Who Comes Out with Their Mental Illness and How Does it Help?

Abstract

Coming out with mental illness may be an effective strategy for reducing self-stigma. This study examined predictors and consequences of coming out. Participants (N=106) with severe mental illness who reported being out (n=79) or not out (n=27) endorsed benefits of being out (BBO) and reasons for staying in (RSI). Predictors from baseline measures were self-stigma, insight, and psychiatric diagnosis. Three outcome measures -- basic psychological needs, care engagement, and depression – were also completed at baseline and one-month follow-up. Among participants already out, BBO and RSI were significantly and independently associated with self-stigma, insight, and lifetime affective diagnoses. In terms of consequences, BBO was associated with cross-sectional and one-month measures of engagement for those already out, but not for closeted participants. Among closeted participants, BBO was associated with baseline and one-month measures of basic psychological needs. Implications for strategies meant to promote disclosure in order to decrease self-stigma are considered.

The impact of serious mental illness is worsened by self-stigma (Livingston and Boyd, 2010); i.e., internalizing negative stereotypes of mental illness leads to diminished self-esteem and self-efficacy causing a “why try” effect (Corrigan et al., in press a; Corrigan et al., 2009a). The why try effect represents a behavioral fatalism that results from self-stigma; “Why try to get a job; someone like me is not worthy?” “Why try to live independently; someone like me is not able.” What then might be effective ways to deal with self-stigma? Although one might think keeping secret one’s mental health experiences might protect people from stigma, suppressing important aspects of identity to stay in the closet has egregious effects on mental and physical health, relationships, employment, and well-being (Pachankis, 2000; Smart and Wegner, 2000). Conversely, individuals from varied stigmatized groups who publicly identify with their groups often experience less stress arising from prejudice and better self-esteem, a result, for example, found in African Americans (Branscombe et al., 1999) and gay men and lesbians (Halpin and Allen, 2004). The latter group is especially relevant for understanding the experiences of identity and mental illness because LGBTQ orientation, like mental illness, are conditions that Goffman (Goffman, 1963) called discreditable, which can be kept hidden from the public. Indication of discreditable group membership is not readily obvious to the public as compared to skin color for ethnicity. Hence, effects of group identification are influenced by public disclosure -- coming out. Strategic disclosure of closeted secrets not only diminishes hurtful effects of self-stigma, but often leads to a sense of personal empowerment and improved self-esteem (Beals et al., 2009). In different ways, this strategy is incorporated into programs meant to diminish the self-stigma of mental illness (Yanos, Lucksted, Drapalski, Roe, & Lysaker, 2015.

If mental illness identity has potential for a positive impact, then disclosure of that identity might yield health and other life benefits. In a previous study, researchers conducted qualitative interviews with gay men and lesbians to identify specific attitudes and behaviors that exemplify the costs and benefits of staying in the closet or coming out (Corrigan et al., 2009b). In a subsequent study, investigators transposed these findings into a quantitative assessment of coming out with mental illness (Corrigan et al., 2010). An exploratory factor analysis of responses provided by 85 people with serious mental illnesses yielded a 2-factor structure: the benefits of being out versus the reasons for staying in. Being out proved to be a protective factor against self-stigma’s effects on quality of life and augmented a sense of personal empowerment.

The current study sought to further understand coming out. Self-stigma may be an important predictor given that decisions to come out may be in reaction to this experience. Explicit self-stigma has been described as a progressively harmful process starting with being aware of public stereotypes (I know the public thinks people with mental illness are incompetent), agreeing with it (I agree; people with mental illness are incompetent), applying it to one’s self (I have a mental illness so I must be incompetent) leading to harm (because I am incompetent, I am a less capable person) (Corrigan & Rao, 2012). We expect to show awareness and agreement of stereotypes will motivate people leading to perceived benefits of coming out. However, applying stereotypes with its resulting harm to self-esteem may actually motivate people to stay in the closet.

