Search tips
Search criteria 


Logo of yjbmLink to Publisher's site
Yale J Biol Med. 2016 June; 89(2): 247–254.
Published online 2016 June 27.
PMCID: PMC4918875
Focus: Sex and Gender Health

Conflations of Marital Status and Sanity: Implicit Heterosexist Bias in Psychiatric Diagnosis in Physician-Dictated Charts at a Midwestern Medical Center

Jonathan M. Metzl, MD, PhD,a,* Sara I. McClelland, PhD,b and Erin Bergner, MPH, MAa


This paper discusses the role of gender role conformity in psychiatric determinants of well-being after of the depathologization of homosexuality from the DSM. In order to examine the heterosexualizing of sanity in U.S. psychiatric and popular cultures, we analyze archived psychiatrist-dictated patient charts from outpatient psychiatric clinics from a Midwestern medical center (n = 45). We highlight ways physicians deployed heteronormative gender expectations to describe and treat women’s and men’s depressive illness and implicitly construed troubled female-male relationships and sexual encounters as indices of psychopathology. We theorize how evolving connections between the heteronormal and the psychiatric normal performed some of the same regulatory functions, as did the DSM, coding particular gender performances and partner choices as mentally healthy while relegating others to the realm of disease. Only here, focusing on the mainstream instead of the marginalized kept the ideological work of these scripts hidden from view.

Keywords: Gender, Psychiatric diagnosis, Implicit bias, Heterosexuality


On December 16, 1973, a New York Times headline announced that, “Psychiatrists, in a Shift, Declare Homosexuality No Mental Illness” above a news story that explained how, after intense debate, the governing body of the American Psychiatric Association (APA) voted to change its century-long position that defined homosexuality as a mental illness. The article detailed how the board of trustees of the 20,000-member group approved a resolution proclaiming, “by itself, homosexuality does not meet criteria for being a psychiatric disorder…we will no longer insist on a label of sickness for individuals who insist they are well and demonstrate no generalized impairment in social effectiveness” [1].

Removal of homosexuality from the Diagnostic and Statistical Manual of Mental Disorders (DSM)-II was the first of a series of diagnostic and professional actions that shifted same-sex desire ever farther away from the realm of perversion in official psychiatric discourse. The DSM-III expunged the term “homosexuality” in 19871, and the APA prohibited so-called “conversion” therapy soon thereafter [3]. Certain voices from within the profession argued that homophobia was more constitutive of mental disorder than was the object of homophobia’s obsessive fear [4].

Given this trajectory, it is understandable that much of the academic literature after the 1973 decision and the subsequent removal of homosexuality from the DSM focused on the implications of the depathologization of homosexuality for queer identity, normativity, and mental health [5-9]. However, an important set of questions was frequently overlooked in the post-1973 debate: if there was indeed a loosening of associations between a person’s sexual identity and mental illness, how might these loosened associations have impacted diagnostic assumptions, imagined expectations, and professional practices regarding gender in heterosexual relationships? Did psychiatry also liberalize or relax its approach to various iterations of female-male pairings? Or, did the removal of homosexuality from the DSM unintentionally fortify normative assumptions about “proper” gender roles in the interactions between mental symptoms, women, and men?

In this paper we theorize that the notion that psychiatry eliminated homophobic bias in the post-DSM-II-era simply by removing “homosexuality” from the text was complicated by the emergence of two American cultural scripts. We wonder whether, as homosexuality became less associated with psychopathology, heteronormativity, heterosexual relational imperatives, and female-male relational norms became increasingly aligned with mental health. And whether, as it became less acceptable to overtly pathologize homosexuality, it became increasingly acceptable to diagnose threats to female-male relationships, pairings, or object choices as conditions that required psychiatric intervention.

Our mixed-method study begins with a brief overview of psychiatric linkages between normative gender roles and normative mental health. We discuss how psychiatry’s age-old associations between heteronormativity and sanity were given new life in the post-DSM-II era by the advent of cultural and pharmaceutical discourses such as mother’s little helpers that overly linked happy heterosexuality with mental health. We then detail content analysis of a set of archived, psychiatrist-dictated patient charts from outpatient psychiatry medication clinics from a particular mental health system from the year 2000 — a time when notions of “pathological” homosexuality were long removed from diagnostic codes, but when cultural depictions of mentally-healthy heterosexuality flourished, particularly in relation to antidepressant medications. Our analysis pays particular attention to the ways that physicians deployed heteronormative gender expectations to observe, describe, diagnose, and treat depressive illness over time. We conclude by discussing how milestones of liberatory promise represent landmarks but not endpoints in the path toward the eradication of homophobic stigma. Diagnostic codes and social mores change in progressive ways. But beneath the headlines, underlying assumptions about “normalcy” and “abnormality” remain invisibly recalcitrant, while bias reformats and rearticulates in ways that needs to be continually recognized if it is to be effectively addressed.

