Behavioural addiction - Overview

A “behavioural addiction” is an addictive disorder involving an activity or behaviour, rather than a psychoactive substance.

Grand format
Ecrans de portable, de tablette et de smartphone

For an overview on gambling, please see the dedicated page.

The observation of behaviours similar to those of substance addictive disorder, but centred around a given activity, sharing the same risk factors and evolving patterns, has contributed to the emergence of the concept of “an addiction to an experience” and minimising the physiological role of substances in the development of an addiction. Substance or activity, combined with contextual elements, will in some way allow personal predispositions and pre-existing psychological problems to lead to addictive disorders. The physiological impact of substances is, for its part, at the root of the dependence expressed by withdrawal symptoms, which may exist in the absence of an addiction. However, behavioural addictive disorders can also give rise to withdrawal symptoms upon cessation (nervousness, anxiety, irritability...), the origin of which is more complex.

Behaviours in question

Extensive research on gambling has shown the functional and anatomical alterations of brain circuits similar to those that characterise substance addiction. Gambling disorder is inserted in DSM-5 (2013, updated in 2022) under the heading “Addictive disorders” and in ICD-11 (2018 in English) as a “Disorder due to addictive behaviour”.

Other practices are usually discussed as potentially leading to addiction. These are mainly gaming disorders and other disorders due to online activities (social media, streaming...), sexual practices (particularly online pornography), compulsive shopping, exercising, work, and certain eating disorders. Certain addictive behaviours are also attributed to the media of these activities, namely smartphones, screens, the Internet, or information and communication technologies, with the role assigned to these media (against the online activities themselves) in the genesis of an addiction not always very clear.

These are common activities, sources of satisfaction, generally practiced by most of the population. The proportion of people suffering from a pathological addiction, among those conducting the activity, remains a small minority and appears when several vulnerability factors are present.

Manifestations

All these behaviours may give rise to symptoms similar to those of addiction: the repetition of a behaviour or an activity which takes up more and more of your time, at the expense of other activities (work, sleep, social life, hobbies...), the loss of control over the practice, inability to reduce or stop this activity, despite the attempts and the significant negative impacts on physical, mental, social, and financial well-being. People with behavioural disorders experience suffering upon cessation, and impulses described as irrepressible urges to repeat the practice (craving). This is precisely this loss of control that differentiate an addiction from other excessive behaviours or from a passion (very high level of investment in a particular activity).

However, with the exception of gambling, there is no consensus on recognising each of these behaviours as a genuine addiction and certain are heavily disputed. This situation is mainly attributable to the criteria considered sufficient or necessary to establish what an addiction is. Clinicians mainly rely on clinical criteria and manage these disorders with cognitive behavioural therapies, and often drug prescriptions used in the treatment of addiction.

In addition, not every problematic practice is of an addictive nature (for example, bullying via social media, or eye strain following a significant amount of time spent in front of a screen). At the same time, although an addiction problem is very often accompanied by dedicating a very significant amount of time to an activity, it has been shown that the same cannot be said for the reverse: a long-standing practice does not necessarily indicate the existence of addiction.

Heterogeneous use of terms

The words “addict”, “addictive”, and “addiction” or “addictive behaviour” describe situations with different levels of severity. There is a spectrum between normal use and pathological addiction. Therefore, the term “addiction” can refer to anything from a catchy leisure to a severe disorder, including problematic but reversible behaviours, depending on its context.

On the contrary, to break away from the uncertainty of whether all these disorders are genuine addictions, problematic practices are very often not referred to by the term “addiction” in scientific literature. There is a variety of terms for each of them: dependence; disorders due to gaming, exercising…, addictive disorders; excessive, problematic, compulsive, addictive-like, pathological, or maladaptive use, etc. It is not always easy to determine what precisely these terms cover.

Measurement of phenomenon

The range of prevalence rates published for disorders related to each kind of such activity is generally very broad, so the possibilities for comparison are limited. These results depend on what exactly the screening tool used measures (level of severity of disorders, importance placed on one criterion over another, scope of activities observed), the specific sample surveyed and its representativeness of a larger population.

Online activities (Internet)

Internet addictive behaviours have grown rapidly, particularly with the spread of smartphones, and the efforts of content providers to draw users in and retain them for as long as possible, thus fostering excessive usage times. In this regard, problematic practices primarily focus on gaming, then on social media practice.

Some professionals speak of “internet addiction”, or “addictions to information and communication technologies”.