We hypothesize that people who are aware of their mental health challenges are better able to understand benefits of disclosure. Also, we examine two aspects of diagnosis on coming out decisions. Because self-stigma harms self-esteem, we expected diagnoses leading to diminished self-esteem, such as defined as a major affective disorder, would be associated with coming out. In addition, because psychosis can undermine insight, we expect those with a history of psychotic disorders to perceive fewer barriers to being out.

Previous research suggested coming out has positive effects on quality of life and personal empowerment (Corrigan et al., 2010; Yanos et al., 2015). Extending these findings, we expect coming out to enhance related aspects of empowerment and self-determination. Deci and Ryan’s (2000) self-determination theory specifies three relevant basic psychological needs: autonomy, experiencing choice and feeling like the initiator of one’s actions; competence, succeeding at challenging tasks to attain desired outcomes; and relatedness, establishing a sense of mutual respect and reliance with others. Of its many aspects, self-determination includes decisions about service engagement (Corrigan et al., 2012), and we expect coming out to enhance attitudes about services and care seeking. Finally, we expect to show enhanced psychological needs and care seeking will diminish clinical depression.

Methods

Data and analyses in this observational study are from a larger investigation on predictors and correlates of the treatment decision process among people with serious mental illness that was approved by the Institutional Review Board at the Illinois Institute of Technology. At baseline, participants completed a measure of coming out as well as assessments of hypothesized predictors of disclosure: self-stigma (both explicit and implicit), cognitive insight, and lifetime diagnosis. At baseline and one-month follow-up, research participants completed measures of expected outcomes pertinent to coming out: basic psychological needs (autonomy, competence, and relatedness), service engagement (attitudes about psychiatric medications and intentions to seek care), and clinical depression.

Participants

Research participants were solicited from two large psychiatric rehabilitation programs serving people with serious mental illness in the Chicago area. One hundred and six research participants completed the protocol at baseline and one-month follow-up; demographics are summarized in Table 1. There were no dropouts. The sample was about 45% female and three quarters heterosexual. Participants were about 75% African American, more than 70% had earned a high school diploma, and a majority of participants were unemployed. Seriousness of mental illness of participants is evident from several demographic indicators. About 85% of research participants received SSI or SSDI because of their psychiatric illness. About 70% were unemployed and 94% reported annual incomes of $25,000 or less. Psychiatric diagnoses were determined through administration of the Mini International Neuropsychiatric Interview (MINI) by trained interviewers (Lecrubier et al., 1997). The MINI is a short, semi-structured interview that provides current and lifetime diagnoses of serious mental illness and substance use disorders. MINI ratings have been shown to be significantly correlated with DSM and ICD diagnoses (Sheehan et al., 1998). As can be seen from Table 1, all research participants had diagnoses of major affective disorder (major depression or bipolar disorder) and/or psychotic disorder.

Table 1.

Demographics of research participants

Variable	M(SD) or %
Age	50.5 (7.2)
Gender
Female	44.9%
Male	53.3
Transsexual	1.9
Sexual orientation
Straight	76.6%
Gay	6.5
Lesbian	3.7
Bisexual	7.5
Other	2.8
Prefer not to answer	2.8
Race/Ethnicity1
African/African American	76.6%
Asian/Asian American	0.9
European/European American	20.6
Latino	6.5
Native American	3.7
Pacific Islander	0.0
Education
Some high school	28.0%
High school diploma	17.8
Some college	39.3
Associate’s degree	6.5
Bachelor’s degree	3.7
Some graduate/professional school	0.9
Graduate degree	0.9
Other	1.9
Employment
Full time	2.8%
Part time	7.5
Retired	6.5
Attending school	9.3
Unemployed	69.2
Volunteer	17.8
Other	6.5
Income
$0–$24,999	94.4%
$25,000–$49,999	4.7
$50,000–$74,999	0.9
Disability status: Receive SSI or SSDI?
Yes	84.1%
Marital Status
Single	51.4%
Married or partnered	5.6
Widowed	6.5
Separated or divorced	35.5
MINI Lifetime Diagnosis2
Major Depressive Disorder	58.3%
Bipolar Disorder	41.7
PTSD	31.1
Alcohol Abuse or Dependence	28.2
Other Substance Abuse or Dependence	24.2
Psychotic Disorder	35.0

Note.