The Heterosexualizing of Sanity

Healthy heterosexuality has long served as an index of sanity in psychiatric discourse. “Partiality to heterosexuality,” as Katz [10] describes it, underlies Freudian assumptions about normal and pathological psychological development. Indeed, psychoanalytic formulations, from castration anxiety to the Electra complex, long connected female-male gender pairings with mental distress. Feminist psychoanalyst Chodorow [11] defines heterosexuality as a “compromise formation” whose normalcy is “taken as a given” in much of psychoanalytic thought.

Of equal relevance for this paper, links between normative gender roles and normative mental health also appeared as central components of late twentieth-century American medical and popular rhetoric surrounding prescription psychopharmaceuticals. For instance, Valium and other minor tranquilizers became widely known as “mother’s little helpers” during the same era that psychiatry expunged homosexuality from the DSM. Doctors in the 1960s and 1970s prescribed tranquilizers to help women care for their husbands, play with their children, and cope with the pressures of white, middle class housewifery [12-14]. One oft-cited ad from the Archives of General Psychiatry from the early 1970s showed a woman named Jan whose inability or refusal to attract a man by the age of 35 led her to seek psychiatric treatment [12,15].

Starting in the late 1980s, rhetoric surrounding serotonin uptake inhibitor (SSRI) antidepressants promoted the notion that psychoactive drugs restored productive white maternity. Pharmaceutical advertisements featured white women whose successful pharmacotherapy was marked by wedding rings and children, or that quoted happy women who claimed, “I got my playfulness back” while playing with infants [12,16].2 The SSRI era also saw the expansion of gender normativizing scripts about men and masculinity, though to a lesser extent than those that linked psychiatric treatment with women’s gender roles.3 These and other representations reinforce critiques of depression as a stereotypically white illness in the U.S. whose construction potentially excludes, silences, or effaces the “suffering” or “shame” of women of color [21].

Similar themes also suffuse analyses of late twentieth-century U.S. popular cultural representations of psychotropic medications [22,23].4 These linkages between normative gender, whiteness, and normative mental health can also be understood through theoretical work that conceptualizes how practices and institutions legitimize and privilege heterosexuality and shape cultural conceptions of health and normality as a result [24-26]. Heteronormativity in clinical practice, however, has until recently functioned as an under-discussed topic [27].

Materials & Methods

This is the second in a series of studies that assess how heterosexualizing themes evolved in physician-dictated charts from archived outpatient psychiatric clinic of a large Midwestern U.S. medical center over roughly the first 15 years of the SSRI era, 1985 and 2000. In the first study [28], we systematically coded the contents of 270 randomly selected records of patients diagnosed with depression at 5 year intervals, with particular attention to how clinicians’ descriptions offered insight into the evolving relationships between compulsory or assumed heterosexual binaries and mental health. In the first study, we found that charts of female patients seen by psychiatrists and prescribed SSRI drugs increasingly referenced expressly heterosexual marriage or relationships, single motherhood, or motherhood in heterosexual relationships. Clinicians’ descriptions of male patients showed nearly opposite trends.5 Extensive sampling of clinic charts from the decades before same-sex marriage became common parlance, admittedly from a single clinic system, uncovered no descriptions of a non-heterosexual relationship, tryst, desire, or family configuration. (Figure 1).

Figure 1
Change in frequency of non-DSM terminology used in depressed female and male patients’ medical charts, 1985-2000.

The current study looks more closely at charts from the year 2000, with a deeper ethnographic focus on descriptions of patients’ intimate relationships.6 Our inquiry focused on the linguistic detail used by doctors to describe patients’ sexual and partner relationships, particularly as they pertained to presenting symptoms and treatment recommendations. We specifically compared themes of heterosexual marriage and anxieties about female-male relationships, gender roles/expectations, and sexual behaviors in physician dictations of patients diagnosed with mood disorders and prescribed antidepressant medications in outpatient psychiatry clinics.

