1 Introduction

Recent years have seen a surge of research on epistemic injustice. The concept was initially conceived by Miranda Fricker in her seminal book Epistemic Injustice: Power and the Ethics of Knowing and describes the injustices that epistemic subjects experience qua knowers (Fricker, 2007). According to Fricker, there are two distinct forms of epistemic injustice: testimonial and hermeneutical injustice. Testimonial injustice refers to instances of a knower’s knowledge and expertise being unfairly dismissed due to negative prejudices, say, racism, sexism, ableism, classism, or other unjust stereotypes concerning a person’s identity. Hermeneutical injustice, on the other hand, refers to harms that knowers sustain due to the dominant hermeneutical frameworks of their culture lacking appropriate terms and understandings for certain predicaments, leaving them silenced or marginalized. One striking example of this was the harm that especially women incurred before the concept of sexual harassment entered the vocabulary and collective consciousness of society. Since Fricker’s original account, her theory has increasingly been adapted and applied to epistemic injustices in health care. Philosophers such as Carel and Kidd have been instrumental in spearheading this movement. In several articles, they have argued how the ill are vulnerable to epistemic injustices and routinely risk stigmatization, marginalization, and mistreatment within healthcare contexts (Carel & Kidd, 2014; Kidd & Carel, 2017).

Epistemic injustices are equally – perhaps especially – pertinent for patients within psychiatry (Abdi & Michalis, 2015; Büter, 2019; Crichton et al., 2017; Kidd et al., 2022; Scrutton, 2017). Persons with mental disorders face great risk of being either misdiagnosed or underdiagnosed due to stereotypes of the mentally ill as violent or lacking willpower (Gosselin, 2018, p. 9) or of not being taken seriously by doctors because their complaints of physical symptoms are perceived as expressions of delusions (Büter, 2023), and so on. Exposing and ameliorating epistemic injustices of misdiagnosis and underdiagnosis are imperative to the lives and well-being of the mentally ill, but in this article, I wish to investigate the converse issue, namely that of overdiagnosis within psychiatry. This issue has received insufficient attention within the field of epistemic injustice in health care,Footnote 1 though I will argue that it presents an equally important challenge with serious repercussions for those that it affects. Similarly, the medical literature has traditionally framed the injustices of overdiagnosis mostly in terms of iatrogenic harms and issues of distributive justice rather than as an epistemic injustice. It is this omission that this article attempts to correct. Specifically, I will argue that the tendency towards overdiagnosis within psychiatry constitutes an instance of hermeneutical injustice, that is, of collective hermeneutical resources classifying cases of mental distress that do not benefit from treatment as mental disorder in a structural way to the detriment of the excessively diagnosed person. This I will show to be the consequence of both an epistemic framework that lacks the appropriate means of distinguishing between disorder and distress and of the widening scope of psychiatric classifications.

Whether or not psychiatry is especially prone to overdiagnosis is partly a matter of interpretation. Apologists argue that psychiatry is just now catching up to the true prevalence of mental disorder, while sceptics maintain that Western societies are undergoing a pervasive medicalization, that is, the “…process by which nonmedical problems become defined and treated as medical problems, usually in terms of illness and disorders” (Conrad, 2007, p. 4). The positions are, however, not mutually exclusive; the tendency to overdiagnose conditions that do not stand to benefit from treatment is entirely consistent with and could potentially arise due to increased awareness and detection of many previously unacknowledged disorders. Though inconclusive, certain figures at least give reason to think that the increased rates of diagnoses and treatments are partly the result of overdiagnosis. For example, Beeker et al. argue that western societies are undergoing a process of psychiatrization: worldwide, increasing amounts of psychotropic medication are prescribed, increasing amounts of people use in- or outpatient mental health services, and the prevalence of mental disorders is on the rise (2021, p. 2). Simultaneously, the prevalence of a disorder such as autism spectrum disorder has been relatively stable even though amounts of diagnoses have increased fivefold from 2004 to 2014 in Sweden (Rydell et al., 2018). A recent cohort study establishes that the lifetime incidence of either receiving a diagnosis of mental disorder or psychotropic prescriptions within as wealthy and healthy country as Denmark was 82,6% cumulatively (Kessing et al., 2023), making it much less statistically probable to not suffer from or be treated for a mental disorder during one’s lifetime. Despite increasing attention to and treatment of mental health issues, rates of mental distress seemingly only increase (Foulkes & Andrews, 2023). Foulkes and Andrews hypothesize that increased awareness might contribute to rather than alleviate the rise in mental problems as it leads to persistent self-monitoring of one’s emotional life and, accordingly, overinterpretation of normal variations in mood as somehow wrong or pathological.

