The Overdiagnosis of Bipolar Disorder Within Marginalized Communities: A Call to Action

This paper argues that the overdiagnosis of bipolar disorder (BD) is an urgent and underrecognized problem within the U.S., threatening to expose vulnerable Americans to heightened stigma and harmful drug effects while disguising the environmental and traumagenic roots of their distress. The paper traces BD overdiagnosis to biomedical assumptions about mental illness and to the decline of social welfare policies over the past twenty-five years. It calls on policymakers to address BD overdiagnosis by revising criteria in the DSM 5, developing psychosocial models of mental illness, and reintroducing protective social welfare programs. Finally, the paper urges social workers to educate themselves about the harms of BD overdiagnosis as well as to recognize their own role in medicalizing their clients’ distress.

This paper argues that the overdiagnosis of bipolar disorder (BD) is an urgent and underrecognized problem within the U.S., threatening to expose vulnerable Americans to heightened stigma and harmful drug effects while disguising the environmental and traumagenic roots of their distress. The paper traces BD overdiagnosis to biomedical assumptions about mental illness and to the decline of social welfare policies over the past twenty-five years. It calls on policymakers to address BD overdiagnosis by revising criteria in the DSM 5, developing psychosocial models of mental illness, and reintroducing protective social welfare programs. Finally, the paper urges social workers to educate themselves about the harms of BD overdiagnosis as well as to recognize their own role in medicalizing their clients' distress.

THE OVERDIAGNOSIS OF BIPOLAR DISORDER WITHIN MARGINALIZED COMMUNITIES: A CALL TO ACTION
Researchers have been sounding the alarm about the overdiagnosis of bipolar disorder (BD) for more than a decade (Ghouse et al., 2013). For example, in 2008, Goldberg et al. found that only 33% of a cohort diagnosed with BD actually met clinical conditions, and in 2010, Ruggero et al. discovered that 60% of patients studied had been misdiagnosed as having BD. A meta-analysis by Mitchell (2012) reported rampant overdiagnosis of BD driven by flawed screening tools and studies inflating the prevalence of BD. One year later, another literature review found BD misdiagnoses reaching rates of 67% (Ghouse et al., 2013).
The scope of the problem is vast, with some researchers pressing for even greater rates of diagnosis and setting BD prevalence as high as 27% (Mitchell, 2012). While anyone can be subject to misdiagnosis, those who suffer from other mental health conditions, such as major depressive disorder (MDD) and substance use disorder (SUD), are especially at risk (Mitchell, 2012). BD diagnoses are also disproportionately applied to people with limited social support, low socioeconomic status, and a history of traumatic experiences (Fusar-Poli et al., 2017). Given that BD is thought to be a lifelong affliction, the consequences of overdiagnosis extend far into each individual's future.
Some of the most serious consequences of misdiagnosing BD are pharmacological. Lithium, a prescription mood stabilizer medication and first-line treatment for BD, has been shown to increase risks of hypothyroidism and chronic kidney disease (Littrell, 2012;Livingstone & Rampes, 2006;Presne et al., 2003). A 2012 systematic review discovered renal, cardio-vascular, and neurotoxicity in older adults using lithium (Sun, Hermann & Shulman, 2017). Huxley & Baldessarini (2007) report that lithium's neurotoxic effects can lead to greater impairment and disability among those diagnosed with BD, such as memory loss and decreased executive functioning. Severe risks associated with antipsychotics, another common treatment for BD, include cardiovascular disease, osteoporosis, and significant reductions in brain tissue over time (Voineskos et al., 2020;Littrell, 2012;Dorph-Petersen et al., 2005).
Another harmful consequence of BD overdiagnosis is the risk of exposure to social and internalized stigma, which can lead to social exclusion, occupational failure, and reduced functioning (Hawke et al., 2013). Researchers suggest that the stigma imparted by BD diagnosis, similar to that of schizophrenia, may reflect the condition's longevity and genetic roots (Hawke et al., 2013). Unfortunately, the harms of BD stigma appear to be highest for those most at risk of misdiagnosis: those who lack support systems and carry stigma brought on by other diagnoses (Fusar-Poli et al., 2017).
