A Verdict Void of Standard: The Instability of Perceived Normalcy in Modern Psychiatric Diagnosis

Introduction

In 1973, a crossroads was reached. 58% of the American Psychiatric Association (APA) voted to confirm the removal of homosexuality from the Diagnostic and Statistical Manual of Mental Disorders (DSM), prompting the process of its complete removal in 1987 (Drescher, 2015). This change did not occur due to newfound neuroimaging, a reversal of findings, or pristine neuroscience; the change was in the cultural definition of “normal”. This moment established a truth in psychiatry’s core — diagnosis has never possessed a stable definition of normal to draw deviations from (Conrad & Schneider, 1992). Even the definition of a mental disorder itself, “a significant disturbance in an individual’s cognition causing significant distress” (APA, 2013), is applied through cultural filters over time; the question of modern over-diagnosis or better recognition is the wrong question to ask at first. The right addition to the question is: to what baseline?

This essay will argue that diagnostic, cognitive, and sociocultural factors each contribute toward better recognition of previously dismissed disorders, with nuances of over-diagnosis due to a constantly fluctuating foundation in psychiatry that outruns the evidence that justifies it.

Changes in the Classification of Mental Illness

The core of any evaluation of diagnostic rates is the criteria by which a diagnosis is made. In Western psychiatry, the DSM, published by the American Psychiatric Association (APA), is the primary instrument of classification. The transition from DSM-IV to DSM-5 in 2013 marked a significant transformation in psychiatry: the shift from aetiological assessment to a predominantly symptom-based measure (APA, 2013). APA argued that this improved consistency and clinical accessibility in providing diagnoses, leading to assessments being more attainable for previously outlier populations, eliciting better recognition. While the newfound reliability in symptom-based identification prevails, its validity can be questioned: could a diagnosis accurately reflect a disorder, rather than simply classifying one consistently? This criticism raises concerns for modern over-diagnosis and has proven institutional effects; the National Institute of Mental Health (NIMH) declared a redirection of funding away from DSM-based research due to a lack of validity in genetic, imaging, and cognitive data (Insel, 2013). An example of the reasoning behind this is the consideration of bereavement within major depressive episodes. The DSM-IV included the bereavement exclusion for diagnosing depression, whilst the DSM-5 removed this exclusion entirely — one of the most contested changes to the manual’s history (Pies, 2014). Thus, a change in the baseline perspective of normality may prompt overclassification as the manual changes its views on circumstances influencing disorders, allowing symptoms following bereavement to be classified as depression.

While APA argued that bereavement does not immunise a patient against genuine depression, the restructuring of diagnosing Attention-Deficit Hyperactivity Disorder (ADHD) provides a contrasting standpoint. The DSM-IV required six out of nine symptoms per domain, with onset before age seven for an adequate diagnosis, while the DSM-5 reduced the symptom threshold for adults from six to five and extended the onset criterion from age seven to twelve (APA, 2013). This led to a 27% increase in expected ADHD prevalence among young adults, compared to the DSM-IV (Matte et al., 2015), representing a narrowing of what we perceive as a “normal brain” in today’s fast-moving society. Furthermore, a study of 445 students concluded a 65% increase in the number of those meeting the new cutoff point compared to the old DSM-IV cutoff; thus, explicit caution is placed that the new criteria should be applied with a broader clinical context to prevent over-diagnosis (Rigler et al., 2016). While it can be argued that the general loosening of criteria in the modern DSM has captured underdiagnosed adults, it may subsequently increase the risk of over-diagnosis due to the dismissal of contextual normalcy, providing a paradoxical argument of increased reliability accompanied by decreased validity. This diagnostic criteria expansion has thus increased the risk of over-diagnosis.

Therefore, the updated DSM-5 is a tool that suggests neither over-diagnosis nor better recognition, but instead re-evaluates the boundaries of both. Ultimately, the outcome that applies to the modern-day cannot be perceived from the manual itself, but from the cognitive and sociocultural aspects that shape the “verdicts” imposed upon the population.

