The Creation of an Illness: Video Games and Defining Addiction

“If every gratified craving from heroin to designer handbags is a symptom of ‘addiction,’ then the term explains everything and nothing.”

— Amanda Heller (Boston Globe, 11/02/08)

In his 2010 book, Unhinged: The Trouble with Psychiatry, Dr. Daniel Carlat¹ (pp. 53–54) shares an interview with Robert Spitzer (chair of the American Psychiatric Association’s DSM-III taskforce), discussing the rationale used to arrive at what would become the guidelines for the diagnosis of major depressive disorder by all mental health professionals in the United States:

Carlat: How did you decide on five criteria as being your minimum threshold for depression?
Spitzer: It was just consensus. We would ask clinicians and researchers, “How many symptoms do you think patients ought to have before you would give them a diagnosis of depression?”
And we came up with the arbitrary number of five.
Carlat: But why did you choose five and not four? Or why didn’t you choose six?
Spitzer: Because four just seemed like not enough. And six seemed like too much.

For obvious reasons, much of what we think we know about the world comes to us by means of trust. Trust that wherever there exists a gap in our knowledge, and where we do not possess the relevant qualifications to engage with or even understand the relevant science dealing with that subject, there will always be an authority whose views we can trust as the received wisdom of that scientific discipline. This is simply a necessary fact about the world we navigate. With limited amounts of time, and the impossibility of receiving the necessary training required for engagement with every major scientific discipline in existence, we simply have to rely on the expertise of those more qualified than us to supply us with the knowledge we are incapable of producing ourselves.

The problem is that sometimes, as above, we discover how certain decisions that affect the lives of millions of individuals, decisions which also shape our understanding of the world and of a phenomenon like depression, are made on the basis of an arbitrary consensus that doesn’t strike us as particularly scientific. If a professional psychiatrist like Dr. Carlat could be unaware of the way the diagnostic criteria for depression were developed within his own profession, how likely is it that the general public would ever come to know about this either?

To make things worse, this kind of decision-making process is only revealed to us (and members of the discipline itself) two decades after the publication of the DSM-III. Millions of individuals have been diagnosed and treated for a mental illness that, upon closer inspection, was codified into the most influential medical text in all of mental health, on the basis of what seemed to be the appropriate number of symptoms.

And yet, 3 decades and 3 major revisions to the DSM later, the process of how mental disorders become codified is even less transparent than before. Spitzer himself was one of the first psychiatrists to sound the alarm when the DSM-5 was in the process of being drafted. To his shock, those involved in the creation of this new edition were under strict non-disclosure agreements preventing them from discussing anything related to the process of creating the latest DSM. A non-disclosure agreement that is legally binding to this day:

DSM-V participants have been required to sign a confidentiality agreement that prohibits them from divulging any confidential information about the DSM-V revision process. That process is broadly defined as “all work product, unpublished manuscripts and drafts and other prepublication materials, group discussions, internal correspondence, information about the development process and any other written or unwritten information in any form that emanates from or relates to my work with the APA Task Force or Workgroup.”

Remarkably, that agreement extends beyond the time of the publication of DSM-V. Even with the exception that allows the participant to discuss DSM matters if “necessary to fulfill the obligations” of his or her appointment, this agreement forces the participant into the awkward position of having to decide whether providing information about the DSM is part of his job. In those likely frequent situations in which providing information is not deemed part of the job, this hardly results in a transparent DSM-V. (Spitzer, Psychiatric Times)

This brief background on the general secrecy behind the decision-making process of how mental illnesses come to be formally recognized is important to keep in mind when we consider the latest condition to be accepted into the canon of mental health: Gaming Disorder.

Defining the Condition

According to the World Health Organization, Gaming Disorder is:

A pattern of gaming behavior (“digital-gaming” or “video-gaming”) characterized by impaired control over gaming, increasing priority given to gaming over other activities to the extent that gaming takes precedence over other interests and daily activities, and continuation or escalation of gaming despite the occurrence of negative consequences.

For gaming disorder to be diagnosed, the behaviour pattern must be of sufficient severity to result in significant impairment in personal, family, social, educational, occupational or other important areas of functioning and would normally have been evident for at least 12 months.²

What makes this debate so interesting is that we can see, in real-time, the intellectual process by which a new type of illness becomes formally recognized by the relevant qualified authorities of the mental health profession. Unlike the often secretive DSM process, the International Classification of Diseases (the international standard for the identification of all diseases and health conditions) is a document with a far more open process that provides fascinating insight into how decisions about classifications are made, and why.