The relationships between gender roles and clinical interactions are highly complex — indeed far more complex than our study begins to capture — and heteronormative does not necessarily mean gendered. Again, historians show how heterosexuality and heteronormativity are “invented” [10] and unstable concepts, subject to numerous configurations and resistances. So too, sexuality and normativity are not synonymous [30], and heteronormative lifestyles are not solely the domain of heterosexual couplings, as Duggan’s [31] work on “homonormativity” rightly points out [32,33].

Given such fluidity, the aim of our study was neither to prove that “heterosexuals” appeared with increasing frequency at psychiatrists’ offices, nor to quantify whether such persons “improved” after treatment. Our Institutional Review Board (IRB) approval for the content analysis portion of the study covered only the initial visit charts from January 1, 2000 to December 31, 2000, and we were unable to discern which patients ultimately took medications or returned for follow-up visits. More important, we made no attempts to discern between the revelations patients shared in doctors’ offices and the intimate practices of their daily lives.

Instead, we see this collection of psychiatric notes as an archive of a particular snapshot in time, roughly three decades after depathologization of homosexuality in the DSM and one decade before the notion of “freedom to marry” [34] entered mainstream discourse. In this sample at least, heterormativity functioned as a common set of assumptions between doctors and patients that, appeared to shape seemingly objective “truths.” These assumptions appeared to impact patients' experiences of their depression as well as their rationales for seeking treatment. And they shaped physicians’ dictated recollections of physician-patient interactions in ways that interpolated these interactions into a heteronormative clinical framework. By so doing, the charts explicitly reinforced linkages between heterosexuality and mental health, while at the same time implicitly rendering other types of pairings to realms masqueraded or unacknowledged.

Sample Selection

We reviewed randomly drawn records of 45 adult patients diagnosed with “major depressive disorder,” “major depressive episode,” or “dysthymic disorder,” via DSM-IV [35] diagnostic codes on initial medication evaluation visits to outpatient psychiatric clinics from the year 2000. Again, the charts represented a subset of a larger, 15-year sample, pulled from a potential pool of roughly 2,500 charts, for the aforementioned analysis of psychiatric and medical records from the outpatient clinics of a large, university-based Midwestern hospital system. All charts were selected using a spin search that electronically generated chart lists at random from the hospital archive.

Resident and attending psychiatrists conducted psychiatric medication evaluations at four suburban outpatient clinics linked to the hospital system. The standard protocol for these types of interactions included 45-minute to one-hour diagnostic conversations organized through sections such as “presenting problem,” “past psychiatric history,” “mental status exam,” “diagnosis,” “treatment plan,” and so forth. At some point after the interactions, clinicians dictated their notes in standard format to phone dictation services. Resident physicians dictated roughly 60 percent of the evaluations, and in these instances a supervising psychiatrist read and approved the final reports. In order to ensure diverse clinical relationships, we analyzed no more than four charts from any single psychiatrist. In order to protect anonymity, we did not collect demographic information about physicians. Instead, we focused on the discourses in the medical charts as a means of investigating the circulation of observations and details outside of DSM criteria, including aspects of gender performativity, object choices, and romantic relationships.

In accordance with IRB protocol, we completely removed all identifying information linking chart material to particular patients (e.g., names, hospital ID numbers, locales), assigned each chart a random ID number unrelated to unique patient identifiers (e.g., #09 M(ale)), and developed methods for confidential storage during the study and material disposal after the study was complete.


An interdisciplinary research team comprised of the three authors of this paper, representing psychiatry, psychology, and sociology, systematically analyzed all records using a thematic coding strategy [36]. As the sample already preselected for physician descriptions of female-male relationships, we focused on trends linked to how physicians described connections between psychiatric symptoms and heterosexual marriage, relationships, gender roles/expectations, sex, and other variables.

Our analysis focused on five primary themes: (1) patient relationships, including relational stressors, family relationship roles, and physician’s queries about patient’s relationship status; (2) home life, including patients’ social and intimate relationships, activities in home, feelings and opinions about patients’ home life, responsibilities within the home, social activities, family descriptors, and children; (3) work, including a patient’s employment or lack thereof; (4) sex, including descriptions of sexual activity, sexual relationships with partner or other individual, as well as references to same-sex desire or potential “experimentation with subjectivity” [37]; and, (5) relationships between gender “norms” [31,38,39] and symptoms, diagnoses, and treatments, including the ways that physicians did or did not frame presenting issues via discussion of gender roles or gender expectations.

