is unusual yet addicts usually improve, often gradually, over years, though they are not fixated on abstinence.
There are many clinics but few massive epidemiologic studies, so we only periodically get the broad view the latter reveal. The results of such studies are stunning, even to professionals, yet they are generally quickly forgotten. In 2005 the National Institute on Alcohol Abuse and Alcoholism (NIAAA) published the results of its 2001–2002 National Epidemiologic Survey on Alcohol and Related Conditions (NESARC), a comprehensive survey focused on Americans’ current and lifetime alcohol and drug use (Dawson et al., 2005). NESARC replicated a similar study, the National Longitudinal Alcohol Epidemiologic Survey, conducted a decade earlier that yielded comparable results (cf. Dawson, 1996). NESARC conducted 43,093 inperson interviews with a national sample of adults (18 or older), among whom, at some point prior to the past year, 4,422 were classifiable as alcohol dependent according to DSM-IV. (Because it remains an evocative and popular term, we will refer to this group as “alcoholics.”) The NESARC results are presented in Table 20.1.
Table 20.1. NESARC Past-Year Drinking among Alcoholics
Past-Year Status Treated
(n = 1,205) Untreated
(n = 3,217)
Remission, risk drinking
Remission, drinking without risk
Source: Dawson et al. (2005).
The first startling finding from NESARC is that only about a quarter of those ever identified as having been alcohol dependent according to DSM-IV criteria were ever treated (including attending Alcoholics Anonymous [AA]). In addition, a higher percentage of treated than untreated alcoholics currently continue to be alcoholic! Twenty-eight percent of those who have received treatment remain dependent, as opposed to 24 percent of those who were never treated. This could partly be accounted for in that treated subjects tend to be more severely alcohol dependent than untreated alcoholics (though the populations overlap). Nonetheless, it is surely embarrassing to an American agency responsible for ameliorating alcoholism that people treated for their alcoholism are no more likely to overcome their alcohol dependence than alcoholics who are not treated.
Those who are treated are substantially more likely to be abstinent (by a threetoone ratio). The nonabstinence categories in NESARC require some explanation. “Partial remission” refers to alcoholics who had at least one drinking problem over the prior 12 months but who do not qualify for a DSM-IV diagnosis of alcohol dependence. There are, furthermore, two remission-with-drinking categories. The NESARC report identifies one group as “risk drinkers in remission.” They have displayed no drinking problems over the prior year, yet they have continued to drink regularly (averaging more than 14 drinks per week for men; 7 drinks per week for women), or they had a single day (or more) in the past year in which they had 5 or more drinks, for men; 4 or more, for women.
Considering a former alcoholic who has had four drinks once in the past year—a year with no alcoholrelated work, health, family, or psychological problems—to be a risk drinker might be considered importing a clinical idea into the epidemiological realm. That is, if someone who has been alcohol dependent is now able to get through an entire year with no drinking problems, but periodically (even if extremely rarely) consumes a number of drinks, the person might seem extremely well inured against a return to alcoholism. A clinician, on the other hand, sees the person dancing on the precipice of a fall into an uncontrollable resumption of alcoholism. In fact, it seems likely that opposition from those with a clinical perspective within and outside the NIAAA prompted NESARC investigators to create this risk category (which was not a part of the analysis of remission in NLAES; see Dawson, 1996).
Discarding this addon to DSM-IV criteria, the NESARC results become even more startling and controversial. Categorizing NESARC respondents purely in terms of their DSMIV remission categories (Table 20.2) makes clear that untreated alcoholics are more than twice as likely to be in remission while continuing to drink as are treated alcoholics. In fact, the typical form remission takes for the large majority of Americans who do not seek treatment for their alcoholism is controlled drinking (by a margin of almost three to one versus abstinence). At the same time, many might be surprised to see that even 16 percent of treated alcoholics achieve nonabstinent remission, or almost half as many as abstain.
Table 20.2. NESARC Past-Year Remission among Alcoholics
Past-Year Status Treated
(n = 1,205) Untreated
(n = 3,217)
Source: Dawson et al. (2005).
Table 20.3. NESARC Past-Year Improvement among Alcoholics
Past-Year Status Treated
(n = 1,205) Untreated
(n = 3,217)
Drinking without dependence
Source: Dawson et al. (2005).
