Extending the Minority Stress Model to Understand Mental Health Problems Experienced by the Autistic Population

Abstract

Research into autism and mental health has traditionally associated poor mental health and autism as inevitably linked. Other possible explanations for mental health problems among autistic populations have received little attention. As evidenced by the minority disability movement, autism is increasingly being considered part of the identities of autistic people. Autistic individuals thus constitute an identity-based minority and may be exposed to excess social stress as a result of disadvantaged and stigmatized social status. The authors test the utility of the minority stress model as an explanation for the experience of mental health problems within a sample of high-functioning autistic individuals (n = 111). Minority stressors including everyday discrimination, internalized stigma, and concealment significantly predicted poorer mental health, despite controlling for general stress exposure. These results indicate the potential utility of minority stress in explaining increased mental health problems in autistic populations. Implications for research and clinical applications are discussed.

Keywords

autism social stress discrimination Asperger syndrome

Over the past two decades, the meaning of the diagnoses of Asperger syndrome (AS) and high-functioning autism (HFA) has changed, with more autistic individuals considering autism central to their identities as opposed to a disorder (Aylott 2000; Bagatell 2007; Elliman 2011). The vast majority of research into HFA and AS has tended toward researching the autism spectrum through the biomedical model, specifically its etiology and a possible cure (Pellicano, Dinsmore, and Charman 2014) rather than recognizing HFA and AS as normal forms of variation in human psychology. Less than 1 percent of autism research funding in both the United States and the United Kingdom has gone into researching adults on the spectrum or any social factors that may contribute to the high rates of mental health problems experienced by people on the autism spectrum (Pellicano et al. 2014). Increasingly, autistic people themselves are beginning to consider AS and HFA as forms of neurodiversity and a key part of their identities (Bagatell 2010; Kapp et al. 2013), as central as other social identities such as sexuality or race (Brown 2017). Under a minority model of disability, HFA and AS represent a distinct sociopolitical experience as neurominorities with wide-ranging diversity (Altman 2001; Jaarsma and Welin 2012). Neurominority is a relatively new term coined to describe those who fall under the neurodiversity model (Bascom 2012). In this study we examine how stress related to social stigma (e.g., Frost 2011) contributes to heightened rates of mental health problems experienced by the autistic community. We highlight the utility of social stress models (Meyer 2003; Meyer, Schwartz, and Frost 2008) in understanding mental health and well-being in autism.

Autism as Identity

The biomedical model aims to cure disability (Rioux, Bach, and Roeher Institute 1994). Understanding disease from this model is only logical considering the need to treat and cure life-threatening illness, but it creates tension when considering disorders such as autism, dyslexia, and dyspraxia (Ward and Meyer 1999). The biomedical model relies on identifying disease and creating meaningful interventions to cure the person who is suffering (Bagatell 2010; Rioux et al. 1994). The traditional idea of autism is one in which a person does suffer (Kanner 1971). Viewing autism from a biomedical model has been opposed, as it leads to dehumanizing research and treatment of autism (Cowen 2009). For example, it has been claimed that an autistic community cannot exist, because of a central aspect of autism: a lack of theory of mind, meaning that autistic individuals are too introspective to want to or be able to form community connections (Barnbaum 2008). Similarly, it has resulted in work in which they are compared less favorably with brain damaged monkeys (Bainbridge 2008). The minority model of disability formed partly as opposition to these notions.

The rise of the minority model of disability specifically challenged the medical model’s notions of disability (Rioux et al. 1994; Smart 2006). The minority model of disability is underpinned by the notion that one can have a condition the medical model considers a disability, but in actuality, it is a society with restrictive notions of normal that creates disability (Altman 2001; Smart 2006). The Deaf community is an example of a community considered disabled by the medical model who reject that status, considering themselves to be a cultural group defined by the use of sign language (Smart 2006). The Autism Network International was the first self-advocacy group created by and for autistic individuals, in part to combat the biomedical view of autism (Ward and Meyer 1999). Narratives of autistic self-advocacy are challenging the notions put forward from the biomedical model and under the banner of neurodiversity, claiming minority status (Kapp et al. 2013).

Individuals who are affected have come to consider autism an intrinsic part of an identity (Bagatell 2010; Kapp et al. 2013). In fact, autism is sometimes as central to the identity of autistic individuals as race, ethnicity, sexuality, gender, or nationality (Brown 2017), a claim often put forward by the neurodiversity movement (neurodiversity itself falling under a minority model of disability). Thus, the minority model of disabilities may provide a basis on which autistic individuals can consider themselves within their own terms, and with dignity (Dunn and Andrews 2015). It has even been proposed that those with AS and HFA form an ethnocultural minority akin to the Deaf community (Jaarsma and Welin 2012).

Thus, in the past two decades, what biomedical researchers originally considered a disorder has come to be a central feature of identity to some. Therefore, it is important to understand whether these minority identities leave autistic individuals vulnerable to the extra social stresses suffered by other minority communities as a result of social stigma and disadvantaged social status.

Stigma as Stress Resulting from Labeling Processes

The process by which stigmatization occurs begins with the process of labeling. A label is a definition that categorizes a person by his or her characteristics (Link and Phelan 1999). Labeling in the case of AS and HFA involves a comparison of autistic individuals against their nonautistic peers and the assignment of meaning to those differences (Bagatell 2010; Elliman 2011; Hacking 2012). Most adults with HFA or AS will be aware of the differences between themselves and those not on the autism spectrum (Aylott 2000). Being aware of these differences is not an issue until a value is assigned to them: whether it is perceived as a positive, neutral, or negative difference. Labels absorb the meaning society gives them, and thus assigning value-based meanings to labels can often perpetuate stigma (Link and Phelan 1999); for example, a stereotype attached to autism is “loner” (Aylott 2000). In a study investigating the stereotypes non-AS peers held toward autistic individuals, 9 of the top 10 terms used to describe AS individuals were negative (Wood and Freeth 2016). A separate study found that the behaviors central to autism were stigmatized (Butler and Gillis 2011).