The research team marked additional demographic information about each patient, including age, race, gender, and sexual orientation, and double-checked this information for accuracy. We marked gender using physician notes: when female pronouns were used, a patient was noted as female; when male pronouns were used, patient was noted as male. Race and sexual orientation were marked when explicitly noted.


Patients in the sample ranged from 18 to 65 years old and were on average 37 years old (SD = 11). Only nine of the charts expressly recorded race/ethnicity — four as “African American,” four as “white,” and one as “Asian.” We noted a trend for physicians to mark race/ethnicity in charts of non-white patients, while race/ethnicity seemed to function as assumed in charts of patients who doctors assumed, or later in the charts identified, as white — thereby reinforcing the problematic assumption that whiteness implies the lack of race. Thirty-six charts identified marital status — five as “divorced, two as “engaged,” three as “single,” twenty-two as “married,” one as “separated,” and one as “widowed.” The breakdown of women (n = 29) to men (n = 16) approximated what are argued as epidemiological gender differences in depression diagnoses [40].

“The patient discovered that she missed him more and more and became very depressed”

Physicians’ decisions about what details to include in the introductory comments when describing patients offered one clue to the priority they afforded to relationship status. We found, for example, that physicians consistently framed women’s presenting problems and concerns in terms of their expressly heterosexual intimate relationships. The majority of women’s charts identified patients’ marital status as a matter of course and then qualified this status through reference to a male partner. For example, “Mrs. A. is a 46-year-old married woman. She and her husband have four children…” (#26W). Charts of women identified as single, widowed, or unmarried also frequently referenced heterosexual relational statuses and specific other-sex partners: “Patient is a 56-year-old widow whose husband died of a heart attack…” (#35W).

Charts suggested that psychiatrists also frequently queried women about their relationship statuses when attempting to explain the onset of their symptoms, even when relationships were not central to case presentations. “She is single. She has never been married…She is not sexually active” read the chart (#18W) of a woman who was diagnosed with depression. Another chart (#40W) detailed how,

She states that her relationship with her boyfriend and her family are going well…She states that she began therapy three weeks ago…and that her therapist recommended that she be evaluated for antidepressant medication...

Descriptions of patients’ reasons for seeking psychiatric care often linked women’s decisions to seek treatment, or their depressive symptoms, to problems stemming from complex relationship problems. Examples included, “The patient describes problems with a romantic relationship…” (#44W). In another case,

The patient states that her husband died four years ago. Within five weeks, she started seeing another man…The relationship with this man has been a source of conflict…the patient has been suffering from a number of depressive symptoms (#31W)

Another chart (#37W) described a woman who sought treatment after,

She broke up with her boyfriend, George, 8 months ago. She now feels that is was all her overreaction and all her fault…The patient discovered that she missed him more and more and became very depressed.

Another chart (#29W) described a woman who presented for an urgent medication assessment with feelings of “sadness and helplessness” accompanied by “anxious symptoms” stemming from “marital instability” that followed her husband’s “extramarital affair.” And another chart (#17W) described a woman referred for psychiatric evaluation after failed marital therapy:

…depression and anxiety…She and her husband were in marital therapy for six months…though she has been reluctant to discuss medications in the past, shse is willing to do so now, as she has recently read articles in journals regarding antidepressants.

Charts also consistently connected women’s symptoms with unfulfilled relationship or gender roles. A chart (#30W) described how a woman’s symptoms and life stressors threatened the functioning of her family:

She reports that her marriage is ‘lousy’ but does not elaborate on the difficulties at home…She states that she has had difficulty keeping with her daily tasks at home.

Relational “problems” and stressors also served as explicit factors in diagnosis and treatment. Women’s relationship issues often appeared as psychosocial stressors under the so-called “Axis-IV” subsection of the diagnostic formulation, the section intended for reporting how “psychosocial and environmental problems” may affect the prognosess of mental disorders (American Psychiatric Association 1994). One chart (#38W) read,

Axis I: Major depressive disorder, single episode….Axis IV: Stressors include conflicted relationship with a man and family issues.

Such formulations frequently led to prescriptions for SSRI antidepressants. For instance, this same chart described the treatment plan as follows:

I discussed with the patient my diagnostic impression. I discussed with her the possibility of medications. We decided that she would start Celexa 20 mg a day…I would like to see her again in three to fourweeks [sic].