Indeed, the NESARC data in Table 20.3—subjects who remain dependent, who cease drinking altogether, or who improve beyond dependence while continuing to drink—represent results that the NIAAA (along with virtually all American clinical authorities) would disavow. Even among those who receive treatment, as many alcoholics improve by drinking with fewer problems as achieve abstinence. Among alcoholics who do not seek treatment, almost two-thirds make such progress. Considering both the treated and untreated populations, the typical outcome of alcoholism in the United States is to improve while continuing to drink. The good news is that the majority of alcoholic Americans ignore the disease theory’s prescription of abstinence, and they gain benefits from doing so. Alcoholics who have been in treatment, however, are significantly impaired in their ability to moderate their drinking, whether because of their drinking problems or attitudes prior to being in treatment or as a result of treatment. Rudy (1986) describes the process of AA members learning, and imitating, the symptoms of fullblown alcoholism that many did not originally display.
Through the Substance Abuse and Mental Health Services Administration (SAMHSA), the United States also conducts a periodic comprehensive examination of Americans’ drug use, the National Survey on Drug Use and Health. According to this survey, only a small percentage (fewer than 10 percent) of Americans who have ever consumed heroin, crack, or cocaine continue to take these drugs, even as infrequently as once a month (Table 20.4). Presumably, only a small percentage of these “current users” use the drugs addictively. These broad population data defy the commonplace classification of notoriously “addictive” drugs as being distinctively addictive relative to other activities people engage in or substances they consume.
Table 20.4. Lifetime, Past-Month Use of “Addictive” Drugs, United States, 2002
Percentage of Americans Using (figures rounded)
Lifetime Previous Month Previous Month/Lifetime
Source: Data from SAMHSA (2003, Table 1.1B).
Other epidemiological research, for example with cocaine, indicates that in untreated populations those whose cocaine use leads to physical (e.g., nasal bleeding), psychological (paranoia), and behavioral (e.g., sleep disruption) problems typically respond by quitting or cutting back (Erickson et al., 1987; Peele and DeGrandpre, 1998). None of these data discount that some individuals undergo extended periods of excessive, even compulsive use, but this is not true for the large majority of users, even after they encounter substantial problems.
How Belief Systems Inﬂuence Levels of Addiction
Attitudes about whether addiction (and alcoholism) is a disease and what that means do not occur in a vacuum; they reﬂect larger cultural attitudes and belief systems that actually impact whether people control intoxicant-related behavior (Peele, 1985). Cultures in which a drug experience is seen as being overwhelming and insurmountable in fact make addiction to that experience more likely (Peele, 1985). For some time, substantial cultural differences in drinking patterns have been reported in observational studies, often anthropological (cf. Heath, 2000), but because these distinctions have been drawn qualitatively observers have frequently questioned their validity. Now, with the advent of multinational epidemiological surveys across Europe, cultural differences in drinking—and their impact on outcomes—have been verified quantitatively. Cultures in which alcohol is most feared to lead to problems and uncontrollable use —as a result of which drinking is more regulated (e.g., limitations on outlets and times where and when alcohol may be consumed)—paradoxically provoke more binge drinking, negative social and psychological outcomes, and addiction. The European Comparative Alcohol Study (ECAS), for example, found a negative correlation between the amount of alcohol consumed within society and the prevalence of alcoholrelated harm in that society (Table 20.5). The United States has not been included in these international surveys as yet, but the drinking data provided in the National Survey on Drug Use and Health, particularly ethnic differences in rates of binge drinking, indicate that similar patterns prevail among ethnic groups in this country (Table 20.6).
As a result, epidemiologists who participate in such surveys have validated anthropological analyses, now supported by hard data:
In the northern countries, alcohol is described as a psychotropic agent. It helps one to perform, maintains a Bacchic and heroic approach, and elates the self. It is used as an instrument to overcome obstacles or to prove one’s manliness. It has to do with the issue of control and with its opposite—“discontrol” or transgression. (Allamani, 2002, p. 197)
In the southern countries, alcoholic beverages—mainly wine—are drunk for their taste and smell, and are perceived as intimately related to food, thus as an integral part of meals and family life. Actually, wine tends to be considered as a food item.. . . It is traditionally consumed daily, at meals, in the family and other social contexts.. . . Typically, Mediterranean people’s drinking is still characterized by relatively even weekly consumption, while in northern cultures drinking is concentrated to rare occasions with high intake per session. (Allamani, 2002, p. 200)
Thus, “safety” messages about the dangers of alcohol in temperanceoriented cultures (mainly Englishspeaking and Scandinavian countries) have unintended consequences reﬂected in the greater alcoholrelated harm experienced in these countries (Hemström et al., 2002). Indeed, ECAS actually found a negative correlation between national levels of alcohol consumption and alcoholrelated mortality due to cultural variations in drinking (Ramstedt, 2002). Addiction is not just a danger these cultures are dealing with; rather, these cultures are iatrogenically creating the conditions that enhance the likelihood of addiction.