The effects of stigma can be long lasting and limit the quality of life available to the stigmatized group (Markowitz 1998). A 2-year study on mental health and stigma showed that exposure to stigma designated the self-worth individuals afforded to themselves (Wright, Gronfein, and Owens 2000). Considering that 9 of 10 stereotypes afforded to autism are negative, there is the possibility of high exposure to stigma. Similarly, this exposure has consequences.

The stigma afforded to autistic individuals likely explains why multiple studies have found a high risk for victimization in the HFA and AS community, including physical, verbal, and sexual victimization across the life span from childhood (Little 2002), to adulthood (Rosenblatt and National Autistic Society 2008). Similarly, autistic individuals are more likely to face workplace discrimination in terms of unfair dismissal, workplace harassment, underemployment, and unemployment (Baldwin, Costley, and Warren 2014; Barnard et al. 2001; National Autistic Society 2012). Social rejection can also be internalized and self-perpetuating. For example, as a result of experiences of rejection, neurominorities may become embroiled in negative self-concepts built upon the foundation of social rejection (Link et al. 1989). Although the previously discussed research into victimization and discrimination documents high rates of exposure among autistic populations, researchers have yet to focus on the impact of victimization on the well-being of autistic individuals.

The Potential Utility of Minority Stress Theory

The primary aim of the minority stress model is to explain disparities in health between majority and stigmatized minority groups (Meyer 2003). Social stress theory hinges on the idea that social disadvantage can translate into health disparities (Schwartz and Meyer 2010). Researchers hypothesize that decreased social standing leads to stigmatized minority groups’ being exposed to more stressful life situations, with simultaneously fewer resources to cope with these events. Social structure facilitates this process through acts of discrimination and social exclusion, which are added stress burdens that socially advantaged groups are not equally exposed to.

The minority stress model has most frequently been used to explain both mental and physical health disparities (Meyer 2003). Studies have consistently shown sexual minorities to have higher stress burdens, while experiencing higher rates of poorer physical and mental health (Cochran and Mays 2000; Frost, Lehavot, and Meyer 2015; Gilman et al. 2001; Herek, Gillis, and Cogan 1999; Meyer 2003; Meyer and Dean 1998; Zietsch et al. 2011). To a lesser degree, the minority stress model has been used to highlight disparities in added stress and negative health outcomes in African American populations (Feagin 1991; Williams and Williams-Morris 2000). Most pertinent for the study at hand, minority stressors have also been shown to affect the severity of depressive symptoms experienced by those with physical disabilities (Brown 2017).

Four premises underpin the minority stress model. The first premise is that not all differences are discrepancies; an increase, for example, of certain illnesses with age is a difference to be expected and is therefore not a discrepancy (Schwartz and Meyer 2010). The second premise is that the theory is based on the law of averages and average effects. Although it is hypothesized that the social disadvantage influences the group in the entirety, if a subgroup remains unaffected, it does not falsify the overall theory (Schwartz and Meyer 2010). The third premise is that the social stress theory applies causally to overall health rather than specific disorders. The fourth premise is that social stress theory is specifically about the sociological category of disadvantage produced by exclusionary social hierarchies, rather than anything specific about that group (Schwartz and Meyer 2010). In essence, the social group is devalued on the basis of societal norms, in keeping with the minority model of disability, which posits that disability (and comorbid psychological outcomes) stem from an inflexible society that has a preference for nondisabled individuals (Smart 2006). As shown in the previous paragraph, support for this was shown by Brown (2017), who demonstrated the utility of understanding perceived stigma and coping in populations with physical disabilities and how it may enhance our understanding of mental health outcomes.

Schwartz and Meyer (2010) pointed out that it is unreasonable for any researcher to extend the model when there is no existing documentation for disparities between populations. A health disparity exists between autistic and nonautistic individuals, with those on the spectrum regularly found to have higher rates of physical and mental health problems (Baldwin and Costley 2016; Baldwin et al. 2014; Gillberg et al. 2016; Hirvikoski et al. 2016; Kamio, Inada, and Koyama 2013; Locke et al. 2010; Shefcyk 2015). A recent study showed an elevated risk for premature mortality for autistic individuals by on average two decades compared with nonautistic peers (Hirvikoski et al. 2016). The predominant cause of early death in individuals with HFA and AS was suicide (Hirvikoski et al. 2016). Rates of depression, suicidality, post-traumatic stress disorder, and poorer mental health are all higher in autistic populations than nonautistic populations (Gillberg et al. 2016; Kerns, Newschaffer, and Berkowitz 2015; McManus 2009; Mikami et al. 2009).

The Present Study

Applying the minoity stress model to understaning social factors relevant to health in the context of HFA and AS could begin to account for the additional stress burden faced by the autistic community and potentially redefine what is known about autism and psychological well-being. In the present study we extended the minority stress model to examine the extent to which stigma-related stressors are associated with diminished well-being experienced by the HFA and AS population. We hypothesized that there would be a relationship between minority stressors and poorer mental health outcomes, such that greater amounts of reported minority stress would be associated with poorer mental health and well-being. In testing the potentially unique contribution minority stress makes to mental health among individuals with HFA and AS, we further hypothesized that associations between minority stress and mental health and well-being outcomes would persist above and beyond the contribution of general stressful life events known to affect the health of everyone, regardless of disadvantaged social status (Frost et al. 2015).

Method

Participants

An online survey was used to test the present study’s hypothesis. Conducting the study using the Internet allowed a method consistent with the way in which autistic individuals communicate regularly; it has been noted that the Internet allows communication unfettered by social interaction (Bagatell 2010; Benford and Standen 2009; Hacking 2012; Jordan 2010). The survey, which is detailed below, was circulated to autistic individuals via the Qualtrics survey system. Inclusion criteria were a minimum age of 18 years and to consider oneself autistic. An official diagnosis was not necessary to participate. This decision was made to ensure that those who have been unable to access a diagnosis because of cost or personal circumstance, but still feel part of the autistic community, could participate (as has been done in other studies, such as Kapp et al. 2013).