The relatively few charts that listed women’s work or employment (n = 7) did so under the “social history” sections of evaluations, rather than as components of the central case formulations, as was frequently the case in men’s charts (see below).

“Recently his level of intimacy with his wife has declined”

Men’s charts also frequently referenced relationships or marital status, yet these charts rarely qualified relationships through naming a specific partner. Instead, basic identifying information was frequently followed by employment status:

Mr. A. is a 41-year-old Caucasian male who works as a janitor… (#14M), or,

The patient is a married 33-year-old married PhD candidate who presents with the chief complaint of ‘feeling depressed’… (#2M).

This early introductory material in the charts suggested that heterosexual and work relationships were weighed differently for male and female patients and that this demographic detail was central to physician assessments.

Work and work life figured prominently in discussions of men’s reasons for seeking psychiatric attention — in sharp contrast to the centrality of relational stress in the formulations in women’s charts. A psychiatrist noted that one man with depressive symptoms and poor concentration,

…does not seem to be at significant risk of injuring himself at work, and says that he has at least ten more years to work before it is possible for him to retire (#9M ).

Work dominated case descriptions about men even when home and family life figured centrally in case formulations. Thus the chart (#5M) of a man who, “…states that he and his wife have been married for 32 years. He indicates that he has six children…” was followed by the claim that, “the patient describes his main problem as difficulty driving his truck…”

Similarly, the case of the “33-year-old married PhD candidate” explained that, “…He primarily complains of having difficulty focusing on his PhD work, having some increased fatigue, and difficulty concentrating…” (#2M).

Relationship concerns did appear in a smaller number of men’s charts. One case described a man whose emotional frustrations with his family led to physical outbursts. “He reports being frustrated that he cannot show his wife his feelings, either emotionally with words or physically via hugs…yelling at kids, hitting walls” (#13M). Overall, however, men’s charts were far less likely than women’s charts to connect mental symptoms to such relationship issues as emotional support or commitment. Instead, men’s charts expressly linked depression and anxiety to concerns about sex. Heterosexual sexual encounters were a common theme in these charts, as were anxieties about sexual performances with women. A psychiatrist described a man who sought psychiatric treatment (#16M) as, “Recently his level of intimacy with his wife has declined from a couple of times a week to every other week…”

As with women’s charts, these heterosexual sex themes coursed through diagnostic and treatment decisions. For instance, the psychiatrist in the “sexual performance” case above (#16M) dictated a lengthy formulation that framed the patient’s diagnosis in relation to his anxieties about heterosexual sexual encounters:

…it was my objective impression that the patient is currently suffering from anxiety disorder. Much of the focus of his current anxiety is related to his insecurities regarding his sexual performance…has had marked difficulties with penile erection….I feel that he will benefit from an antidepressant…

Similarly, the case file (#4M) of a patient that a psychiatrist described as a “45-year-old married gentleman” linked the man’s “…tremendous anxiety, difficulty sleeping, feelings of inadequacy, and depressive symptoms…” to his, “…sexual/erectile dysfunction.”

Treatment included an SSRI plus a new medication that was, in late 2000, only just beginning to appear in psychiatric treatment decisions: “Viagra, 100 mg tablets…samples of Serzone provided.”


Our analysis suggests that outpatient psychiatric charts from the year 2000 linked concerns about depressive illness to seemingly extra-clinical tensions regarding heteronormativity and its discontents. The pressures of attaining or maintaining heterosexual relationships — but not other types of relationships — functioned as common modes for describing depressive symptoms. Normative codes featured prominently in treatment deliberations, and particularly deliberations about antidepressant medications.

The charts also signal ways that clinicians weighed heteronormative gender expectations differently for women and men. Psychiatrists couched women’s depressive illness almost entirely in relation to the men in their lives. “She reports that her marriage is ‘lousy,’” or “she and her husband were in marital therapy for six months,” or “sadness and helplessness…anxious symptoms” related to “marital instability” followed a husband’s “extramarital affair.” Dictations frequently deployed the language of affect and emotion to connect women’s relationship stressors to psychological distress. “Since then she has not felt like socializing with her colleagues.” Or, “Patient now feels that it was all her overreaction and all her fault.” Or, “Patient reports that for at least a year she has felt more emotional.”