Ever since Thomas Kuhn’s The Structure of Scientific Revolutions, historians, sociologists, and philosophers of science have produced a voluminous literature detailing how scientific knowledge is produced and disseminated throughout culture. By providing a sketch of what this process looks like in the context of mental health, we will have a glimpse into how newly created (or “discovered,” depends on how you think about these things) disease entities gradually become entrenched into the public consciousness.

As defined, the idea of a Gaming Disorder is not remotely as absurd as the social media backlash portrays it to be:

But as the ICD makes clear:

Studies suggest that gaming disorder affects only a small proportion of people who engage in digital- or video-gaming activities.

The key to understanding who qualifies for the diagnosis and who doesn’t is the presence of significant impairment:

For gaming disorder to be diagnosed, the behaviour pattern must be of sufficient severity to result in significant impairment in personal, family, social, educational, occupational or other important areas of functioning and would normally have been evident for at least 12 months.

Most gamers, even hardcore ones, do not suffer from significant impairment in these major life areas to such an extent as to meet the criteria for having a gaming disorder. The ICD makes clear that what it’s concerned with are specific types of consequences characteristic of gaming addiction:

People who partake in gaming should be alert to the amount of time they spend on gaming activities, particularly when it is to the exclusion of other daily activities, as well as to any changes in their physical or psychological health and social functioning that could be attributed to their pattern of gaming behaviour.

If most of us are comfortable with the idea that something like gambling addiction is a real phenomenon worthy of treatment, then it’s not a massive stretch to think video games can lead to similarly addictive behavior, especially considering the fact that many videog ames feature reward loops that possess many structural and psychological similarities with gambling.³

So, the idea of a gaming disorder is not absurd, and certainly not a cause for alarm as loud and vociferous as the social media one has played it up to be. There is, however, good cause for skepticism and caution about enshrining Gaming Disorder into the ever-expanding list of mental health conditions.

The Problem of “Significant Impairment”

The ICD’s criteria for Gaming Disorder makes it clear that you may only be diagnosed with the condition if your gaming habits cause “significant impairment” in major life areas. In fact, all psychiatric disorders include this criterion. In the United States, the DSM also requires that this criterion is met before making any diagnosis, though the term used in the DSM is something called “clinically significant distress.” The problem they both face is vagueness.

All mental illnesses, by the definition agreed upon within the profession, must cause at least some harm to an individual. If there is no harm, there is no illness. This is why, for instance, despite Donald Trump possessing all the symptoms of narcissistic personality disorder, he does not meet the full criteria for a diagnosis. Because despite his narcissism, his condition does not cause him any clinically significant distress. Dr. Allen Frances, a towering figure in psychiatry who himself chaired the DSM-IV and DSM-IV-TR, has publicly argued as much against his fellow mental health professionals who think we should diagnose Trump with NPD. He writes that those wanting to diagnose the president:

…ignore the further requirement that is crucial in defining all mental disorders — the behaviors also must cause clinically significant distress or impairment.

Trump is clearly a man singularly without distress and his behaviors consistently reap him fame, fortune, women, and now political power. He has been generously rewarded, not at all impaired by it.⁴

A mental illness does not count as a genuine illness unless it impairs or causes clinically significant distress to a person. Harm of some sort must exist for something to count as an illness. This value judgment is built-in to the very idea of illness. Without harm, there is nothing to diagnose. And because the idea of what is harmful is a value judgment which is often either unclear or subject to widespread disagreement within and between cultures, the entire process of diagnosing someone is never a wholly scientific one. At some point, the mental health professional must always make a decision that necessarily includes a value judgment about whether harm exists in this particular client’s life because of their symptoms. Most of the time, the answer is quite clear and uncontroversial.

For instance, if a client is unable to maintain steady employment due to a condition like insomnia, which in turn causes friction at home leading to an unsuccessful marriage, and in desperation the client seeks help from a mental health professional, then the presence of real harm because of the condition is rather clear.

Where theory and practice come into conflict are cases where the very idea of what constitutes a “harm” aren’t nearly so clear. In these cases, the DSM, and science in general, offer little to no guidance. As Gary Greenberg writes:

Psychiatrists have gotten better at agreeing on which scattered particulars they will gather under a single disease label, but they haven’t gotten any closer to determining whether those labels carve nature at its joints, or even how to answer that question. They have yet to figure out just exactly what a mental illness is, or how to decide if a particular kind of suffering qualifies.