As the topic is highly sensitive and divisive, I wish to preface the article by saying that I am neither denying the validity of mental disorders nor attempting to invalidate individuals who have received great help from being diagnosed. I fully recognize that the tendency to depathologize and “water down” mental disorders can have highly harmful effects and constitute instances of epistemic injustice in their own right (Spencer & Carel, 2021). However, the fact that the validity of many disorders has been – and still are – ignored or relativized should not make us blind towards the harms of overdiagnosis. The issue calls for a balanced approach that recognizes the validity of psychiatric disorders while simultaneously remaining critical of excessive diagnoses that will not benefit those who receive them.

The article is structured as follows. First and foremost, I briefly delineate how overdiagnosis is originally conceptualized within somatic medicine and what it means in a psychiatric context. Afterwards, I outline the differences between the debates concerning epistemic injustice of overmedicalization versus overdiagnosis. Next, I attempt to probabilize that the rise in psychiatric disorders can partly be attributed to the tendency of overdiagnosis, and, lastly, I highlight how these tendencies constitute instances of hermeneutic injustice that ought to be remedied.

2 The meaning of overdiagnosis in psychiatry

There is no broad consensus about the exact definition of overdiagnosis (Carter et al., 2015), however, several scholars define it as the diagnosis of abnormalities that, all things equal, would not lead to increased mortality, overt symptoms or reduced well-being (Brodersen et al., 2018; Hofmann, 2022). To give an example, screening for cancer is standard practice among the elderly where cancerous abnormalities are sometimes detected, which might never cause symptoms, because the variant is mild, will go into remission on its own, or the patient will die of other causes before the cancer evolves. After detection, however, health care systems are obliged to expend resources on examinations and treatments, while the awareness of being at risk for a more serious form of cancer might cause the patient unnecessary mental distress. There are two primary drivers of overdiagnosis, namely overdetection and overdefinition (Brodersen et al., 2018; Jønsson & Brodersen, 2022). Overdetection refers to the discovery of abnormalities that will not cause symptoms or increased mortality, while overdefinition refers to “(…) [firstly] lowering the threshold for a risk factor without evidence that doing so helps people feel better or live longer and [secondly] by expanding disease definitions to include patients with ambiguous or very mild symptoms” (Brodersen et al., 2018, p. 1).

As stated, there is disagreement about the precise nature and definition of overdiagnosis in somatic medicine, but the disagreements seem even larger within psychiatry where the term is used inconsistently and is often conflated with related concepts like misdiagnosis, false positives, medicalization, among others (Thombs et al., 2019). This is perhaps partly due to the differences between diagnostic practices in somatic medicine and psychiatry. Overdefinition is more straightforwardly comparable between somatic medicine and psychiatry as it entails the widening of disease definitions that is either of questionable utility or validity. Overdetection, however, differs somewhat. The key difference is that many overdetected cases within somatic medicine are asymptomatic as they often consist in early screening results of pathophysiological occurrences or risk factors, but as psychiatry operates with a symptom-based diagnostic system, asymptomatic cases are, definitionally, impossible in cases of true disorder. The key similarity between cases of overdetection from both domains is that they are correctly diagnosed according to the formal criteria but will not benefit the patient. Therefore, both overdetection and overdefinition, despite certain differences, characterize diagnostic practices in psychiatry and are elaborated on in this order below.

Following the diagnostic ethos established with the third edition of the “bible of psychiatry”, the Diagnostic and Statistical Manual of Mental Disorders (DSM) (Mayes & Horwitz, 2005), psychiatry operates with operationalistic, polythetic, and atheoretical diagnoses (Hyman, 2010, pp. 159–160; Tsou, 2015). With the exception of conditions such as post-traumatic stress disorder, brain trauma, substance abuse, among others, diagnosis of mental disorders does not require the establishment of an aetiology. Provided the patient exhibits a certain number of subjectively assessed and behaviourally observed symptoms of a certain disorder, they live up to the formal requirements for diagnosis. Consequently, establishing a cause is in most cases indifferent to the diagnosis, at least in principle. This is where the crux of the problem lies according to critics (Wakefield, 2010). Many conditions, which are non-pathological at core, exhibit similar symptoms as clinical disorders. A case of major depressive disorder might resemble a person undergoing a persistent bout of sadness due to, for instance, a difficult separation or an existential crisis. In such cases, the intuition of many is that such a condition constitutes a normal albeit distressful reaction to external circumstances that every person experiences at some points in their lives and therefore should not be treated like a medical issue, although it lives up to the formal criteria for a diagnosis of major depressive disorder. Overdetection therefore happens when conditions technically live up to the formal criteria for diagnosis, but symptoms might be transient, “normal”, or mild. The patient therefore does not stand to gain from diagnosis and treatment, as the condition might resolve itself given time or might be better treated through non-medical means. This is, for example, the case in certain instances of major depressive disorder where response and recovery rates are nearly as high among those who are not treated as those who are treated with medication and therapy (Cuijpers, 2018).

