Behavioural addictions: A challenge for psychopathology and psychiatric nosology

Abstract

Disorders characterised by repetitive and problematic behaviours and poor impulse control have been increasingly conceptualised as behavioural addictions. This article examines the concept of behavioural addiction and argues that the addiction framework is only one approach to these behavioural disturbances. It cautions against a tendency to regard many activities that are performed with an extraordinary intensity or frequency and that have some negative consequences as behavioural addiction. There is a need for more research to better understand the links between repetitive and problematic behaviours and other psychopathology, as well as the function of these behaviours and factors that maintain them.

Behavioural addictions have been receiving increasing attention. Recent articles refer to them as a ‘plague of our era’ (Billieux et al., 2015: 119) and a ‘rising tide’ (Chamberlain et al., 2016). In 2012, a journal entirely devoted to behavioural addictions (Journal of Behavioral Addictions) was launched. At least three edited books on behavioural addictions have been published between 2014 and 2016 (Ascher and Levounis, 2015; Petry, 2016; Rosenberg and Feder, 2014), which is remarkable for disorders that are not officially recognised as a distinct class. A decision to move pathological gambling, the prototypical behavioural addiction, from the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) group of impulse-control disorders and to classify it among ‘substance-related and addictive disorders’ in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) has bolstered a notion that behavioural addiction is a valid concept.

Along with these developments, there have been calls for caution that have in common a concern that too many behaviours, if indulged in excess, might end up being conceptualised as behavioural addictions (e.g. Billieux et al., 2015; Frances and Nardo, 2013; Starcevic, 2013). In addition, the most appropriate conceptualisation and classification of behavioural addictions remains controversial. The aim of this article is to contribute to this debate.

Brief history

In 1990, Isaac Marks (1990) introduced the term ‘behavioural (non-chemical) addiction’. He described behavioural addictions as repeated urges to engage in counter-productive behaviours and proposed that they encompass obsessive-compulsive disorder (OCD), ‘compulsive spending’ (including gambling), ‘overeating’ (bulimia), ‘hypersexuality’ and kleptomania. A few years later, the concept of obsessive-compulsive-related disorders or obsessive-compulsive spectrum disorders was introduced (Hollander, 1993) to refer to the conditions characterised by a difficulty controlling or delaying repetitive behaviours. This concept, similar to behavioural addictions, allowed for disorders lying on a continuum from compulsivity to impulsivity to be included. As a result, many psychiatric and neurological disorders were postulated to belong to this group: OCD, body dysmorphic disorder, hypochondriasis, impulse-control disorders, behavioural addictions, eating disorders, repetitive self-injurious behaviours (e.g. skin picking), some personality disorders, substance use disorders, autistic and Asperger’s disorders, chronic tics, Tourette’s disorder, stereotypic movement disorders and others.

These developments were accompanied by further elaboration of the concepts of impulsivity and compulsivity. Impulsivity is a multidimensional construct that refers to thrill seeking, greater importance attached to immediate rewards than to long-term goals, difficulty delaying gratification, inability to resist urges, quick and poorly planned response to stimuli, unreflective decision making and a tendency to ignore the negative consequences of one’s actions. Compulsivity denotes an unpleasantly repetitive and functionally impairing behaviour, which is performed in a habitual or stereotyped way in order to prevent a perceived negative consequence and persists despite being harmful. Although impulsivity and compulsivity appear to be quite different constructs phenomenologically, research has demonstrated much overlap between their neurobiological correlates (e.g. Fineberg et al., 2014; Grant and Kim, 2014).

The occurrence of impulsivity and compulsivity at different times was postulated to characterise both substance use disorders (substance addictions) and behavioural addictions, and the notion of ‘impulsive-compulsive disorder’ was proposed to account for these conditions (Cuzen and Stein, 2014). According to this model, in the early stages of addictions, impulsivity plays a prominent role because of the reinforcement of behaviour that is motivated by reward seeking, whereas in the later stages, compulsivity ‘takes over’ and behaviour that is motivated by avoidance of the perceived negative consequences (e.g. withdrawal symptoms) is reinforced.