Perhaps the most insidious result of inflating BD diagnoses, and the one which may be of most concern to social workers, is the masking of oppressive social conditions. Like all disorders involving psychosis, BD and its symptoms have been linked to childhood abuse, perceived stress, and poor physical health, all of which disproportionately affect poor communities and communities of color (Fusar-Poli et al., 2017). Environmental stressors including poverty, crime, and racial discrimination are correlated with depressive symptoms, psychosis, and the dysregulation of the limbic system; research suggests that they may also be erroneously attributed to BD (Cogan et al., 2020;Gómez, 2015;Jackson et al., 2010). As a result, there are serious concerns that the rise in BD diagnoses, with their biological underpinnings, is concealing increasing social and environmental distress among marginalized Americans.

POLICY RESPONSE
According to critics, the biggest drivers of BD overdiagnosis are its diagnostic criteria and estimates of prevalence, both of which have shifted dramatically over the past three decades (Ghouse et al., 2013;Mitchell, 2012;Littrell, 2012;Burrows, 2010). In 1994, the DSM IV introduced bipolar II by expanding mania, the central feature of bipolar I disorder, into the less severe category of hypomania, despite mania's PAUL DOYEN THE OVERDIAGNOSIS OF BIPOLAR DISORDER WITHIN MARGINALIZED COMMUNITIES high concordance and limited evidence for its expansion (Yutzy et al., 2013;Singerman et al., 1981). Following the introduction of bipolar II, a small group of researchers developed the idea of "subthreshold" or "subclinical" BD, which further expanded criteria and prevalence estimates (Mitchell, 2012). Between 1994 and 2007, BD diagnoses doubled in adults and multiplied by 40 in children, while estimates of BD's prevalence rose from 1.6% to 24.2% of the population (Burrows, 2010;National Institute of Mental Health, 2007).
Critics regard the DSM IV's changes to BD's nosology as arbitrary and unsupported by evidence (Yutzy et al., 2013;Mitchell, 2012). They reject that surges in BD diagnoses capture an increasing incidence of the disorder, which, given BD's biogenetic roots, should remain stable over time (Almeida et al., 2020). Instead, these researchers argue that BD's expanded diagnostic criteria and inflated prevalence estimates have led to an epidemic of overdiagnosis (Ghouse et al., 2013;Mitchell, 2012).
Despite these criticisms and growing evidence of overdiagnosis, researchers and policymakers have done little to address the problem. In 2013, the newly released DSM 5 rebuffed warnings of overdiagnosis by expanding subsyndromal criteria for BD, a change that may significantly increase the risk of depressive patients being misdiagnosed with BD (Carta & Angst, 2016;Cerimele et al., 2014). Other issues related to BD's overdiagnosis, such as diagnostic overlap with schizophrenia, schizoaffective and borderline personality disorders, comorbidity with anxiety and unipolar depression, and reliance on overly sensitive screening instruments were not addressed within the DSM 5, even though they were well-documented at the time of its publication (Cosgrove & Suppes, 2013;Mitchell, 2012;Vieta & Philips, 2007).
Of the factors contributing to BD overdiagnosis, one of the most serious is the DSM 5's failure to clarify the relationship between BD and post-traumatic stress disorder (PTSD). The co-incidence of PTSD in BD patients has been cited at 35%; the two disorders share multiple symptoms and affect similar regions of the brain (Carmassi et al., 2020;McCormack & Thompson, 2017;Rakofsky et al., 2011). Despite calls from some researchers, the need to screen and control for trauma disorders in potential BD patients has been ignored, leaving marginalized communities, who experience disproportionate rates of trauma and decreased access to trauma-informed care, at risk of misdiagnosis (Etaine et al., 2008). Evidence suggests that misdiagnoses of BD leave trauma survivors less likely to receive appropriate care, with BD patients on Medicaid being offered less therapy and more unsupervised drug treatments than other patients (Busch et al., 2007;Fontanella et al., 2015).