Table 20.5. Percentage of Males Drinking Daily, Binge Drinking, and Experiencing Drinking Harms, Selected European Countries
(%) Binge Drinking per Drinking Occasion
(%) Experiencing Alcohol Harms
Source: Data from Ramstedt and Hope (2003), Hemström et al. (2002).
Table 20.6. Percentage of American Adults Who Binge Drink, by Ethnic/Racial Group
Racial Group Currently Drink*
(%) Percentage Who Binge/
Source: Data from SAMHSA (2003, Table 2.54B).
* Drank in the past month
† Binge = five or more drinks on a single occasion in the past month
The Treatment Implications of Disease Models of Addiction
Leading clinicians in the United States have fully embraced the disease model of addiction as reﬂecting the reality of their clinical practices and research (McClellan et al., 2000), and it is impossible for them to escape the limitations of their clinical perspective. Yet even within the clinical framework, the disease model has severe drawbacks. That is, a long tradition of psychological research has indicated the value of beliefs and treatments that enhance selfefficacy for overcoming problems. The disease model undercuts— denies the possibility of—such selfefficacy. Along with encouraging selfefficacy, psychological treatment models emphasize modifying environmental conditions and reinforcement in order to change behavior. What are the implications for outcomes of differences in treatment approaches?
One team of investigators has catalogued controlled clinical trials of alcoholism treatment over several decades (Miller et al., 2003). Their metaanalyses summarize the “cumulative evidence (of efficacy) score” for 48 popular modes of treatment (negative scores indicate relative ineffectiveness when compared in studies with other treatments— or no treatment). Table 20.7 lists selected results for effective (and ineffective) alcoholism therapies, with special reference to whether those therapies accept or reject the disease model of alcoholism.
Brief interventions (BIs), as their name indicates, are shorter in duration than conventional alcohol therapies. At their briefest, such an intervention can simply be an emergency care physician or general practitioner who notes a significant medical trauma or condition (e.g., poor liver function) and proceeds to instruct the patient to reduce or stop drinking. BIs are commonly used preventively in the absence of observed medical complications, where the physician just asks about the patient’s drinking and may give advice based on the patient’s answer. This first element of the BI is to provide feedback to the patient. In some BIs, providing this feedback and linking it to the patient’s drinking and instructions to cut back or to quit comprise the entire intervention.
Motivational enhancement (ME) is a treatment geared to help people sort out their ambivalence about change, and to direct their own efforts to reducing their drinking. ME therapists assist people to change by, initially, asking openended questions about their drinking, to assist clients in recognizing for themselves how their drinking is hurting them and violating values they hold. The key to ME is never to confront people—confrontational therapies have been shown to have little success and to produce other negative consequences.
Table 20.7. Efficacy of Alcohol Treatments
Treatment Cumulative Evidence Score
Community reinforcement approach
General alcoholism counseling
Education (tapes, etc.)
Source: Data from Miller et al. (2003).
The community reinforcement approach (CRA) is a package of behavioral techniques designed to encourage sobriety in more severely alcoholic (or dependent) patients than those typically treated with BIs and ME. CRA begins with a socalled functional analysis—that is, the when, where, and whys of the individual’s drinking. Among the components of CRA are a job club (or vocational training), a leisure club (or recreational counseling) to direct the individual to nondrinking venues for entertainment, and relationship (or marital) therapy aimed at enhancing intimate relationships and ensuring that a spouse’s behavior supports sobriety. If work, social time, and family or intimate life all reinforce the individual’s ability to avoid drinking, his sobriety will be supported throughout his or her life space. Alcoholics are also trained in skills they must have to conduct themselves successfully in these situations.