¹Total greater than 100% because participants might check more than one ethnic group.

²Total greater than 100% because some participants received multiple diagnoses.

MINI: the Mini International Neuropsychiatric Interview

Measures

The Coming Out with Mental Illness Scale (COMIS) was developed using mixed methods and community based participatory research to assess aspects of disclosure and mental illness (Corrigan et al., 2010). Results of qualitative interviews yielded items that represent perceived benefits of being out (BBO) and reasons for staying in (RSI) (Corrigan et al., 2009b). An item analysis of subsequent quantitative data yielded a COMIS that begins with a yes/no question about whether the respondent is “out” about his or her mental illness with” out” defined as “you decided to tell most of your family, friends, and acquaintances that you have a mental illness?” People who replied in the affirmative then rated their agreement with different BBOs (7 items) and RSIs (14 items) on a 7-point agreement scale (7=strongly agree). Alphas for the two scales of participants in this study were 0.73 and 0.95 respectively. Sample items include “I came out of the closet to gain acceptance from others” and “In the past, I stayed in the closet to avoid being labeled as a person with mental illness.” Those responding negatively to the question about whether they are out described BBO (7 items) and RSI (14 items) in terms of future perspective about coming out. Alphas for these scales were 0.90 and 0.89 respectively. Results of an earlier study with 85 people with serious mental illness support the reliability and validity of the COMIS (Corrigan et al., 2010).

Predictors of Coming Out

Several scales have emerged to describe the explicit self-stigma of mental illness (Boyd et al., 2013). We selected the Self-Stigma of Mental Illness Scale (SSMIS) because scores correspond with four stages of a progressive model of self-stigma: being aware of the stereotypes of mental illness, agreeing with them, applying them to one’s self, thereby harming one’s sense of self-esteem (Corrigan et al., 2006). The SSMIS comprises 40 items (10 per stage) representing endorsement of mental illness stereotypes identified in mixed methods research across the four stages: aware, agree, apply, and harm. Participants endorse items on a nine-point agreement scale (9=strongly agree). Research has supported its reliability and validity (Corrigan et al., 2013; Corrigan et al., 2011; Watson et al., 2007). Baseline alphas for the four SSMIS subscales from this study ranged from 0.74 to 0.90.

Explicit self-stigma has been distinguished from implicit self-stigma, namely, the degree to which stigmatizing evaluations occur outside awareness. Implicit stigma seems to independently predict clinical decision-making and related health from explicit self-stigma (Peris et al., 2008). It was assessed in this study using the Brief Implicit Associated Test for Self-Stigma (BIAT-SS). The BIAT-SS is a shorter, validated version of the standard Implicit Association Test (Greenwald et al., 2009) comprising two blocks of stimuli. Stimuli categorized more rapidly during a BIAT block that presents a target-attribute pairing (e.g., Me-Good versus Me-Bad) matches the respondent’s automatic associations and represents implicit self-stigma. A participant with more negative implicit self-stigma will respond more rapidly to the BIAT block (i.e., Me-Bad) than the corresponding block (i.e., Me-Good).

In terms of scoring, participants with more than 30% BIAT errors were excluded from the analysis (Teachman and Woody, 2003). Order of BIAT blocks is counterbalanced across administrations to diminish order effects. BIAT score computations are made by D-score calculations by computing a mean latency for each block of a BIAT (Greenwald et al., 2003). The BIAT’s D-score is computed by subtracting the block means (M1 – M2) and dividing by an inclusive standard deviation (Nosek and Sriram, 2007).

Illness awareness and insight were assessed using the Beck Cognitive Insight Scale (BCIS; Beck et al., 2004). The BCIS is a self-report instrument comprised of nine items assessing self-reflectiveness (“Some of my experiences that have seemed very real may have been due to my imagination.”) and six items measuring self-certainty (“I can trust my own judgment at all times.”) Items are endorsed on a four-point agreement scale (3 = agree completely). Higher self-reflectiveness total scores represent greater openness to consider another person’s perspective and/or objective data. Higher self-certainty represents greater willingness to consider feedback about personal beliefs and conclusions. Cronbach’s alpha of 0.65 and 0.57 were obtained for the baseline self-reflectiveness and self-certainty scales, respectively. Diagnosis was also examined as a predictor of coming out using the MINI.

