The purposes of this chapter are to dispel the myth that psychiatric diagnosis is value free and based on sound scientific evidence and to show how the client’s sex and the therapist’s biases about gender often fill parts of the vacuum left by the absence of science when therapists attempt to identify, categorize, and label people’s emotional suffering. We contend that women and men who experience psychological distress are better served if they are aware that the alleviation of mental suffering is not always achieved, and sometimes has been hindered, by reliance on the Diagnostic and Statistical Manual of Mental Disorders (DSM; American Psychiatric Association, 2000), the book that has come to dominate psychiatry, psychology, and their allied professions.

In any consideration of gender bias in psychiatric diagnosis, it is essential to keep in mind that the categories described in the DSM are not “real.” These diagnostic categories are invented, defined, and often re-invented by a relatively small group of people, most of whom are White, American, male psychiatrists who are positioned to describe what is normal and what is abnormal (Caplan, 1995; Cohen & Jacobs, 2007). So, we ask the question: Who is served by this categorization process? First, the existence of the DSM provides psychiatry and clinical psychology with the appearance of scientific credibility. Second, the DSM is part of an enormously profitable mini-industry that consists of the manual and its various supplements, casebooks, sourcebooks, and software packages. The profits swell the coffers of the DSM’s publisher, the American Psychiatric Association, which is a lobby group as well as a professional association. Questions about the validity and scientific basis of the DSM put such profits in jeopardy and are, therefore, often unwelcome among mental health professionals closely connected with the handbook itself; many therapists from various disciplines also worry that questioning the diagnostic categories threatens their livelihood. However, questioning is necessary if we are to understand the risks for clients when DSM labels – and their attendant stigmata and consequences – are applied.

The most commonly chosen standards of psychiatric diagnoses warrant a critical thinking approach, including a challenge to the very notion of objectivity as it is applied to psychiatric diagnosis. In this chapter, we examine the ways in which stereotypic conceptions of women and men powerfully influence diagnostic decision making and, indeed, the very process through which diagnostic criteria are derived and then put into clusters called diagnostic categories. In our discussion, we use the accepted definition of gender as encompassing cultural expectations for women and girls (i.e., femininity) and men and boys (i.e., masculinity) and the term sex to denote biological maleness or femaleness (Caplan & Caplan, 2009; Lips, 2001). We discuss stereotypes concerning men and women in relation to psychiatric diagnosis, but for three reasons we focus especially on women. First, women represent the majority of clients who seek psychotherapy and thus are the majority of people to whom the bulk of the most damaging diagnoses are applied (Caplan, 1995; Worell & Remer, 2003). Second, in both society in general and diagnostic and other psychiatric discourse, women in particular are saddled with debasing stereotypes of psychological weakness, theories of hormonal susceptibility to disorders, and strict societal expectations about appropriate and inappropriate behavior (Brown, 1994; Caplan & Cosgrove, 2004). Third, in the creation of many diagnostic categories and criteria, the DSM authors historically have shown a marked lack of consideration for the realities of women’s lives, life contexts, and experiences of victimization. Of course, men, especially racial, ethnic, and sexual minorities, experience many of the same forms of victimization as women do, including poverty, violence, and discrimination. In this chapter, we examine the ways that gender stereotypes in diagnoses affect men and women differently.

We present a critical framework for the examination of gender bias in selected diagnostic categories. Where we use DSM terms and quote specific DSM criteria, we do so not because these categories have been proven to be valid, but because it is specifically those categories and labels that we want to critique. We first outline the assumptions that guided our consideration of gender-related biases in the DSM; we also describe the fundamentally flawed beliefs that together form the basis from which various gender biases are derived. Next, we systematically consider the groups of diagnostic categories that have been most closely scrutinized by critics of the DSM with respect to gender bias: personality disorders, mood disorders, and anxiety disorders. We then consider three key over-arching issues that cut across diagnostic categories and are broadly relevant to gender bias in psychiatric diagnosis: poverty, abuse, racism, and sexual orientation and identity. Last, we consider the future of psychiatric diagnoses and the dangers, both societal and individual, that result from gender bias in diagnosis.

Fundamental Assumptions Concerning Gender Bias in Diagnosis

In this section we present two sets of assumptions: (1) the beliefs that informed us as authors in considering gender and gender-related bias in the DSM and (2) the beliefs and stereotypes about women and men that have led to diagnostic bias. We begin with the premise that equality ought to underlie all practices applied to men and women, that both sexes ought to receive equal scientific attention and access to help, and that standards of care should be equal for all. Although this assumption appears to be so fundamental that it might scarcely seem worth stating, we will show that current-day diagnosis is replete with examples characterized by considerable disregard for the best interests of women and, in some instances, of men.

A second assumption is that it is important to acknowledge that a psychiatric label often confers on its recipient a cascade of negative effects that can include personal, interpersonal, and, sometimes, medical and legal consequences. Although therapists often assert that their decision making is objective and untainted by sociocultural influences, diagnostic decisions are made about real people, and they are never fully devoid of social stigma or interpersonal challenges. So, for instance, when a person is labeled mentally ill, we need to ask ourselves how that diagnosis (and by this we mean the therapist’s assignment of a diagnosis, not the patient’s emotional difficulties) came about and how living with that diagnosis will alter the person’s life circumstances, the person’s self-perception, and the perception of that person by others.

Our final assumption is that not every person who is assigned a psychiatric diagnosis is mentally ill. There are two main reasons for this contention. First, sometimes clients are given a diagnosis so that their therapy will be covered by their health insurance. The second, more sinister reason is, we argue, that certain aspects of human experience, which may actually be normal responses to abusive or traumatic situations, are labeled as psychiatric abnormality in clients when they seek help. This labeling can place abused women in a vulnerable position by emphasizing their supposed individual, psychological weakness rather than the traumatic nature of the abuse. Moreover, although practitioners who adopt the DSM approach generally assume that applying a diagnostic label to a person’s suffering is essential to helping that person, this presumption is highly questionable, especially given that the changes that are often needed must take place not in the mind of the client but rather in the host of damaging and disempowering forces that can surround traumatized women and men, including not only threats of violence but also harassment and discrimination (Caplan, 1995; Worell & Remer, 2003). By focusing on the minds of traumatized individuals rather than on the traumatic context in which they have had to function, we dismiss the sometimes horrific circumstances of their trauma. Caplan (2006) pointed out that such dismissal often occurs in the experiences of individuals returning from war; she stated that “by pathologizing and privatizing their suffering, we add to their burdens the belief that they should have endured the war emotionally unscathed and that no one other than a therapist wants to listen” to their stories (p. 20).