UNDERLYING ASSUMPTIONS
Diagnostic psychology has been slow to recognize the problem of BD overdiagnosis in part due to its underlying assumptions about individuals with BD. One such assumption is that BD and its symptoms are biogenetic in origin. This claim is supported by family studies of bipolar I, but has not been substantiated for the diagnostic labels critics believe are driving overdiagnosis: bipolar II and subclinical BD (or "bipolar not otherwise specified") (Almeida et al., 2020;Mitchell, 2012). In fact, inconclusive findings have led some researchers to question whether bipolar I and bipolar II are genetically related conditions (Littrell, 2012). Despite this lack of evidence and a wealth of literature identifying BD symptoms with over 51 unique environmental stressors, diagnostic psychology continues to regard bipolar II and sub-clinical BD as biogenetic disorders (Bortolato et al., 2017).
Another assumption obscuring BD overdiagnosis is that people diagnosed with BD have little hope of recovery. Once again, evidence of poor prognosis is significant for bipolar I but is less clear for bipolar II and subclinical BD, both of which have garnered less research (Almeida et al., 2020). One historically important variable in determining BD's prognosis is unemployment, which involves social factors as much as it reflects individual impairment. For instance, data showing that around 57-65% of BD patients were unemployed in 2007 compared to only 15% in the 1970s is difficult to explain without recourse to other factors, such as deindustrialization and declining job security (Huxley & Baldessarini, 2007). Indeed, opponents of overdiagnosis argue that social distress due to rising work insecurity has been systematically mislabeled as mental illness over the past four decades, disguising the problem and worsening outcomes (Buffel et al., 2017;Wong, 2016).
A third assumption about those diagnosed with BD is that their condition is universal and "colorblind." In light of this assumption, some researchers have taken the low prevalence of BD among Black Americans as a sign of underdiagnosis, dismissing findings that the prevalence of mood disorders is consistently lower in Black communities than in white communities (Alvarez et al., 2018;Boyd et al., 2011;Breslau et al., 2008;Neighbors et al., 2003;Woodward et al., 2011). Calls for increased screening among Black Americans carry a notable threat of overdiagnosis, given that actual prevalence appears to be low, and the depressive and trauma-related symptoms associated with anti-Black oppression are frequently mislabelled as mental illness (Jarvis, 2007;Stevenson et al., 1997). Meanwhile, the DSM's colorblind approach to mental health treatment, which has failed to account for how psychological diagnoses impact racial groups differently, means that the risks of BD overdiagnosis within communities of color have not been monitored by mental health policymakers (Green et al., 2012).
While these assumptions have all likely contributed to the failure to address BD overdiagnosis, diagnostic psychology's neglect of social and environmental problems is rooted in its assumption that mental illness represents discrete, biologically based diseases, a claim unsupported by evidence despite over 40 years of influence (Jacob et al., 2014;Timimi, 2014). Critics of the "bio-medicalization" of mental health, stemming from the publication of the DSM III in 1980, complain that it has led to "tunnel vision" about mental illness, even as evidence has increasingly characterized mental illness not as distinct clusters of biogenetic symptoms, but as transdiagnostic distress linked to trauma, deprivation, and social inequality (Pilgrim, 2014;Rimke, 2016;Timimi, 2014).
One explanation for the persistence of the biomedical model of mental illness, despite a lack of evidence to support it, is its shared assumptions with neoliberalism, which holds that individuals are responsible for their own health, have equal agency, and resolve their needs and problems through the consumption of private goods (Rimke, 2016). Detractors argue that neoliberal assumptions about mental distress have pathologized poor and minority groups, who have limited agency to address their problems, and have helped medicalize social welfare, with government support becoming increasingly contingent on diagnoses of physical and mental disability (Wong, 2016). From this perspective, the medicalization of social welfare and the overdiagnosis of mental illness are mutually reinforcing, expressing the same neoliberal logic of privatization and growth that has dominated U.S. policy since the 1980s (Rimke, 2016).