Likewise, overdefinition can be found within psychiatry as the boundaries of what counts as mental disorder are continually widened (Rose, 2006) with sparse evidence that it benefits patients. Throughout the years, the number of diagnoses contained within the DSM, for example, has increased dramatically from its first edition that listed 47 conditions to the fourth that included over 300 (Haslam, 2016, p. 7). Both tendencies involved in overdefinition of somatic conditions also hold for psychiatric conditions. Firstly, psychiatry continually “lowers the thresholds” whereby diagnostic criteria for certain disorders come to include new demographics, such as the inclusion of adults in the diagnosis of attention-deficit hyperactivity disorder (“ADHD”) (Conrad & Potter, 2000). Secondly, definitions of illness have been expanded to include phenomena hitherto considered normal parts of or reactions to life’s hardships. Examples of this tendency include how grief in severe cases can be considered pathological (Granek, 2010), how severe shyness can be diagnosed as social anxiety (Lane, 2006), sadness as major depressive or similar mood disorders (Horwitz & Wakefield, 2007), serious worries and fears as anxiety disorders (Horwitz & Wakefield, 2012), and problems of concentration and misbehaviour as ADHD (Searight & McLaren, 1998). Whether we consider this development right or wrong, the pertinent point is that such states of being in their more severe forms can now in certain cases be considered pathological, and therefore diagnosed and treated, rather than negative but non-pathological conditions of life as previously. Haslam argues that such processes are instances of concept creep where concepts of harm are expanded to include experiences previously thought of in non-pathological terms (Haslam, 2016), which is also known as diagnosis creep (Moynihan, 2016).

3 The medicalization versus overdiagnosis debate

Before turning to the epistemic frameworks that enable overdiagnosis, I wish to address an objection. Even if granted that overdiagnosis within psychiatry is an issue, it might be questioned why this tendency is so problematic. Diagnoses provide many positive things such as recognition of distress, explanations of “what is wrong”, and access to treatment. People have fought fiercely to have certain predicaments such as infertility medicalized (Becker & Nachtigall, 1992). As pointed out by several researchers (Degerman, 2023; Gagné-Julien, 2021; Reiheld, 2010; Wardrope, 2015), one should be cautious about making blanket statements about the problematic nature of medicalization. Denying certain predicaments medical and pathological status can entail great injustices, and diagnoses can become rallying flags that create communities, identities, and enables resistance against societal injustices and what Kidd has termed pathophobia (2019). It is therefore important to emphasize that medicalization and pathologization are neither harmful nor morally objectionable tendencies per se. Many mental disorders are undoubtedly genuine and therefore rightfully medicalized and pathologized.

While I am sympathetic to such lines of argument that nuance the traditional debate concerning overmedicalisation that can be highly polemical, I believe that it misses the mark if levelled at overdiagnosis. Although the debates between overdiagosis and medicalization in psychiatry are related, they are also distinct (Hofmann, 2016). The overdiagnosis debate stems directly from a medical discussion regarding excesive diagnoses that turn out to be unbeneficial to those who receive them. As stated, this can happen in psychiatry when symptoms are transitory, or because a certain state of being was misinterpreted as a disorder, and so on. The medicalization critique, however, often stems from outside the medical field and usually proffer external critiques of what they perceive to be an unwarranted expansion of medicine to various walks of life. Medicalisation may contribute to overdiagnosis, as it increases the amount of diagnoses and awareness of mental illness, and so on, but overdiagnosis as such is not dependent on (over)medicalization since it may occur even in healthcare systems that are not overly medicalized.

It is therefore not psychiatric diagnoses as such, that I wish to target with my analysis, but specifically overdiagnosis, which are instances of distress that live up to the formal criteria for diagnosis, but will not benefit from treatment. This is different to the researchers who show that medicalization is warranted and even beneficial, as these usually refer to clinically valid and relevant cases of disorder amenable to psychiatric intervention, where the benefits of diagnosis and treatment accordingly outweigh the harms. This will per definition not be the case with overdiagnosis. There is nothing to gain by treating overdiagnosed cases as they refer to excessively diagnosed cases unamenable to medical intervention. Therefore, all treatment of overdiagnosis is overtreatment. This similarly entails that harms are felt much more acutely for this subset of patients since the injustices are not offset by the benefits that treatment otherwise might entail. While the critics of the tradtional medicalization debate therefore raise insightful points that are salient to valid diagnoses, it seems unlikely that there is much to gain by medicalizing cases unamenable to medical intervention. In fact, medicalization of such cases only seems to constitute an epistemic harm that reduces the overdiagnoseds’ abilities to make sense of their situation and generates hermeneutical marginalization, which I argue below.

