“I’m Not Mentally Ill”

Abstract

Mental illness identity deflection refers to rebuffing the idea that one is mentally ill. Predictors of identity deflection and its consequences for well-being were examined for individuals with mental disorders in the National Comorbidity Study–Replication (N = 1,368). Respondents more often deflected a mental illness identity if they had a nonsevere disorder, had low impairment in functioning, had no treatment experience, viewed possible treatment as undesirable, and held multiple social roles, consistent with theory about stigma resistance. Persons who deflected a mental illness identity had lower distress and more positive affect than those who accepted it, even net of disorder severity, impairment level, and treatment experience. Among those who had ever been in treatment, deflection buffered the negative effects of serious impairment but exacerbated the effects of having a severe disorder on well-being, suggesting more complex consequences of formal labeling (greater stigma but helpful services), consistent with previous research.

Keywords

mental illness psychological well-being stigma stigma resistance

Modified labeling theory (Link 1987; Link et al. 1989) proposes that negative cultural stereotypes about persons with mental illness become relevant and threatening to individuals after they have been diagnosed and treated for a psychiatric condition. According to the theory, labeled persons anticipate devaluation or rejection by “normals” (Goffman 1963) and attempt to protect themselves by keeping their condition secret, withdrawing from social interaction, distancing themselves from other patients, educating people about mental illness, or challenging mental illness stereotypes (Corrigan and Watson 2002; Link et al. 2002; Shih 2004; Thoits 2011). Some of these coping strategies—secrecy, withdrawal, avoiding others—can have harmful consequences for labeled persons’ quality of life and psychological well-being (Kleim et al. 2008; Link, Mirotznik, and Cullen 1991; Link et al. 2001, 2002; Ritsher and Phelan 2004; Thoits and Link 2016; Vauth et al. 2007; Yanos et al. 2008). Other strategies are viewed as potentially beneficial to well-being, in particular, attempts to confront or challenge stigma (Corrigan and Watson 2002; Thoits 2011). Challenging promotes self-esteem, lower depressive symptoms, better quality of life, and a sense of empowerment (Link et al. 2002; Thoits and Link 2016; Wahl 1999), all of which are considered key outcomes in the recovery literature (e.g., Corrigan et al. 1999; Markowitz 2001, 2005).

Among the array of tactics for coping with the stigma of mental illness, the strategy of deflection has received the least attention and has inconsistent empirical support. Deflection is a form of cognitive resistance to stigma—one rejects, rebuffs, or refutes the idea that one has a mental disorder or that one holds the identity of “mental patient” (Thoits 2011). Belief that one is not mentally ill makes the existence of cultural stereotypes inapplicable and thus minimally threatening to the self. For this reason, deflecting a mental illness identity should maintain or perhaps even enhance a person’s psychological well-being. But findings have been mixed. One study showed that hospitalized patients who rejected the idea that they had mental illness had higher self-esteem and higher perceived quality of life than patients who accepted this idea (Thoits and Link 2016). However, two studies found no relationship between patients’ endorsement of this coping strategy and their well-being (Link et al. 2002; Ritsher and Phelan 2004), although in these two studies, measures of identity deflection were indirect.¹

This article reexamines identity deflection as a self-protective coping strategy. Specifically, both the determinants and the well-being consequences of deflection are investigated to assess ideas about this type of identity resistance (Thoits 2011). To date, the studies that examine this concept (variously termed deflecting, distancing, or stigma resistance) have been based on samples of patients who suffer from psychoses. Such samples are unfortunate because individuals with nonsevere, typically less impairing conditions should be more likely than patients with psychoses to rebuff the idea that they have a mental illness (Thoits 2011). The prevalence of this type of stigma resistance has probably been underestimated in research that includes only patients with severe mental illness. Moreover, the advantages of deflection for individuals’ well-being are harder to detect when variance in symptom severity is curtailed due to sample composition. To overcome these limitations, this article utilizes data from a nationally representative epidemiological survey (the National Comorbidity Survey–Replication [NCS-R]; Kessler et al. 2004), which allows comparisons between persons who have mild to moderately serious versus severe disorders.

Background

Antecedents and Consequences of Identity Deflection

Deflection involves blocking the application of a mental illness identity to the self, where identity refers to self-definitions in terms of one’s social roles and social attributes. Self-definitions are answers to the question, “Who am I?” (e.g., “I am a father, a financial consultant, Hispanic, a successful person”) (McCall and Simmons 1978; Stryker 1980). People who claim a mental illness–related identity, then, define themselves in terms of a psychiatric condition—“I am mentally ill/a mental patient” or “I have a mental disorder” (i.e., “I am a person who has a mental disorder”).² Those who refute the identity say the opposite: “I’m not mentally ill.”

Several factors should raise the likelihood that a person will deflect a mental illness–related identity (Thoits 2011). The first involves the relative severity of a person’s symptoms. Individuals with less serious disorders are likely to resist a mental illness identity because their symptoms (e.g., anxiety, suicidal thoughts, excess drinking or drug use) simply do not correspond to public stereotypes of “crazy” persons as unpredictable, uncontrollable, irrational, violent, incompetent, and unkempt (Link et al. 1999; Martin, Pescosolido, and Tuch 2000; Pescosolido et al. 1999, 2010; Phelan et al. 2000; Rabkin 1980; Stout, Villegas, and Jennings 2004; Wahl 1995). Even persons with severe disorders may perceive cultural stereotypes as extreme and thus inapplicable to themselves: their symptoms may match stereotypes in some ways (e.g., having irrational thoughts) but not in others (having no hallucinations, no violent tendencies, being fully capable of self-care) (Estroff et al. 1991; Gove 2004; Quadagno and Antonio 1975). Nevertheless, the distorted portrayals of mental illness in the media (Wahl 1995) make it likely that persons with conditions that are mild to moderate in severity will deflect a mental illness identity much more frequently compared to those with psychoses.³

A second factor influencing identity deflection concerns the person’s history of treatment. Individuals with no recent or prior history of mental health treatment should be more inclined to refute a mental illness identity than those who have been diagnosed and treated before. For people with new-onset conditions, mental illness stereotypes simply “don’t fit me.” Additionally, those with new onsets are likely to believe (often rightly) that their problems are time limited, recoverable, and/or manageable with the right medications or therapy. In contrast, individuals who have been ill for a prolonged period of time, who have repeatedly relapsed, or whose regimens have failed to reduce impairment should find it more difficult to believe that “I don’t have a mental illness” or “I’m not mentally ill.” Such deflecting assertions become increasingly implausible in the face of mounting evidence to the contrary.

The third factor involves the degree of stigma an individual perceives to be attached to mental illness. Stigma is a characteristic that is “deeply discrediting” (Goffman 1963:3) and elicits stereotyping, status and power loss, social rejection, and discrimination (Link and Phelan 2010). The greater the mental illness stigma perceived, the greater the potential threat to the self-conception of persons who have been given a psychiatric diagnosis or been in mental health treatment—they know they are at risk of being viewed by others as “mentally ill” or as a “mental patient” (Link et al. 1989). Perceived threat should motivate the rejection of such characteristics as self-descriptive. Individuals should self-protectively assert, “I am not mentally ill,” “I do not have a mental illness,” “I’m not like them,” or “That’s not me” (McCall 2003; Snow and Anderson 1987).