Depression linked to somatic action (or inaction) in men’s charts. Anxiety symptoms emerged when erstwhile psychoanalytic symbols of embodied power met the pressures, “doubts,” and “insecurities” of imagined sexual performance and its imagined expectations. The centrality of these expectations suffused treatment decisions, particularly regarding the relationships between sexual performance anxieties and the sexual side effect profiles of particular medications. Dictating clinicians also frequently assumed that symptoms arose in men as a result of their failures to live up to expectations about being breadwinners for their families. Thus the tensions of “32 years” of marriage and providing for a large family linked to depressive symptoms that manifest in a man’s “difficulty driving his truck.” These gender themes were roughly though in no way causally suggestive of those seen in popular representations of depression and psychotropic drugs over the same time period [12,22].

Again, our IRB mandate limited the scope of our analysis to charts from a single clinic system, surely leading to various forms of homogeneity in our source materials, and particularly so regarding race. Our reading unpacks silent assumptions about the gendered expectations associated with depressive illness, but by so doing we in no way wish to reinforce the ways that American society can, for instance, stereotypically validate the psychological suffering of white women while failing to recognize or acknowledge the suffering of women of color [21]. We found this year-2000 lack of diversity notable, if not wholly surprising. The hospital system served a racially diverse patient population, but such diversity appears to not have extended to outpatient psychiatric clinics. As we expand our sample over time, we aim to further address potential linguistic differences in the ways that doctors treated, conceptualized, and described white and non-white charts.

Hospital archiving methods further narrowed our textual analysis to charts from a single year, although as detailed above, the charts represent a subset of a larger, 15-year sample that demonstrates expanded use of gender stereotypes over time. With support from IRBs and studies that can attend to laws surrounding patient privacy concerns, future research from our project will sample clinical materials from several geographical locales and from more recent time periods, so that we might address how the stereotypes we uncover might (or might not) vary in charts of clinics that serve more racially or ethnically diverse patient populations and to compare the clinical attitudes about LGBTQI relationships in early twenty-first-century charts with those in the present day.

These limitations acknowledged, our findings open several potentially productive threads of inquiry for study of the relationships between gender norms, cultural aesthetics, market forces, assumptions about whiteness, and psychiatric diagnoses. The sample of charts examined for this study did not include any descriptions of patients that suggested same-sex desire, gender nonconformity, or ambiguity. This finding stands out only to mark a set of questions that are worthy of noting in their absence. It may mean, for example, patients in this sample did not discuss same-sex desires or relationships with their physician (as a result of not wanting or needing to) or perhaps that the physician did not note these in the charts.

Perhaps most important, our study suggests the need for a more developed psychiatric understanding of the diagnostic implications of heterosexuality and heteronormativity. Psychiatry enacts a long tradition of diagnostic expertise regarding matters it deems sexually deviant. But the materials we study suggest that psychiatry’s observational theories and practices regarding pairings or racial or gender formations that society assumes as mainstream, normal, or normative are potentially less developed at best, and function as blind spots at least. The need for attention to the mainstream became particularly salient in the time period of our analysis, 1985 to 2000, when, for instance, expanded pharmaceutical advertising and expanded use of SSRI medications also subtly impacted diagnostic interactions, while at the same time psychiatric diagnostic manuals eliminated the term “homosexuality” once-and-for-all.

To be clear, normative tropes regarding heterosexual relationships and role stressors provide common language that allows patients and doctors to more readily describe, listen to, and ultimately communicate about the impact of illness on people’s lives. This language often understandably represents the pressures, realities, and expectations of daily experience for many people. Patients, or their loved ones, often identify depressive symptoms by the strains placed on primary interpersonal relationships, or by the ways that such symptoms render the expectations of functioning as mothers, fathers, lovers, daughters, or employees ever-more-difficult. And it is likely the case that patients who rotely detail their symptoms using the unimpassioned language of the DSM (“Doctor I have two weeks of depressed mood and I suffer from significant appetite disturbance”) would elicit less of an empathic response from physicians than would patients who describe the anguish of not being able to attain desired familial, social, or professional fulfillment.

At the same time, the charts function as historical object lessons illustrating how normative and normativizing tropes and stereotypes, as they pertain to diagnostic practices, are both inclusive and exclusive. At the manifest level, mainstream constructions make it potentially easier to identify, empathize with, and treat the pain of the mainstream. But at the latent level, the same gendered and racialized cultural norms construe what are deemed non-normative or minority identities or pairings as abject, uncoded, or invisible. In this sense, by one reading, the charts might be argued to represent a victory for DSM revisions and reversals: we uncovered no instances of doctors who attempted to diagnose homosexuality. Yet by another reading, the gendered politics of diagnosis were anything but value-free, in as much as clinical interactions in the year 2000 still held white heteronormativity as mental health and treated it as such.