Measurements

Whether they concern online activities as a whole or certain ones specifically, studies are, for the most part, conducted among young populations, often composed of younger or senior students, on the basis of non-representative samples.

The most popular measurement tools are the Internet Addiction Test (IAT) and the Compulsive Internet Use Scale (CIS).

The prevalences measured show very significant disparities between studies (from 1% of adolescents in Norway to around 50% according to a study carried out in 2013 in numerous Asian countries), due to a lack of consensus on the definition of disorders, and on measurement grids.

There is a convergence of data indicating that European countries are the least affected by these problematic uses, particularly in comparison with Asian countries. Young adults appear to be the most affected group. Data on gender is heterogeneous.

Among European teenagers, the average prevalence of problematic internet use stands between 1.2% and 4.4% (Lopez-Fernandez et al., 2023). One of the international studies, carried out in 2009 in 11 European countries, reported an average prevalence of pathological internet use among school-aged teenagers, of 4.4% (2.6% for France) (Durkee et al., 2012).

Among adults, Internet addiction affects around 2% of the population across the globe (Lozano-Blasco, 2022). An international study carried out in 2015 identified that the prevalences of problematic use ranged from 0.7% in Italy to 5.4% in France (except for the United Kingdom, which is most unusual at 10.1%) (Lopez-Fernandez et al., 2023). Other research supports this scale, but some international studies find prevalences to be in the range of 15% to 20%.

Gaming

The existence of an addiction to video games is only discussed at the margins. The gaming disorder is inserted in ICD -11 and appears in DSM-5 in the chapter Disorders proposed by supplementary studies under the heading Internet gaming disorder.

Two points of attention

Most professionals highlight the positive benefits of gaming, and the fact that this practice should be regulated and controlled, rather than stigmatised for fear of potentially addictive behaviours. Gaming addiction must not be mistaken with temporary periods of excessive gaming, which may affect a large proportion of adolescents at some point in time.

In the 2010s, an online gaming economic model emerged, which consisted of making profits from the player during gameplay, and not just from the act of purchase. This monetisation of gaming involves repeated (direct or implicit) incentives aimed at players, to spend real money to improve playing conditions, or speed up their progress. To make games more addictive and maximise player spending, game developers integrate gambling mechanics for which there is a well-documented impact on loss of control.

Measurements

Among the multiple measurement scales, those which appear to be used to the most are derived from the Young Internet Addiction Test (YIAT), followed by GAS-7 (Game Addiction Scale-7), for which there is a version with 9 questions (GAS-9).

The majority of prevalences of the online gaming disorders, measured on large samples of adolescents (1 000 to 20 000 people), published since 2015, and using the most common screening tools, are concentrated between 1.2% and 1.8% (more widely between 0.3% and 2.3%). In France, a measurement of problematic gaming use (using GAS) among 434 adolescents in the second grade (international classification, age around 13) from several junior high schools resulted in a prevalence of 8.8% (Bonnaire and Phan, 2017).

Sexual practices

Since the 2010s, clinicians have been increasingly faced with requests for assistance from people who have lost control of their practices: particularly viewing pornographic contents online, but also compulsive masturbation for example, or repeatedly seeking multiple sexual partners via specific online applications. The frequency of these problems has increased with the spread of smartphones (also known as cybersex).

Sexual addictive disorders are most often referred to as compulsive behaviour rather than addiction. They appear in DSM-5 and ICD-11, but in chapters not dedicated to addictive behaviours.

Measurements

There are numerous screening tools which do not completely overlap, neither in terms of activities measured, nor in terms of criteria, and very few studies have been carried out in the general population.

The prevalences of compulsive sexual behaviours reported in bibliographical reviews, measured particularly in the United States are mainly in the range of 1% to 6% (in different time frames). Aside from the issue of tools, the reliability of responses may also be questioned, in light of the potential reluctance of respondents to speak on these topics (Derbyshire and Grant, 2015).

Problematic practices may be at their peak among young adults (26–40-year-olds). Among students, prevalence data for problematic use of online pornography present extreme disparities ranging from 0.17% to 80% depending on the study. It was established that the prevalence of disorders is more significant among men than women (around 3 to 5 men for every woman).

No French data has been found.

Screens

The word screen, as an object of a problematic use, appears very little in scientific literature on addictions, in favour of all the terms denoting activities which are accessible via the Internet. These may be referred to as addictive, in the sense that they share the characteristic of drawing users in and keeping them interested, and thus foster prolonged usage times.