A final factor that should influence deflection is the number of conventional role identities that an individual holds (Thoits 2011). Identity theorists assume that people occupy multiple role identities that are organized hierarchically.⁴ Some theorists argue that identities are organized by their salience, that is, their likelihood of being invoked in interpersonal interaction (Stryker and Serpe 1982). Others posit that identities are organized by their personal importance or psychological centrality to the individual (McCall and Simmons 1978; Rosenberg 1979). Because of its discrediting nature, a mental illness identity should rank low in an individual’s salience hierarchy; it is unlikely to be called up and performed voluntarily in interaction with other people. Alternatively, because a mental illness identity offers few intrinsic or extrinsic rewards, exacts greater costs, and attracts little positive support from other people, it should rank low in a person’s importance or centrality hierarchy (McCall and Simmons 1978). In general, the more conventional role identities an individual holds, the lower a mental illness–related identity should rank relative to other more salient and/or important aspects of the self, essentially making illness an “irrelevant, unimportant part of me.” Although a person may acknowledge that he or she has a mental health problem, “it doesn’t define who I really am” relative to other aspects of self-conception (Howard 2006; Thoits 2011). In short, individuals with multiple conventional role identities should be more likely to deflect a mental illness identity.

To summarize the main argument, individuals who have a mild to moderately serious rather than a severe condition, who have no recent or prior treatment experience, who perceive a high degree of mental illness stigma, and who possess many conventional role identities are more likely to believe that they are not mentally ill, that is, to deflect a mental illness identity (Hypothesis 1). That belief enables individuals with mental health problems to dismiss the personal relevance of widespread cultural stereotypes about “mental patients.” Because the stigma of mental illness seems inapplicable to the self, individuals’ self-esteem, confidence, and positive affect (aspects of psychological well-being) should remain high compared to people who see themselves as mentally ill and who fear devaluation and rejection by other people (Link et al. 2014) (Hypothesis 2). Therefore, identity deflection should help to explain (i.e., mediate) the greater well-being of individuals with less severe illness, no history of psychiatric treatment, awareness of the potential for stigma, and more conventional social roles (Hypothesis 3).

Figure 1 illustrates the model underlying these hypotheses and guides the analyses in the following sections.

Data and Methods

Sample

The NCS-R drew a nationally representative sample of community-dwelling, English-speaking adults 18 years or older who were living in the continental United States.⁵ Respondents (N = 9,282) were selected through multistage, clustered-area probability sampling of households (Kessler et al. 2004). They were interviewed in person with a structured survey instrument between 2001 and 2003, with a response rate of 71%.

All respondents participated in Part 1 of the NCS-R survey, which assessed the lifetime and 12-month prevalence of the most common and the most severe mental disorders. Each disorder included in the survey was assessed on the basis of symptom criteria outlined in the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV; American Psychiatric Association 1994). A subsample of 5,692 respondents participated in Part 2 of the survey, which focused on risk factors for mental health problems and respondents’ treatment experiences. The Part 2 sample consisted of all participants who met lifetime criteria for any disorder that was included in the survey and a random sample of other respondents with no lifetime disorder. This article draws data only from the subsample, weighted to adjust for biases due to selection within households, nonresponse, differences between the sample and the 2002 Current Population Survey, and oversampling from the Part 1 respondents (Alegria et al. 2008).

This article analyzes only respondents who met DSM-IV criteria for one or more mental disorders in the past year (n = 1,368; 24% of the Part 2 sample). Restricting the analysis to respondents with a recent mental disorder allows a distinction between those who currently believe they do and do not have a mental illness (this variable is described below). Participants who do not have a disorder but believe that they do (the “worried well”) were too few in number for meaningful analysis (n = 93; 1.6% of the Part 2 sample). Participants who were without disorder and also believed they were well are not at risk of mental illness stigmatization so are excluded from analysis.

Measures

Mental Disorders

Part 1 of the NCS-R assessed all core mental disorders measured by the World Health Organization–Composite International Diagnostic Interview (WHO-CIDI), which was adapted for the World Mental Health Survey Initiative (Kessler and Ustun 2004). Criteria for each disorder were translated into a series of straightforward survey questions that could be administered by trained lay interviewers. Once data were collected, project algorithms combined respondents’ answers into dichotomous variables indicating the occurrence or non-occurrence of each disorder during the respondent’s lifetime and during the past 12 months.

As mentioned above, I analyzed only participants with disorders occurring in the past year. This is because respondents were asked in the survey whether they “currently” have a mental or emotional condition (described below). I included 16 disorders in this analysis: bipolar I and II, mania, major depressive episode, hypomania, dysthymia, generalized anxiety disorder, posttraumatic stress disorder (PTSD), panic disorder, specific phobia, social phobia, agoraphobia, alcohol abuse, alcohol dependence, drug abuse, and drug dependence. All disorders were assessed without hierarchy.⁶ A dummy variable was constructed from this set of diagnoses to indicate whether the respondent had any disorder during the past year (1 = yes, 0 = no). There were 1,368 respondents with a 12-month disorder.

Severity of Illness

I employ two measures of the severity of the respondent’s illness in the analyses. The first indicates whether respondents experienced a severe mental disorder in the past year (1 = yes, 0 = no). I consider those with no severe disorder to have a mild to moderately serious disorder. Severe disorders included bipolar I, bipolar II, mania, and major depressive episode. Mild to moderately serious disorders consisted of dysthymia, generalized anxiety disorder, panic disorder, PTSD, agoraphobia, social phobia, specific phobia, hypomania, and substance-use disorders (alcohol abuse, alcohol dependence, drug abuse, and drug dependence). Among respondents with any 12-month disorder, 39% had a severe disorder and 61% a mild to moderately serious disorder.

The second measure was of serious impairment (coded 1 = present, 0 = not present) based on the K6 scale (Kessler et al. 2002, 2003). The K6 was developed to discriminate quickly and efficiently in surveys between individuals who do and do not have a serious mental illness, defined by its developers as having any DSM-IV disorder (other than substance use) accompanied by significant impairment, indicated by Global Assessment of Functioning scores less than 60 (Kessler et al. 2003). The K6 scale asks respondents,

Think of the one month in the past 12 months when you were at your worst emotionally in terms of being anxious, depressed, or emotionally stressed. If there was no month like this, think of a typical month in the past 12 months. How often did you feel: nervous, hopeless, restless or fidgety, so depressed that nothing could cheer you up, that everything was an effort, worthless?

Possible responses varied from 0 = “none of the time” to 4 = “all of the time.” Responses were summed, with scores ranging from 0 to 24. Cronbach’s alpha for the scale for the Part 2 sample was .86. The optimal cut point is a score of 13 or higher; this cut point distinguishes serious from nonserious conditions with high accuracy (Kessler et al. 2003). Using this cut point, 19% of respondents with a 12-month disorder had serious impairment.⁷

Among participants with one or more disorders in the past year, severe disorder and serious impairment were positively but only moderately associated (phi = .30, p = .001). This is because the former measure is based solely on diagnostic categories while the latter taps the adequacy of daily functioning. Individuals might have met criteria for a severe disorder, such as major depressive episode, in the past year but had its functional consequences under control with therapy or medication, for example.

Treatment Experience

I determined whether respondents had a history of mental health treatment by canvassing their age at first hospitalization for “your emotions, nerves, or use of alcohol or drugs”; age at first contact with any of a list of professionals for these problems; and age at first use of prescription medication for mental health or substance use problems.⁸ When the difference between respondents’ current age and their age at first treatment was two years or greater, I coded them as having had previous treatment experience (1 = yes, 0 = no); in other words, respondents had treatment that occurred prior to the current 12-month period. Among respondents with any disorder in the past year, 66% had had prior treatment experience involving hospitalization, contacts with service professionals, or medications for their emotional or substance-use problems.

Respondents with a history of treatment may or may not have used mental health services in the past year. To indicate recent treatment experience, participants who had been hospitalized, saw a mental health professional, or took prescription drugs for their “emotions, nerves, or alcohol or drug use” in the past 12 months were coded 1; otherwise, 0. Among participants with any 12-month disorder, 46.5% had utilized these mental health services in the past year. Past and recent treatment experience were only moderately associated (phi = .32, p < .001).