Having considered the assumptions that guide our analysis in this chapter, we now turn to the main assumptions that lead to gender bias in psychiatric diagnosis. One key assumption is a widely held and long-standing belief in the “natural” (i.e., biological) inferiority of women. Although some would insist that psychiatry and medicine in general have moved beyond such beliefs, we present examples of current biases that clearly reflect this assumption; most – but not all – of these examples pertain to the host of psychiatric “disorders” that involve women’s susceptibility to hormonal factors. As we discuss below, those who make treatment decisions based on biased diagnoses frequently discount the role of social factors in women’s lives and, instead, emphasize and pathologize women’s biology.

A second assumption underlying gender bias is the belief that knowledge and expertise lie solely in the domain of medical professionals. Because the medical profession has historically been dominated by men and by androcentric viewpoints (Willard, 2005), women’s knowledge of their own bodies and minds can play a secondary role to medical authority. Psychiatry’s overreliance on narrowly defined notions of “normal” behavior can also be seen as symptomatic of the field’s aim to broaden the range of problems considered as appropriate targets for medical intervention. Furthermore, as we discuss later, definitions of normalcy are never value free.

A final assumption that informs gender bias in diagnosis is the widely propagated notion that many common emotional and behavioral problems are endogenous (i.e., stem from an internal cause) in nature, rather than reactive (i.e., a response to an external cause). It is this dichotomy, and the corresponding preference for endogenous explanations for suffering, that bolsters the belief that diagnosis can occur in an ideological vacuum, uninfluenced by cultural stereotypes and social norms, including sexist ones. However, stereotypes are common elements of the derivation and application of diagnostic criteria, and stereotypic ideas about sex and gender can moderately or severely impair clients’ emotional well-being.

Diagnostic Categories Replete with Gender Stereotypes

Personality Disorders

We begin our critique of various clusters of DSM diagnostic categories with personality disorders. The psychiatric categorization of personality disorders is an attempt to formalize definitions of extreme forms of personality patterns that exist in the general population. These diagnoses are questionable largely because they categorize thoughts and actions as disordered by virtue of their degree of deviation from what are assumed to be normal patterns of behavior. We all have personality traits, but the question for the practitioner is presumably whether a personality pattern is abnormal enough to be considered a disorder. As Caws (1994) stated, the word diagnosis “suggests a process of inference: from symptoms to underlying condition. [However], one difficulty with psychiatric diagnoses is that disorders almost never have unambiguously pathognomonic symptoms” (pp. 205–206). Therefore, if a person is diagnosed as having a personality that is in some way “disordered,” the diagnosing clinician must infer from the person’s thoughts and actions some abnormality that warrants a psychiatric label. From a perspective of gender bias, the DSM categories of personality disorders that have been most criticized are dependent personality disorder (DPD), histrionic personality disorder (HPD), borderline personality disorder (BPD), and antisocial personality disorder (APD).

Numerous critics of the DSM have argued for decades that a diagnosis of DPD or a diagnosis of HPD is simply applied to a woman’s strict conformity to the stereotypic feminine gender role (e.g., Brown, 1992; Gibson, 2004; Rivera, 2002; Walker, 1993). The diagnostic criteria for DPD mainly involve “a pervasive and excessive need to be taken care of that leads to submissive and clinging behavior and fears of separation” (American Psychiatric Association, 2000, p. 721). The diagnostic criteria for HPD mainly involve “pervasive and excessive emotionality and attention-seeking behavior” (American Psychiatric Association, 2000, p. 711). It is not surprising that these diagnostic labels are far more commonly applied to women than to men (American Psychiatric Association, 2000; Cosgrove & Riddle, 2004). These diagnoses are problematic partly because the determination of whether observed behaviors and needs are “excessive” is left up to the individual practitioner’s decision about what is normal and what is abnormal. Research has demonstrated that clinicians’ diagnostic impressions are influenced by gender-role stereotypes (Cook, Warnke, & Dupuy, 1993; Cosgrove & Riddle, 2004). In particular, studies have shown that clinicians assign diagnoses differentially depending upon a client’s gender, such that different diagnoses are applied to women and men whose symptoms are the same. For example, in a classic study of 65 clinicians, Hamilton, Rothbart, and Dawes (1986) used a series of written case histories that described identical symptoms in male and female clients and found that the diagnosis of HPD was applied significantly more often to the women’s case histories than to the men’s. It seems that gendered traits, such as dependency and emotionality, not only are stereotypically associated with women but also lead disproportionately to the diagnosis of certain psychiatric disorders among women.

The extension of traditional feminine stereotypes into abnormality can be seen as an overt example of the “double bind” in which women often find themselves. Women are required to fulfill the cultural mandates of femininity through self-sacrifice, passivity, and deference to others, but these same characteristics are regarded as signs of weakness, vulnerability, and even mental illness (Caplan, 1995; Walker, 1993). For example, the diagnostic criteria for DPD include going “to excessive lengths to obtain nurturance and support from others, to the point of volunteering to do things that are unpleasant” (American Psychiatric Association, 2000, p. 725), but the cultural imperative for women to volunteer their time in the service of others is a defining facet of the feminine gender role (Bepko & Krestan, 1990; Jack, 1991; Worell & Remer, 2003). Furthermore, many responsibilities that women are typically expected to perform (e.g., housecleaning, diaper changing) are unpleasant but also necessary.