The DSM instructs users to determine not only that a patient has the symptoms listed in the book (or, as psychiatrists like to put it, that they meet the criteria), but that the symptoms are “clinically significant.” But the book doesn’t define that term, and most psychiatrists have decided to stop fighting about it in favor of an I-know-it-when-I-see-it definition (or saying that the mere fact that someone makes an appointment is evidence of clinical significance).

Instead, they argue over which mental illnesses should be admitted to the DSM and which symptoms define them, as if reconfiguring the map will somehow answer the question of whether the territory is theirs to carve up.⁵ (p. 11)

In a different cultural climate, one where we’re not so concerned about strict boundary policing within academia and squabbling over what questions belong to which field and which are off-limits, we might all agree without much controversy that the reason psychiatrists have such a difficult time with defining “clinically significant distress” is because it’s ultimately a philosophical question. To answer that, you must first provide a comprehensive account of what “health” and “illness” mean. And to answer those, we need a comprehensive philosophical account of what behaviors really, ultimately, lead to human flourishing. And on the flip-side, what kinds of things are ultimately truly harmful for us.

I know many readers may think these questions aren’t that difficult. No doubt we can all immediately name things that are obviously good and obviously bad for us, and in doing so we’re tempted to believe that the question of well-being shouldn’t be too hard to resolve. This would be a mistake. Explaining why would require a detour into the history of ethical philosophy. For that, I can only point elsewhere and hope the reader can see that answers to this are too important and require too much nuance to be responsibly condensed into a digestible soundbite.

What the DSM and ICD do is ignore, or at least sidestep, the fundamental questions of what a good life looks like, and simply assume, out of practical necessity, the answers to these questions as settled. As Warren Kinghorn (2015) notes, in doing so, the DSM:

…functions in our culture as a kind of moral document that, by designating certain forms of experience and behavior as disorder, displays personal and cultural judgments about the shape of a life well lived.

In defining normative parameters outside of which medical intervention and medical technology are ‘clinically indicated’, the DSM demarcates certain outer limits of the good life and, therefore, displays certain conceptions about how we ought to live, if one is influenced by prevailing cultural judgments.⁶ (p.66)

A Social Construction by Any Other Name

Since neither the DSM nor the ICD provide a standard as to what kinds of things count as “significant impairments,” it is always up to the clinician to make a determination based on their personal assessment of the situation. And it’s here where a scientific enterprise becomes much more of an art. Suppose a college student is trying to balance a full-time course load while working part-time on the side. The combined stress of balancing school, work, and somehow also maintaining a healthy social life becomes so stressful that this student begins to play games regularly as a coping mechanism. The amount of gaming hours played leads to the student missing several assignments and doing poorly on tests, and as a consequence their final GPA suffers, going from a 4.0 to a 3.0. Is this an addiction? Does this count as significant distress? There’s no principled way to answer this question. It’s up to the clinician.

What if a clinician decides that, while unfortunate, the consequences of the student’s gaming do not quite reach such a level as to be considered significant distress?

Suppose the clinician also sees another client with an identical life circumstance, but instead of the GPA lowering from a 4.0 to a 3.0, it goes down to a 2.5 GPA. This leads the student to lose their scholarship, eventually having to leave the university due to an inability to pay. Is that enough to meet the diagnostic criteria? Again. There is no standard.

In every case the clinician has to make a judgment call that is not at all a scientific one. It’s a judgment call that will be influenced by all sorts of external non-scientific factors such as personal beliefs, values, judgments, moods, personality, and a million other possible variables.

Most of the time this is fine. In ideal clinical settings, a clinician will work closely with a client, building a therapeutic relationship that allows a decision about the existence of significant impairment to be made through a mutual convergence of opinion. Most of the time we just know whether an activity we engage in frequently harms us in other areas of our lives, and there’s no need for philosophy to come in and challenge something so obvious.

The problem is what this kind of judgment call says about the scientific status of calling something an illness. Think back to the cases above of the two college students with identical gaming habits. What does it mean to say that one of them has a mental illness and the other doesn’t, when the only difference between the two is that one lost his scholarship while the other only slightly suffered a drop in GPA? Can the existence of a medical illness really boil down to something like losing a scholarship? Doesn’t this imply that the existence of an illness depends on an external social factor, such as the way that society responds to your situation?

Or phrased in a slightly different way, it seems like the social consequence of losing a scholarship is the crucial determining factor of whether you receive the medical diagnosis of Gaming Disorder. It is precisely this that sociologists and philosophers of science mean when they say something is socially constructed. Part of what we mean when we say something is socially constructed is that the existence of an entity, in this case a specific medical condition, partly or wholly depends on certain social attitudes, beliefs, or reactions towards that entity. In this particular case, a mental illness exists if and only if it causes certain types of distress that we get to define.