Perceived stigma

No directly worded questions about the stigma of mental illness were available in the survey. Three questions in the survey implicitly tapped the stigma that individuals attach to seeking help for mental health problems. The three questions were asked of all participants in the survey—both persons who did and did not have a disorder or treatment experience. Respondents were asked, “How embarrassed would you be if your friends knew you were getting professional help for an emotional problem?” (1 = “not at all embarrassed” to 4 = “very embarrassed”). Possible embarrassment attached to treatment suggests that individuals view mental health problems as stigmatizing. Among participants with a 12-month disorder, 35% said they would be “somewhat to very embarrassed.”

Respondents were also asked, “People differ a lot in their feelings about professional help for emotional problems. If you had a serious emotional problem, would you definitely go for professional help, probably go, probably not go, or definitely not go for professional help?” (coded 4 = “definitely go” to 1 = “definitely not go”) and “How comfortable would you feel talking about personal problems with a professional?” (coded 4 = “very comfortable” to 1 = “not at all comfortable”). These latter two variables were correlated .48 (p < .001). I presume here that opposition to and discomfort with the idea of using mental health services reflect a person’s underlying view that entering treatment is undesirable and thus potentially stigmatizing. Participants’ responses to the two questions were reverse coded and summed, then divided by the number of items answered, so scores range from 4 (strongly perceive treatment seeking as undesirable) to 1 (strongly perceive treatment seeking as desirable). The average score among respondents with a 12-month disorder was 1.9 (SD = .75), indicating that the typical respondent’s orientation was favorable to treatment seeking. Because perceived embarrassment and perceptions of treatment entry as undesirable were only moderately correlated (r = .25, p < .001), I keep these stigma perception variables in the analyses as separate constructs.

Number of Social Roles

Sociodemographic questions indicated whether participants were married, had children living in the household, and were employed part-time or full-time. Each of these roles were coded 1 = yes or 0 = no and then summed to range from 0 to 3. Among respondents with a 12-month disorder, the mean number of roles was 1.4 (SD = .87), essentially between one and two roles. Forty-six percent were married, 43% had children at home, and 65% were employed.

Identity Deflection

All respondents in the NCS-R were asked, “Do you have any of the following conditions?” The last item in the list of queried health conditions (e.g., blindness, deafness, hearing problem, physical disability) was, “Any emotional or mental disability?” (1 = yes, 0 = no). I assumed here that respondents who said yes were in effect replying, “Yes, I have a mental illness.” I reverse coded responses to this question so that 1 indicated identity deflection: “I do not have a mental illness.” Among respondents with any 12-month disorder, 83% said they did not have a mental illness (i.e., they deflected a mental illness identity), and 17% said that they did (i.e., they accepted that identity).

There are several potential problems with this indicator of identity deflection that need to be addressed. First, the term emotional or mental “disability” might seem an odd euphemism for mental illness. However, the term was appropriate in the context of prior questions about the respondent’s health conditions, which included learning disability and physical disability. “Emotional or mental disability” might be a softer phrase than “mental illness,” but “disability” still signals very serious impairment, consistent with stereotyped images of mental illness (e.g., irrational, incompetent).

Second, the question asked about having a mental health condition and not about how respondents actually view or define themselves; the survey does not include direct questions about self-conception. I viewed this question as an indicator of identity deflection because having a condition can be easily generalized as a personal attribute or descriptor. For example, the observation that “I have diabetes” can become “I am diabetic” or “I am a diabetic.” I assume that most individuals are aware of such elisions in meaning and are prone to such elisions themselves, so claiming “I do not have a mental illness” is equivalent to “I am not mentally ill.”

Third, respondents who say they do not have a mental illness may be giving a socially desirable answer. However, on a 10-item scale of socially desirable responding (Zuckerman et al. 1993) that was included in the NCS-R interview, the desirability scores of respondents who said they did and did not have a mental illness were the same.

Finally, it is possible that people who say they do not have a mental illness are unaware of their psychological problems. They may be in denial or even delusional about the state of their mental health, so this deflection measure may not be capturing a conscious rejection of a mental illness condition or identity. However, 75% of those respondents who said they do not have a mental illness recognized that they in fact had a mental health problem: On questions asked later in the interview, they agreed they had felt a need to see a mental health professional in the past year because of “problems with their emotions, nerves, or use of alcohol or drugs,” or they admitted to having such problems but said they preferred to handle them on their own rather than see a professional. So a large majority of people who deflect the idea that they have a mental illness appear to be in touch with reality, that is, not in denial or delusional about the state of their mental health.

It is worth noting that 89% of respondents who had a severe disorder and 94% of those with a serious impairment recognized that they had problems with their “emotions, nerves, or use of alcohol or drugs.” Yet, 69% of respondents with severe disorders who acknowledged having mental health problems and 59% of those with serious functional impairment who acknowledged having mental health problems said they did not have a mental illness. Consistent with the definition of deflection, then, respondents rebuffed the idea that they had a mental illness or were mentally ill, even though they in fact had very major mental health problems (Thoits 2011). I presumed they reject this idea because mental illness is a stigmatized status and potentially threatening to self-conception.

Well-being

I employ two measures of psychological well-being in the analyses, psychological distress and positive affect. Psychological distress refers to a subjective state of malaise accompanied by emotional and somatic symptoms of anxiety and depression (Mirowsky and Ross 2003). Participants were asked,

During the past 30 days, how often did you: feel “trapped” or “caught,” feel suddenly scared for no reason, blame yourself for things, feel lonely, feel blue, worry too much about things, feel no interest in things, feel frightened, feel hopeless about the future, have trouble concentrating, feel tense or keyed up, feel everything was an effort, feel worthless, feel exhausted for no good reason?

Possible responses on these 14 items ranged from 4 = “often” to 1 = “never.” Each respondent’s responses were summed and divided by the number of items that he or she answered to reduce missing values, so scores ranged from 4 (high distress) to 1 (low). Cronbach’s alpha for the scale for the Part 2 sample was .92. The mean distress score for respondents with a 12-month disorder was 2.1 (SD = .74), indicating that they experienced all distress symptoms on average “sometimes” during the past 30 days. None of the distress items overlap with the serious impairment scale (the K6).⁹

Positive affect was measured with four questions: “During the past 30 days, how often did you feel: confident, optimistic, happy, full of life?” Possible responses were “all of the time” (5), “most of the time” (4), “some of the time” (3), “a little of the time” (2), or “none of the time” (1). To reduce missing values, respondents’ summed responses were divided by the number of items they answered, so scores ranged from 5 (high positive affect) to 1 (low). Cronbach’s alpha for the Part 2 sample was .83. On average, participants with any disorder in the past year had positive feelings “some of the time,” a mean score of 3.1 (SD = .8).

It is important to note that distress symptoms and positive affect were reported for the past month, in contrast to the variables for clinical disorder, serious impairment, and recent treatment experience, which were reported for the past year. The different time frames for these measures allow illness and treatment experiences to be viewed as probable antecedents to psychological well-being as an outcome.

Control Variables

Female is coded 1, male 0. Age and age squared are measured in years. Race-ethnicity is categorized as black (0, 1), Hispanic (0, 1), and other race (0, 1), with whites serving as the omitted comparison group. Education is measured as years of schooling, top coded at 17 years or more. Income is based on the ratio of household income to the Census 2001 poverty threshold for each family size, coded originally into 17 categories. Following Wang et al. (2005), the 17 categories were collapsed into four: 1 = low income, 2 = low average income, 3 = high average income, 4 = high income. Having health insurance, either private or public insurance, was coded 0 = no, 1 = yes.

Table 1 summarizes all measures, including the control variables.

Table 1.

Distributions of Weighted Study Variables for Respondents with a 12-Month Disorder (n = 1,368), National Comorbidity Survey Replication, 2001–2003.