Another factor contributing to the rise of behavioural addictions has been a broadening of the definition of addiction. Thus, the ‘core elements’ of addiction were described as follows: (1) a state of craving or an urge that immediately precedes behavioural engagement; (2) poor control over behavioural engagement; (3) continued behavioural engagement despite negative consequences (Potenza, 2006). With addiction defined via ‘behavioural engagement’, regardless of whether it pertains to substance use or performance of certain activities, the door was wide open for various behaviours to be considered addictive and to potentially lead to behavioural addiction. A 2011 official definition of addiction by the American Society of Addiction Medicine (‘inability to consistently abstain, impairment in behavioural control, craving, diminished recognition of significant problems with one’s behaviours and interpersonal relationships and a dysfunctional emotional response’) has cemented this broad approach to addiction.

The range of behavioural addictions

Despite the broadening of the concept of addiction, behavioural addictions are often defined more restrictively, via components associated with substance addiction, such as tolerance and withdrawal symptoms. Thus, one influential model (Griffiths, 2005) has postulated the presence of all of the following components in any behavioural or substance addiction: salience, mood modification, tolerance, withdrawal symptoms, conflict and relapse. Attempts to transpose tolerance and withdrawal symptoms from substance addictions to behavioural addictions might have been driven by a need to ‘legitimise’ behavioural addictions as bona fide addictions, but they have been controversial. Tolerance, defined as an increase in various aspects of the activity to achieve the desired effects, is not equivalent to dose escalation in substance addiction. Likewise, the nonspecific and negative emotional and behavioural responses to activity deprivation are very different phenomena from the relatively specific physical withdrawal symptoms that accompany a sudden cessation of the use of a substance that one is addicted to.

Many repetitive and problematic behaviours have been proposed as behavioural addictions, regardless of whether this is based on a broad or a more restrictive concept of addiction. It has been suggested that the conceptualisation of certain disorders as behavioural addiction might help overcome their neglect, raise awareness and attract research funding. The most consistently mentioned behavioural addictions are pathological gambling (renamed gambling disorder in the DSM-5), ‘sex addiction’, ‘compulsive buying’, ‘Internet addiction’, Internet gaming disorder and kleptomania. The second ‘tier’ of behavioural addictions, discussed in the literature somewhat less often, comprises exercise addiction, eating addiction (or food addiction), ‘love addiction’, work addiction and addictions to social networking, using mobile phones and tanning. The final ‘tier’ is represented by disorders or behavioural patterns that have been proposed only occasionally as behavioural addictions: binge-eating disorder, pyromania, trichotillomania, skin-picking disorder, orthorexia (preoccupation with eating healthy food), muscle dysmorphia (as an addiction to body image), musicorexia (addiction to musical activity) and addictions to self-mutilation, dancing, studying, fortune telling, stock markets and so on.

The list of behavioural addictions could be expanded ad infinitum. However, only gambling disorder has found a place in the DSM-5 as an addiction disorder. Most of the other putative behavioural addictions have been left out, and the Internet gaming disorder was introduced in the DSM-5 only as a condition for further study. The remaining behavioural addictions are classified in the DSM-5 as ‘disruptive, impulse-control and conduct disorders’ (kleptomania, pyromania), ‘obsessive-compulsive and related disorders’ (trichotillomania, skin-picking disorder, muscle dysmorphia as a specifier for body dysmorphic disorder) and elsewhere (binge-eating disorder). This reflects a lack of agreement as to how various repetitive and problematic behaviours should be conceptualised.

Problems with behavioural addictions

While most behavioural addictions refer to the activities that are a part of normal, everyday life (e.g. shopping, sexual activities, using the Internet, exercising, eating, working and using mobile phones), others do not extend to normality and are always pathological (e.g. kleptomania) or extend to normality in a very limited sense (e.g. gambling disorder). Addiction to eating or food is different from other behavioural addictions and perhaps more akin to substance addictions because of the ingested ‘substance’, i.e., food. Therefore, there appears to be an excessive conceptual heterogeneity within behavioural addictions.