A total of 142 participants completed the survey. All participants who had extensive missing data (i.e., multiple variables were missing data) were removed (n = 31), resulting in a final sample of 111 participants. Table 1 presents the demographic information for the final analytic sample. Potential demographic limitations (e.g., gender) are discussed later.

Table 1.

Sample Demographics (n = 111).

Characteristic	Percentage	n
Gender
Female	72.1	82
Male	21.7	22
Other	6.2	7
Race/ethnicity
British white	45.0	50
Other white	39.6	44
Mixed multiple	7.2	8
Asian	1.8	2
Other	6.3	7
Official Diagnosis
Yes	70.3	78
No	29.7	33
Autism type
Asperger syndrome	82.9	92
Classic autism	10.8	12
Pervasive development disorder	6.3	7
Age in years	M = 35.8	SD =11

Procedure

A survey was developed using the measures described below to assess minority stress experiences in the autistic community. Nonprobability sampling techniques were used. Recruitment was conducted within the following online groups: Aspergers Reality, Autistic Women’s Association, The Aspie Cloud, Asperger Safe Room, Adults with Asperger Syndrome, Wrong Planet, Neurodiverse UK, Autism Action NZ, Autism Worldwide, Autistics UK, and Heart for Autism. It was distributed to AS and HFA community pages with permission from community moderators. Sampling evolved into a snowball technique as participants referred other people from outside these groups to the survey. After each participant had consented, he or she completed a 14-minute survey, with the chance of winning a £50 voucher to a popular online retailer. This research received a favourable ethical opinion from the University of Surrey ethics committee prior to the commencement of data collection.

Measures

Demographic Information

Participants reported their gender, age, and ethnicity. The options presented for gender were “male,” “female,” and “other.” When participants selected “other,” they were asked for a descriptor of gender they felt comfortable with. Age was reported by participants in a numerical entry box in the survey. Ethnicity was recorded in line with British census categories (as the research was based primarily in Britain). If none of the categories presented were relevant, participants could select the “other” box and were consequently asked to provide descriptors for their ethnicity and race. Diagnosis was self-reported, with participants reporting if they had an official diagnosis and, if so, providing the details of it following the procedure used by Kapp et al. (2013).

General Stressful Life Events

A stressful life events inventory (adapted from Slopen et al. 2011) was used to assess the impact of stressful life events on well-being. The inventory is a yes-or-no inventory. The measure is not related to minority stress but rather to general life stress. The measure was coded in such a way that higher scores reflected more stressful life events in the 12 months prior to taking the survey. Questions included items such as “You recently ended a long-term relationship.”

Victimization and Discrimination Events

Victimization and discrimination measured the extent to which participants have faced discriminatory events in the last 12 months (Cronbach’s α = 0.72, eight items). The four-point scale ranges from “never” (0) to “three times or more” (3). Scores were coded (summed) in a way that higher scores reflected higher frequencies of victimization and discrimination. Questions included items such as “You were hit, beaten, physically attacked, or sexually assaulted.”

Everyday Discrimination

Experiences of everyday discrimination were measured using the Everyday Discrimination Scale, which specifically measures covert discrimination (Williams et al. 1997; Cronbach’s α = .87, eight items). The measure used a four-point scale ranging from “often” (3) to “never” (0) and asked questions such as “In your day-to-day life over the past year, how often did any of the following things: People acted as if they thought you were not smart.” The scale was coded (summed) in a way that higher scores reflected greater everyday discrimination.

Expectation of Rejection

Experiences of expecting rejection were measured using theExpectation of Rejection Scale (Link 1987; Meyer et al. 2008; six items, Cronbach’s α = 0.90). It asked respondents to consider their disability, gender, and race and then presented items such as “Employers will not hire a person like you.” Participants responded on a scale that ranged from “strongly agree” (4) to “strongly disagree” (0). Scores were coded (summed) so that higher scores reflected a higher expectation of rejection.

Outness

The outness scale measured the degree to which people on the spectrum disclosed to peers, colleagues, nonautistic friends, health care providers, or family members (adapted from Meyer et al. 2002; Cronbach’s α = 0.71, four items). Responses scaled from “out to all” (4) to “out to none” (1) The scale was coded (summed) so that higher scores reflected higher outness. The wording of “outness” was still used with regard to autism and disclosure because it is a term the community has adopted to describe disclosure (Jones, n.d.).

Physical Concealment

The physical concealment scale was designed specifically to measure the extent to which participants physically conceal behaviors associated with autism (Cronbach’s α = 0.83, five items). It asked participants to recall whether they had had certain experiences in the past 12 months. The measure contained questions such as “I have purposefully avoided disclosing being autistic on official documents (job applications etc.).” Participants responded on a scale ranging from “never” (0) to “always” (3) The measure was coded (summed) in such a way that higher scores reflected higher behavioural concealment.

Internalized Stigma

An adapted scale (specific to autism) was used to measure the extent to which individuals reject their status on the autism spectrum (adapted from Meyer and Dean 1998; Cronbach’s α = 0.84, eight items). It had items such as “You have felt alienated from yourself because of being on the Autism Spectrum.” It contained adjusted questions from an internalized homophobia scale, of which the language was changed to relevant terms, but the concept remained the same. It also added two novel questions addressing certain unique aspects of HFA and AS. It was measured on a scale ranging from “strongly disagree” (1) to “strongly agree” (5). It was coded (summed) in a way in which higher scores reflect more intense feelings of internalized stigma with regard to being autistic.