Variable	M or %	SD
Female	60.6	—
Male	39.4	—
Age (18–90)	39.5	15.1
White	72.3	—
Black	11.1	—
Hispanic	12.1	—
Other race-ethnicity	4.5	—
Years of education	13.0	2.3
Low income	25.5	—
Low average income	30.1	—
High average income	27.5	—
High income	16.9	—
Have health insurance	81.9	—
No health insurance	18.1	—
Possible embarrassment (1–4)	2.0	1.0
Treatment is undesirable (1–4)	1.9	.7
Number of social roles (0–3)	1.5	.9
Severe disorder	38.8	—
Mild to moderately serious disorder	61.2	—
Serious impairment	19.3	—
Nonserious impairment	80.7	—
Past history of treatment	65.8	—
No history of treatment	34.2	—
Recent treatment experience	46.5	—
No recent treatment experience	53.5	—
Deflection of mental illness identity	82.6	—
Acceptance of mental illness identity	17.4	—
Psychological distress (1–4)	2.1	.7
Positive affect (1–5)	3.1	.8

Analysis

Because of the complex design of the study, survey design methods are necessary to correct variances, standard errors, and confidence intervals. I employed the regression and logistic regression survey procedures of Stata/SE 10.0 (StataCorp 2007) and the Taylor series linearization method for variance estimation in the analyses.

Results

Predictors of Mental Illness Identity Deflection

According to Hypothesis 1, respondents who have less serious illness, who have not been in mental health treatment, who perceive mental health treatment as stigmatizing, and who hold many conventional social roles should be more likely to deflect a mental illness identity. This hypothesis was tested with logistic regression. Holding a deflection belief was regressed on indicators of each predictive factor separately, with background variables controlled in each equation. Then in a final step, all predictors were entered together. The results of these analyses appear in Table 2.

Table 2.

Predictors of Mental Illness Identity Deflection among Respondents with a 12-Month Disorder, National Comorbidity Survey Replication, 2001–2003.

	Identity Deflection
	(1)	(2)	(3)	(4)	(5)
Predictors
Severe disorder (0, 1)	−.97***	—	—	—	−.80***
Serious impairment (0, 1)	−1.15***	—	—	—	−.89***
History of treatment (0, 1)	—	−1.29***	—	—	−1.30***
Recent treatment (0, 1)	—	−1.61***	—	—	−1.29***
Possible embarrassment	—	—	.05	—	−.02
Treatment is undesirable	—	—	.38***	—	.18†
Number of social roles	—	—	—	.48***	.25**
Control variables
Female (0, 1)	−.13	−.00	−.14	−.26†	−.01
Age	−.10***	−.08***	−.12***	−.17***	−.09**
Age squared	.001***	.001***	.001***	.002***	.001**
Black (0, 1)	.54*	.15	.44†	.54*	.23
Hispanic (0, 1)	.40†	.31	.65**	.55*	.02
Other race-ethnicity (0, 1)	.29	.02	.14	.21	.15
Education	.00	.05	.04	.04	.03
Income	.36***	.42***	.44***	.35***	.29**
Have insurance (0, 1)	−.31	−.04	−.34†	−.43*	−.13
Constant	4.01***	3.75***	2.00***	3.49***	4.20***

Note: Logistic regression coefficients are reported. N = 1,344.

†

p < .10, *p < .05, **p < .01, ***p < .001.

Model 1 of Table 2 shows that individuals with a severe disorder and those with serious functional impairment are significantly less likely to deflect a mental illness identity compared to persons who have a mild to moderate disorder or less impairment. The converse, of course, is that persons with less serious problems are more likely to reject the idea that they have a mental illness, consistent with expectations.

In Model 2, respondents who previously or recently have been in mental health treatment are significantly less likely to deflect a mental illness identity. Conversely, those who have not been in treatment view themselves as not mentally ill, again supporting predictions.

In Model 3, individuals’ possible embarrassment about being in treatment is unrelated to identity deflection, contrary to expectations regarding the effects of perceived stigma. However, perceiving psychiatric treatment as undesirable (and presumably more stigmatizing) is associated with a higher likelihood of identity deflection, consistent with expectations.

Model 4 shows that respondents who hold more conventional social roles are more likely to define themselves as not mentally ill, also as hypothesized.

When all predictor variables are entered simultaneously in Model 5, all of the significant effects hold, although two coefficients decreased noticeably in size from Models 3 and 4 to Model 5: the coefficients for viewing treatment as undesirable and for the number of roles held. Exploratory analyses showed that these drops in coefficient size were attributable to the illness variables and to past and recent treatment experience. But the significant coefficients in Model 5 (marginally significant for aversion to treatment) clearly indicate that the associations of the predictor variables with identity resistance are not simply the spurious products of preexisting illness or prior treatment for illness.

Well-being Consequences of Identity Deflection

Hypothesis 2 suggests that persons who resist a mental illness identity should have lower psychological distress and higher positive affect than those who accept the identity, or self-label (Thoits 1985). To test this hypothesis, distress and positive affect were regressed on mental illness identity deflection and the set of control variables. In a second step, severe disorder, serious impairment, past treatment experience, and recent treatment experience were added to the equations to assess whether links between deflection and well-being were spurious products of preexisting illness or mental health service use. Table 3 reports the results of these analyses.

Table 3.

The Effect of Mental Illness Identity Deflection on Respondents’ Psychological Well-Being, National Comorbidity Survey Replication, 2001–2003.

	Psychological Distress		Positive Affect
Variable	(1)	(2)	(1)	(2)
Identity deflection (0, 1)	−.53***	−.19***	.55***	.27***
Severe disorder (0, 1)	—	.24***	—	−.33***
Serious impairment (0, 1)	—	.66***	—	−.45***
Past history of treatment (0, 1)	—	.14**	—	−.12**
Recent treatment (0, 1)	—	.11*	—	−.06
Control Variables
Female (0, 1)	.10**	.06†	−.04	−.00
Age	.02***	.01*	-.02**	−.01†
Age squared	−.00**	−.00	.00**	.00
Black (0, 1)	.03	.05	.10	.08
Hispanic (0, 1)	−.05	.06	.15†	.06
Other race-ethnicity (0, 1)	.11	.09	−.10	−.08
Education	−.04***	−.03**	.00	−.00
Income	−.10***	−.07***	.06**	.04*
Have insurance (0, 1)	−.07	−.10†	.00	.02
Constant	2.82***	2.20***	2.90***	3.43***
R ²	.15	.33	.09	.19

Note: Unstandardized regression coefficients are reported. N = 1,351.

†

p < .10, *p < .05, **p < .01, ***p < .001.

In the columns for Model 1 in Table 3, deflecting the idea that one is mentally ill is associated with significantly lower distress and higher positive affect, as hypothesized. When illness conditions and treatment experiences are controlled in each Model 2, the influences of identity resistance on the well-being outcomes remain significant, but the deflection coefficients reduce dramatically in size (66% for psychological distress and 51% for positive affect). The reduced size of these coefficients indicates that the relationship between deflection and well-being is in large part a product of respondents’ prior mental health and treatment history. Nevertheless, because the deflection coefficients remain significant in each Model 2, the hypothesis that resisting a mental illness identity is associated with positive well-being is supported.

Does Identity Deflection Mediate the Effects of Predictor Variables on Psychological Well-being?

We know from Table 2 that people who have less serious illness, no history of psychiatric treatment, discomfort about entering treatment, and multiple conventional social roles are more likely to deflect the idea that they are mentally ill. From Table 3, we know that deflecting a mental illness identity is associated with positive well-being. Hypothesis 3 proposes that identity deflection should therefore help to explain (i.e., mediate) the greater well-being of individuals with less illness, no treatment history, aversion to mental health treatment, and multiple roles.

To test this hypothesis, well-being variables were first regressed on the predictor variables, with control variables held constant in each equation. In a second step, identity deflection was added to the equations to test for mediation. Table 4 summarizes the results of these two steps.

Table 4.