Again, such attention to the mainstream surely enabled increasing numbers of people to seek psychiatric treatment for highly painful life events. But in an aggregate sense, evolving connections between the heteronormal and the psychiatric normal suggest the possibility that the rhetoric of psychiatry and psychopharmacology performed some of the same regulatory functions as once did the DSM, coding particular gender performances, races, or object choices as mentally healthy while relegating others to the realm of disease. Only here, focusing on the mainstream instead of the marginalized kept the ideological work of these scripts hidden from view. The end result, one might theorize, was a particular imperviousness to the types of political mobilizations and critiques against psychiatric diagnosis that marked earlier decades.


Diagnostic and Statistical Manual
American Psychiatric Association
Lesbian, Gay, Bisexual, Trans, Queer, and Intersex
Selective Serotonin Reuptake Inhibitor
Institutional Review Board
Health Insurance Portability and Accountability Act


1Even then, although homosexuality per se was expunged from the third edition of the DSM [2], the manual still contained the diagnosis of “ego-dystonic homosexuality,” an ill-defined condition in which a person’s same-sex desire caused psychic distress.

2So too, physician-authored books such as Listening to Prozac [17] and Prozac Backlash [18] described case studies of white women who returned to their husbands or boyfriends after treatment.

3Popular novels such as Faye Weldon’s Splitting [19] depicted elderly, heterosexual, white men rendered suddenly virile after treatment with SSRI drugs. Meanwhile, public health campaigns sought to overcome men’s reluctance about psychiatric care by highlighting the impact of depressive illness on men’s roles as fathers and boyfriends. One campaign by the National Institute for Mental Health [20] urged treatment for men such as “Rene,” a police officer, who explained that, “I lost interest with the kids and doing the things that we used to do, going to the movies, just simple things…nothing… you know that families do. Ya know, they'd ask their mother, “Why is Daddy not getting up and not wanting to do anything with us?”

4Sampling a database of thousands of articles between 1985—two years prior to the advent of SSRIs—and 2000, the analysis uncovered no explicit descriptions of LGBTQI (lesbian, gay, bisexual, trans, queer, intersex) persons as depressed or as requiring treatment with Prozac or other SSRIs.

5Fatherhood and marriage appeared with regularity over time, but did not approach the frequency seen in women’s charts. In contrast to the relational signifiers found in female patient charts, clinicians regularly described male patients in terms of work, aggression, and athletics.

62000 was the first year of electronic record keeping in the medical center, and was thus the only year in our window that we were allowed to print chart materials and search and analyze them in depth, while charts after 2000 were HIPAA [29] protected, meaning they were compliant with the Health Insurance Portability and Accountability Act (HIPAA).