The diagnostic criteria for HPD include the following: “consistently uses physical appearance to draw attention to self” (American Psychiatric Association, 2000, p. 714). Because girls and women in our society are socialized to use physical appearance as a means of asserting self-worth (Bepko & Krestan, 1990), this criterion too can be seen as an extension of the stereotypical feminine role. Certain sub-groups of men are also socialized to use physical appearance to draw attention to themselves. For example, male bodybuilders are expected to seek such attention, but most lay people and most therapists would not readily consider bodybuilders to be histrionic. Moreover, in her study of gender-related stereotypes in personality disorders, Landrine (1989) found that nearly all study participants reported that a histrionic person was most likely to be a woman.

The diagnosis of BPD, also made far more frequently in women than in men (American Psychiatric Association, 2000; Cosgrove & Riddle, 2004), has been criticized for its lack of specificity and validity and for its use as a pathologizing label applied to women’s understandably distraught responses to abuse. The diagnostic criteria for BPD include “a pervasive pattern of instability of interpersonal relationships, self-image, and affects, and a marked impulsivity” (American Psychiatric Association, 2000, p. 706). In their study of gender bias in diagnoses, Becker and Lamb (1994) presented over 1,000 clinicians with descriptions of either a male or a female case with an equal number of symptoms from the criteria for both borderline personality disorder and post-traumatic stress disorder. The study revealed that the clinicians were more likely to diagnose the female clients with borderline personality disorder and the male clients with post-traumatic stress disorder even though the symptoms for the male and female clients were identical. These findings are important for a number of reasons, most notably that they demonstrate a lack of reliability in the diagnosis of these conditions and that they show that the diagnosis that involves a reaction to external causes (PTSD) was applied more frequently to male clients, whereas the more de-contextualized diagnosis (BPD) was applied more frequently to female clients. People whose emotions are considered to be reactions to external causes are less likely than others whose emotions are not as clearly linked to external causes to be regarded as qualitatively different from most people. Thus, they are less likely to be pathologized and isolated.

Perhaps the most problematic element of bias in the diagnosis of BPD concerns the extremely high percentage of clients with this label who have a history of physical and/or sexual abuse (Fish, 2004; Paris, 1994; Perry, Herman, van der Kolk, & Hoke, 1990; Westen, Ludolph, Misle, Ruffine, & Block, 1990). Because practitioners often neglect to ask – or choose not to ask – whether clients have ever been abused (Brown, 1991; Gallop, McKeever, Toner, Lancee, & Lueck, 1995), there is a high likelihood that the kinds of behavior used to arrive at a diagnosis of BPD are regarded as intrinsic to the client rather than as reactions to abuse. We explore this issue in more depth in a later section in reference to PTSD. Readers should keep in mind that borderline personality disorder is the most commonly diagnosed of all personality disorders (Frances & Widiger, 1987; Kroll, 1988), and it is also one of the most stigmatized (Nehls, 1998; Simmons, 1992). Therefore, there are probably large numbers of abused women who carry this derogatory label and who may be unaware that the problems they are experiencing stem from past and/or current abuse. The damage that biased diagnostic categories can create in women’s lives is obvious, as they are deprived of the treatment, support, advice, and empathy they sorely need.

It should be noted that the diagnostic criteria for certain personality disorders are biased against stereotypical masculine characteristics as well. Although less attention has been paid to bias against men, it has been argued that the diagnosis of antisocial personality disorder is an example of this. Just as dependent and histrionic personality disorders are seen by some to be extreme forms of the feminine gender role, antisocial personality disorder (APD) is seen as the extreme form of the masculine gender role. The diagnostic criteria for antisocial personality disorder include “a pervasive pattern of disregard for, and violation of, the rights of others” (American Psychiatric Association, 2000, p. 701). Because women are stereotypically expected to play the role of looking after others, the corresponding expectation for men is that they do not have to attend to others’ needs and that they can expect women to attend to their needs (Bepko & Krestan, 1990). In APD, a condition that is more commonly diagnosed in men than in women (American Psychiatric Association, 2000; Cosgrove & Riddle, 2004), we see an exaggeration of that expectation. One of the specific criteria for this condition includes extreme aggressiveness, a hallmark trait of hypermasculinity (Lips, 2001). Because of the apparently gendered nature of this diagnostic category, it is possible that clinicians may be “primed” by their gender stereotypes to expect that simply being male makes some clients likely candidates for the antisocial personality diagnosis. For example, in one study (Belitsky et al., 1996), 96 male and female resident psychiatrists were presented with case descriptions of either male or female clients who reported sets of identical symptoms. The results revealed that significantly more male than female clients were given the diagnosis of APD by both male and female psychiatric residents. In her study of gender stereotypes and personality disorders, Landrine (1989) presented participants with verbatim descriptions of DSM disorders without the diagnostic labels and asked them to predict for each disorder the demographic characteristics of the person who would fit that description. Twenty-one of the 23 participants stated that the person with APD was male. Landrine explained that the reason for this finding, and for similar findings of stereotypes in diagnosis, “may be that gender-role categories and personality disorder categories are simply flip sides of the same stereotyped coin” (p. 332). As Cale and Lilienfeld (2002) noted, the gender stereotyped nature of the APD category is also evidenced by the fact that its criteria are more reflective of men’s typical criminal behavior, such as rape, robbery, and pedophilia, than of women’s typical criminal behavior, such as child abuse, shoplifting, and prostitution. It is clear that gender bias in the diagnosis of personality disorders can affect both male and female clients.