Well-being

The Mental Health Continuum was used in its three subscales: social (five items), emotional (three items), and psychological (six items) well-being, with respective α ratings of .84, .91, and .87 (Keyes et al. 2008). The subscales were used individually to capture quintessential aspects of various forms of well-being (social, emotional, and psychological) and examine how different forms of well-being interrelated to minority stress variables. They were coded (summed) in a way that lower scores reflected poorer well-being.

Psychological Distress

The psychological distress scale (K6) was originally developed by the U.S. National Center for Health Statistics (Kessler 2003; Cronbach’s α = 0.84). The K6 was designed to be sensitive around clinical thresholds for mental health disorders, with the short form (6 items) being “as sensitive” as the 10-item survey (Wittchen 2010). Items on it included “How often during the past month did you feel . . . nervous? . . . fidgety? . . . worthless?” The five-point response scale ranged from “all the time” to “none of the time.” It was coded (summed) in such a way that higher scores reflected higher psychological distress.

Results

Descriptive and Bivariate Analyses

SPSS 24 was used to conduct all analysis. Chi-square tests were carried out to identify whether any differences of note (gender, ethnicity, autism type, age at diagnosis, age of identifying as autistic, diagnosis status, autistic symptoms, and mental health outcome variables) existed between those included and those excluded, but none were detected (p ≥ .082). Where there was a single missing value per case, the mode was computed and input; there were only 25 values missing across all cases, and exclusion from the sample on the basis of one missing value would be extreme (all means and standard deviations can be found in Table 2).

Table 2.

Bivariate Correlations, Means, and Standard Deviations of Independent and Dependent Variables (n = 111).

	1	2	3	4	5	6	7	8	9	10	11	12	13	14
1 Ethnicity	1
2 Gender	−.03	1
3 Diagnosis status	.06	.06	1
4 Stressful life events	.05	.15	.06	1
5 Victimization and discrimination	.05	−.02	.06	.42***	1
6 Everyday discrimination	−.09	−.03	.02	.27**	.44**	1
7 Expectation of rejection	−.21*	−.11	−.15	.13	.27**	.59***	1
8 Outness scale	−.02	.10	.46*	.02	.03	.06	−.06	1
9 Behavior concealment of autism	.06	−.02	.21*	.29**	.19*	.185	.08	.34***	1
10 Internalized stigma	−.05	−.19*	.11	.09	.19*	.28**	.18	.18	.42***	1
11 Social well-being	.14	−.13	.04	−.02	−.06	−.32**	−.49***	−.13	−.12	−.06	1
12 Emotional well-being	.13	.01	.18	−.06	−.01	−.37***	−.40***	−.01	−.09	−.24*	.65***	1
13 Psychological well-being	.14	−.07	.03	.01	−.01	−.35***	−.38***	−.22*	−.08	−.18	.72***	.71***	1
14 Psychological distress	−.06	.06	−.23*	.16	.11	.42***	.42***	.11	.10	.30**	−.56***	−.71***	−.68***	1
M				13.9	11.7	21.3	19.6	16.3	22.6	22.9	6.66	7.29	12.8	13.1
SD				2.07	3.72	6.82	6.41	4.40	8.12	6.98	5.94	4.44	7.89	5.20

p < .05. **p < .01. ***p < .001.

Data were examined to identify whether distributions met parametric standards. All variables were normally distributed apart from victimization and discrimination, which were skewed. To correct for skewness, this variable was transformed into a binary variable (Walters 2009). The variable was divided on the basis of exposure to the specific items of the inventory (no exposure = 0, any exposure = 1).

Bivariate analyses are presented in Table 2. Correlations of variables raised some concern regarding multicollinearity, with independent variables having medium to high correlations. Regression models with diagnostic information for multicollinearity were performed to further examine multicollinearity. Variance inflation factors were all less than 2, originally suggesting little to no multicollinearity. However, upon further inspection, multicollinearity was identified within eigenvalues (≤.01), condition indexes (≥15), and variance proportion scores (≥.85). Theoretically, it is likely that minority stressors may not be independent but rather have a relation to one another, causing multicollinearity. To address this issue, ridge regression (RR) was used to test the study hypotheses.

RR is an extension of linear regression. When there is a problem with multicollinearity, RR can be preferable to ordinary least squares (OLS) regression (Helwig 2017; Jacobucci, Grimm, and McArdle 2016; McNeish 2015). This is because OLS regression performs poorly with highly correlated variables or when there are many predictor variables, as it causes large prediction intervals, making the model uninterpretable (Helwig 2017). RR has been shown to be more effective at providing accurate results than other forms of nonpenalized regression (OLS, stepwise, etc.) when multicollinearity is present (Abram et al. 2016; Eledum 2016; Firinguetti, Kibria, and Araya 2017; Zhang and Ibrahim 2005). Introducing a small increase in bias can result in a large decrease in prediction error. It is a process of trade-off between bias and variance (Marquaridt 1970). The small “penalty” (λ) on the OLS estimators will reduce the variability of the estimators, making them more stable, easier to interpret, and more likely to transfer to new samples (Helwig 2017). Penalized regression has been highlighted as a good tool available to psychologists to increase the replicability of their research (Helwig 2017). Thus RR was chosen as the appropriate statistical method to approach the data with.

The penalty coefficient ranges from 0 (no penalty) to 1, with an increment of .01 (Helwig 2017). The higher the penalty terms, the less variance but also the smaller the β coefficients. As such, it is a process of balancing. SPSS uses an iterative approach, which runs multiple versions of the model using different penalties to find that which best balances bias, variance, and error. Using either cross-validation or the bootstrap .632 method of resampling is standard practice when using RR (hence it is built into SPSS as part of the RR algorithm) (a comprehensive explanation of the .632 method can be found in Efron and Tibshirani 1997). We use the .632 estimator bootstrap method because research has shown it to be more reliable (Efron and Tibshirani 1997; Linting et al. 2007). A randomized selection of 50 different cases were included in each iteration (with 1,000 iterations run), and a mean standard error (MSE) computed from that. Similarly, the estimate of standard error on the standardized coefficients was calculated using bootstrapping with 1,000 samples. This acts as a form of confidence intervals on the standardized β coefficients. No unstandardized β values are calculated, because standardization of all variables is undertaken before RR is computed.