OVERDIAGNOSIS AND WELFARE REFORM
While few, if any, peer-reviewed studies have explored the relationship between social welfare reform and the overdiagnosis of BD, evidence suggests that the two are closely related. Following the 1996 Personal Responsibility and Work Opportunity Act's (PRWORA) dismantling of traditional welfare, BD diagnoses skyrocketed, along with enrollment in supplemental security income (SSI), which rose fourfold between 1996 and 1998 (Burrows, 2010;Jans et al., 2004). As impoverished Americans flocked to SSI as a source of financial stability and Medicaid access, mood disorders became the largest and fastestgrowing drivers of enrollment, with a 100% increase in children applying to SSI and Medicaid under  Findings of rampant BD overdiagnosis among SSI recipients indicate that these surges in SSI enrollment reflect welfare-related needs rather than shifts in BD's true prevalence (Ghouse et al., 2013). Multiple studies have connected rising BD diagnoses to gaps in post-welfare services. For example, a 2006 study found that impoverished Americans were pursuing BD diagnoses in order to receive substance-related treatment, which the PRWORA purged from coverage in 1996 (Stein et al., 2006). Meanwhile, multinational studies on the medicalization of unemployment suggest that increasing rates of joblessness among SSI recipients with BD diagnoses reflect unmet needs for stable employment, with many frustrated SSI enrollees expressing a desire to work (Buffel et al., 2017;Frank, 2013;Holmqvist, 2009).
As with most neoliberal policies, the PRWORA's transfer of poverty services from welfare to disability programs has benefitted private interests, including pharmaceutical companies and a growing industry of mental health professionals, all of which reap profits from psychiatric overdiagnosis, expanded categories of mental illness, and reallocations of government spending (Mitchell, 2012;Rimke, 2016). Vulnerable Americans, including those applying for disability under misapplied BD diagnoses, suffer the greatest losses as unmet needs for employment, housing support, healthcare access, neighborhood investment, and antidiscrimination policies are met with irrelevant and often harmful mental health interventions (Shepherd & Wilson, 2018;Rimke, 2016;Mills, 2015;Hansen et al., 2014).

ADVOCACY ORGANIZATIONS
According to prominent psychiatrist and DSM 5 critic Allen J. Frances, American mental health advocacy groups have consistently failed to push back against the overdiagnosis of psychiatric disorders (Frances, 2010). Instead, advocacy groups such as the American Psychological Association (APA), the National Alliance of Mental Illness (NAMI), and the National Association of Social Workers (NASW) have issued demands for heightened mental health screenings within marginalized communities, calls which have drawn frequent support from the pharmaceutical lobby (American Psychiatric Association, 2015; Davis & Williams, 2020;Frances, 2010). These campaigns, which express the influence and assumptions of America's powerful mental health industry, reflect that there are currently few, if any, advocacy groups addressing the problem of BD overdiagnosis within the U.S. The majority of organizations recognizing overdiagnosis now operate in other countries.
One of the most influential of these organizations is Mental Health Europe (MHE), a pan-European, non-governmental organization (NGO), which has pushed for a demedicalized, psychosocial approach to mental health policy over the past decade. In 2012, MHE set up a task force to investigate the development of the DSM 5, releasing a statement that denounced its biomedical assumptions, expanded diagnoses, and promotion of "unnecessary and harmful" drug treatments (Mental Health Europe, 2014;Mental Health Europe, 2013). In a follow-up report, MHE called on the European Union and World Health Organization to ignore expanded DSM diagnoses unsupported by evidence, to offer diagnostic alternatives, and to cease promoting psychiatric medications as a first line response to distress (Mental Health Europe, 2015). MHE's criticisms of the DSM 5 have been influential across Europe, with the Superior Health Council of Belgium declaring in 2019 that the DSM would no longer be at the nation's "center of [mental health]care planning" (Simons, 2019).