Mood Disorders

The most commonly diagnosed psychiatric disorder is major depressive disorder (MDD). The diagnostic criteria for major depression include “depressed mood or the loss of interest or pleasure in nearly all activities” (American Psychiatric Association, 2000, p. 349). It is estimated that twice as many women as men are diagnosed with depression (Kessler, 2003; Stoppard, 1993; see also Chapter 7), and despite the lack of research evidence to support the notion that women’s hormones increase their likelihood of experiencing depressed mood, hormonal factors are very often named among the posited explanations for the higher depression rate in women (Hankin, & Abramson, 2001; Whiffen, 2006). Feminist authors have argued against the medicalization of women’s emotional distress through the label of “depression” and, instead, support an approach in which women’s distressing experiences and life circumstances are more fully explored and articulated (Gammell & Stoppard, 1999; Jack, 1991). For example, Gammell and Stoppard (1999), in reporting their findings from interviews with women who had been diagnosed with depression and were prescribed anti-depressants, concluded that medical practitioners’ pathologizing of the women’s reactions to stressful situations served to disempower the women and led them to believe that they had little control over their life circumstances. The authors stated that “a medicalized understanding and treatment of women’s depressive experiences cannot readily coexist with personal empowerment” (p. 112). The disempowerment of women whose everyday lives are fraught with stress and trauma is particularly problematic, given that such women need to recognize their own sources of strength and resilience in order to alter their life situations. Therefore, the label of depression not only draws attention away from contextual factors in women’s lives but can also compromise their abilities to make positive changes that could support their psychological well-being. Furthermore, as Cohen and Jacobs (2007) pointed out, research on treatments for depression by examining only the symptoms of depression and not clients’ own personal histories or life circumstances – which usually focuses on drugs – further disempowers depressed individuals. The authors stated that “in the conventional medical framework of psychiatric drug treatment research, the patient’s own voice is either eliminated or relegated to a distinctly inferior position” (p. 46). They further argued that depression researchers should study clients’ individual stories rather than reducing their experiences to specific symptom profiles. Many women’s stories include histories of abuse and other forms of victimization, as well as social factors, such as poverty and discrimination, all of which can serve as precipitating factors for psychological distress (McGrath, Keita, Strickland, & Russo, 1990). Labeling a woman as “depressed” serves to mask the presence of possible factors in her life that should be the focus of work in psychotherapy.

Depression is also important to any discussion of gender bias because it can be argued that the large number of women in the general public who can potentially be labeled as suffering from depression represents a segment of the population that is fervently targeted by pharmaceutical marketing. In the 1960s, psychotropic drugs were marketed not to consumers but to physicians in the form of print advertisements in medical journals. In these advertisements, the diagnostic criteria for depression were portrayed in part through the visual representations of depressed women. The gender-typed nature of these representations is striking in its exploitation of stereotypes of happy and unhappy women. In his book Prozac on the Couch, Metzl (2003) examined these representations and illustrated how, by 1970, the stereotype of the unmarried woman came to portray the typical unhappy person in need of medication. He cited as an example the 35 and Single advertising campaign for Valium in which “Jan,” one of the “unmarrieds with low self-esteem,” “realizes…that she may never marry” and comes to her physician for treatment – a treatment that undoubtedly includes Valium. For physicians, the women portrayed in these advertisements served as prototypes of the mentally ill patient and of the benefits of appropriate treatment for mental illness. Drug therapy increasingly became the treatment of choice for depression among psychiatrists and general practitioners alike (Metzl, 2003).

Similar manifestations of the depressed person now appear in direct-to-consumer advertising on television and in magazines. Here the symptoms necessary for the diagnosis of depression are represented as people – most often women – who suffer not only from sadness but also, as is explicit in recent advertising campaigns, from vague physical symptoms, such as pain and fatigue. Consumers are expected to conclude that not only might they be depressed, but the solution to their depression is a pharmaceutical drug. Clients and practitioners, therefore, have become (often unwittingly and unknowingly) complicit in the propagation of drug therapy for treating suspected depression. The emphasis on anti-depressants is a serious concern in light of evidence of the negative effects of the antidepressants currently on the market, effects that include dizziness, headaches, gastrointestinal problems, insomnia, difficulties in sexual functioning, risks during pregnancy, and suicidal ideation (Breggin, 2007; Chrisler & Caplan, 2002; Gartlehner et al., 2008; Healy, 2004).

Depression that occurs during the postpartum period has been the focus of much attention of feminist researchers. The mainstream media and the general public presume a biological cause – namely hormones – for women’s experience of depression following the birth of a child (Martinez, Johnston-Robledo, Ulsh, & Chrisler, 2000). However, there has yet to be any consistent evidence of a direct link between postpartum depression and women’s hormone levels (Whiffen, 2004, 2006). Moreover, research has demonstrated that new fathers also experience an elevated rate of depression (Ballard, Davis, Cullen, Mohan, & Dean, 1994; Vandell, Hyde, Plant, & Essex, 1997). In light of these findings, Whiffen (2004) suggested that our understanding of depression that occurs after the birth of a child should be informed by an analysis of the social system and relational dynamics of the couple rather than by an assumption of maternal hormonal imbalance.

The diagnosis of premenstrual dysphoric disorder (PMDD) has also been criticized for gender bias. PMDD appears in the appendix section of the DSM-IV-TR as a diagnosis listed for further study. However, it is also named in the section “Mood Disorders” as a Depressive Disorder Not Otherwise Specified, despite the fact that the diagnostic criteria for PMDD can be met without the presence of depressed mood. The criteria described for PMDD include marked lability of mood, decreased interest in usual activities, difficulty in concentrating, fatigue, appetite changes, difficulties in sleep, feeling out of control, and physical symptoms, such as breast tenderness, headaches, and bloating sensations, all of which occur premenstrually (American Psychiatric Association, 2000, p. 771). The diagnostic description states that “[t]he presence of the cyclical pattern of symptoms must be confirmed by at least 2 consecutive months of prospective daily symptom ratings” (American Psychiatric Association, 2000, p. 772), which implies that there is a sound, systematic grounding to the derivation of the PMDD criteria. However, very little research exists to support the notion that PMDD is a confirmed clinical entity (Caplan, 1995, 2004; Chrisler & Caplan, 2002; Cosgrove & Caplan, 2004; O’Meara, 2001), and the European Union’s Committee for Proprietary Medicinal Products has formally reached the conclusion that it is not a recognized “disorder” (Moynihan, 2004). Furthermore, research calls into question the existence of PMDD. In a study by Gallant and colleagues (Gallant, Popiel, Hoffman, Chakraborty, & Hamilton, 1992), when three groups of research participants (women who reported premenstrual problems, women who did not report such problems, and men) completed a checklist of symptoms for late luteal phase dysphoric disorder (the DSM-III term for PMDD), no differences in symptom ratings were found between the three groups. Moreover, the DSM authors themselves have stated that “[t]here has been very little systematic study on the course and stability of this condition” (American Psychiatric Association, 2000, p. 772).