The demographics used in the analysis (ethnicity, gender, and diagnosis status) were included as binary variables. They were coded into majority/minority cases (as seen in Meyer et al. 2008). Gender was coded as male = 1 and female and other = 2. Ethnicity was coded as white British and other white = 1 and mixed/multiple, black British, Asian, and other = 2. Having a diagnosis was coded as 1, while no official diagnosis was coded as 2. Codes used were 1 and 2 because the RR function in SPSS reads zero as missing data. Similarly, because of this same issue, cases for which a variable computed to a true zero had to be recoded, as advised by the software manual (IBM n.d.), to a very small nonzero value (1 × 10⁻⁶). This allowed it to be included in the analysis without adverse consequences on the results.

Results of RR Analyses

The results of RR models predicting each of the mental health and well-being outcomes are shown in Table 3.

Table 3.

Minority Stress Predictors of Mental Health and Well-being Outcomes.

	Social Well-being				Emotional Well-being				Psychological Well-being				Psychological Distress (K6)
Variable	β	ESE	df	F	β	ESE	df	F	β	ESE	df	F	β	ESE	df	F
Ethnicity	0.01	0.06	1	0.04	0.10	0.06	1	2.60	0.08*	0.04	1	4.18	0.03	0.05	1	0.51
Gender	0.20*	0.09	1	4.76	0.03	0.08	1	0.17	0.06	0.07	1	0.87	0.02	0.05	1	0.28
Diagnosis status	0.07	0.06	1	1.45	0.19	0.10	1	4.01	0.04	0.05	1	0.67	0.14*	0.07	1	4.14
Stressful life events	0.18	0.13	4	2.06	−0.12	0.17	2	0.46	0.06	0.13	3	0.18	0.08	0.09	2	0.80
Victimization and discrimination	0.19	0.18	3	1.09	−0.20*	0.12	3	3.05	−.14*	0.07	2	3.93	−0.10	0.09	2	1.44
Everyday discrimination	−0.22	0.21	5	1.12	−.24*	0.13	3	3.34	−.16*	0.10	4	2.56	0.23**	0.12	5	3.58
Expectation of rejection	−.44***	0.10	6	19.81	−.35***	0.15	7	5.21	−.50***	0.13	7	13.65	0.48***	0.17	5	8.30
Outness scale	−.20	0.14	5	1.98	−0.16	0.18	4	0.81	−.15*	0.09	6	2.63	0.15**	0.08	7	3.86
Physical concealment	−.22*	0.15	8	2.24	0.19	0.17	6	1.15	0.02	0.17	2	0.02	−0.06	0.10	6	0.38
Internalized stigma	0.15	0.18	4	0.71	−.26*	0.16	3	2.77	−0.15	0.16	4	0.88	0.26**	0.13	6	4.09

Note: Estimated standard error (ESE) was calculated using bootstrap method (based on 1,000 bootstrap resamples). Lower on well-being reflect poorer well-being, and higher distress scores reflect higher distress (n = 111).

p ≤ .05. **p ≤ .01. ***p ≤ .001.

The regularization penalty applied to the social well-being model was λ = .12. Social well-being was significantly predicted by the behavioral concealment of autism and expectation of rejection. The model accounted for 58 percent of the variance in social well-being, F(38, 72) = 2.55, p < .001, R² = .58, MSE = .87. The significant standardized coefficients showed that lower levels of social well-being were associated with higher levels of expectation of rejection and behavioral concealment of autism. Gender was also associated with social well-being, but the sample size difference between men and women (and other) meant that this could not be explored further.

The regularization penalty applied to emotional well-being was λ = .12. Emotional well-being was significantly predicted by internalized stigma and diagnosis status. The model accounted for 48 percent of the variance in emotional well-being, F(32, 78) = 2.40, p = .001, R² = .48, MSE = .89. The significant standardized coefficients indicated that lower levels of emotional well-being were associated with higher levels of victimization and discrimination, everyday discrimination, expectation of rejection, and internalized stigma.

The regularization penalty applied to the psychological well-being model was λ = .26. Psychological well-being was significantly predicted by expectation of rejection, outness, stressful life events, and everyday discrimination. The model accounted for 58 percent of the variance in psychological well-being, F(32, 78) = 3.57, p = .000, R² = .57, MSE = .85. The significant standardized coefficients indicated that lower levels of psychological well-being were associated with higher levels of victimization and discrimination, everyday discrimination, expectations of rejection, and outness. Psychological well-being was also associated with ethnicity. The difference based on ethnicity could not be explored because the size of the sample differed too extremely.

The regularization penalty applied to the psychological distress model was λ = .22. Psychological distress was significantly predicted by everyday discrimination, expectation of rejection, outness, internalized stigma, and diagnosis status, and the model accounted for 72 percent of the variance of psychological distress F(36, 74) = 6.15, p< .001, R² = .72, MSE = .73. The significant standardized coefficients indicated that higher levels of psychological distress were associated with higher levels of everyday discrimination, expectation of rejection, outness, and internalized stigma, and having a diagnosis status.

Discussion

The aim of this study was to investigate the utility of the minority stress model in understanding how stigma-related stressors contribute to mental health and well-being problems in the autistic population. Originally designed to investigate sexual minorities and ethnic minorities, the minority stress model (Meyer 2003) has provided a novel way to consider the experience of having HFA or AS. The findings suggest that autistic individuals experience an added stress burden in the form of minority stress. This stress burden is a potentially preventable factor in the mental health and well-being disparity seen in the autistic population. Minority stressors such as victimization and discrimination, everyday discrimination, expectation of rejection, outness, internalized stigma, and physical concealment of autism consistently predicted diminished well-being and heightened psychological distress.