Psychological and Cognitive Factors in Diagnosis

Neurobiology follows in a similar vein to the DSM-5 reclassifications, with technological advances paving the way for better recognition of disorders; these advances were not available during earlier decades.

In major depressive disorder (MDD), positron emission tomography (PET) scans found abnormally reduced blood flow and glucose metabolism in the subgenual prefrontal cortex (sgPFC) in patients with familial unipolar or bipolar depression, in addition to a 48% reduction in mean grey matter volume in unipolar depressive subjects, an area associated with cognitive decline and emotional instability (Drevets, 1997). These physiological changes differ from those without the disorder — the sgPFC is responsible for processing emotions, mood, and the body’s autonomic stress responses. Reduced activity in the sgPFC causes behavioural symptoms to appear in the individual, prompting an adequate symptom-based diagnosis, such as the DSM-5. Thus, it can be suggested that neurobiological changes in brain structure can translate to observable symptoms, which leads to overall better recognition as criticisms of symptom-based diagnoses are contradicted by empirical, replicable evidence produced by neuroimaging. Consequently, later research suggested the sgPFC’s activity levels are bidirectionally implicated in MDD (Mayberg et al., 1999), contrary to earlier findings of decreased activity. This led to the following treatment: deep brain stimulation for treatment-resistant depression. Findings reported that electrical stimulation of subgenual cingulate white matter adjacent to the sgPFC can effectively reverse symptoms of otherwise treatment-resistant depression, with four out of six patients reporting symptom remissions (Mayberg et al., 2005). The localisation of brain functions (Phillips, Zeki & Barlow, 1984) and subsequent treatment to localised areas, related to symptomatic presentations, suggests a causal relationship between differences in neurobiology and disorders, promoting improved recognition of disorders through nomothetic, objective, and scientific means.

Conversely, it is vital to recognise the limitations of neuroscientific research as access to electronic laboratory results is often unavailable to the majority of psychiatric clinicians and clients, ranging from 6% to 14% of psychiatrists being able to access electronic results produced from laboratory testing (Skelton et al., 2013). Hence, diagnosing patients comes with the responsibility of judgment; however, due to the subjective nature of individuals, there is likely to be a degree of uncertainty present in each psychiatric case (Strauss, 2017).

Anchoring — the overweighting of initial information in subsequent judgment (Tversky & Kahneman, 1974) — arguably plays a crucial role in day-to-day diagnosis. Often, a single symptom, a reason to visit, can establish a centre of gravity around subsequent diagnostic processing employed by the psychiatrist. Additionally, confirmation bias — the selective seeking, weighting, and recall of evidence aligned with a previously existing hypothesis (Nickerson, 1998) — can manifest as leading questions probing symptoms aligned with the working diagnosis, whilst dismissing alternatives. This dynamic has proven to be bidirectional: studies demonstrated that psychiatrists exposed to tentative, preliminary diagnoses subsequently searched for confirmatory information rather than disconfirmatory information at significantly higher rates (Mendel et al., 2011). Furthermore, anchoring and confirmation bias are identified as the most consistently documented contributors to diagnostic error across medical specialties (Saposnik et al., 2016), although more research is suggested. Therefore, it is appropriate to consider the role of cognitive bias in decreasing recognition of disorders. Instead of recognising what the diagnostic system previously misses, misdiagnosis due to cognitive factors consistently causes double distortions; in conditions with high public salience — depression, anxiety, ADHD (Freckelton, 2018) — the anchor lands disproportionately compared to other possible conditions, stimulating over-diagnosis. On the other hand, in less culturally-legible presentations — male depression, adult autism, bipolar disorder often mistaken as a unipolar depressive episode (Swetlitz, 2021; O’Nions, 2023; Hirschfeld, 2001) — the same mechanism produces under-recognition or an entire misdiagnosis. The diagnostic gap left by the culturally unstable criteria is filled by individual clinicians’ intuition, shown as susceptible to biases and cultural visibility rather than reality. Yet, this distortion depends on criteria that are now more empirically anchored than before. Categories now draw on contemporary neurobiological evidence, which psychiatry once lacked, promoting more genuine pathology as a whole.