The inclusion of PMDD in the current DSM (along with the inclusion of late luteal phase dysphoric disorder in an earlier edition) has been much criticized. Critics have pointed not only to the lack of research support for the existence of this condition but also to the influence of the pharmaceutical industry on the invention and development of the PMDD diagnosis (Caplan, 1995, 2004; Chrisler & Caplan, 2002; Cosgrove & Caplan, 2004). These criticisms involve the close connection between the positioning of PMDD as a psychiatric condition and the Food and Drug Administration’s 1999 approval of Eli Lilly’s drug Sarafem as an appropriate treatment for PMDD. Sarafem is actually Prozac; they are simply two different labels for the identical drug (i.e., fluoxetine hydrochloride). In 1999, Eli Lilly’s exclusive patent on Prozac was about to expire; unless the drug company could find a psychiatric condition different from depression for which Prozac could also be approved, the company would face a loss of profits. At that time, Eli Lilly sponsored a roundtable discussion on the topic of PMDD, which was the basis for a journal article entitled Is Premenstrual Dysphoric Disorder a Distinct Clinical Entity? (Endicott et al., 1999). The authors of that article reported that very recent evidence showed not only that PMDD is a “real” disorder but also that Prozac is an effective treatment for it. In fact, little of the evidence cited in that report was very recent, and none of the studies in question showed clear support for the existence of PMDD as a clinical entity (Caplan, 2004), let alone a psychiatric disorder that is best treated with an antidepressant medication. Nonetheless, based on the recommendations of the roundtable members – many of whom were also on the DSM subcommittee for PMDD and had received research funding from Eli Lilly – the FDA deemed Prozac to be an effective and safe treatment for PMDD.

We can presume that many women who are prescribed Sarafem do not know that it is Prozac, as that fact is not made clear in the direct-to-consumer advertisements that women see; perhaps some would choose not to take the drug if they were aware that it is an antidepressant. Furthermore, labeling and treating these women as though they are mentally ill (by prescribing for them a drug that is identical to Prozac in every way except that Sarafem is packaged in pink and purple) suggests that their bodies and minds – presumably due to the effects of some hormonal abnormality – are defective and that their emotional suffering is not grounded in daily life. However, research has shown that women who report premenstrual problems are significantly more likely than other women to be in upsetting life situations, such as enduring domestic abuse or being mistreated at work (Caplan, 1995; Golding & Taylor, 1996; Golding, Taylor, Menard, & King, 2000; Taylor, 1999). A focus on prescribing a drug can lead therapists to ignore external sources of problems and could put women at risk of remaining in unsafe situations.

Men are also subject to diagnostic labels related to the effects of hormones, especially labels concerning sexual function and dysfunction. The creation of the term andropause, the male version of menopause in which men’s hormones decrease with age, demonstrates society’s discomfort with men who do not fit the hypersexual masculine role. Testosterone treatment is the drug therapy usually recommended for andropause, and the research concerning testosterone has shown mixed results. Although some studies have shown that testosterone is not an important determinant of cognitive, psychological, or sexual functioning, others suggest that many psychiatric symptoms accompany a decrease in testosterone (Caplan, 1995). Furthermore, research on the validity of the andropause diagnosis is lacking, and no studies to date have examined whether men themselves are comfortable with accepting this label.

A similar example of the medicalization of the need for hypermasculinity is the diagnosis of muscle dysmorphia. Muscle dysmorphia does not appear in the DSM, but it is considered by some to be a form of body dysmorphic disorder, a DSM diagnosis characterized by preoccupation with a perceived defect in physical appearance (Pope, Phillips, & Olivardia, 2000). In muscle dysmorphia, usually seen in men, misperceptions and obsessive thoughts concern the individual’s desire for muscularity. Pope et al. (2000) suggested a strong role of culture in the development of these thoughts. “Society and the media preach a disturbing double message: A man’s self-esteem should be based heavily on his appearance, yet by the standards of modern supermale images, practically no man measures up” (p. 13). Similarly, Cafri et al. (2005) reported that media influences play a key role in precipitating muscle dysmorphia and also that the adolescent boys most at risk for muscle dysmorphia are those who experience teasing and are striving for increased peer popularity. There are considerable similarities between muscle dysmorphia in men and eating disorders in women. For example, McCreary and Sasse (2000) found, in their study of adolescent boys and girls, that the boys with a high drive for muscularity were also the boys with the lowest self-esteem and highest depression in their sample, a finding that parallels psychological correlates of disordered eating in girls. In his conceptual model of muscle dysmorphia, Grieve (2007) described perfectionism as a contributing psychological factor in the development of this condition; perfectionism has also been found in empirical studies to correlate with symptoms of disordered eating in girls (Hewitt, Flett, & Ediger, 1995). The risks associated with what has been termed muscle dysmorphia, including the use of anabolic steroids and unhealthy eating patterns, can severely compromise an individual’s health (Pope & Katz, 1994). However, if muscle dysmorphia is formally added to the DSM, it is unclear how helpful its designation as a mental disorder would be. Not only would a psychiatric diagnosis add a stigma to the experiences of the men who meet the diagnostic criteria but there would also be an increased likelihood that some drug therapy would be encouraged for them. As with other conditions, attention should be paid to the broader social and cultural factors at play rather than to a presumed mental defect within individuals.