4 What causes overdiagnosis in psychiatry? Some tendencies

Diagnostic practices within psychiatry carry great risk of overdiagnosis for several reasons. In terms of overdetection, one significant challenge is the difficulty of making proper demarcations and distinctions between disorder and distress. The current diagnostic systems, DSM and the International Classification of Diseases (ICD), both operate with a system of discrete categories that represent distinct disease entities of mental disorders. However, there is a rising consensus that discrete diagnostic categories do not correspond with natural taxa (Hengartner & Lehmann, 2017; Hyman, 2010; Kendell & Jablensky, 2003). Instead, diagnoses are rather to be perceived as heuristic clusters that organize certain and often co-occurring symptoms, but the boundaries between disorders are highly fluid as the problem of comorbidity indicates (Kendler & Sullivan, 1998). Several researchers suggest that mental disorders are rather dimensional and located on a continuum between normality and disorder (Kotov et al., 2017). This has the unfortunate consequence of making demarcations difficult in practice, for example, in distinguishing between disordered and non-disordered behaviour (Bortolotti, 2020) or between intrinsic distress and social deviance (Aftab & Rashed, 2021), which is a formal requirement for diagnosis. However, the diagnostic system is dichotomous and binary, which spells trouble in mild and grey-zone cases. The exception to this rule is autism spectrum disorder that allows for different degrees of severity but even this diagnosis operates with a cut-off point above or below which a person can be considered to have autism or not. Diagnoses play an important gate-keeping function in so far as they demarcate those who are ill and in need of treatment from those who are not. When approached by a person experiencing distress, the health professional has a moral obligation to help and therefore an incentive to err on the side of diagnosis, which is a driver of overdetection.

A further issue that primes overdetection is the one briefly touched upon above, namely the lack of etiological explanations, which makes diagnoses reliant on behavioural observations and subjective assessments of symptoms. A diagnosis may therefore in certain cases be made without taking the context of the distress into account. As many “normal” albeit painful reactions to the adversities of life manifest themselves similarly to clinical disorders (Frances, 2009; Wakefield, 2010, 2015; Wakefield & First, 2003), distinguishing between disordered and non-disordered distress becomes inherently difficult in a symptom-based diagnostic system. Experiencing a traumatic event, for instance, may understandably make a person anxious, or a difficult separation may give rise to diminished mood, hopelessness, exhaustion, and so on. Behaviourally and experientially, these conditions can resemble generalised anxiety disorder and major depressive disorder, but are, at least according to some critics, either expected or even appropriate reactions to stressors.

Now, there is no reason to assume that merely because a condition entails an expected reaction, it is not pathological. Traumatic events and difficult periods of life often can contribute to or even cause mental disorders. It therefore seems, prima facie, that a distinction should be drawn between normal reactions understood as not clinically significant conditions of distress versus expected but nonetheless pathological reactions to external stressors. When a reaction is and is not appropriate or expected in given contexts is a discussion that unfortunately cannot be done justice in this paper. The main point is that the diagnostic system of psychiatry to a large degree carries a blind spot towards contextual dimensions as diagnosis can technically be made without ascertaining what the symptoms are symptoms of in most cases, which can blur the distinction between disorder and distress. This potentially leads to diagnosis of assorted conditions that fit the formal criteria but do not benefit from medical treatment. As Foulkes and Andrews hypothesize (2023), increased awareness can become especially problematic when “ordinary” suffering is viewed through a medicalizing lens since doctors are incentivized to diagnose – and patients to seek diagnoses – because diagnoses are a prerequisite for recognition of distress and professional help.

Several issues lead to the potential overdefinition of disorder. Despite the great aspirations of the “decade of the brain” in the 1990’s to tie mental disorders to some sort of biological substratum (Andreasen, 1997), many of these ties are often either unproven, unknown, or merely partially substantiated (Hyman, 2010) – regarding the validity of the DSM-5, former head of the National Institute for Mental Health, Thomas Insel, stated: “biology never read that book” (Belluck & Carey, 2013). Psychiatry has a long history of attempting to ground mental disorders in biology either to no avail or only limited success (Harrington, 2019). Seemingly, diagnoses do not reflect natural kinds (Hyman, 2021). Initiatives such as the Research Domain Criteria (Insel et al., 2010) seek to delineate the underlying pathophysiology of mental illnesses understood as brain disorders, but so far, such initiatives are tentative, research-focused, and have no bearing on clinical practices that still utilize the DSM and ICD.