Sociocultural Changes Impacting Diagnosis

It can be argued that diagnostic categories are not descriptions of pre-existing disorders. Instead, they are human terminology — categories that, when applied, feed into the experience of being classified as such. Once a category begins to culturally circulate, people shape their distress around it, leading to subsequent internalised presentation in order to conform to said category. This “looping” effect (Hacking, 1995) depicts how increasing diagnostic rates are produced from the wide circulation of a classification, generating ill populations as opposed to counting them.

The Hong Kong anorexia nervosa (AN) trajectory provides an exemplative case to support the theory. Longitudinal studies before the 1990s displayed a culturally distinct illustration of AN in Hong Kong — patients often reporting complaints such as bloating and loss of appetite, contrary to fat-phobic cognitions and body-image distortion core to the DSM criteria (Lee, 1991; Lee, Ho & Hsu, 1993). Suffering persisted physiologically, but did not conform to the traditional Western template for AN. A turning point was reached in November 1994 as fourteen-year-old Charlene Hsu Chi-Ying collapsed and died of AN on a Hong Kong street. In the years following this incident, symptomatology shifted: fat-phobia became central while somatic complaints decreased and the prevalence of AN increased. It could be argued that collectivist conceptions of AN had changed following Chi-Ying’s death, diagnostic criteria undergoing rapid westernisation (Watters, 2010) in conjunction with Hong Kong’s 1994 media cycle. This increased symptom prevalence could reflect broader social change, including the westernisation of disorders. Nevertheless, the outcome of the national paradigm shift (Kuhn, 1962) demonstrated better overall recognition as the clinical focus for AN shifted from overgeneral complaints to cognitive thinking and distortions.

This looping theory is seen in modern media as well, with effects at heightened speeds. Algorithmic platforms amplified cultural transmissions of diagnostic categories. TikTok’s #ADHD hashtag has accumulated over four billion views by 2022 (Yeung, Ng & Abi-Jaoude, 2022). An analysis of its contents revealed that 52% of ADHD-related videos contained misleading information surrounding the disorder’s presentation, diagnosis, and/or treatment, while 27% were based solely on personal experience. Similar inaccuracy could be found in autism content, often demonstrating personality variation and expected social discomfort as diagnostic symptomatology; statistics including 41% inaccuracy and 32% overgeneralisation across 133 analysed media (Aragon-Guevara et al., 2023). Likewise, the “AuDHD” compound identity (comorbidity of ASD and ADHD) has emerged, contributing to social formations, shared vocabulary, mutual recognition “rituals”, and self-diagnostic checklists without clinical input. In NHS clinical services, a corresponding rise in adolescent self-diagnosis is found, exacerbating waiting times (Gilmore et al., 2022). Thus, the looping effect becomes involved in a wide online network, with patients shifting from category-to-subject looping to a wider network loop: diagnostic identity gives way to social capital, which drives further identification and encourages clustering individuals into communities idolising a shared symptom template. This counters the Hong Kong case study’s thesis for better recognition, as algorithms push categories to populations whose underlying symptomatology may not align with them. Due to rapid algorithmic speed and social incentive in modern times, it can be argued that the looping effect can contemporarily produce over-diagnosis.

Conclusion

The 1987 decision to remove homosexuality from the DSM was not a newfound discovery: rather, it was a re-evaluation of the boundaries between disorder and normality; this boundary continues to shift across every dimension within this essay. The recalibration of the DSM, the cognitive biases affecting clinicians, and the cultural implications of psychiatry all indicate one underlying instability: psychiatry has never possessed a stable definition of “normal”, against which better recognition or over-diagnosis could be measured. By examining how diagnostic, cognitive, and sociocultural factors interlock, it can still be argued that the overall direction of modern psychiatry is one of better recognition, with pockets of over-diagnosis emerging due to an expanding framework. However, until psychiatry can ground its concept in a stable definition of normality, as opposed to institutional and cultural consensus, the verdict on whether mental illness is over-diagnosed or better recognised remains tentative.

References