Identity Deflection as a Possible Mediator between Predictor Variables and Respondents’ Psychological Well-being, National Comorbidity Survey Replication, 2001–2003.

	Psychological Distress		Positive Affect
Variable	(1)	(2)	(1)	(2)
Severe disorder (0, 1)	.25***	.23***	−.35***	−.32***
Serious impairment (0, 1)	.69***	.66***	−.49***	−.44***
Past history of treatment (0, 1)	.17***	.15**	−.16***	−.13***
Recent treatment (0, 1)	.16**	.13*	−.11*	−.08
Possible embarrassment	.02	.02	.01	.01
Treatment is undesirable	.07**	.07***	−.12***	−.13***
Number of roles	−.04*	−.03†	.05*	.04†
Identity deflection	—	−.20***	—	.27***
Constant	1.75***	1.97***	4.05***	3.77***
R ²	.34	.35	.19	.20

Note: Unstandardized regression coefficients are reported. Each equation controls for female, age, age squared, black, Hispanic, other race-ethnicity, education, household income, and health insurance. N = 1,344.

†

p < .10, *p < .05, **p < .01, ***p < .001.

In each Model 1 of Table 4, we see that most predictor variables are related to well-being in ways that we would expect. Persons with severe disorders and those with serious functional impairment were significantly more distressed and had fewer positive emotions, compared to respondents with mild to moderate disorders and nonserious levels of impairment, respectively. Individuals who had used mental health services earlier in their lives or in the past year reported higher levels of current psychological distress and lower levels of positive affect. Although anticipated embarrassment about mental health treatment was unrelated to distress and positive affect, individuals who viewed mental health treatment seeking as undesirable were more distressed and had fewer positive emotions compared to persons who viewed treatment use as acceptable and positive. Finally, respondents holding more conventional social roles were less distressed and had more positive affect than those with few social roles.

Each Model 2 in Table 4 tests whether deflecting a mental illness identity intervenes in the relationships that appeared in each Model 1. If deflection mediates the effects of the antecedent variables on individuals’ well-being, we should see previously significant coefficients in each Model 1 become nonsignificant and near zero in size when identity deflection is added to each Model 2. But when the Model 1 and Model 2 coefficients are compared, it is clear that identity resistance does not mediate those associations, contrary to expectations: all Model 2 coefficients remain significant (or marginally significant) and similar in size to those that appear in Model 1. Deflecting a mental illness identity therefore has beneficial effects on participants’ well-being independent of the variables that predict the use of this resistance strategy: deflection helps to offset or counterbalance the distress and unhappiness connected to severe symptoms, serious functional impairment, having been in treatment, viewing treatment as undesirable, and holding few roles.

Exploration: Does Identity Deflection Moderate the Effects of Predictor Variables on Well-being?

It was puzzling that mental illness identity deflection did not serve as a mechanism through which predictor variables had influences on individuals’ well-being, especially given that most antecedent variables were linked to deflection and to well-being in expectable ways. Stress theory suggests an alternative possibility to explore: rejecting a mental illness identity may function to buffer or moderate the well-being effects of the predictor variables rather than mediate them. From a stress theory perspective, symptoms of disorder, impaired daily functioning, viewing treatment entry as undesirable, and holding few social roles (i.e., social isolation) can be considered stigma-related stressors with which a person must cope (Thoits and Link 2016). Stress theory posits (and considerable research shows) that coping strategies can reduce or buffer the harmful effects of stressors on individuals’ well-being (Lazarus and Folkman 1984; Pearlin and Bierman 2013). Thus, we might expect mental illness identity deflection to buffer (moderate) the impacts of these stressors on respondents’ distress and positive affect.

To explore this possibility, identity deflection was interacted with each of the predictor variables (seriousness of illness, treatment experiences, views of treatment as stigmatizing, number of social roles). These interactions were added one by one to the second equations of Table 4. Only two interaction terms were significant: Deflection × Severe Disorder and Deflection × Serious Impairment. Nonsignificant interactions were dropped, and the two significant interaction terms were added simultaneously to the Model 2 regressions of Table 4. The results of this analysis are presented in the first two columns of Table 5.

Table 5.

Identity Deflection as a Moderator of the Relationship between Respondents’ Illness Conditions and Their Psychological Well-being, for All Respondents and for Respondents Who Have Ever Been in Treatment, National Comorbidity Survey Replication, 2001–2003.

	All Respondents (N = 1,344)		Ever-treated Respondents (n = 996)
	(1)	(2)	(3)	(4)
Variable	Psychological Distress	Positive Affect	Psychological Distress	Positive Affect
Severe disorder (0, 1)	.12*	−.10	.15*	−.12
Serious impairment (0, 1)	.80***	−.62***	.79***	−.61***
Identity deflection (0, 1)	−.21***	.35***	−.22***	.38***
Severe Disorder × Deflection	.13*	−.27*	.11†	−.24*
Serious Impairment × Deflection	−.20**	.25*	−.21*	.23†
R ²	.35	.19	.34	.20

Note: Unstandardized regression coefficients are reported. Each equation controls for background variables, past history of treatment, recent treatment experience, possible embarrassment, perception of treatment as undesirable, and number of social roles.

†

p < .10, *p < .05, **p < .01, ***p < .001.

If identity resistance buffers the effects of severe disorder and functional impairment on well-being, then the interaction terms in the first two columns of Table 5 (for all respondents) should be negative in sign for psychological distress and positive in sign for positive affect. Unexpectedly, however, respondents with severe disorders who refute having a mental illness identity are significantly more distressed and feel less positive affect compared to respondents who accept the identity. But consistent with expectations, individuals with serious impairment who assert they are not mentally ill report less distress and more frequent positive emotions than persons who accept it (i.e., do not deflect). So identity deflection appears harmful to well-being among individuals with a severe disorder but beneficial to well-being among those with serious difficulties in day-to-day functioning, a mixed set of findings.

Further exploration revealed that these effects were confined to individuals who had ever been in treatment either previously or in the past year. Almost all respondents who had never been in treatment (n = 355) said they were not mentally ill: 98% deflected a mental illness identity (not shown). For these respondents, deflection was essentially a constant, and thus it simply could not function as a counter to nor a buffer of disorder or impairment. Not surprisingly, in a regression, the deflection and interaction coefficients were all nonsignificant for this subgroup of respondents (not shown). In contrast, among persons who had past or recent treatment experience (n = 1,013), 77% rejected a mental illness identity while 23% accepted it (not shown). Contact with mental health professionals (through hospitalization, counseling/therapy, or psychoactive medication use) seems to be a condition under which individuals take more seriously the possibility that they are mentally ill and either reject or accept that idea—a labeling effect (Link et al. 1989; Scheff 1966).

The second set of columns in Table 5 reports the main and interactive effects of identity deflection for this ever-treated subgroup of respondents. We see again that those who had a severe disorder and refuted mental illness had marginally higher distress and significantly lower positive emotions than those who accepted an illness identity—stress exacerbation effects. Respondents with serious functional impairment who refuted illness were significantly less distressed and experienced marginally more positive feelings than those who accepted illness—stress buffering effects. It is possible that having an official severe (and stigmatizing) disorder while believing oneself not mentally ill creates cognitive dissonance (Anspach 1979), exacerbating the stress due to disorder.¹⁰ At the same time, receiving professional assistance with symptom management and day-to-day functioning while believing oneself not mentally ill might enable hope and aid in recovery, buffering the stress of impairment.

Discussion and Conclusion

Research attention recently has turned to stigma resistance—how people who have been labeled “mentally ill” resist internalizing stigma and thus sustain their well-being and quality of life (Link et al. 2002; Thoits and Link 2016). One strategy of resistance, identity deflection, has been studied only in samples of patients who have psychoses, decreasing the chances of observing a form of resistance that may be far more frequent in samples with a broader range of disorders. This limitation was overcome here with data from the NCS-R for respondents who met clinical criteria for a number of mild-to-moderate and severe mental disorders. These data enabled examination of both the proposed determinants and the well-being consequences of identity deflection.