Coping and Addiction

“It is impossible to understand addiction without asking what relief the addict finds, or hopes to find, in the drug or the addictive behaviour.”

― Gabor Maté, In the Realm of Hungry Ghosts: Close Encounters with Addiction

What is the distinction between someone with a gaming disorder and someone who plays excessive amounts of video games as a coping mechanism for life related problems?⁷ From an outside perspective, two clients may seem to be engaging in identical behavior, though the internal reasons that explain that behavior may be totally different. Many of us, myself included, play games as a means to deal with work and life related stress. The hours we put into games are often difficult to justify even to ourselves, let alone to others. These are hours spent at the expense of spending time with loved ones, building new social relationships, and often comes at the cost of neglecting work or general life-related responsibilities.

Because of life stressors, we engage in behaviors like excessive gaming that themselves create new problems in major life areas like relationships. It’s all too common to see something like:

Stress → Games → Damaged Relationships → Increased Stress → Doubling Down on Unhealthy Coping Behaviors

And yet, there do not exist any good screening tools to distinguish these cases from one another. That’s because there is still no widespread agreement as to whether or not these are two separate phenomena.

The problem is in fact much bigger than this. There is no widespread agreement about any of the major details on gaming disorder: How is it defined? How widespread is the problem? What kinds of treatment are useful? What exactly are the symptoms and cut-off points? We don’t know.

And this is acknowledged as such even by those defending its inclusion into the ICD and DSM:

With regard to the prevalence of the Gaming Disorder, the authors of the debate paper are absolutely right: we have no idea about the exact magnitude of the problem, because studies were conducted in different populations, and with different disorder definitions and different assessment instruments and procedures. As a consequence, in general population studies, the prevalence of “gaming problems” ranges between a low of 0.5% and a high of 10% (e.g., Petry, Rehbein, Ko, & O’Brien, 2015). Moreover, very little is known about the long-term course of “gaming problems” and the probability of spontaneous recovery.

But what about the lack of international and interdisciplinary consensus about the definition of a Gaming Disorder? Indeed, there is no general consensus on the symptoms and cut-off points that are most suitable for the definition of the disorder; a point that is explicitly recognized by the authors of the DSM-5.⁸

The Invention of an Illness

If there is no widespread agreement about any of the major details,⁹ why do we include a condition like Gaming Disorder in a book alongside viruses like Ebola, Influenza, Dengue, Parkinson’s, HIV, Cancer, and countless other far more empirically validated conditions?

The answer is as boring as it is illustrative of how science works in the real world: technical trivialities coupled with societal concerns. Notice, for instance, how in a published paper defending the inclusion of Gaming Disorder into the ICD, a major reason offered supporting the decision is that the book doesn’t have a category specific to tentative disorders in need of further research:

Unfortunately, ICD-11 does not have a special category of tentative disorders and the decision has to be made to either not include a certain problem as a diagnostic category or to present it as a definitive disorder with a fixed set of criteria. This is unfortunate, but that seems to be the reality that WHO has to deal with in finding the most optimal solution.

This is the trivial technicality explanation. A new disease entity is formally recognized on the basis that the book where such diseases are categorized does not have a section for up-and-coming diseases that may or may not eventually make the cut. So it is deemed better to over-include than to adopt a conservative stance towards the inclusion of new diagnoses.

The social reason for its inclusion is financial. People who allegedly have the condition want treatment. But the only way for insurance to cover it is if it’s for a condition formally recognized in the ICD and DSM. We should therefore include it so that insurance companies will help clients pay for such treatment:

In many countries, treatment is only reimbursed when it concerns an officially recognized disorder, that is, treatment is only reimbursed for disorders mentioned in the ICD or DSM classification. Therefore, it is important that Gaming Disorder will be included in ICD-11 and that Internet Gaming Disorder has already been recognized as a tentative diagnosis in DSM-5.

And there it is. A mental disorder is born.

One Last Thing

If everything talked about until this point isn’t enough to make you pause, there is one last caveat to the ICD’s diagnostic criteria worth noting. I’ll let the reader draw their own conclusion about its possible implications in a clinical setting:

The pattern of gaming behaviour may be continuous or episodic and recurrent. The gaming behaviour and other features are normally evident over a period of at least 12 months in order for a diagnosis to be assigned, although the required duration may be shortened if all diagnostic requirements are met and symptoms are severe.