However, the term “screen addiction” is misleading. Most of the discourse and research focuses on the behaviours of children and adolescents, when screen time is deemed excessive. This stems from the fact that screen time takes away the time spent on other activities (which are necessary for cognitive development, sleep, physical activity, academic engagement...) and negatively impacts on physical health (eye strain, obesity...), but also because exposure to screens, or certain activities and online contents, are age-inappropriate for children. However, impacts which are not related to addiction are often attributed to it (such as delays in cognitive acquisition among young children), and a negative nature attributed to the object itself.

Literature shows that the problematic nature of online activities among children is dependent on their characteristics (age-appropriateness for child, diverse risks...), on the context (parental involvement, interactivity...), and on the proposed alternative activities.

There is extensive data on exposure to and use of screens by children and adolescents, and on their potential negative impacts, but seemingly little on addiction.

Social media

Beyond the addictive nature of social media (difficulty in detaching from it) it seems that the questions are more so centred around their misuse, particularly among young people (bullying, misinformation, rumours...), and its potential negative impacts (for example on self-image and self-esteem during adolescence), than on a real pathological addiction. For example, depending on the person, the same amount of time spent using the Internet can reflect a state of malaise or a state of wellbeing, for someone with a rich social network, online, but also in the real world.

Measurements

There are diverse screening tools used for behavioural disorders related to social media. It is difficult to assess which are the most widely used. A meta-analysis conducted on 63 independent samples, originating from 32 countries, observed a heterogeneity in the prevalence of social media addiction. The studies were divided into three groups, based on the chosen measurement tools, and interpreted as varying degrees of disorders severity. The average prevalences stand at 5% for an addiction identified as “very severe”, 13% for a “severe” level, and 25% for a “moderate” level (Chen et al., 2021).

There is seemingly no French data.

Smartphones

The spread of smartphones is accompanied by the rise in all addictive disorders related to online practices, due to their constant availability, and the growing number of applications available. Nevertheless, professionals most often link these disorders to the activities and content to which the smartphone give access, and many consider the latter as merely an enabler of these behaviours.

However, certain behaviours, which are maybe not addictions, appear to arise from the phone itself: notably the excessive checking of notifications or messages (overchecking), potentially because this behaviour has turned automatic - or a decrease in awareness of others, placing more importance on your phone than those present (phubbing, which is a contraction of “phone”, as in “telephone”, and “snubbing” as in “to snub”).

Measurements

Of the numerous scales used, two have been validated in France: the short version of the Problematic Mobile Phone Use Questionnaire (PMPUQ-SV) and the Internet Addiction Test–smartphone version (IAT-S).

The prevalences recorded at international level (from around 10% to 60%) depend on the scales used and the differentiation between significant, excessive, and problematic use, and addiction, which is often implicit. The relevance of those data is heavily criticised.

Eating disorders

Regarding eating habits, the addiction model is mainly proposed for a form of hyperphagia, [binge eating disorder]”. The object of the disorder is not yet clearly identified, borrowing from both behavioural addictions and substance addiction: addiction to high-calorie, high-sugar, and high-fat substances which are very easy to binge on (food addiction).

The concern surrounding these hyperphagic behaviours is closely linked to that of obesity, to the extent of creating misunderstandings which incorrectly tend to assimilate them. The application of the hyperphagia addiction model meets two major challenges: the first is testing the treatment developed for addiction, among people suffering from this form of hyperphagia. The second is societal. One of the stakes of the recognition of an addictive disorder linked to fatty and sugary foods, which are often cheap and widely available, is the search for opportunities to influence their composition.

An addictive component (addiction to hunger or fasting, and control) is also raised for eating disorders characterised by restriction or avoidance, and for anorexia nervosa.

Eating disorders are present in DSM-5 and ICD-11, but also in other chapters than those on addictive behaviours.

Measurements

The Yale Food Addiction Scale (YFAS) is the most widely used scale. There are also other tools, particularly the Eating Disorder Inventory (EDI), which is validated and widely used in France, particularly in its short form.

Prevalence data is dependent on targeted practices. Food addiction may affect between 20% and 40% of people being treated for obesity. It affects more men than women.

Exercising

There is evidence of the existence of withdrawal symptoms due to physiological dependence, upon cessation of a very regular sporting activity. The existence of problematic practices associated with addiction-like symptoms is relatively well-known. It is difficult to identify an addictive behaviour, given that sport is an activity which is highly valued by society.