The analyses showed that mental illness identity deflection was indeed very common among persons with a wide range of disorders: 83% of respondents with any 12-month disorder did not see themselves as mentally ill. Even among those who had a severe disorder or serious functional impairment, deflection was common—72% and 60%, respectively. Because the majority of individuals with severe disorder or serious impairment knew they had problems with their “emotions, nerves, or use of alcohol or drugs,” assertions that they did not have a mental illness suggest that they perceived mental illness stereotypes and stigma as inapplicable to themselves.

A number of factors thought to encourage deflection (Thoits 2011) were confirmed in the analyses. Persons with mild to moderately severe disorders and those with low levels of functional impairment were more likely to reject a mental illness identity, compared to individuals with severe disorders or a high level of impairment. Similarly, individuals who had no previous or recent history of psychiatric treatment were more likely to dismiss the idea that they were mentally ill. (Conversely, having a severe disorder, exhibiting serious impairment, and having had mental health treatment was linked to higher rates of identity acceptance—essentially, self-labeling [Thoits 1985].) Individuals with several conventional role identities were more likely to refute mental illness, suggesting that they did not view their psychological problems as relevant or important to self-definition. Only concerns about stigma produced mixed effects. Identity resistance did not covary with anticipated feelings of embarrassment attached to mental health treatment, contrary to expectations. However, consistent with predictions, persons who saw mental health service use as undesirable (potentially stigmatizing) were more likely to reject a mental illness identity.

As a stigma resistance strategy, identity deflection should protect an individual’s psychological well-being. The findings showed that people who believed they were not mentally ill indeed had significantly lower psychological distress and higher positive affect, even net of the severity of their disorder, the impairment caused by their condition, and their past and recent experiences of mental health service use. What people believe to be true about themselves appears to be true in its consequences, consistent with Thomas and Thomas’s (1928) well-known dictum.

However, resisting a mental illness identity did not mediate relationships between predictor variables and psychological well-being, as I proposed. Deflection beliefs simply counteracted or offset a portion of the negative well-being consequences of individuals’ illness conditions, treatment experiences, stigma perception, and low social role involvements.

Further exploration grounded in the stress process model revealed that identity resistance moderated the effects of illness conditions on individuals’ current distress and positive affect. Specifically, identity deflection buffered the negative consequences of seriously impaired daily functioning, consistent with stress theory. However, individuals with a severe disorder who resisted thinking of themselves as mentally ill were more distressed and had fewer positive emotions, an unexpected stress exacerbation effect. These mixed effects of identity deflection occurred only among respondents who had previously or recently been in mental health treatment, that is, only among individuals who had undergone formal psychiatric labeling (Link et al. 1989; Scheff 1966). Persons who had never been in treatment almost uniformly dismissed the idea that they were mentally ill, so for these individuals, rejecting a mental illness identity could neither counteract nor interact with their illness conditions to influence their well-being.

Why did identity deflection heighten distress among treated individuals with severe disorder and reduce distress among treated persons who had serious functional impairment? Entering treatment means that one acquires a clinical diagnosis and one receives therapy or medications. Acquiring a severe diagnosis (e.g., bipolar disorder, major depression) puts one at high risk of stigma. On the other hand, mental health professionals dispense expert social support (Thoits 1985) or drug prescriptions that can improve patients’ subsequent functioning and relieve their symptoms. Rosenfield (1997) demonstrated that perceived stigma and received services had opposing effects on the life satisfaction of participants with psychosis in a psychiatric rehabilitation program. Specifically, greater perceived stigma from having a psychiatric label was associated with lower life satisfaction among participants, while at the same time, benefits derived from psychological, social, and vocational therapies were related to higher satisfaction. In essence, treatment entry exposes individuals to stigma and helpful services, not just one or the other kind of influence (Rosenfield 1997). Thus, it seems plausible that individuals who reject the idea that they have a mental illness despite clear indicators that they do (i.e., a severe diagnosis from an authoritative source, participation in treatment) may struggle with a heightened sense of potential threat that raises their distress and dampens positive emotions. At the same time, obtaining effective coping assistance or medication-assisted symptom control while holding onto the belief that one is not (severely, stereotypically) ill may expedite improved functioning that lowers distress and promotes positive feelings. But to further assess these possible processes, longitudinal data would be required.

This observation points to one of several limitations of this study that should be recognized. First, these are cross-sectional data, so relationships of the predictor variables with identity deflection and well-being potentially could be reversed. For example, people who refuse to see themselves as mentally ill may be more likely to view illness as stigmatizing and avoid entry into mental health treatment, rather than these factors producing deflection, as argued here. Similarly, persons with greater well-being may have less serious disorders and few functional impairments, prompting them to rebuff a mental illness identity. Fortunately, the timing of the illness and treatment variables (in the past year) and the well-being measures (in the past month) help to mitigate these concerns. Still, panel data are necessary to better establish causal ordering among variables.

Second, there was no direct measure of mental illness identity in these data. Respondents reported that they did or did not have an emotional or mental “disability.” Asking about a “disability” was in keeping with other conditions about which participants were being queried at that point in the interview. “Disability” is a strong term although perhaps more tactful than “mental illness.” Respondents were not asked directly whether they “thought of themselves” as having a mental illness, which would have provided a clear marker of identity avowal or disavowal. I assumed here that admitting or refuting that one has an emotional or mental “disability”/illness was equivalent to saying that “I am/am not mentally ill,” an assumption that may not be warranted. Future work will need to verify whether “I do not have a mental illness” and “I’m not mentally ill” carry equivalent meaning.

Further, it is possible that respondents who claimed they did not have a mental illness were giving a socially desirable answer rather than stating a true belief about themselves. Alternatively, respondents may have been in denial, reflecting unconscious motivation or lack of insight into their condition. Distinguishing with certainty among social desirability, defensive denial, and genuine belief is difficult with these survey data. Nevertheless, three observations are relevant. First, identity deflection and socially desirable responding were unrelated in this sample. Second, the vast majority of persons with severe disorders (89%) or with serious functional impairment (94%) acknowledged that they had problems with their “emotions, nerves, or use of alcohol or drugs” on later interview questions; this certainly suggests that respondents with the most serious problems had insight into their condition. Third, patterns of identity deflection and identity acceptance indicated that respondents were making fairly realistic assessments of their mental health status: if participants had a severe disorder, were seriously impaired, or had been in mental health treatment, they were more likely to acknowledge having an emotional or mental “disability”/illness (i.e., they self-labeled). These systematic patterns suggest a rough match between reality and individuals’ claims about their mental health. In general, the available evidence suggests that respondents knew they had a mental health problem; when they refuted having a mental illness, then, they were likely protecting themselves from the cultural stigma attached to having such problems.

Importantly, respondents may have viewed having an emotional or mental “disability” as a more severe and stigmatizing condition than having an emotional or mental “illness.” If this were so, more individuals with serious problems would refute having a disability (i.e., they would deflect the identity), along with those with only mild to moderate problems. If more respondents with serious disorder or impairment were included among deflectors, observing a significant positive relationship between identity deflection and psychological well-being would be less likely. Thus, although far from ideal as an indicator of mental illness identity resistance, the disability question may have provided a more conservative test of the main hypothesis of this study.

As a final limitation of the study, items measuring perceived stigma focused on the stigma of potential mental health treatment rather than the stigma of having a mental illness per se. One could plausibly argue that the perceived stigma items used here tap a general aversion to treatment seeking, rather than an aversion to treatment for mental health problems specifically. More direct indicators of perceived mental illness stigma unfortunately were not available.