Anxiety Disorders

Women more often than men receive diagnoses of panic and phobic disorders (American Psychiatric Association, 2000; Cosgrove & Riddle, 2004). The cluster of symptoms labeled “agoraphobia” is one category diagnosed far more frequently in women than in men (American Psychiatric Association, 2000; Cosgrove & Riddle, 2004). The criteria for agoraphobia include “anxiety about being in places or situations from which escape might be difficult…or in which help may not be available” (American Psychiatric Association, 2000, p. 429). In agoraphobia, according to the DSM, the afflicted individuals’ anxiety causes them to avoid certain situations and to restrict their behavior. Although the focus of this diagnosis in the DSM system is the set of symptoms, theorists have argued that the central issue in understanding supposed agoraphobic behavior in women is an analysis of why there are so many fear-inducing spaces and situations confronting women. For instance, Holmes (2008) emphasized the need to normalize the agoraphobic response by considering the avoidance of certain spaces and situations to be self-protective for women who feel unsafe and vulnerable among strangers and away from familiar people who could help them should a threat arise. Such normalization could be a useful re-framing of the responses of women who have been victims of violence and, thereafter, fear being in public on their own. Indeed, research indicates elevated rates of physical and sexual abuse among women who receive the diagnosis of “agoraphobic” (Langeland, Draijer, & van den Brink, 2004; Raskin et al., 1989).

In her critique of dominant conceptualizations of agoraphobia that pathologize women who fear to enter public spaces, Bankey (2004) argued that we should focus instead on “a spectrum or continuum of experiences that connect our anxieties and fears to the external worlds in which we live” (p. 348). She favored an approach that attends to the realities that confront women whose fears of venturing into the outside world alone may be rooted in previous trauma. Similarly, McHugh and Cosgrove (2004) stated that we must consider both history of trauma and women’s gender-role socialization. They pointed out that being fearful is part of the stereotypical feminine gender role and that women’s socialization to be fearful, along with the real threat of possible victimization, can predispose women to adopt behaviors that match the DSM symptoms of agoraphobia. However, as those authors pointed out, labeling such fear as a mental illness is not a productive way to help women who are concerned for their safety because it discounts the seriousness of threats that could confront them in their daily lives.

The diagnosis of post-traumatic stress disorder (PTSD) has also been critiqued for gender bias. The diagnostic criteria for PTSD include “intense fear, helplessness, or horror… and numbing of general responsiveness” following exposure to an extreme traumatic stressor, as well as “persistent reexperiencing of the traumatic event” (American Psychiatric Association, 2000, p. 463). Post-traumatic stress disorder has been a controversial diagnosis since it was introduced prior to its addition to the DSM in 1980. The category was first introduced as a means of drawing attention to the suffering of Vietnam War veterans who were experiencing a host of trauma-related problems (Linder, 2004). The diagnosis was later broadened to apply to a large range of traumatic experiences, including violent assault, natural disasters, car accidents, and diagnosis of a life-threatening physical illness.

The main controversy about PTSD concerns its use to label those individuals who have suffered extreme trauma as mentally ill. The psychiatric diagnosis of a traumatized person focuses attention on the supposed weaknesses within the individual rather than on the traumatizing situation itself (Becker, 2004; Caplan, 2006). This critique is often discussed in reference to abused women, but it applies to combat veterans of both sexes as well. Theorists have argued that focusing on the psychopathology of victims of violence not only discounts the potentially adaptive – and indeed fundamentally human – nature of their stress responses (e.g., the diagnostic criteria include such responses as “persistent avoidance of stimuli associated with the trauma,” “increased arousal,” and “hypervigilance,” all of which could adaptively help one to avoid further trauma) but also may be detrimental to their recovery (Caplan 2006; Linder, 2004). One consequence of a label of mental illness is that it may compound a victim’s sense of self-blame (e.g., “If I am mentally and emotionally flawed, I may have brought this situation upon myself,” “I am not feeling fine right now because I am inadequate or sick”). This self-perception can, in turn, erode people’s sense of agency and, thereby, neutralize the inner resources they require to escape continued abuse. Furthermore, when a practitioner labels such responses as mental illness, there is implicit in that label the assumption that, regardless of the severity of the stressor, the client should be able to get “back to normal” and to leave the effects of the trauma in their past. Because mental illness denotes abnormality, a diagnostic label carries with it a judgment that deems the thoughts and behaviors in question to be out of the range of normal responses – even when those thoughts and behaviors were precipitated by extreme trauma. That implicit judgment keeps some male combat veterans from seeking a psychotherapist’s assistance because it suggests that they are less masculine or less effective at their work than they should be. After all, a “real man” or a “good soldier” should be able to cope bravely with any war-related experience.

From a broader perspective, an emphasis on labeling trauma victims’ responses as psychiatric symptoms rather than the results of trauma can distract clients, practitioners, and society in general from the social injustices of victimization. Such social ills as the brutality of war, torture, and violence against women go unexamined, and, instead, victims’ mental and emotional states are scrutinized and discussed. The public is thus excused from the responsibility of rectifying the various forms of oppression and violence that lead to psychological trauma. The internal focus on the mind of the sufferer indicates that treatment of that mind is the necessary remedy, and the need for social and political change remains largely invisible.

Some Issues That Cut Across Diagnostic Categories

In addition to those sources of gender bias that are specific to particular diagnostic categories, there are major problems in the lives of women and men that are relevant to many diagnostic categories and to our consideration of the DSM in general. These include poverty, various forms of violence, and issues related to race and racism.

Poverty

Because the DSM is written from a mainstream psychiatric vantage point, it presents a medicalized view of experience, thus psychological distress is explained as pathology that originates within the individual rather than in social ills. However, there is consistent evidence that social and economic factors have a tremendous impact on psychological suffering. Living in poverty has been identified as an important factor that can increase an individual’s likelihood of emotional distress (Belle & Doucet, 2003; Lorant et al., 2003), and poverty itself disproportionately affects women and children (U.S. Census Bureau, 2002). Because poor women in the United States usually have no health insurance, they are generally not seen in psychiatric settings until their distress is severe (Chalifoux, 1996). Possibly as a consequence of this delay, as well as because many clinicians do not believe that poor clients benefit from talk therapy, poor clients are more likely than clients who are not poor to be given psychoactive drugs to manage their distress (Bullock, 2004; Judd, 1986; Killian & Killian, 1990).