In addition to curtailing the influence of the DSM 5, MHE has targeted several other drivers of psychiatric overdiagnosis. Since 2013, MHE has called for legislation to expose the financial ties between mental health policymakers, professional organizations, and pharmaceutical companies (L'Ecluse, 2019). MHE has also launched an investigation into the influence of chronic unemployment on rising "psychosocial disability," a term MHE uses in place of mental illness (Mental Health Europe, 2016). MHE's proposal of work programs for those labeled mentally ill was approved by the Council of the European Union in 2016, suggesting that Europe is moving away from the medicalization of unemployment, as well as from biogenetic models of mental illness in which disability is considered incurable (Finn, 2017;Mental Health Europe, 2016). Meanwhile, the ascendency of MHE's psychosocial approach to mental health was powerfully expressed in a 2017 report from the United Nation General Assembly, which declared that the "neurological paradigm [of mental illness] causes more harm than good," and called for a shift "from focusing on chemical imbalances to focusing on power imbalances and inequalities" (Kinderman, 2020).

A CALL TO ACTION: POLICY AND PRACTICE
To address BD overdiagnosis within the U.S., policymakers must first study the hundreds of thousands of Americans receiving SSI benefits under BD diagnoses, a growing population at high risk of childhood trauma, poverty, and homelessness (Cerimele et al., 2014;Etain et al., 2008;Huxley & Baldessarini, 2007). New and focused research is needed to determine if these are in fact "risk factors" for BD, as is usually assumed, or if they are generating transdiagnostic symptoms which are driving overdiagnosis. There is also an urgent need for studies exploring the socioeconomic deprivation behind BD diagnoses, and the extent to which underemployment and poor access to healthcare are motivating Americans to seek SSI enrollment (Ghouse et al., 2013;Wong, 2016). Finally, researchers need to examine the impact of BD diagnoses within communities of color and the intersections of racial stigma with those brought on by DSM diagnoses and disability status (Hawke et al., 2013).
As researchers develop a clearer image of BD overdiagnosis, the mental health community can begin taking steps to address it. One strategy is to demand that researchers stop using inflated figures to characterize BD and instead adhere to conventional, evidence-based estimates of BD's prevalence, which have ranged from 1% to 2% of the population (Mitchell, 2012). Another step is to discourage or eliminate the use of screening tools like the Mood Disorder Questionnaire (MDQ) and return to clinical interviews as the "gold standard" in BD diagnosis (Ghouse et al., 2013). The most direct path to reducing overdiagnosis is to revise expanded criteria for BD by removing the "bipolar not otherwise specified" category from the DSM 5, and by toughening criteria surrounding hypomanic episodes, the key symptom in bipolar II (Mitchell, 2012). Some researchers, citing findings that bipolar II has no clear genetic profile and is not phenotypically distinctive from unipolar depression, have explored eliminating the bipolar II diagnosis altogether, which would lead to substantially fewer diagnoses of BD (Gitlin & Malhi, 2020;Parker & Fletcher, 2009).
Others insist that addressing the roots of BD overdiagnosis will require more dramatic changes to the DSM. BD's high comorbidity, expanding criteria and subtypes, and failure to control for environmental distress point to a broader crisis in post-DSM III psychiatry, which has failed to successfully classify mental illness into distinct conditions, find biogenetic markers for core diagnoses, or curtail its own expansion (Ghaemi, 2014;Timimi, 2014). Thus, stemming the flow of overdiagnosis may require the diagnostic system established with the DSM III, which has consistently prized reliability over external validity, to be abandoned (Dutta et al., 2007;Pilgrim, 2014;Vieta & Philips, 2007). In its place, researchers propose the DSM be reorganized into a dimensional system with less arbitrary divisions between symptoms and fewer pathologizing diagnoses (Dutta et al., 2007;Vieta & Philips, 2007). Others demand that the DSM adopt a "traumagenic" model of mental illness that is based on shared experiences of adversity and encourages environmental over drug-focused interventions (Pilgrim, 2014).