There are, however, also strong currents within established psychiatry that is committed to the biopsychosocial rather than biomedical model (Tripathy et al., 2019). The biopsychosocial model was originally envisaged by Engel (1977) in response to the dominant biomedical models of psychiatry that were perceived by Engel to be needlessly reductive. Instead, psychiatry should adopt a multifactorial understanding of mental disorders capable of integrating biological, psychological, and social factors in its understanding. The commitment to this multidimensional and holistic model shows that established psychiatry is often more nuanced than critics tend to give it credit for (Bolton, 2023). But the biopsychosocial model has issues of its own that relate, for example, to difficulties in determining causal interactions between these three dimensions within given disorders, or that psychiatrists in clinical and research settings sometimes seemingly revert to a more biomedical conception despite formal avowals to the biopsychosocial model (Benning, 2015).

Neither model provide satisfying answers to the conceptual issues that drive overdefinition. Establishing the boundary between normality and disorder depends to a large degree on conventions, which makes them vulnerable to normative judgments. It is notoriously difficult to delineate when distress is clinically significant (Bortolotti, 2020) and when it is merely the expression of “unhealthy” social norms and external pressure on “non-conforming deviants” (Aftab & Rashed, 2021). Moreover, there is a strong risk of social contingencies affecting clinical judgment (Phelan & Link, 1999). The boundary between distress and disorder is therefore fuzzy, and the criteria for many disorders are often up for negotiation, which opens the risk that psychiatric classifications may verve into the territory of misinterpreting non-pathological distress that does not or only limitedly benefits from medical intervention as disorder.

Discussing the politics of health care and psychiatry lies outside the scope of this article, I will merely mention that powerful societal forces such as the pharmaceutical industry push for increasing classification and treatment of distress. There have been – and still are – close ties and financial interests between psychiatry and the pharmaceutical industry. The industry often seeks to influence the pathologization of disorders by, for example, supporting research conducive to its own aims or ingratiating themselves financially with psychiatrists involved in the development of the DSM (Cooper, 2014, pp. 14–15). Despite efforts to curtail such influences, new reports still find that 60% of the DSM-5 taskforce received payments of some sort from industry amounting to 14.2 million dollars (Davis et al., 2024), which creates pro-industry biases. All things considered, I believe it is highly plausible that the rise in disorders and subsequent treatment thereof is not merely a consequence of psychiatry catching up to the true prevalence of mental illness, but partly the consequence of overdiagnosis as well, which is a structural injustice leading to hermeneutical marginalization for those overdiagnosed.

5 Hermeneutical injustices of overdiagnosis

Fricker defines hermeneutical injustice as: “(…) a gap in collective interpretive resources [which] puts someone at an unfair disadvantage when it comes to making sense of their social experiences” (Fricker, 2007, p. 1). That is, collective epistemic frameworks putting someone at a disadvantage by distorting their ability to make sense of their predicament in a structural fashion. This is exactly what happens with overdiagnosis, I claim. The overdiagnosed are wronged by being classified, perceived, and treated as sick by themselves, healthcare, and society due to problematic diagnostic practices within psychiatry, though their distress might be non-pathological, or they do not stand to gain from medical treatment. Consequently, the overdiagnosed experience difficulties in being understood, making themselves understood, and understanding themselves, which is highly against their interest. They experience both cognitive disablement and hermeneutical marginalization (Fricker, 2007, 151–153). That is, they experience both hermeneutical injustices of the semantic and performative kind (Medina, 2017, pp. 45–46). The former refers specifically to lacunas or deficient understandings within collective epistemic frameworks; the latter to narrow or oppressive understandings of proper conduct and expressive styles that limit what can meaningfully be expressed within given contexts. Rather than a lacunae or gap in collective epistemic resources, hermeneutical injustice here signifies a certain epistemic framework becoming dominant to the marginalization or even virtual exclusion of other relevant interpretations and understandings. These deficient and insufficient understandings within collective epistemic resources can be as damaging and wrongful as the lack of proper concepts and understandings (Dotson, 2014).

The structural classification of cases of distress that do not benefit from medical intervention due to problematic epistemic frameworks is the primary harm of epistemic injustice of overdiagnosis. But, as Fricker argues, such primary epistemic injustices leads to secondary harms as well (2007, p. 162). Below, I outline both structural issues and individual consequences that follow from the hermeneutical injustice of overdiagnosis. I draw on several well-known critiques (Tekin, 2011) that, however, bear repeating and explication in this context because the overdiagnosed are more acutely affected by these issues due to the epistemic benefits of diagnosis not outweighing the harms, thereby constituting an epistemic injustice.