In their encounters with the psychiatric establishment, poor women must also contend with widely held stereotypes of low-income people as lazy, morally weak, and not intelligent enough to benefit from psychotherapy (Bullock, 1995; Lott, 2002; Seccombe, 1999). As Bullock (2004) pointed out, mental health professionals often do not understand the realities of poverty, including the absence of safe living conditions and difficulties associated with transportation and basic survival, such as providing for one’s children. Consequently, therapists may be unaware that the anger, hostility, and shame they see in the behavior of poor clients may be directly caused by poverty. The shame associated with being poor may be further compounded by a client’s awareness that her therapist comes from a position of privilege. This subtext of shame can further influence the client’s behavior and leave her vulnerable to diagnostic labels that pathologize her reactions.

In her book Poor-Bashing: The Politics of Exclusion, Swanson (2001) described the insidious ways in which policies and public attitudes concerning the poor perpetuate the myth that poverty is a social ill caused by the poor themselves. This victim blaming on the part of the middle and upper classes contributes to self-blame among low-income individuals and to their heightened awareness of their personal failures and shortcomings. Swanson argued that one of the greatest threats to the well-being of poor individuals is the belief that the poor are somehow less deserving than others of the services commonly associated with privilege; these include preventive services that promote psychological wellness. As long as mental healthcare for the poor is construed only as a means of responding to severe psychiatric crises, the biases against them will persist, and their care will remain oriented to drug treatment rather than to supportive services and the need for social and structural change.

Violence and Abuse

As discussed earlier, one risk of using the DSM system involves applying a label of mental illness to those who have been victims of violence. Certainly, if people who have experienced rape, incest, or other forms of abuse were excluded from the number of individuals given a psychiatric diagnosis, the reported prevalence of many psychiatric disorders would drop dramatically. Abused women are at risk of receiving psychiatric diagnoses partly because they tend to express anger more readily than non-abused women do (Jack, 1999). However, psychiatric labels are also applied frequently to women who have experienced abuse even when such women are seeking help solely for the mental suffering that results from having been a target of violence. Their abusers, on the other hand, are often seen in psychiatric settings only if mandated to be there by the courts. Consequently, the supposed mental illnesses of abused women rather than the mental states of their abusers are more commonly discussed. Moreover, because severely distressed women may be institutionalized after having sought help, whereas their abusers are institutionalized only if charges are pressed against them, the individual who is deemed unfit to reside in society might be the victim rather than the perpetrator. We should also bear in mind that a woman who is institutionalized for mental illness is viewed as being defective as a person, whereas a man who is imprisoned or paroled for violent abuse is punished because of his particular actions. This distinction is all the more damaging when we consider that women are socialized to internalize labels that are assigned to them by others and not to question medical and other authorities (Bepko & Krestan, 1990). Therefore, the application of a psychiatric label to an abused woman can present a risk to the woman’s sense of self. In the case of men who are victimized, a psychiatric label is particularly damaging because victimization runs contrary to the masculine gender role and is experienced as especially shameful. There is a need for greater recognition among practitioners that men can also be victims of abuse, and those victims must deal with a constellation of physical and psychological effects in a social climate that requires them to act strong at all times and not express their suffering. Thus, for both women and men, there are risks associated with shame and self-blame when attention is paid to their presumed mental weakness rather than to the effects of trauma itself.

Race and Racial Discrimination

As much as bias in psychiatric diagnosis affects female clients in general, the problems of bias are often even more pronounced for racialized women. Because of the dominance of a White, Western viewpoint in psychiatry, the actions and thoughts that are most likely to be considered normal by the DSM authors and others in the mental health establishment are those that are acceptable within mainstream White society (Caplan, 1995; Russell, 1994). Thus, kinds of behavior that may be commonplace among people of color and within non-dominant cultures are often labeled “abnormal” (Ali, 2004). At the same time, there is a risk that real emotional suffering is overlooked by mental health professionals when they attempt to apply psychiatric labels to women of color. For instance, a practitioner who holds a stereotype of Asian women as quiet and passive may miss signs of extreme sadness and emotional suffering if an Asian woman is quiet and withdrawn; quiet behavior may be attributed to cultural norms rather than to suffering. More broadly, the very idea of psychiatric categorization, treatment, and institutionalization runs counter to the traditions and beliefs of cultures that value a holistic and broadly encompassing view of wellness for men and women. However, in psychiatric settings, women of color who resist treatment are at risk of being viewed as difficult patients (Greene, 1995).

We also must consider the interaction between sexism and racism in the construction and use of the DSM categories. Examples of this interaction have been found in the DSM-IV Casebook (Spitzer, Gibbon, Skodol, & Williams, 2002), a widely used book designed for training in the DSM approach. In the Casebook, women of color are described in more sexualized terms than are men or White women (Cermele, Daniels, & Anderson, 2001). These representations are evidence not only of racist bias within the DSM but also of the potential for the DSM to shape trainees’ stereotypes of women of color.

Another risk associated with race in the use of the DSM system is its lack of consideration of the psychological damage that can be inflicted by racism. As with poverty and abuse, racism encountered by women who are also psychiatric clients can be readily “explained away” by the clients’ diagnoses. A client who is deemed mentally ill – especially a female client – is in danger of having her every perception and estimation of her own experience challenged (Caplan, 1995). Therefore, when a person of color describes the emotional distress of racist encounters, the focus is often not on the racism itself but on the underlying pathology that is causing him or her to feel distressed (Ali, 2004; Pauling & Beaver, 1997; Root, 1996). Racist bias, therefore, inflicts on individuals from non-dominant backgrounds an additional threat of harm to their emotional well-being when they seek help within the psychiatric system.