MHE's campaign for financial transparency and protective social policies provides a potential blueprint for curbing psychiatric overdiagnosis within the U.S. Legislation preventing researchers with drug industry ties from serving on DSM panels could reduce pressures to further widen diagnostic criteria, while housing and employment programs for those at risk of being labelled mentally ill would likely lead to fewer diagnoses and less biogenetic pessimism about mental illness (Elinson et al., 2007;Huxley & Baldessarini, 2007). By following Belgium's lead and decentering the DSM within its healthcare system, the U.S. could begin to explore diagnostic models that are more sensitive to the social, economic and racial determinants of mental health. Finally, retiring the stigmatizing term "mental illness," which locates the roots of mental distress within dysfunctional individuals, may help cleanse U.S. mental healthcare of neoliberal assumptions that are driving overdiagnosis and the privatization of poverty (Mills, 2015;Rimke, 2016;Shepherd & Wilson, 2018).
Social workers will need to think critically about how to respond to the problem of BD overdiagnosis, both as clinicians and as advocates for change. As clinicians, social workers need to familiarize themselves with the harms that BD diagnoses can impose on clients, such as heightened stigma, adverse drug effects, and enrollment in disability programs that can further reduce functioning (Hawke et al., 2013;Huxley & Baldessarini, 2007;Littrell, 2012). Similarly, social workers should be aware of the pressures they face to apply DSM diagnoses, including organizational needs for efficiency, growth, and specialization, directives which often benefit social workers and their status as health professionals (Rimke, 2016). Finally, clinical social workers need to look beyond biological theories and identify the environmental factors driving clients to seek BD diagnoses, such as childhood trauma, housing instability, underemployment, and a lack of welfare related services. They must bear in mind that a BD diagnosis is a mark of lifelong biogenetic dysfunction, one that may render clients less likely to receive the help that they need.
As advocates, social workers must challenge calls to reduce mental health disparities through indiscriminate screening and warn policymakers about the differential impact of BD overdiagnosis within minority communities. They must resist the increasing bio-medicalization of mental health and be prepared to defy a system of diagnostic psychology which many researchers say has lost its claim to scientific validity (Dutta et al., 2007;Ghaemi, 2014;Jacob et al., 2014;Pilgrim, 2014;Timimi, 2014). Finally, social workers will need to join researchers in psychiatry and psychology to develop new systems of diagnoses that link mental distress to environmental risk factors and harmful public policies (Rimke, 2016). Social workers will need to decide for themselves whether a public conception of mental health is possible within the U.S.'s privatized healthcare system, and within an unregulated market economy that prioritizes innovation and profits over social protections.
Whatever they decide, social workers can no longer afford to ignore the overdiagnosis of BD, which has mischaracterized distress, bolstered stigmas, and placed the burden of change on those who are often least able to effect it.

CONCLUSION
BD overdiagnosis is a growing problem within the U.S., especially among marginalized Americans seeking social support through SSI enrollment. Consequences of misdiagnosis include heightened stigma, adverse drug effects, and the concealment of environmental distress within poor communities and communities of color. A new frontline in the medicalization of poverty, the overdiagnosis of BD has been poorly monitored by mental health organizations, which have launched uncritical campaigns for increased mental health screening amidst widespread reports of invalid diagnoses (American Psychiatric Association, 2015). BD overdiagnosis is perpetuated by clinical assumptions that BD is undertreated, prone to subclinical presentations and underrecognized in communities of color, claims which have failed to amass significant evidence (Littrell, 2012;Mitchell, 2012). Assumptions about BD's poor prognosis and biogenetic origins are rooted in the unsubstantiated claim that mental illness involves discrete biologically based conditions, and from neoliberal assumptions about the sources of and solutions to mental distress, which have helped to medicalize and dismantle social welfare programs over the past four decades (Timimi, 2014;Wong, 2016). Policymakers are encouraged to follow Mental Health Europe's example by passing financial transparency laws, decentering the DSM within mental healthcare, and introducing protective social programs that offer employment, stable housing, and expanded access to healthcare. Recommendations targeting BD overdiagnosis include narrowing diagnostic criteria in the DSM 5 and developing new traumagenic models of mental distress that are sensitive to social inequalities. Social workers are called to educate themselves about the harms of BD overdiagnosis, to press for a psychosocial and welfare-based approach to mental healthcare, and to maintain a healthy skepticism about DSM diagnoses and their status as evidence-based guides to practice.