Impact of Coming Out

Impact was assessed using Deci and Ryan’s (2000) Basic Psychological Needs Scale (BPNS). The BPNS is a measure of three basic psychological needs (autonomy, competence, and relatedness) which participants rate on a 7-point scale representing the degree to which participants believe the item is true of their life (7 = very true). The 21-item version of the BPNS yields scores for each psychological need; alphas ranged from 0.64 to 0.73 for the three baseline scores and 0.59 to 0.75 for one month.

Service engagement was assessed in two ways: attitudes about the effectiveness of psychiatric medications and intention to seek care. The Attitudes Towards Psychiatric Medications (ATPM) Scale was administered at baseline and one-month to assess the former (Croghan et al., 2003). ATPM items represent opinions about effectiveness and side effects of psychiatric medications using a 5-point Likert scale (5 = strongly agree). The side effects factor yielded very low alphas (0.17 at baseline and 0.48 at one-month) and was excluded from the remaining analyses. A sample item of effectiveness is, “Taking these medications helps people control their symptoms.” Four effectiveness items were summed with higher total scores representing a more positive opinion toward effectiveness. Alphas for effectiveness were acceptable: 0.77 and 0.82 at baseline and one month respectively.

Although several studies have examined relationships between stigma and care seeking (Clement et al., 2015), no measure that examines the pursuit of psychiatric and other services has emerged in the research literature. Previously, we partnered with a community based research team to generate sample services and providers including: primary care doctors, psychiatrists, counselors, ministers, and friends (Corrigan et al., 2014). Research participants were asked about the degree they agree (9 = strongly agree) with statements like, “I would speak to a psychiatrist if I were significantly anxious or depressed.” Results of an exploratory factor analysis of baseline and one-month data from research participants in this study (principal component analysis with varimax rotation) yielded a two-factor solution with eigenvalues greater than 2.0. The first factor, called professional care seeking, included items that represented a person’s engagement with primary care doctors, psychiatrists, and counselors to address anxiety or depression. The second, called extraprofessional care seeking, represented engagement with ministers or other clergy; friends or family, and peers. Alphas for the professional care seeking factor were 0.70 and 0.70 for baseline and one-month respectively; alphas for the extraprofessional factor were 0.57 and 0.53.

Decreased self-stigma and greater engagement in care should have positive effects on psychiatric symptoms, especially depression. Depression was assessed using the 20-tem Center for Epidemiological Studies Depression Scale (CESD) (Radloff, 1997; Eaton et al., 2004). Alphas were 0.57 and 0.53 for baseline and one-month respectively.

Results

Seventy nine of the 106 research participants (74.5%) reported affirmatively that they were out with their mental illness on the COMIS; 27 (25.5%) said they were not. Table 2 summarizes Pearson Product Moment Correlations for COMIS factor scores (BBO and RSI) by self-reported out group (yes/no). First, we consider the pattern of associations for those who reported being out. Benefits of being out (BBO) at baseline were not associated with implicit self-stigma. Otherwise, BBO was significantly associated with proxies of self-stigma, insight, and diagnosis. Agreement about the BBO was correlated with the earlier stages of the self-stigma model and significantly with agreement. This might infer that participants who reported being out and endorsed greater awareness and agreement of stigma were more likely to agree with the BBOs. BBO was also associated with two indices of cognitive insight: significantly with self-reflectiveness. This might suggest that research participants who were better aware of their illness and its impact endorsed more benefits of being out. BBO ratings were significantly associated with lifetime diagnosis of major depressive disorder.

Table 2.

Correlations between COMIS indices representing coming out and staying in the closet with key predictors. Correlations are summarized by whether participant self-reported as “out.”