Despite these limitations, this study generally supports the argument that identity resistance protects individuals’ well-being. Those who deflect a mental illness identity enjoy better psychological well-being than persons who view themselves as mentally ill, even when their disorder is severe and their day-to-day functioning is impaired. What people believe to be true about themselves seems to be true in its consequences. Rebuffing a mental illness identity may prevent internalized stigma (devaluing oneself because one has a mental disorder), which has been linked to lower self-esteem, social isolation, and higher depressive symptoms, among other undesirable outcomes (Corrigan, Watson, and Barr 2006; Link et al. 2002; Thoits and Link 2016; Yanos et al. 2008). Reinforcing a person’s idea that he or she is not mentally ill or that mental illness is only a part of who he or she is may help to prevent self-stigma, although paradoxically, reinforcing that idea might also discourage treatment seeking. Clearly, a more subtle understanding of mental illness identity deflection will be desirable, in particular, how and when individuals simultaneously can believe themselves not mentally ill while still recognizing that they are in need of mental health services.

Footnotes

Notes

Author Biography

Peggy A. Thoits is the Virginia L. Roberts Professor of Sociology at Indiana University–Bloomington. Her interests are in physical and mental illness; stress, coping, and social support processes; self and identity; and emotion. Her research focuses on the psychological determinants and consequences of holding multiple role identities, the role of similar-other support in reducing ill health and distress, the conditions under which individuals label themselves or are labeled by others as mentally ill, and factors promoting resistance to the stigma of mental illness.

References

Alegria

Margarita

Jackson

James S.

Kessler

Ronald C.

Takeuchi

David

. 2008. Collaborative Psychiatric Epidemiology Surveys (CPES), 2001-2003 (ICPSR20240-v5) [Computer file]. Ann Arbor, MI: Institute for Social Research, Survey Research Center [Producer], Inter-University Consortium for Political and Social Research [Distributor].

American Psychiatric Association. 1994. Diagnostic and Statistical Manual of Mental Disorders, 4th ed. Washington, DC: Author.

Anspach

Renee R.

1979. “From Stigma to Identity Politics: Political Activism among the Physically Disabled and Former Mental Patients.” Social Science & Medicine 13:765–73.

Barke

Megan

Fribush

Rebecca

Stearns

Peter N.

2000. “Nervous Breakdown in 20th Century American Culture.” Journal of Social History 33(3):565–84.

Corrigan

Patrick W.

Giffort

Daniel

Rashid

Fadwa

Leary

Matthew

Okeke

Iheoma

. 1999. “Recovery as a Psychological Construct.” Community Mental Health Journal 35(3):231–39.

Corrigan

Patrick W.

Watson

Amy C.

2002. “The Paradox of Self-stigma and Mental Illness.” Clinical Psychology: Science and Practice 9(1):35–53.

Corrigan

Patrick W.

Watson

Amy C.

Barr

Leah

. 2006. “The Self-stigma of Mental Illness: Implications for Self-esteem and Self-efficacy.” Journal of Social and Clinical Psychology 25(8):875–84.

Estroff

Sue E.

Lachicotte

William S.

Illingworth

Linda C.

Johnston

Anna

. 1991. “Everybody’s Got a Little Mental Illness: Accounts of Illness and Self among People with Severe, Persistent Mental Illnesses.” Medical Anthropology Quarterly 5(4):331–69.

Goffman

Erving

. 1963. Stigma: Notes on the Management of a Spoiled Identity. Englewood Cliffs, NJ: Prentice Hall.

10.

Gove

Walter R.

2004. “The Career of the Mentally Ill: An Integration of Psychiatric, Labeling/Social Construction, and Lay Perspectives.” Journal of Health and Social Behavior 45(4):357–75.

11.

Howard

Jenna

. 2006. “Expecting and Accepting: The Temporal Ambiguity of Recovery Identities.” Social Psychology Quarterly 69(4):307–24.

12.

Kessler

Ronald C.

Andrews

Colpe

L. J.

Hiripi

Mroczek

D. K.

Normand

S.-L. T.

Walters

E. E.

Zaslavsky

A. M.

2002. “Short Screening Scales to Monitor Population Prevalences and Trends in Non-specific Psychological Distress.” Psychological Medicine 32(6):959–76.

13.

Kessler

Ronald C.

Barker

Peggy R.

Colpe

Lisa J.

Epstein

Joan F.

Gfroerer

Joseph C.

Hiripi

Eva

Howes

Mary J.

Normand

Sharon-Lise T.

Manderscheid

Ronald W.

Walters

Ellen E.

Zaslavky

Alan M.

2003. “Screening for Serious Mental Illness in the General Population.” Archives of General Psychiatry 60(2):184–89.

14.

Kessler

Ronald C.

Berglund

Patricia

Chiu

Wai Tat

Demler

Olga

, et al 2004. “The US National Comorbidity Survey Replication (NCS-R): Design and Field Procedures.” International Journal of Methods in Psychiatric Research 13(2):69–92.

15.

Kessler

Ronald C.

Chiu

Wai Tai

Demler

Olga

Merikangas

Kathleen R.

Walters

Ellen E.

2005. “Prevalence, Severity, and Comorbidity of 12-month DSM-IV Disorders in the National Comorbidity Survey Replication.” Archives of General Psychiatry 62(6):617–27.

16.

Kessler

Ronald C.

Ustun

T. Bedirhan

. 2004. “The World Mental Health (WMH) Survey Initiative Version of the World Health Organization (WHO) Composite International Diagnostic Interview (CIDI).” International Journal of Methods in Psychiatric Research 13(2):93–121.

17.

Kleim

Birgit

Vauth

Roland

Adam

Gundula

Stieglitz

Rolf-Dieter

Hayward

Peter

Corrigan

Patrick

. 2008. “Perceived Stigma Predicts Low Self-efficacy and Poor Coping in Schizophrenia.” Journal of Mental Health 17(5):482–91.

18.

Lazarus

Richard S.

Folkman

Susan

. 1984. Stress, Appraisal, and Coping. New York: Springer.

19.

Link

Bruce G.

1987. “Understanding Labeling Effects in the Area of Mental Disorders: An Assessment of the Effects of Expectations of Rejection.” American Sociological Review 52(1):96–112.

20.

Link

Bruce G.

Cullen

Francis T.

Struening

Elmer

Shrout

Patrick E.

, and Bruce P. Dohrenwend. 1989. “A Modified Labeling Theory Approach to Mental Disorders: An Empirical Assessment.” American Sociological Review 54(3):400–23.

21.

Link

Bruce G.

Mirotznik

Jerrold

Cullen

Francis T.

1991. “The Effectiveness of Stigma Coping Orientations: Can Negative Consequences of Mental Illness Labeling Be Avoided?” Journal of Health and Social Behavior 32(3):302–20.

22.

Link

Bruce G.

Phelan

Jo C.

2010. “Labeling and Stigma.” Pp. 571–87 in A Handbook for the Study of Mental Health: Social Contexts, Theories, and Systems, 2nd ed., edited by Scheid

T. L.

Brown

T. N.

New York: Cambridge University Press.

23.

Link

Bruce G.

Phelan

Jo C.

Bresnahan

Michaeline

Stueve

Ann

Pescosolido

Bernice A.

1999. “Public Conceptions of Mental Illness: Labels, Causes, Dangerousness, and Social Distance.” American Journal of Public Health 89(9):1328–33.

24.

Link

Bruce G.

Struening

Elmer L.

Neese-Todd

Sheree

Asmussen

Sara

Phelan

Jo C.

2001. “The Consequences of Stigma for the Self-esteem of People with Mental Illnesses.” Psychiatric Services 52(12):1621–26.

25.

Link

Bruce G.

Struening

Elmer L.

Neese-Todd

Sheree

Asmussen

Sara

Phelan

Jo C.

2002. “On Describing and Seeking to Change the Experience of Stigma.” Psychiatric Rehabilitation Skills 6(2):201–31.

