HJAR May/Jun 2022

HEALTHCARE JOURNAL OF ARKANSAS I  MAY / JUN 2022 25 definitions of “disease” [10]. He suggested that the addiction field needed to follow the rest of medicine in moving away from view- ing disease as an “entity”, i.e., something that has “its own independent existence, apart from other things” [11]. To modern medi- cine, he pointed out, a disease is simply a label that is agreed upon to describe a clus- ter of substantial, deteriorating changes in the structure or function of the human body, and the accompanying deterioration in biopsychosocial functioning. Thus, he concluded that alcoholism can simply be defined as changes in structure or function of the body due to drinking that cause dis- ability or death. A disease label is useful to identify groups of people with commonly co-occurring constellations of problems— syndromes—that significantly impair func- tion, and that lead to clinically significant distress, harm, or both. This convention allows a systematic study of the condition, and of whether groupmembers benefit from a specific intervention. It is not trivial to delineate the exact cat- egory of harmful substance use for which a label such as addiction is warranted (See Box 1). Challenges to diagnostic categoriza- tion are not unique to addiction, however. Throughout clinical medicine, diagnos- tic cut-offs are set by consensus, com- monly based on an evolving understand- ing of thresholds above which people tend to benefit from available interventions. Because assessing benefits in large patient groups over time is difficult, diagnostic thresholds are always subject to debate and adjustments. It can be debated whether diagnostic thresholds “merely” capture the extreme of a single underlying population, or actually identify a subpopulation that is at some level distinct. Resolving this issue remains challenging in addiction, but once again, this is not different from other areas of medicine [see e.g., [12] for type 2 diabetes]. Longitudinal studies that track patient tra- jectories over time may have a better ability to identify subpopulations than cross- sec- tional assessments [13]. By this pragmatic, clinical understanding of the disease concept, it is difficult to argue that “addiction” is unjustified as a disease label. Among people who use drugs or alco- hol, some progress to using with a quantity and frequency that results in impaired func- tion and often death, making substance use a major cause of global disease burden [14]. In these people, use occurs with a pattern that in milder forms may be challenging to capture by current diagnostic criteria (See Box 1), but is readily recognized by patients, their families and treatment providers when it reaches a severity that is clinically significant [see [15] for a classical discus- sion]. In some cases, such as opioid addic- tion, those who receive the diagnosis stand to obtain some of the greatest benefits from medical treatments in all of clinical medi- cine [16, 17]. Although effect sizes of avail- able treatments are more modest in nicotine [18] and alcohol addiction [19], the evidence supporting their efficacy is also indisput- able. A view of addiction as a disease is justified, because it is beneficial: a failure to diagnose addiction drastically increases the risk of a failure to treat it [20]. Of course, establishing a diagnosis is not a requirement for interventions to be mean- ingful. People with hazardous or harmful substance use who have not (yet) devel- oped addiction should also be identified, and interventions should be initiated to address their substance-related risks. This is particularly relevant for alcohol, where even in the absence of addiction, use is fre- quently associated with risks or harm to self, e.g., through cardiovascular disease, liver disease or cancer, and to others, e.g., through accidents or violence [21]. Interventions to reduce hazardous or harmful substance use in people who have not developed addiction are in fact particularly appealing. In these individuals, limited interventions are able to achieve robust and meaningful benefits [22], presumably because patterns of misuse have not yet become entrenched. Thus, as originally pointed out by McLel- lan and colleagues, most of the criticisms of addiction as a disease could equally be applied to other medical conditions [2]. This type of criticism could also be applied to other psychiatric disorders, and that has indeed been the case historically [23, 24]. Today, there is broad consensus that those criticisms were misguided. Few, if any healthcare professionals continue to main- tain that schizophrenia, rather than being a disease, is a normal response to societal conditions. Why, then, do people continue to question if addiction is a disease, but not whether schizophrenia, major depressive disorder or post- traumatic stress disorder are diseases? This is particularly troubling given the decades of data showing high co- morbidity of addiction with these conditions [25, 26]. We argue that it comes down to stigma. Dysregulated substance use contin- ues to be perceived as a self-inflicted con- dition characterized by a lack of willpower, thus falling outside the scope of medicine and into that of morality [3]. Chronic and relapsing, developmentally- limited, or spontaneously remitting? Much of the critique targeted at the con- ceptualization of addiction as a brain dis- ease focuses on its original assertion that addiction is a chronic and relapsing condi- tion. Epidemiological data are cited in sup- port of the notion that large proportions of individuals achieve remission [27], fre- quently without any formal treatment [28, 29] and in some cases resuming low risk substance use [30]. For instance, based on data from the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC) study [27], it has been pointed out that a significant proportion of people with an addictive disorder quit each year, and that most afflicted individuals ultimately remit. These spontaneous remission rates are argued to invalidate the concept of a chronic, relapsing disease [4]. Interpreting these and similar data is complicated by several methodological and conceptual issues. First, people may appear to remit spontaneously because they actually do, but also because of limited test–retest reliability of the diagnosis [31]. For instance, using a validated diagnostic