These successful therapies are characterized by two principal underlying features: (1) each encourages people to take responsibility for their behavior—and to feel they can control these behaviors—and (2) alcoholism is not seen as a disease, but as a condition the individual can modify by changing behavior, environment, and effort. These results support a finding by Miller and colleagues in a prospective study of treatment outcomes. Two primary factors predicted relapse: “lack of coping skills and belief in the disease model of alcoholism” (Miller et al., 1996).
The disease model of addiction is disproved by basic evidence about the incidence and course of addictive substance use and is counter-productive in both the prevention and treatment of addiction. The key issues for addiction treatment in the twenty-first century are the persistence of ingrained cultural beliefs in the face of scientific refutation and demonstrations that they are counter-productive and whether new neuroscientific props will be used to support already demonstrably counter-productive attitudes and approaches to treating addiction (cf. McClellan et al., 2000).
Allamani, A. 2002. Policy implications of the ECAS results: a southern European perspective. In T. Norstöm, ed., Alcohol in Postwar Europe: Consumption, Drinking Patterns, Consequences and Policy Responses in 15 European Countries. Stockholm: National Institute of Public Health.
Dawson, D.A. 1996. Correlates of past-year status among treated and untreated persons with former alcohol dependence: United States, 1992. Alcoholism: Clinical and Experimental Research 20:771–79.
Dawson, D.A., Grant, B.F., Stinson, F.S., Chou, P.S., Huang, B., and Ruan, W. J. 2005. Recovery from DSM-IV alcohol dependence, United States, 2001–2002. Addiction 100:281–92.
Erickson, P.G., Adlaf, E.M., Murray, G.F., and Smart, R.G. 1987. The Steel Drug: Cocaine in Perspective. Lexington, Mass.: Lexington.
Heath, D.B. 2000. Drinking Occasions: Comparative Perspectives on Alcohol and Culture. Philadelphia: Brunner/Mazel.
Hermström, O., Leifman, H., and Ramstedt, M. 2002. The ECAS survey on drinking patterns and alcohol-related problems. In T. Norström, ed., Alcohol in Postwar Europe: Consumption, Drinking Patterns, Consequences and Policy Responses in 15 European Countries. Stockholm: National Institute of Public Health.
McLellan, T.A., Lewis, D.C., O’Brien, C.P., and Kleber, H.D. 2000. Drug dependence, a chronic medical illness: Implications for treatment, insurance, and outcomes evaluation. Journal of the American Medical Association 284:1689–95.
Miller, W.R., Westerberg, V.S., Harris, R.J., and Tonigan, J. S. 1996. What predicts relapse? Prospective testing of antecedent models. Addiction 91 (Suppl.):S155–171.
Miller, W.R., Wilbourne, P.L., and Hettema, J.E. 2003. What works? A summary of alcohol treatment outcome research. In R. K. Hester and W. R. Miller, eds., Handbook of Alcoholism Treatment Approaches: Effective Alternatives, 3rd ed. Boston: Allyn and Bacon.
Peele, S. 1985. The Meaning of Addiction: Compulsive Experience and Its Interpretation. Lexington, Mass.: Lexington.
Peele, S., and DeGrandpre, R.J. 1998. Cocaine and the concept of addiction: environmental factors in drug compulsions. Addiction Research 6:235–63.
Ramstedt, M. 2002. Alcohol-related mortality in 15 European countries in the postwar period. In T. Norström, ed., Alcohol in Postwar Europe: Consumption, Drinking Patterns, Consequences and Policy Responses in 15 European Countries. Stockholm: National Institute of Public Health.
Ramstedt, M., and Hope, A. 2003. The Irish Drinking Culture: Drinking and Drinking-Related Harm, A European Comparison. Dublin, Ireland: Report for the Health Promotion Unit, Ministry of Health and Children.
Room, R. 1980. Treatment seeking populations and larger realities. In G. Edwards and M. Grant, eds., Alcoholism Treatment in Transition. London: Croom Helm.
Rudy, D. 1986. Becoming Alcoholic: Alcoholics Anonymous and the Reality of Alcoholism. Carbondale: Southern Illinois University Press.
Substance Abuse and Mental Health Services Administration. 2003. Results from the 2002 National Survey on Drug Use and Health. Rockville, Md.: U.S. Department of Health and Human Services.