26.

Link

Bruce G.

Wells

Jennifer

Phelan

Jo C.

Yang

Lawrence

. 2014. “Measuring and Understanding the Importance of ‘Symbolic Interaction’ Stigma.” Unpublished manuscript, Mailman School of Public Health, Columbia University.

27.

Martin

Jack K.

Pescosolido

Bernice A.

Tuch

Steven A.

2000. “Of Fear and Loathing: The Role of ‘Disturbing Behavior,’ Labels, and Causal Attributions in Shaping Public Attitudes toward People with Mental Illness.” Journal of Health and Social Behavior 41(2):208–23.

28.

McCall

George J.

2003. “The Me and the Not-me: Positive and Negative Poles of Identity.” Pp. 11–25 in Advances in Identity Theory and Research, edited by Burke

P. J.

Owens

T. J.

Serpe

Thoits

P. A.

New York: Kluwer Academic/Plenum.

29.

McCall

George J.

Simmons

J. L.

1978. Identities and Interactions. New York: Free Press.

30.

Markowitz

Fred E.

2001. “Modeling Processes of Recovery from Mental Illness: Relationships between Symptoms, Life Satisfaction, and Self-concept.” Journal of Health and Social Behavior 42(1):64–79.

31.

Markowitz

Fred E.

2005. ‘‘Sociological Approaches to Recovery.’’ Pp. 85–100 in Recovery and Mental Illness: Consumer Visions and Research Paradigms, edited by Ralph

R. O.

Corrigan

P. W.

Washington, DC: American Psychological Association.

32.

Mirowsky

John

Ross

Catherine E.

2003. Social Causes of Psychological Distress, 2nd ed. New York: Aldine de Gruyter.

33.

Pearlin

Leonard I.

Bierman

Alex

. 2013. “Current Issues and Future Directions in Research into the Stress Process.” Pp. 325–40 in Handbook of the Sociology of Mental Health, 2nd ed., edited by Aneshensel

C. S.

Phelan

J. C.

Bierman

Dordrecht, Netherlands: Springer.

34.

Pescosolido

Bernice A.

Martin

Jack K.

Long

J. Scott

Medina

Tait R.

Phelan

Link

Bruce G.

2010. “‘A Disease Like Any Other’? A Decade of Change in Public Reactions to Schizophrenia, Depression, and Alcohol Dependence.” American Journal of Psychiatry 167(11):1321–30.

35.

Pescosolido

Bernice A.

Monahan

John

Link

Bruce G.

Stueve

Ann

Kikuzawa

Saeko

. 1999. “The Public’s View of the Competence, Dangerousness, and Need for Legal Coercion of Persons with Mental Health Problems.” American Journal of Public Health, 89(9):1339–45.

36.

Phelan

Jo C.

Link

Bruce G.

Stueve

Ann

Pescosolido

Bernice A.

2000. “Public Conceptions of Mental Illness in 1950 and 1996: What Is Mental Illness and Is It to Be Feared?” Journal of Health and Social Behavior 41(2):188–207.

37.

Quadagno

Jill S.

Antonio

Robert J.

1975. “Labeling Theory as an Oversocialized Conception of Man: The Case of Mental Illness.” Sociology and Social Research 60(1):33–45.

38.

Rabkin

Judith

. 1980. “Determinants of Public Attitudes about Mental Illness: Summary of the Research Literature.” Pp. 15–26 in Attitudes toward the Mentally Ill: Research Perspectives, edited by Rabkin

Gelbland

Lazer

National Institute of Mental Health, Science Reports Branch, Washington, DC: Government Printing Office.

39.

Ritsher

Jennifer Boyd

Phelan

Jo C.

2004. “Internalized Stigma Predicts Erosion of Morale among Psychiatric Outpatients.” Psychiatry Research 129(3):257–65.

40.

Rogers

Joseph W.

Buffalo

M. D.

1974. “Fighting Back: Nine Modes of Adaptation to a Deviant Label.” Social Problems 22(1):101–18.

41.

Rosenberg

Morris

. 1979. Conceiving the Self. New York: Basic Books.

42.

Rosenfield

Sarah

. 1997. “Labeling Mental Illness: The Effects of Received Services and Perceived Stigma on Life Satisfaction.” American Sociological Review 62(4):660–72.

43.

Scheff

Thomas

. 1966. Being Mentally Ill: A Sociological Theory. Chicago: Aldine.

44.

Shih

Margaret

. 2004. “Positive Stigma: Examining Resilience and Empowerment in Overcoming Stigma.” Annals of the American Academy of Political and Social Science 591(1):175–85.

45.

Snow

David A.

Anderson

Leon

. 1987. “Identity Work among the Homeless: The Verbal Construction and Avowal of Personal Identities.” American Journal of Sociology 92(6):1336–71.

46.

StataCorp. 2007. Stata Statistical Software: Release 10.0. College Station, TX: Stata Corporation.

47.

Stout

Patricia A.

Villegas

Jorge

Jennings

Nancy A.

2004. “Images of Mental Illness in the Media: Identifying Gaps in the Research.” Schizophrenia Bulletin 30(3):543–61.

48.

Stryker

Sheldon

. 1980. Symbolic Interactionism: A Social Structural Version. Menlo Park, CA: Benjamin/Cummings.

49.

Stryker

Sheldon

Serpe

Richard T.

1982. “Commitment, Identity Salience, and Role Behavior: Theory and Research Example.” Pp. 199–218 in Personality, Roles, and Social Behavior, edited by Ickes

Knowles

New York: Springer-Verlag.

50.

Thoits

Peggy A.

1985. “Self-labeling Processes in Mental Illness: The Role of Emotional Deviance.” American Journal of Sociology 91(2):221–49.

51.

Thoits

Peggy A.

1992. “Identity Structures and Psychological Well-being: Gender and Marital Status Comparisons.” Social Psychology Quarterly 55(3):236–56.

52.

Thoits

Peggy A.

2011. “Resisting the Stigma of Mental Illness.” Social Psychology Quarterly 74(1):6–28.

53.

Thoits

Peggy A.

Link

Bruce G.

2016. “Stigma Resistance and Well-being among Persons in Treatment for Psychosis.” Society and Mental Health 6(1):1–20.

54.

Thomas

William Isaac

Thomas

Dorothy Swain

. 1928. The Child in America: Behavior Problems and Programs. New York: Knopf.

55.

Vauth

Roland

Kleim

Birgit

Wirtz

Markus

Corrigan

Patrick W.

2007. “Self-efficacy and Empowerment as Outcomes of Self-stigmatizing and Coping in Schizophrenia.” Psychiatry Research 150(1):71–80.

56.

Wahl

Otto F.

1995. Media Madness: Public Images of Mental Illness. Piscataway, NJ: Rutgers University Press.

57.

Wahl

Otto F.

1999. “Mental Health Consumers’ Experience of Stigma.” Schizophrenia Bulletin 25(3):467–78.

58.

Wang

Philip S.

Lane

Michael

Olfson

Mark

Pincus

Harold A.

Wells

Kenneth B.

Kessler

Ronald C.

2005. “Twelve-month Use of Mental Health Services in the United States: Results from the National Comorbidity Survey Replication.” Archives of General Psychiatry 62(6):629–40.

59.

Yanos

Philip T.

Roe

David

Markus

Keith

Lysaker

Paul H.

2008. “Pathways between Internalized Stigma and Outcomes Related to Recovery in Schizophrenia Spectrum Disorders.” Psychiatric Services 59(12):1437–42.

60.

Zuckerman

Marvin

Kuhlman

D. Michael

Joireman

Jeffrey

Teta

Paul

Kraft

Michael

. 1993. “A Comparison of Three Structural Models for Personality: The Big Three, the Big Five, and the Alternative Five.” Journal of Personality and Social Psychology 65(4):757–68.