VARIABLES	Yes, I am out with my mental illness. N=79		No, I am not out with my mental illness. N=27
	Benefits of being out BBO	Reasons for staying in RSI	Benefits of being out BBO	Reasons for staying in RSI
BIAT: self-stigma	.12	.02	.58**	−.03
SSMIS: aware	.22	.18	−.09	.26
SSMIS: agree	.26*	.08	−.11	−.01
SSMIS: apply	.12	.26*	.03	.03
SSMIS: harm	−.08	.20	.02	−.04
BCIS: self-reflectiveness	.25*	.25*	−.15	−.13
BCIS: self-certainty	.20	−.12	.30	.00
MINI Lifetime Diagnosis: Major depressive disorder	.30**	.33	−.12	−.28
MINI Lifetime Diagnosis: Bipolar Disorder	.18	.31**	.27	.10
MINI Lifetime Diagnosis: Psychotic disorder	.07	.11	−.42*	.12

Note.

^*p<.05

^**p<.01

BIAT, Brief Implicit Association Test; SSMIS, Self-Stigma of Mental Illness Scale; BCIS, Beck Cognitive Insight Scale; MINI, the Mini International Neuropsychiatric Interview

Table 3 summarizes findings from a multiple regression with the largest BBO correlates from self-stigma, cognitive insight, and diagnosis included as independent variables. The analysis was conducted as a simultaneous model. Results show the three variables were independently associated with BBO at a significant level (p<.05). R for the entire model was 0.43 accounting for 18.5% of the variance.

Table 3.

Predictors of benefits of coming out versus reasons for staying in the closet. These are findings for only the subsample of research participants who said they were “out” (n=79).

INDEPENDENT VARIABLE	BETA	t-TEST	p
Benefits for being out (BBO) from the COMIS
SSMIS: Self-stigma: agree	.23	2.16	<.05
BCIS: self-reflectiveness	.24	2.16	<.05
MINI diagnosis: major depressive disorder	.22	1.96	<.05
	R=.43
Reasons for staying (RSI) in from the COMIS
SSMIS: Self-stigma: apply	.29	2.69	<.01
BCIS: self-reflectiveness	.19	1.79	<.10
MINI diagnosis: Bipolar disorer	.32	2.95	<.005
	R=.48

Note. COMIS, Coming Out with Mental Illness Scale; SSMIS, Self-Stigma of Mental Illness Scale; BCIS, Beck Cognitive Insight Scale; MINI, the Mini International Neuropsychiatric Interview

A slightly different pattern of correlations was found between reasons for staying in (for those 79 participants who reported they were out) and measures of self-stigma, cognitive insight, and diagnosis. Self-stigma was significantly associated with RSI, but this time at the more harmful stages of the progressive model; a significant association was found with applying stigma to one’s self. RSI was significantly correlated with self-reflectiveness from the BCIS. RSI was associated with a lifetime diagnosis of affective disorder, in this case bipolar disorder. The multiple regression examining RSI predictors is in Table 3, once again entering the largest correlates from self-stigma, cognitive insight, and diagnosis as independent variables into a simultaneous model. Results showed the three variables independently accounted for RSI variance. Two of three variables were significant at p<.05. R for the regression analysis was 0.48 accounting for 23.0% of the variance.

The right hand columns of Table 2 shows correlations between BBO and RSI for those who self-reported they were not out with their mental illness. This sample was only 27 thus diminishing the statistical power of these analyses. Note for example, that none of the self-stigma, cognitive insight, or diagnostic indices were significantly associated with RSI for the “not out” sample. Still, two significant and robust associations were found between correlations of self-stigma and diagnosis with BBO. People who scored relatively higher on implicit self-stigma on the BIAT were more likely to endorse benefits of being out, accounting for 33.6% of the variance. Moreover, people with lifetime diagnoses of psychoses were less likely to endorse benefits of being out: r=−0.42.