A common critique is that psychiatry individualizes a complex social issue. The systematic interpretation of distress as a fault within the patient’s biological or psychological constitution in need of correction is arguably also a problematic solution to genuine mental disorders (Conrad, 1975, p. 19; Cooper, 2014, p. 4) since social determinants and structural challenges have continually proven to be intricately connected to mental distress (Kirkbride et al., 2024), but it becomes especially problematic in cases of overdiagnosis. For instance, treatment within psychiatry often involves psychotropic solutions, many of which carry serious side effects and can create dependencies among many other harmful iatrogenic effects (Fava & Rafanelli, 2019). What is uniquely epistemically unjust about this is the fact that the overdiagnosed subject’s utterances and predicament are perceived through a pathologizing lens that accords individual factors higher relevance than social or environmental factors. This effectively limits the overdiagnoseds’ ability to make sense of an issue, whose cause might not be disordered, and whose solution might lie beyond psychiatric intervention. While social and individual interventions are not mutually exclusive, society and healthcare will often focus their efforts on the individual that is easier to change than larger structural tendencies.

The above tendency is intimately connected to another, namely that despite formal avowals towards more holistic models of medicine within certain branches, psychiatry in practice occasionally reverts to a more biomedical view of disorders, and the understanding of disorder as “diseases of the brain” is deeply ingrained within folk conceptions. Kidd and Carel have argued that not just moral and epistemic agents can be the carriers of vices and virtues, but abstracta like theories and conceptions can be considered either epistemically virtuous or vicious as well in so far as they enable and facilitate epistemic injustices (Kidd & Carel, 2018, 2019). In this case, naturalistic conceptions of health and illness, such as the biomedical view, influence how mental disorders are perceived and treated in healthcare.Footnote 2 Predominantly biomedical conceptions of distress fail to capture cases of overdiagnosis because these present as disorders while being of a different nature. When biomedical understandings are given primacy, hermeneutical injustices can occur. As Carel & Kidd emphasize, if disorders are solely interpreted as, e.g., chemical imbalances, which is often the case in folk percetions, patients’ experiences and perspectives can be conceived as less relevant to diagnosis and treatment, which might exclude them from taking a more participatory role in their treatment and the co-production of meaning with the health care professional (Spencer & Kidd, 2023). For the overdiagnosed, this is especially problematic because the distinction between disorder and distress requires close attention to factors that extend beyond biomedical facts. Moreover, the privileging of biomedical conceptions might effectively marginalize other pertinent interpretations that would help the overdiagnosed make sense of their situation (Brinkmann, 2014). This includes existential, phenomenological, social, and environmental interpretations, which could elucidate the origins of their distress and ways of coping with it.Footnote 3

Structural issues beget individual repercussions. Wrong classifications of mental illness can, for example, distort self-conceptions by making the overdiagnosed adapt to the sick role and excessively identify with or internalize the classified disorder. The mechanisms, through which these identity-distorting processes take place, can be interpreted as looping effects, originally coined by Hacking (1996, 2007). In this context, I utilize the concept to refer to instances of certain classifications affecting self-conceptions of the overdiagnosed, altering their behaviour in accordance with the classification, which in turn confirms the validity of the original classification, and so on, in feedback loops. To give a simplistic example: post diagnosis, the person might come to perceive affective responses to external stressors as, say, symptoms of depression rather than normal variations in mood. Consequently, they might withdraw from social contexts and activities, receive sick leave, self-isolate and wait for recovery, and so on. Many of these practices have a demonstrable negative impact on one’s well-being, which only further confirms the validity of the diagnosis in the first place, although it was potentially overdiagnosed. Examples from real life will naturally be much more intricate. The point is that classifications are not neutral. They impact how the classified person perceives themselves at a fundamental level and therefore their behaviour. When classifications are unhelpful as in overdiagnosis, they become hermeneutical offenses that not only obstructs and distorts self-perceptions but also the ability to make sense of and find meaning in certain experiences. Through looping effects, overdiagnosed conditions can therefore become self-fulfilling prophecies.

A classification may moreover alienate the overdiagnosed person towards their predicament and impede adequate understandings of their conditions. Though psychiatric diagnoses are syndromes, that is, clusters of symptoms with uncertain aetiologies, in dominant psychiatric understandings they often become reified as distinct disease entities (Hyman, 2010, p. 156). To take an example, within biomedically inclined understandings, psychiatric conditions may be conceived as chronic disorders of the brain with a certain “biological destiny”, a natural course which the patient is passively subjected to and powerless to resist. For instance, folk conceptions of depression as a chemical imbalance shapes: “(…) how people understand their moods, leading to a pessimistic outlook on the outcome of depression and negative expectancies about the possibility of self-regulation of mood” (Moncrieff et al., 2023, p. 11), even though there is no compelling evidence for the chemical-imbalance hypothesis of depression currently. Diagnoses are more than mere labels – they designate “what is wrong” and may thereby override existing narratives and structures of meaning in an individual’s life, which can constitute instances of hermeneutical injustice (Hassall, 2024; Tekin, 2011). Even in cases of mental illness that benefit from treatment, such belief states may lead to reduced feelings of agency, self-efficacy, and hopes of recovery as the diagnosed individuals believes themselves at the mercy of their disorder.