  • Lyons R. Psychiatrists, in a Shift, Declare Homosexuality no Mental Illness. The New York Times. 1973 Dec 16;
  • APA. Diagnostic and Statistical Manual of Mental Disorders (DSM-III) Washington, DC: American Psychiatric Publishing; 1980.
  • Bayer R. Homosexuality and American Psychiatry. New York, NY: Basic Books; 1981.
  • Vedantam S. Psychiatry Ponders Whether Extreme Bias Can be an Illness. The Washington Post. 2005 Dec 15;
  • Sedgwick E. Epistemology of the Closet. Berkeley, CA: University of California Press; 1991.
  • Stein M. Boutilier and the US Supreme Court’s Sexual Revolution. Law and History Review. 2005;23(3):491–536.
  • Friedman R, Downey J. Sexual Orientation and Psychoanalysis: Sexual Science and Clinical Practice. New York, NY: Columbia University Press; 2002.
  • Cahn S. Come Out, Come Out Whatever You’ve Got! Or, Still Crazy After All These Years. Feminist Studies. 2003;29(1):7–18.
  • The Health of Lesbian, Gay, Bisexual, and Transgender People: Building a Foundation for a Better Understanding. Institutes of Medicine [Internet] Available from: .
  • Katz J. The Invention of Heterosexuality. Chicago, IL: University of Chicago Press; 2007.
  • Chodorow N. Heterosexuality as a Compromise Formation: Reflections on the Psychoanalytic Theory of Sexual Development. Psychoanalysis & Contemporary Thought. 1992;15(3):267–304.
  • Metzl J. Prozac on the Couch: Prescribing Gender in the Era of Wonder Drugs. Durham, NC: Duke University Press; 2003.
  • Herzberg D. Happy Pills in America: From Miltown to Prozac. Baltimore, MD: Johns Hopkins University Press; 2009.
  • Tone A. The Age of Anxiety: A History of America’s Turbulent Affair with Tranquilizers. New York, NY: Basic Books; 2009.
  • 35 and Single. Arch Gen Psychiatry. 1970;22:291.
  • Metzl JM. Selling Sanity Through Gender: Psychiatry and the Dynamics of Pharmaceutical Advertising. J Med Humanit. 2003;24(1):79–103.
  • Kramer P. Listening to Prozac. New York, NY: Penguin Books; 1994.
  • Glenmullen J. Prozac Backlash: Overcoming the Dangers of Prozac, Zoloft, Paxil, and Other Antidepressants with Safe, Effective Alternatives. New York, NY: Simon & Schuster; 2000.
  • Weldon F. Splitting. New York: NY: Atlantic Monthly Press; 1996.
  • NIMH Real Men Real Depression [Internet] [cited 2016 May 10]. Available from: .
  • Harris-Perry M. Sister Citizen: Shame, Stereotypes, and Black Women in America. New Haven, CT: Yale University Press; 2011.
  • Blum L, Stracuzzi N. Gender in the Prozac Nation: Popular Discourse and Productive Femininity. Gender & Society. 2004;18(3):269–286.
  • Metzl JM, Angel J. Assessing the Impact of SSRI Antidepressants on Popular Notions of Women’s Depressive Illness. Soc Sci Med. 2004;58(3):577–584. [PubMed]
  • Zita J. Body Talk: Philosophical Reflections on Sex and Gender. New York, NY: Columbia University Press; 1998.
  • Jackson S. In: Sex, Gender, & Sexuality. Ferber A, Holcomb K, Wentling T, editors. New York, NY: Oxford University Press; 2009. Sexuality, Heterosexuality, and Gender Hierarchy; pp. 15–39.
  • Hudak J, Giammattei S. In: American Family Therapy Academy Monograph Series Expanding our Social Justice Practices. Ariel J, Hernández-Wolfe P, Stearns S, editors. Washington, D.C: American Family Therapy Academy, Inc.; 2010. Doing Family: Decentering Heternormativity in Marriage and Family Therapy; pp. 105–115.
  • Lindner P, Martell C, Bergström J. et al. Clinical Validation of a Non-heteronormative Version of the Social Interaction Anxiety Scale (SIAS). Health and Quality of Life Outcomes. 2013;11:209–214. [PMC free article] [PubMed]
  • Metzl JM. In: Globalization of Pharmaceuticals. Jenkins J, editor. Santa Fe, NM: School of Advanced Research Press; 2011. Gender Stereotypes in the Diagnosis of Depression: A Systematic Content Analysis of Medical Records.
  • Health Information Privacy. U.S. Department of Health and Human Services [Internet] Available from: .
  • Ward J, Schneider B. The Reaches of Heteronormativity: An Introduction. Gender & Society. 2009;23(4):433–439.
  • Duggan L. Queering the State. Social Text. 2004;39:1–14.
  • Halperin D. What Do Gay Men Want?: An Essay on Sex, Risk, and Subjectivity. Ann Arbor, MI: University of Michigan Press; 2007.
  • Halperin D. How to be Gay. Cambridge, MA: Harvard University Press; 2007.
  • Freedom to Marry [Internet] Available from: .
  • APA. Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) Washington DC: American Psychiatric Publishing; 1994.
  • Braun V, Clarke V. Using Thematic Analysis in Psychology. Qual Res Psychol. 2006;3(2):77–101.
  • Race K. Pleasure Consuming Medicine: The Queer Politics of Drugs. Durham, NC: Duke University Press; 2009.
  • Jackson S. Interchanges: Gender, Sexuality, and Heterosexuality: The Complexity (and Limits) of Heteronormativity. Feminist Theory. 2006;7:105–121.
  • Warner M. Fear of a Queer Planet: Queer Politics and Social Theory. Minneapolis, MN: University of Minnesota Press; 1993.
  • Piccinelli M, Wilkinson G. Gender Differences in Depression: Critical Review. Br J Psychiatry. 2000;177:486–492. [PubMed]

Articles from The Yale Journal of Biology and Medicine are provided here courtesy of Yale Journal of Biology and Medicine