Impact of BBO and RSI

Table 4 summarizes Pearson Product Moment Correlations between BBO and RSI with impact measures: basic psychological needs, attitudes about medication and care seeking, and depression. Associations represented assessments at baseline as well as COMIS impact on these variables one-month later. Significance in Table 4 was marked both in terms of p<.05 and those that met criteria as Bonferroni corrections.

Table 4.

Correlations between COMIS indices representing coming out and staying in the closet with selected outcomes. Correlations are summarized by whether participant self-reported as “out.”

VARIABLES	Yes, I am out with my mental illness. N=79		No, I am not out with my mental illness. N=27
	Benefits of being out BBO	Reasons for staying in RSI	Benefits of being out BBO	Reasons for staying in RSI
BPNS: autonomy	.20	−.17	.39*	−.28
BPNS: autonomy 1 month later	−.04	−.32*	.40*	−.52**
BPNS: competence	.16	−.04	.30	−.19
BPNS: competence1 month later	.06	−.20	.09	−.37
BPNS: relatedness	.09	−.12	.53***	−.43*
BPNS: relatedness 1 month later	−.10	−.21	.51**	−.70***
ATPM: Medication effectiveness	.37**	.02	.23	.18
ATPM: Medication effectiveness 1 month later	.28*	.11	.25	−.23
Care seeking: prof	.20	.05	.27	.05
Care seeking: prof 1 month later	.16	.05	.34	.03
Care seeking: extraprof	.36***	−.23	−.11	.04
Care seeking: extraprof 1 month later	.23*	.12	−.07	.14
CESD	−.13	.09	−.01	−.07
CESD one month later	−.05	.23	.01	.15

Note.

^*p<.05

^**p<.01

^***p<.005

BPNS, Basic Psychological Needs Scale; CESD, Center for Epidemiologic Studies Depression Scale

Interesting differences in basic psychological needs were observed between research participants who were and were not out. BBO for participants who were out was not associated with BPNS. Inverse relationships were found between RSI and BPNS though representing only one of six associations. Far more significant associations were found between impact measures and BBO and RSI for the sample of not out research participants, despite the much smaller N in these analyses. BBO was positively associated with autonomy and relatedness at baseline and one-month later with the associations between BBO and relatedness meeting the Bonferroni criterion. Similarly, RSI was inversely associated with basic psychological needs; four of six associations met criteria with two of these meeting the Bonferroni corrections.

Table 4 also includes correlation coefficients between BBO and RSI factors and indices of engagement in future care. In this case, the only set of significant associations were between BBO for the group of participants who were out and five indices of engagement. Those who endorsed more benefits to being out were likely to report medication as personally more effective on the ATPM at baseline and one month later, with the latter index meeting the adjusted Bonferroni criterion. In addition, those who endorsed greater benefits to being out were significantly more likely to seek care at baseline and one month. Specifically, BBO was significantly associated with seeking extraprofessional care at baseline and one-month with the cross sectional association meeting the adjusted Bonferroni criterion. None of the other relationships among engaging in care and RSI for the group of out participants -- nor for BBO and RSI of the group of “not out” participants -- was significant. Coming out did not seem to have an effect on depression.

Discussion

This study sought to describe reasons why people perceive benefits for being out versus reasons for staying in. Self-stigma stages were associated with both BBO and RSI. BBO was associated with the earlier stages of self-stigma (being aware of and agreeing with stereotypes) while RSIs were associated with the later stages (apply leading to harm related to lower self-esteem). These findings suggest self-stigma might motivate people to come out when the impact of self-stigma is less personally aversive. When self-stigma leads to personal application and harm, it may lead someone to stay in the closet. BBO and RSI also seem to be impacted by insight. Namely, people who show introspection about their psychiatric challenges and are willing to acknowledge fallibility in their decisions are more likely to agree with both the benefits of coming out as well as the reasons for staying in. Insight increases the person’s understanding of both the positives and negatives of coming out. We need to keep in mind that correlation does not imply causation; subsequent research is needed over time to validate some of these findings.