Stereotypes Concerning Sexual Orientation and Identity

There is evidence that bias in the DSM’s medicalized approach extends beyond social and racial arenas into sexuality. As in other forms of psychiatric bias, elevated rates of diagnosed conditions are presumed to indicate that greater numbers of individuals in a given group are mentally flawed or psychologically ill. In the case of LGBT individuals, in which researchers have found higher rates of depression, suicide, and substance abuse than in members of the general population (Cochran & Mays, 2006), the presumption of psychological weakness is often a reflection of homophobic bias in the psychiatric system. For example, although in the 1970s the DSM authors said that they would remove homosexuality from the next edition (Stein, 1993), in fact, they did not do so. It remained listed as “ego dystonic homosexuality,” a diagnosis that pathologized the consequences of a homophobic, intolerant society that made it difficult for gay men and lesbians to be totally comfortable with their sexual orientation (Caplan, 1995). Despite the ultimate removal of the “disorder” from the DSM, research on psychological issues of LGBT individuals has been skewed by recruitment practices that often draw research participants from help-seeking samples rather than from community samples (Cochran & Mays, 2006). In addition, the focus on diagnostic categories leads clinicians and researchers to ignore the psychological stress that can stem from the lack of social (and/or familial) support, social isolation from peers, and the social stigmatization of being an oppressed minority. Research has shown that the internalized oppression often experienced in the LGBT population is correlated with depression (Majied, 2003). Furthermore, a study (Jones, 2001) of depression, alcoholism, and substance abuse among older LGBT persons revealed that a primary trigger for these conditions is the devastating effect of the AIDS epidemic that has left many LGBT elders with only small support networks, as they have lost many friends and community members. Therefore, there is a need for research on mental health issues among LGBT individuals to address the realities of social, familial, and societal stressors.

Gender-Role Factors in Seeking Treatment

For men in particular, shame and stigma are associated with seeking help in the form of psychiatric treatment. Men are socialized to keep their emotions to themselves and not to express feelings of sadness, anxiousness, or fear (Bepko & Krestan, 1990; Lips, 2001; see also Chapter 10). For both men and women, seeking a psychotherapist’s help is often seen as a sign of mental and emotional weakness and as an indication of an inability to cope with one’s problems. Furthermore, mental illness can be under-diagnosed in men because many forms of mental illness are assumed to affect only women. In a classic study of the influence of gender-related stereotypes on ideas about mental illness, Broverman, Broverman, Clarkson, Rosenkrantz, and Vogel (1970) found that mental health professionals’ ideal of the mentally healthy person was similar to their ideal of the mentally healthy man, but that their descriptions of what constitutes a mentally healthy woman were quite different from their descriptions of what constitutes a mentally healthy person in general. These findings indicate that it is more challenging for the typical woman to be considered mentally healthy than it is for the typical man, and this remains true today, as Seem and Clark (2006) recently replicated the Broverman et al. study.

More recent studies have indicated that both women and men are more likely to be regarded as mentally ill if they behave in ways that are inconsistent with their respective gender stereotypes (Rosenfield, 1982; Waisberg & Page, 1988). In addition, Landrine (1988) found that participants in her undergraduate sample described depressed people in more feminine terms and non-depressed people in more masculine terms. Because masculine and feminine stereotypes differ in important ways, different types of behavior are considered acceptable for men than for women. For instance, as Doyle (1995) described, aggressive behavior on the part of men is less likely to be considered pathological than is women’s aggressive behavior. Women’s own affective states may also be influenced by the awareness of certain diagnostic categories that presume emotional vulnerability in women. For example, Nash and Chrisler (1997) found that undergraduate women who read a description of the PMDD diagnostic criteria reported more negative affect after reading the description than did men who read the same description. In terms of help-seeking behavior, we can assume that women’s exposure to assumptions of women’s psychological weakness (as presented by the media and by the public) leads women to seek psychiatric treatment more readily than do men.

Future Directions

In this chapter we argue that the use of gender-biased psychiatric diagnoses obscures the realities of people’s lives. It is important to consider the risks that may be associated with treatment decisions entailed by the DSM’s de-contextualized emphasis on an individual’s symptoms over the examination of sociocultural factors. The focus on specific psychiatric symptoms in the DSM diagnostic system is an attempt to divorce mental illness from the broader contexts of clients’ lives on the assumption that, regardless of the particular environments they inhabit, individuals who fall within a particular diagnostic category have in common precisely that: their diagnosis. However, the very belief that individuals can be understood outside of their life contexts is highly questionable. Some observers (Caplan, 1995; Kutchins & Kirk, 1997) of the process by which the DSM is created have argued that the DSM itself is a formalization of prevailing norms of acceptability as determined by the dominant groups in society (that is, male, White, heterosexual, economically privileged people). We contend that future research should be directed toward uncovering alternate views of understanding and conceptualizing mental and emotional suffering.

One important area of future research is investigation of the effectiveness of community involvement and advocacy in helping individuals to take an active role in changing the circumstances of their lives and the lives of others. Helping others can be personally empowering, can lead to an increased sense of self-efficacy (McWhirter, 1994), and may be, for some, a favorable option over traditional psychotherapy. In addition, researchers should investigate specific effects of psychiatric labeling with respect to the damage to a client’s sense of self after having received a label, as well as the effects of community-based alternatives to traditional mental health services that do not rely on diagnostic labels. If psychiatry is positioned to dictate what kinds of behavior are acceptable and unacceptable, as well as what means are most effective for attaining normalcy, then both clients and non-clients must accept that its system of defining and remedying problematic behaviors is superior to any other approaches. However, by locating illness in the individual, mental health professionals neglect to acknowledge the realities of people’s lives and the oppression encountered by the disadvantaged. Moreover, when medications are routinely favored over social and community-based interventions and other means of creating personal, interpersonal, and contextual change, clients are taught that their problems lie within them. Thus, no change is needed in the status quo.

Research should also be directed at systematically exploring the influence of social inequality and oppression on psychological well-being. If we shift the focus of research away from diagnostic entities and toward the broad causes of human suffering, we can discover innovative means of alleviating distress. For example, as we consider gender bias in the DSM, we must bear in mind that the over-representation of women across numerous diagnostic categories is the result of several interactive forces. It is not solely scientific and diagnostic bias that is responsible for women being deemed mentally ill more often than are men. If, in our society, women’s lives disproportionately include suffering and high levels of stress, then it should be expected that women will experience more psychological distress than men do. However, what the creators of the DSM ignore is the crucial and necessary consideration of why women’s lives so often are fraught with emotional anguish. It is true that psychiatric treatment has been of benefit to many who have suffered psychological distress. The dangers of the system arise when scientists, practitioners, clients, and the public become blind to – and, therefore, do not challenge – the unscientific and biased nature through which individuals are labeled and their problems are treated.