Another difficulty, particularly when assessing prevalence, is the differentiation between an addiction (with serious negative impacts), and a very high level of investment in exercising. Practising a high-performance sport in fact presents certain characteristics which are close to those of addiction, and may have similar impacts: withdrawal from social relationships, injuries, potential conflicts with inner circle, etc. At the same time, issues concerning diet, weight, and body shape may be requirements of this discipline.

Addictive behaviours concerning exercising are often associated with eating disorders, which they can manifest.

As for the most affected sports, studies yield contradictory results, even though endurance sports are often cited as the most concerned.

Measurements

The most widely used scales are the Exercise Dependence Scale (EDS), which is the reference, and the Exercise Addiction Inventory (EAI). Both have been validated in French. There are other more secondary tools. Some scales are focused on a sport such as the Running Addiction Scale or the bodybuilding addiction criteria, for example.

Currently, there is no truly reliable prevalence data based on statistically representative samples. The reasons for this are the diversity of tools, which identify heterogeneously defined problem behaviours (risk, high-risk, addiction...), and their inappropriateness for certain populations, but also the disparity of populations surveyed (high-performance sports in a training centre, people frequenting a fitness studio, amateur runners or groups where the levels of sport practice are not well differentiated...).

Work

The notion of “work addiction”, most often referred to as dependence, is a recent phenomenon (2000s). The existence of excessive or compulsive behaviours linked to work and leading to burn-out situations, for example, is recognised, but there is no definitive consensus on the addictive nature of these disorders. The term addiction may be applied too broadly to refer to excessive work behaviours or situations, which do not fall within an addictive process. They can respond to various kinds of constraints, those linked to work itself, or to working conditions. On the contrary, it may be difficult to identify the pathological nature of a behaviour which is highly valued by society.

The English term “workaholism” is often used as a synonym of work addiction. Yet, since the 1990s, this concept has been developed through managerial sciences and those involving psychology applied to work or organisations, making no reference to the clinic and the addiction process: workaholism characterises the individual that is working excessively and compulsively, in response to an internal pressure which causes unease and guilt. This approach falls within a behavioural model characterised by the major role given to the individual’s personal characteristics.

Measurements

Measurements of prevalence experience vagueness around the concept of work addiction, a term which is often applied to the criteria characterising workaholism, in which the problematic work relationship greatly exceeds the limits of addiction. None of the three main tools for measuring work addiction: the Dutch Work Addiction Scale (DWAS), the Work Addiction Risk Test (WART), which is already used in France, and the Bergen Work Addiction Scale (BWAS), stem from the field of addiction. Based on bibliographical information, a study deems only the BWA to measure work addiction (Morkevičiūtė et al., 2021).

In light of the diversity of measurement scales and particularly the lack of statistical representativeness of samples in many studies, the international prevalence data for workaholism ranges from 1.5% to almost 45%. A meta-analysis published in 2023 estimates the average prevalence of workaholism at 8% (IC95% = [3.4–17.8]), based on 10 studies conducted on representative national samples (out of 53 selected studies). This measurement rises to 15.2% (IC95% = [12.4– 8.5]), if it is based on all studies (Andersen et al., 2023). When they are primarily based on working hours, data unsurprisingly reflect that people in higher managerial, intellectual and scientific professional position and employees of service-sector companies work more than others, and men more than women (even more so when part-time hours are included in these measurements).

There is seemingly no representative French data.

Compulsive buying disorders

Compulsive buying disorders (sometimes also designated as compulsive shopping disorders) are mostly referred to as “compulsive”. They refer to largely heterogeneous situations.

There are two ways to perceive compulsive buying.

  • The individual approach, which promotes the idea, implicitly or not, that the practice of the people affected are distinct from that of other buyers/shoppers. It focuses on vulnerability factors and specific psychic functioning of individuals.
  • The societal approach which considers the precedence of societal factors (consumer society, integration of the role of the object as compensation...), economic factors, and those associated with marketing practices and/or enabling the act of purchase (online shopping for example) in provoking problematic practices. This point of view presents a problematic buyer/shopper who is not fundamentally different from the general population, with whom they share the same shopping practices, but who is located at the extreme of a continuum of situations.

Measurements

The most used measurement tools are derived from the Compulsive Buying Scale (CBS).

The measurements of prevalence are not comparable and vary considerably (type of practices measured, most often self-selected populations, more or less significant weight placed on emotional factors and/or levels of income...).

In the general population, international studies measure prevalences of around 5% (1% to 10%).

There is seemingly no French data in the general population.

The prevalence of compulsive shopping shifts inversely with age. With regard to gender, the fact that these problems more frequently affect women is not always recognised. However, they would be more likely to seek help.