When behavioural addictions are conceptualised on a continuum with normal behaviours, the criteria for distinguishing between high activity engagement and behavioural addiction need to be clearly formulated. The ‘amount’ of activity, which usually means too much time spent while performing that activity, is not a reliable criterion; this is because normative activity engagement cannot be predetermined, it varies from one person to another and with some activities like use of the Internet and mobile phones, it has been steadily increasing in recent times. Preoccupation with the activity is similarly unreliable as a criterion because of its subjective and not necessarily pathological nature. Mood-modifying effects of behaviours are sought by people who enjoy the respective activities as well as those who may be addicted to them. The negative consequences of repetitive behaviours might be a more useful criterion, but a boundary also needs to be drawn here, and it is uncertain where to do that in terms of the number and type of the consequences and the extent of impairment.

Repetitive and problematic behaviours do not have to be conceptualised as an addiction. This conceptual framework may be limiting in several ways. First, it may impose analogies with substance addiction that are inaccurate or tenuous at best; this has been demonstrated by the aforementioned attempts to ‘borrow’ tolerance and withdrawal symptoms from substance addictions and incorporate them into behavioural addictions. Second, the treatment of substance addictions usually entails abstinence, whereas recommending abstinence in the treatment of behavioural addictions that extend to normality would be unreasonable. Furthermore, addiction has negative and stigmatising connotations, and it does matter whether or not a condition like pathological gambling is regarded as an addiction. Also, the addiction framework may denote ‘moral panic’ with which some of the proposed behavioural addictions are approached.

More fundamentally, the addiction framework presupposes that the function of repetitive and problematic behaviours is to avoid unpleasant emotional states that would occur with the cessation of the particular behaviour; while this may be the case with some behaviours and in certain individuals, the purpose of repetitive and problematic behaviours seems to vary (e.g. Billieux et al., 2015). For example, qualitative studies have suggested that problematic gambling may serve the purpose of coping and escaping (Wood and Griffiths, 2007), as well as emotion management (Ricketts and Macaskill, 2003). Likewise, it has been postulated that negative life situations or unmet needs may motivate some people to use the Internet excessively and that such use may alleviate their emotional turmoil or help them meet their needs (Kardefelt-Winther, 2014). Moreover, the consequences of behaviours such as pathological gambling may have a very different impact from the consequences of other repetitive and problematic behaviours in that debt in pathological gambling often reinforces further gambling. Instead of assuming that the person is addicted to the particular activity, it might be more beneficial to attempt to understand the reasons for it and then tailor the therapeutic approach accordingly. Ultimately, when regarded solely within the addiction framework, behavioural addictions may suffer from the same problem that has plagued the concept of the obsessive-compulsive spectrum disorders (Starcevic and Janca, 2011): focusing on repetitive and problematic behaviours without considering their function prevents their understanding and use of adequate treatments.

The question of the underlying psychopathology is related to the above considerations. For example, the Internet gaming disorder can in many cases be more usefully regarded as a manifestation or a consequence of the underlying attention-deficit/hyperactivity disorder or severe social anxiety disorder. Likewise, a range of factors, such as trauma, insecure attachment, intimacy and relationship difficulties or emotion dysregulation may account for ‘sex addiction’. It would be erroneous to pay attention only to the outward addiction-like expression of psychopathology without addressing the underlying problems.

Yet another problem with behavioural addictions is their apparent longitudinal instability. A recent study examining a 5-year course of several behavioural addictions (involving video gaming, online chatting, exercising, eating, sexual activity and shopping) found that these behavioural patterns tended to be transient for most study participants (Konkolÿ Thege et al., 2015). Considering that addiction is usually a chronic condition, a transient or episodic course of these repetitive and problematic behaviours in a large number of individuals does not support the notion that they should necessarily be conceptualised as addiction. Instead, it appears that many of these behaviours may be better understood as context-dependent, i.e., as having specific triggers, arising in specific situations and lasting for limited periods of time.