Thus, these findings provide the first, albeit preliminary, support that the minority stress model can be usefully extended to understand mental health and well-being problems faced by the HFA and AS population in that greater exposure to minority stressors is associated with poor mental health and well-being. Even further, it is important to note that these associations between minority stressors and mental health indicators persisted above and beyond the effect of general life stress and other demographic factors known to be associated with health and well-being (e.g., gender, race/ethnicity). Thus, these findings support previous research indicating that minority stress has a unique and additive negative effect on health, which is not reducible to general stress alone (Frost et al. 2015).

Everyday discrimination was highly positively correlated with the expectation of rejection, which is something that should be investigated further. With each small act of discrimination, it would, theoretically, make sense that expectations of rejection would increase. This association between smaller events of discrimination and the expectation for rejection is a sentiment that previous papers have expressed (Link and Phelan 1999; Stucky et al. 2011).

Outness, in the case of HFA and AS, was associated with poorer mental health in the form of lower psychological well-being and higher psychological distress. These findings run contrary to some findings from research on outness and well-being among sexual minorities, which indicate that outness is beneficial for well-being (Daley 2010; Legate, Ryan, and Rogge 2017). A potential explanation for this discrepancy is that when individuals with HFA or AS disclose their status on the spectrum, it opens them up to more acts of discrimination. Within the minority stress literature, the safeness of the environment is taken into account; in one situation outness may be therapeutic, and in another, outness could be considered a risk (di Bartolo 2013). Such situational differences are highlighted by the “don’t ask, don’t tell” policy that was for a time implemented in the U.S. armed forces (Davis 2010), where outness of LGBT status threatened one’s career and safety within the army. That policy affected the mental health of soldiers whether or not they disclosed, but more so when they did (Barber 2012). The potential well-being detriment of disclosure in autism could represent the effect of social punishment for being outside of the expected norm of neurotypicality. As previously discussed, the rate of unfair dismissal and bullying in the workplace is high for individuals with HFA and AS (Baldwin et al. 2014). As tolerance and acceptance for neurodiversity and the autistic population increase, the direction of the relationship between outness and mental health may change. Although others have postulated that openness may reduce stigma (Corrigan, Kosyluk, and Rüsch 2013), the present results indicate that it can also be a factor in reduced well-being for the person “coming out.” Eventually it may also represent a therapeutic process that will correlate with lower internalized stigma and better mental health, as it does currently with LGBT communities in most Western situations (di Bartolo 2013).

Similarly, labeling theory (Link and Phelan 1987) may also explain why outness decreases mental health and well-being; after diagnosis, certain labels are attached to the individual and stereotypes often attached to autism are rarely positive, with 9 of 10 being derogatory (Wood and Freeth 2016). However, there was a significant negative relationship between psychological distress and diagnosis, with higher distress experienced among those with a diagnosis. Increased expectation of rejection in the diagnosed group may reflect the stigma that comes from having a proper diagnosis or the stress of then having to hide this aspect of the self.

Clinical Implications

In light of the results, the findings of this study, if upheld in further research, could mark a change in the way we consider mental health and well-being in the autistic community. Previously, poor quality of life (Barnbaum 2008; Kamio et al. 2013) has been intrinsically linked with autism, without the consideration that negative social factors (i.e., minority stress) may play a part in the well-being disparity experienced by autistic individuals. In Kamio et al. (2013), researchers investigating suicidal ideation in those on the spectrum found that three quarters of their sample had suffered from bullying yet still attributed suicidal ideation to the characteristics of autism. This research has expanded the focus to include the wider implications of discrimination on members of the autistic community. These findings may shed light on the experiences of autistic individuals in society and highlight the consequences of the discrimination and victimization highlighted in other research, for example, increased employment discrimination, sexual victimization, bullying, isolation, and homelessness (Baldwin and Costley 2016; Brown-Lavoie, Viecili, and Weiss 2014; Carter 2009; Heinrichs and Myles 2003; Homeless Link 2014). This reframing of perspectives on autism can reflect a movement in which mental health problems such as depression, anxiety, and suicidal ideation are no longer considered inherent to autism (as in Kamio et al. 2013).

Limitations and Directions for Future Research

A possible limitation of the study is the translation of the measures from one minority community to another. Although some of the measures were created for use with all minorities (expectation of rejection and everyday discrimination), others were originally designed for use with sexual minority communities and needed to be translated specifically for this study (outness and internalized stigma). The unique aspects of the autistic community may not have been reflected in these measures, and rather than changing the language of existing measures, attempts need to be made to design new assessments of the unique qualities of the autistic experience “from the ground up.”

Women were disproportionately represented in the study, which may decrease the generalizability of these findings. However, this is something that frequently happens in reverse, with men being overrepresented in autism research. The present sample could reflect the frustration that autistic women feel at the usual exclusion from research (Rynkiewicz et al. 2016; Shefcyk 2015). It is important to note that gender was controlled for in all analyses in order to partially account for this limitation. Future research needs to investigate the potential additive effects of multiple minority identities, such as being both autistic and an ethnic minority, to see whether there are effects related to “double discrimination” (as suggested by Grollman 2014). Similarly, other research (Brown 2014) has found a gender difference in the effects of perceived social devaluation on mental health (albeit of physical disability), which might explain gender affecting social well-being in these results. A more in-depth analysis of that effect in a sample balanced for gender is needed.

It may be important to understand the relationship that everyday discrimination has on expectations of rejection and the place of labeling theory within the experiences of discrimination and expectation of rejection. This would require longitudinal research to understand the causal and cyclical relationships between these aspects of the minority stress experience. Such research has the potential to provide a better understanding of minority stress as the dynamic and situational model that it is theorized to be. Similarly, more research could be done on the meaning of diagnosis to unpick the relationship it has with the autistic person being labeled and the societal context and perception of that label.