BBO and RSI were both associated with lifetime diagnoses of affective disorders, though interestingly, in different directions. BBO was associated with lifetime diagnosis of major depressive disorder (MDD) with those manifesting MDD more likely to agree with the BBO. RSI was correlated with a lifetime diagnosis of bipolar disorder. This pattern may suggest a paradoxical effect. Benefits to coming out emerge when the person suffers the loss of self-esteem consistent with major depression. People suffering a diminished positive sense of self may benefit from the empowerment of being out. Conversely, people experiencing an enhanced sense of personal esteem consistent with mania may endorse reasons for staying in; hence the paradox. Of course, those with an expansive sense of self while manic eventually deflate. Perhaps it is this significant loss of self-esteem that drives people further into the closet. Multiple regression analyses showed self-stigma, insight, and diagnosis independently influenced BBO and RSI, accounting for more than 18.5% of the variance.

Generally, few predictors of perceived benefits of being out and reasons for staying in for those who reported NOT being out were found. BBO for people who are not out was associated with implicit self-stigma measured on the BIAT. This means the unconscious effects of self-stigma, rather than the more manifest impact of the phenomena, influences BBO. In addition, lifetime diagnosis of psychosis was inversely associated with BBO. This seemed to suggest the disorganizing effect of illnesses such as schizophrenia undermines the perceived benefits of being out.

Findings from the study also suggested positive and negative impacts of BBO and RSI. For the most part, BBO and RSI seemed to have no associations with basic psychological needs for the person who is out. However, clear patterns emerged for those still closeted. Autonomy and relatedness were positively associated with BBO cross-sectionally at baseline as well as at one-month follow-up. Conversely, and as expected, RSI was inversely associated with three of six basic needs scales. This pattern suggests basic psychological needs are more poignant for the person who is in the closet, wrestling with the idea of coming out. Relevance of these psychological needs may no longer be pressing once the person is out.

BBO for the person who is out had positive impacts on engagement. Those who agree with BBO were more likely to endorse attitudes about medication effectiveness at baseline and one-month. They were also more likely to seek care, although extraprofessional care seemed to be more appealing that professional care. Despite better service engagement, BBO had no effects on clinical depression.

This study has several limitations that need to be addressed in future research. Further research needs to describe mediators and moderators of predictors and consequences. Longitudinal data were not collected for predictor analyses thereby limiting assumptions about the direction of results. In terms of mediators, other research showed BBO actually mediates the effects of personal empowerment on quality of life (Corrigan et al., 2010). Future research should include larger samples of structural equation models to test mediating and moderating effects. Findings failed to show associations between BBO and depression; one might expect actions that decrease self-stigma would positively impact depression. Research needs to further unpack consequences of disclosure related to the broad construct of depression. Does disclosure and BBO, in fact, enhance self-esteem and self-efficacy? In turn, might these changes be associated with clinical depression, or even anxiety? A third issue is how course of disorder impacts self-stigma and coming out. Course variables might include length of illness, breadth and depth to which illness has been disabling, and experiences of remission and relapse. Equally important here is to track quality of self-stigma interventions (e.g., incorporate fidelity measures). Given that mental health providers can be stigmatizing (Schulze, 2007), future research might show strategies that fail to promote recovery worsen self-stigma and discourage coming out.

Given findings of this study, how might disclosure be used to decrease self-stigma and enhance positive outcomes? Research reviews have grouped approaches to erasing self-stigma into education-based strategies (teaching people facts meant to challenge internalized stereotypes and cognitive behavioral strategies to use these facts) versus strategic disclosure (assist people to come out) (Mittal et al., 2012; Yanos et al., 2015). Coming Out Proud (COP) to Erase the Stigma of Mental Illness is a brief structured program led by people who are out to help peers identify the pros and cons of disclosing in different settings (e.g., work, faith-based community, or with extended relatives), disclose safely, and draft a personally meaningful story. Results of two randomized controlled trials suggest COP decreases the stress related to self-stigma, enhances self-esteem, and diminishes depression (Corrigan et al., in press b; Rusch et al., 2014). Findings from this study may help to further understand the impact of strategic disclosure. Specifically, how do insight and basic psychological needs explain predictors and consequences of strategic disclosure?