From behavioural addictions back to impulse-control disorders

The classification of gambling disorder as an addiction disorder in the DSM-5 has not been supported by the committee preparing the International Classification of Diseases, 11th Revision (ICD-11). Their argument is that such classification is ‘premature’ because gambling disorder/pathological gambling has close links not only with substance use disorders but also with mania, major depressive disorder, other impulse-control disorders and OCD (Grant et al., 2014). Consequently, the committee has recommended a classification of pathological gambling among impulse-control disorders in the ICD-11. The same committee has suggested inclusion of a new condition, ‘compulsive sexual behaviour disorder’, as another impulse-control disorder. This condition, often referred to as ‘sex addiction’ or ‘hypersexual disorder’, is a behavioural addiction for which there is no counterpart in the DSM-5, not even as a disorder for further study. If adopted, this recommendation will lead to an enlargement, not further shrinking, of the group of impulse-control disorders; this is contrary to the trend observed in the transitioning from the DSM-IV to DSM-5.

The different approaches taken by the architects of the DSM-5 and ICD-11 reflect a lack of guidelines that would stipulate precisely how similar or dissimilar the psychiatric disorders need to be in order to be classified together or separately. Thus, there is an ongoing uncertainty, arbitrariness and ambivalence about the most adequate conceptualisation and classification of the disorders characterised by repetitive and problematic behaviours and poor impulse control. Furthermore, the same authors sometimes advocate different approaches and conceptualisations, usually in different publications. This has created even more confusion.

De-emphasising behavioural addictions as a way out of the quagmire?

Various behavioural addictions have been promoted to the extent that they have become a part of popular discourse. An increasing number of publications provide behavioural addictions with a quasi-scientific status and reify the concept. Some prominent authors do not seem disturbed by the embarrassing proliferation of behavioural addictions or are at best somewhat ambivalent about the concept. There is also a sense that behavioural addictions represent a ‘modern’ way of conceptualising certain psychopathology, which makes potential dissenters vulnerable to a criticism that they are not in touch with the latest developments in the field. A notion that behavioural addictions are becoming more and more common may be true to the extent that there are greater opportunities for people in the affluent societies to indulge in behaviours such as shopping and eating and that behaviours mediated by the Internet have the potential to become problematic via anonymity, disinhibition and impulsivity. However, the ‘modernity’ of these behaviours does not imply that they have to be regarded as addictions.

The debate about behavioural addictions should not remain within the academic realm, and it does matter how disorders characterised by repetitive and problematic behaviours and poor impulse control are called, conceptualised and classified. While many terms in the domain of behavioural addictions may be convenient, they often represent a simplification and as such, tend to be misleading. Clinicians should be aware of the danger of considering as addiction many gratifying behaviours that are indulged in excess and cause certain problems. If unchecked, this trend will inevitably lead to a further, unjustified and uncontrolled expansion of the catalogue of behavioural addictions, drastic lowering of the diagnostic threshold and spurious epidemics of all sorts of behavioural addictions. We need to go back only 10–15 years, to the excesses of the obsessive-compulsive spectrum disorders, to be reminded or to learn that the profession has already been in a comparable situation. Indeed, history lessons could help us avoid the same or similar mistakes.

It is important to emphasise that the addiction framework is only one way of conceptualising disorders characterised by repetitive and problematic behaviours and poor impulse control. There is no evidence that other conceptual approaches are inferior. While the addiction framework may be applicable to some individuals with a particular behavioural pattern, it is not applicable to all. Moreover, this framework has not been unequivocally espoused by psychopathologists and psychiatric nosologists. Their scepticism about behavioural addictions should inform clinicians and researchers, while fostering open-mindedness.

A need for better understanding and further studies of behavioural addictions can hardly be overstated. For example, the interplay of impulsivity and compulsivity is still poorly understood, and efforts to assess these constructs dimensionally should be encouraged. This could lead to the formulation of testable hypotheses about each behavioural addiction. Qualitative studies should help us better understand the function of repetitive and problematic behaviours and take into account the contextual factors that are often neglected. Relationships between putative behavioural addictions and various forms of psychopathology are another important area of research. We also need to learn much more about the natural course and complications of these repetitive and problematic behaviours. The concept of addiction itself might benefit from some ‘tightening’ and further elucidation of what it encompasses. Finally, it still remains to be ascertained under what circumstances various disorders characterised by repetitive and problematic behaviours and poor impulse control are better conceptualised as impulse-control disorders, impulsive-compulsive disorders, addictions or in some other way.