Such reification becomes even more problematic in cases of overdiagnosis for several reasons. For example, what might constitute isolated or episodic incidents of mental distress in an individual’s life becomes construed as chronic conditions requiring life-long treatment. Further, disease entity conceptions risk reifying mental distress as being entirely outside the individual’s control. Both can have alienating and disempowering effects on the epistemic subject. Considering mental disorders to be “just like” diabetes or other somatic diseases is undoubtedly an attempt to destigmatize certain conditions and discourage people from moralizing issues of mental health, though it seemingly has the opposite effects (Malla et al., 2015).

However, there are good reasons to not consider mental disorders to be “just like diabetes” (Arpaly, 2005). These reasons pertain not only the issues in adequately determining the nature and causes of mental disorders, but also to the reduction of mental distress to mere pathological mechanisms devoid of intentionality and content. As Arpaly argues, there is an important distinction between the tics of Tourette’s syndrome and the anxiety caused by, say, fear of abandonment. In the former case, the content of the tics is meaning- and contentless because it is the involuntary consequence of a pathological neurological mechanism, in the latter case, the fear of abandonment is different because the content of the mental state is causally efficacious in bringing about the distress; the content is not accidental but meaningful. When distress is reified as “nothing more” than a disease entity with an inevitable fatum, it is deprived of meaning, even though the content of the mental belief is important to determine what is at stake for the overdiagnosed person: is there a good reason for them to be feeling a certain way, is it an adverse response to events or contexts. Not only does this hermeneutical framework misinterpret what is at stake for the overdiagnosed, it also restricts the playing field of what they can meaningfully do to (re)interpret or alter their predicament. It diminishes their epistemic agency.

The hermeneutic injustice of being unfairly classified as sick due to problematic diagnostic practices not only impacts the self-conceptions and identities of the overdiagnosed individual but also the ways that society perceives, and consequently interacts with, said individual. Fricker describes in passing the phenomenon of people adopting the behaviour expected of them according to certain unjust stereotypes (Fricker, 2007, pp. 55–57). Within the field of educational science, this is known as stereotype threat and has been demonstrated to happen when, e.g., testing the intelligence of African-Americans in school; when a test was explicitly described as a test of intelligence, it demonstrably lowered the performance from African-Americans students, but when the test was not flagged as such, they performed better, while there was no difference for Caucasian students in either cases (Steele & Aronson, 1995). The sociology of healthcare has developed an equivalent theory, namely the labelling theory, which “(…) postulated a process in which official labelling through treatment contact and the stigma that accompanies such labelling jeopardize the life circumstances of people with mental illnesses by harming their employment chances, social networks, and self-esteem” (Link & Phelan, 2013, p. 527). To be clear, the mere ascription of a label does not automatically generate behaviour that fits certain stereotypes. Rather, it is the stigmatizing effects of societal institutions and social groupings holding these labels to be true and treating the labelled accordingly, which has the power to alter the behaviour of the labelled to fit what is expected of them, because they internalize problematic norms, censor and isolate themselves, and become targets of discrimination (Link & Phelan, 1999).

Stigmatization and marginalization are huge issues of mental illness and diagnoses of mental disorder, which range from issues in forming and keeping close relationships, differential treatment on the job and housing market, in healthcare, as well as distorted self-perceptions and consequent self-stigmatization (Thornicroft, 2006). There is no reason to assume that the overdiagnosed are immune to such detrimental effects of being labelled. If granted, then labelling effects of the overdiagnosed are an immense source of hermeneutical marginalization, defined by Fricker as: “(…) unequal hermeneutical participation with respect to some significant area(s) of social experiences” (Fricker, 2007, p. 153). Hermeneutical frameworks that apply labels of disorders too liberally can therefore invariably prevent the overdiagnosed from participating fully in society through the structural misinterpretation of “what is wrong with them”. One pertinent example are the negative stereotypes surrounding children with ADHD, which are stereotyped as lazy, less intelligent and face social exclusion and stigma from both parents and teachers (Thomas et al., 2013, p. 3), which is especially problematic given the likelihood that a certain degree of autism spectrum disorders are overdiagnosed (Paris et al., 2015). Negative labelling effects are wrong in any case, although perhaps a necessary evil, but even more harmful in the case of overdiagnosis because the overdiagnosed are much less likely to receive benefit from being classified as mentally ill.

The final critique posited in this context consists in the epistemic injustices that overdiagnosis can generate in interactions with healthcare. A diagnosis of any sort subjects the individual to a process of patientification (Jønsson & Brodersen, 2022, pp. 101–102). Patientification refers both to the adoption of the sick role and the consequent changes to a person’s self-conceptions described above, or to how a person becomes enrolled in the health care system, i.e., how an individual is turned into a health care user with all that this entails. There is nothing inherently wrong with patientification, but in the case of overdiagnosed mental disorders, it can become problematic. It exposes the overdiagnosed individual to potential iatrogenic harms such as mistakes in treatments or examinations, side-effects of medication, and so on, which I will not go into further detail with here.