The prospects for future research stemming from this article are numerous. We found that increased exposure to minority stress was associated with poorer well-being within an autistic sample. By carrying out within-group analysis, we can understand the impact of the actual social stress (Schwartz and Meyer 2010). It provides an opportunity to understand the effect of exposure to minority stress on well-being in the autistic population. Every individual within the group may experience the process of social stress to different degrees. A between-group study is also needed however, to fully test the full minority stress model.

Finally, although the present study contributes to the integration of disability and stress literature (by demonstrating a clear relationship between minority stress and mental health in the autistic population), further work is needed to examine resilience factors that potentially buffer the negative effects of minority stress. Indeed, the minority stress model points to the potential buffering effect community connectedness may have, and the stress, stigma, and disability literatures have been poorly integrated thus far (Brown 2010). Given the increasing community identity emerging in members of the HFA and AS population, theorizing and empirically testing the stress-ameliorating function of community connectedness for autistic people will likely prove useful.

Summary and Conclusions

Although preliminary, this study is the first to examine the applicability of the minority stress model to the autistic community, demonstrating the unique and additive impact of minority stress on mental health and well-being for members of the HFA and AS population. More research is needed to replicate these findings and address questions of causality in the association between minority stress and mental health for autistic individuals, along with stress-ameliorating factors in the autistic population.

Footnotes

ORCID iD

David M. Frost

References

Abram

Samantha V.

Helwig

Nathaniel E.

Moodie

Craig A.

DeYoung

Colin G.

MacDonald

Angus W.

III Waller

Niels G.

2016. “Bootstrap Enhanced Penalized Regression for Variable Selection with Neuroimaging Data.” Frontiers in Neuroscience 10:344.

Altman

Barbara M.

2001. “Definitions, Models, Classifications, Schemes, and Applications.” Pp.97–122 in Handbook of Disability Studies. Thousand Oaks, CA: Sage.

Aylott

2000. “Autism in Adulthood: The Concepts of Identity and Difference.” British Journal of Nursing 9(13):851–58.

Bagatell

Nancy

. 2007. “Orchestrating Voices: Autism, Identity and the Power of Discourse.” Disability & Society 22(4):413–26.

Bagatell

Nancy

. 2010. “From Cure to Community: Transforming Notions of Autism.” Ethos 38(1):33–55.

Bainbridge

David

. 2008. Beyond the Zonules of Zinn: A Fantastic Journey through Your Brain. Cambridge, MA: Harvard University Press.

Baldwin

Costley

2016. “The Experiences and Needs of Female Adults with High-functioning Autism Spectrum Disorder.” Autism 20(4):483–95.

Baldwin

Costley

Warren

2014. “Employment Activities and Experiences of Adults with High-functioning Autism and Asperger’s Disorder.” Journal of Autism and Developmental Disorders 44(10):2440–49.

Barber

Mary E.

2012. “Mental Health Effects of Don’t Ask Don’t Tell.” Journal of Gay & Lesbian Mental Health 16(4):346–52.

10.

Barnard

Judith

Harvey

Virginia

Potter

David

and National Autistic Society. 2001. Ignored or Ineligible? The Reality for Adults with Autism Spectrum Disorders. London: National Autistic Society.

11.

Barnbaum

Deborah R.

2008. The Ethics of Autism: Among Them, but Not of Them. Bloomington: Indiana University Press.

12.

Bascom

Julia

, ed. 2012. Loud Hands: Autistic People, Speaking. Washington, DC: The Autistic Press.

13.

Benford

Penny

Standen

2009. “The Internet: A Comfortable Communication Medium for People with Asperger Syndrome (AS) and High Functioning Autism (HFA)?” Journal of Assistive Technologies 3(2):44–53.

14.

Brown

Lydia

. 2017. “Identity-first Language: The Significance of Semantics: Person-first Language: Why It Matters.” Nothing about Us without Us. Retrieved September 21, 2018 (http://autisticadvocacy.org/about-asan/identity-first-language/).

15.

Brown

Robyn

. 2010. “Physical Disability and Quality of Life: The Stress Process and Experience of Stigma in a Chronically-strained Population.” PhD dissertation, Florida State University.

16.

Brown

Robyn Lewis

. 2014. “Psychological Distress and the Intersection of Gender and Physical Disability: Considering Gender and Disability-related Risk Factors.” Sex Roles 71(3–4):171–81.

17.

Brown

Robyn Lewis

. 2017. “Understanding the Influence of Stigma and Discrimination for the Functional Limitation—Psychological Distress Relationship: A Stress and Coping Perspective.” Social Science Research 62:150–60.

18.

Brown-Lavoie

S. M.

Viecili

M. A.

Weiss

J. A.

2014. “Sexual Knowledge and Victimization in Adults with Autism Spectrum Disorders.” Journal of Autism and Developmental Disorders 44(9):2185–96.

19.

Butler

Robert C.

Gillis

Jennifer M.

2011. “The Impact of Labels and Behaviors on the Stigmatization of Adults with Asperger’s Disorder.” Journal of Autism and Developmental Disorders 41(6):741–49.

20.

Carter

Susan

. 2009. “Bullying of Students with Asperger Syndrome.” Issues in Comprehensive Pediatric Nursing 32(3):145–54.

21.

Cochran

Susan D.

Mays

Vickie M.

2000. “Relation between Psychiatric Syndromes and Behaviorally Defined Sexual Orientation in a Sample of the US Population.” American Journal of Epidemiology 151(5):516.

22.

Corrigan

Patrick W.

Kosyluk

Kristin A.

Rüsch

Nicolas

. 2013. “Reducing Self-stigma by Coming Out Proud.” American Journal of Public Health 103(5):794–800.

23.

