Balancing validity, utility and public health considerations in disorders due to addictive behaviours

The concept of “behavioural (non‐chemical) addictions” was introduced close to three decades ago, and a growing body of literature has emerged more recently on this and related constructs1, 2. Simultaneously, some authors have noted that the classification of behavioural addictions requires further effort3, 4. Here we provide an update on this area, emphasizing recent work undertaken during the development of the ICD‐11, and addressing the question of whether it is useful to have a separate section on disorders due to addictive behaviours in this classification.

Both the DSM and ICD systems have long avoided the term “addiction” in favor of the construct of “substance dependence”. However, the DSM‐5 includes gambling disorder in its chapter on substance‐related and addictive disorders, and provides criteria for Internet gaming disorder, considering it an entity requiring further study, and highlighting its similarities to substance use disorders5, 6, 7. In the draft ICD‐11, the World Health Organization has introduced the concept of “disorders due to addictive behaviours” to include gambling and gaming disorders2, 8. These disorders are characterized by impaired control over engagement in the addictive behaviour, the behaviour occupying a central role in the person's life, and continued engagement in the behaviour despite adverse consequences, with associated distress or significant impairment in personal, family, social, and other important areas of functioning2, 8.

An important focus during the development of DSM‐5 was on diagnostic validators. Certainly, there is some evidence for overlap between substance use disorders and disorders due to addictive behaviours, such as gambling disorder, on key validators including comorbidity, biological mechanisms, and treatment response5, 6, 7. For gaming disorder, there is increasing information on clinical and neurobiological features. For a wide range of other putative behavioural addictions, less evidence exists. Further, several of these conditions may also demonstrate overlap with impulse control disorders (in DSM‐IV and ICD‐10), including comorbidity, biological mechanisms, and treatment response9.

The groups working on ICD‐11 recognize the importance of validators of mental and behavioural disorders, given that a classification system with greater diagnostic validity may well lead to improved treatment outcomes. At the same time, ICD‐11 workgroups have focused in particular on clinical utility and public health considerations in their deliberations, with an explicit focus on improving primary care in non‐specialist settings, consistent with the ICD‐11's emphasis on global mental health. Fine‐grained differentiations of disorders and disorder subtypes, even if supported by empirical work on diagnostic validity, are arguably not as useful in contexts where non‐specialists provide care. However, associated disability and impairment are key issues in this perspective, supporting the inclusion of gambling and gaming disorders in ICD‐112, 8.

There are multiple reasons why the recognition of disorders due to addictive behaviours and their inclusion in the nosology together with substance use disorders may contribute to improving public health. Importantly, a public health framework for prevention and management of substance use disorders may well be applicable to gambling disorder, gaming disorder, and perhaps some other disorders due to addictive behaviours (although the draft ICD‐11 suggests that it may be premature to include in the classification any other disorder due to addictive behaviours outside of gambling and gaming disorders).

A public health framework to considering disorders due to addictive behaviours arguably has a number of specific advantages. In particular, it places appropriate attention on: a) the spectrum from leisure‐related behaviour without any harms to health through to behaviour associated with significant impairment; b) the need for high‐quality surveys of prevalence and costs of these behaviours and disorders, and c) the utility of evidence‐based policy‐making to reduce harm.

Although some may be concerned about the medicalization of ordinary living and lifestyle choices, such a framework overtly recognizes that some behaviours with addictive potential are not necessarily and may never become a clinical disorder, and it emphasizes that prevention and reduction of health and social burden associated with disorders due to addictive behaviours may be achieved in meaningful ways by interventions outside the health sector.

Several other criticisms of the constructs of behavioural disorders or disorders due to addictive behaviours may be raised for discussion. We have previously pointed out in this journal that additional work is needed to make strong claims about diagnostic validity9, and the draft ICD‐11 currently also lists gambling and gaming disorders in the section on “impulse control disorders”. Relatedly, there is a reasonable concern that the boundaries of this category may be inappropriately extended beyond gambling and gaming disorder to include many other types of human activity. Some of these arguments overlap with those which emphasize the dangers of a reductionist medical model of substance use disorders.

While cognizant of the importance of these issues, our view is that the potentially large burden of disease due to behavioural addictions requires a proportionate response, and that the optimal framework is a public health one.

Here we have outlined reasons why a public health framework that is useful for substance use disorders may also be usefully applied to gambling disorder, gaming disorder and, potentially, other health conditions due to addictive behaviours. This argument provides support for including substance use disorders, gambling disorder and gaming disorder in a single section of the chapter on mental, behavioural or neurodevelopmental disorders in ICD‐11.

The authors alone are responsible for the views expressed in this letter and they do not necessarily represent the decisions, policy or views of the World Health Organization. The letter is based in part upon work from Action CA16207 “European Network for Problematic Usage of Internet”, supported by the European Cooperation in Science and Technology (COST).