More importantly for present purposes is the risk of epistemic wrongs that patientification exposes the individual to in interactions with healthcare. Though testimonial and hermeneutical injustice are conceptually distinct, they are also mutually causally efficacious and can compound (Fricker, 2007, pp. 159–160). After overdiagnosis, where dominant epistemic frameworks have already contributed to obscuring “what is wrong” with the individual, they become enrolled as patients in healthcare systems where the mentally ill are much more likely to be attributed with credibility deficits. For instance, the content and form of their testimonies are less likely to be taken seriously, they are much more likely to be stereotyped as irrational, annoying, excessively emotional, dangerous, and so on. Hermeneutic and testimonial injustices are therefore likely to reinforce one another and aggravate the hermeneutical marginalization of the overdiagnosed. Among the overdiagnosed, some will have the capacity to resist such injustices by reframing or reinterpreting their distress and the causes thereof or finding other ways of coping, others will become extended or even life-long users within psychiatry due to problematically epistemic frameworks.

6 Diagnoses and social perceptions

A lingering objection needs to be adressed here, namely that it would be an unfair generalization to say that all people who are overdiagnosed are epistemically harmed to the same degree. If the degree to which overdiagnosis reduces one’s epistemic agency and causes epistemic harm is contingent upon conceptualizations of diagnoses and how these are interpreted by the diagnosed individual and their peers, then it stands to reason that changing these perceptions would also alleviate epistemic harms. The neurodiversity movement, for example, attempts to separate diagnoses from experienced suffering and show that these are not inherently problematic to the individual but might even be the source of greater self-understanding (Chapman, 2023). This also tracks with certain proposals within the literature on overdiagnosis that suggests we should separate diagnoses from experienced suffering to ameliorate the harms that excessive diagnoses may cause (Hofmann, 2019). Perhaps, it could be argued, the solution to epistemic harms of overdiagnosis lies more in adressing social perceptions of diagnoses rather than fundamental changes to diagnostic practices.

This is a valid critique, which should make us cautious of sweeping generalizations concerning the epistemic harms of overdiagnosis. However, I still maintain that overdiagnosis more often than not does more harm than good. While movements such as the neurodiversity movement are growing in size, and public perceptions seem to be gradually changing, it is yet a minority movement. Mainstream psychiatry still operates with an understanding of diagnoses wherein experienced harm or impairment is a formal criterion for disorders. The dominant conceptualization of disorder is therefore as something inherently problematic or negative. An individual might to a certain degree be able to resist internalizing such negative conceptualizations of their conditions, but they would still be vulnerable to the misunderstandings of peers and health care systems. For these reasons, it seems unlikely that excessive diagnoses would leave the overdiagnosed person completely epistemically unscathed.

Moreover, conceiving diagnoses as mere differences in mental functioning could conceivably lead to the converse issue of people construing their genuine mental disorders as harmless variations, thereby forgoing help.Footnote 4 This might in itself constitute another form of epistemic injustice leading to a similar lack of ability to make sense of one’s situation and be understood by others, including healthcare services. Notably, the same epistemic harms described above in relation to overdiagnosis might apply here, such as reduced epistemic agency and self-efficacy. Prima facie, decoupling disorder and distress therefore seems to be insufficient in itself to alleviate the harms of overdiagnosis. Rather, the problem calls for a multi-pronged approach that not only adresses social perceptions of disorders but also reckons with diagnostic practices that enable and proliferate overdiagnoses in psychiatry. However, the important discussion of how to alleviate the (epistemic) harms of overdiagnosis unfortunately lies outside the scope of this article.

7 Conclusion

The amelioration of hermeneutic injustices of overdiagnosis is a complex issue. Calls for revision or reform of current diagnostic practices are numerous (Paris, 2015; Watson, 2019; Whitaker, 2010), but any alterations need to be sensitive to the wider social, political, and ethical role that diagnoses play in contemporary societies. Changes in medical and psychiatric practice must be accompanied by equivalent societal changes with special attention to the social progenitors of mental distress. Nevertheless, hermeneutic injustices of overdiagnosis demonstrate the need for less diagnose-centric healthcare systems where help and recognition does not hinge on the ascription of labels. As such changes are likely to be even more resource-intensive, it is important that they are accompanied by a practice of psychiatry that is more watchful and less interventionistic than current systems. This, in turn, requires a reckoning with the tacit assumption that only diagnoses can legitimate distress. Until such revisions or reforms materialize, healthcare and research must be more mindful of the risk of epistemic injustices that overdiagnosis can entail.