Cowen

Tyler

. 2009. “Autism as Academic Paradigm.” Chronicles of Higher Education. Retrieved September 21, 2018 (https://www.chronicle.com/article/Autism-as-Academic-Paradigm/47033).

24.

Daley

Andrea

. 2010. “Being Recognized, Accepted, and Affirmed: Self-disclosure of Lesbian/Queer Sexuality within Psychiatric and Mental Health Service Settings.” Social Work in Mental Health 8(4):336–55.

25.

Davis

Brandon A.

2010. Don’t Ask, Don’t Tell: Background and Issues on Gays in the Military. Hauppauge, NY: Nova Science.

26.

di Bartolo

Adriana N.

2013. “Is There a Difference? The Impact of Campus Climate on Sexual Minority and Gender Minority Students’ Levels of Outness.” PhD dissertation, Claremont Graduate University.

27.

Dunn

Dana S.

Andrews

Erin E.

2015. “Person-first and Identity-first Language: Developing Psychologists’ Cultural Competence Using Disability Language.” American Psychologist 70(3):255.

28.

Efron

Bradley

Tibshirani

Robert

. 1997. “Improvements on Cross-validation: The .632+ Bootstrap Method.” Journal of the American Statistical Association 92(438):548–60.

29.

Eledum

Hussein

. 2016. “A Comparison Study of Ridge Regression and Principle Component Regression with Application.” International Journal of Research 3(8):283.

30.

Elliman

Louise

. 2011. “Asperger’s Syndrome—Difference or Disorder?” Psychologist 24(2):114–17.

31.

Feagin

Joe R.

1991. “The Continuing Significance of Race: Antiblack Discrimination in Public Places.” American Sociological Review 56(1):101–16.

32.

Firinguetti

Luis

Kibria

Golam

Araya

Rodrigo

. 2017. “Study of Partial Least Squares and Ridge Regression Methods.” Communications in Statistics—Simulation and Computation 46(8):6631–44.

33.

Frost

David M.

2011. “Social Stigma and Its Consequences for the Socially Stigmatized.” Social and Personality Psychology Compass 5(11):824–39.

34.

Frost

David M.

Lehavot

Keren

Meyer

Ilan H.

2015. “Minority Stress and Physical Health among Sexual Minority Individuals.” Journal of Behavioral Medicine 38(1):1–8.

35.

Gillberg

I. Carina

Helles

Adam

Billstedt

Eva

Gillberg

Christopher

. 2016. “Boys with Asperger Syndrome Grow Up: Psychiatric and Neurodevelopmental Disorders 20 Years after Initial Diagnosis.” Journal of Autism and Developmental Disorders 46(1):74–82.

36.

Gilman

Stephen E.

Cochran

S. D.

Mays

V. M.

Hughes

Ostrow

Kessler

R. C.

2001. “Risk of Psychiatric Disorders among Individuals Reporting Same-sex Sexual Partners in the National Comorbidity Survey.” American Journal of Public Health 91(6):933–39.

37.

Grollman

Eric Anthony

. 2014. “Multiple Disadvantaged Statuses and Health: The Role of Multiple Forms of Discrimination.” Journal of Health and Social Behavior 55(1):3–19.

38.

Hacking

Ian

. 2012. “Two Histories of Autism, One by an Outsider, One by an Insider.” BioSocieties 7(3):323–26.

39.

Heinrichs

Rebekah

Myles

Brenda Smith

. 2003. Perfect Targets: Asperger Syndrome and Bullying: Practical Solutions for Surviving the Social World. Shawnee Mission, KS: Autism Asperger Publishing.

40.

Helwig

Nathaniel E.

2017. “Adding Bias to Reduce Variance in Psychological Results: A Tutorial on Penalized Regression.” Quantitative Methods for Psychology 13(1):1–19.

41.

Herek

Gregory M.

Gillis

J. Roy

Cogan

Jeanine C.

1999. “Psychological Sequelae of Hate-crime Victimization among Lesbian, Gay, and Bisexual Adults.” Journal of Consulting and Clinical Psychology 67(6):945–51.

42.

Hirvikoski

Tatja

Mittendorfer-Rutz

Boman

Larsson

Lichtenstein

Bölte

2016. “Premature Mortality in Autism Spectrum Disorder.” British Journal of Psychiatry 208(3):232–38.

43.

Homeless Link. 2014. “Autism and Homelessness—Briefing from the Frontline.” Retrieved September 21, 2018 (https://www.homeless.org.uk/our-work/resources/autism-homelessness)>

44.

IBM. N.d. “IBM SPSS 22 Categories.” Armonk, NY: IBM.

45.

Jaarsma

Pier

Welin

Stellan

. 2012. “Autism as a Natural Human Variation: Reflections on the Claims of the Neurodiversity Movement.” Health Care Analysis 20(1):20–30.

46.

Jacobucci

Ross

Grimm

Kevin J.

McArdle

John J.

2016. “Regularized Structural Equation Modeling.” Structural Equation Modeling: A Multidisciplinary Journal 23(4):555–66.

47.

Jones

2017. “Aspergers and Disclosure and Outness and Passing.” Aspergia. Retrieved (http://www.lettersfromaspergia.com/2014/04/aspergers-and-disclosure-and-out-ness.html).

48.

Jordan

Chloe J.

2010. “Evolution of Autism Support and Understanding via the World Wide Web.” Intellectual and Developmental Disabilities 48(3):220–27.

49.

Kamio

Inada

Koyama

2013. “A Nationwide Survey on Quality of Life and Associated Factors of Adults with High-functioning Autism Spectrum Disorders.” Autism 17(1):15–26.

50.

Kanner

Leo

. 1971. “Follow-up Study of Eleven Autistic Children Originally Reported in 1943.” Journal of Autism and Childhood Schizophrenia 1(2):119–45.

51.

Kapp

Steven K.

Gillespie-Lynch

Kristen

Sherman

Lauren E.

Hutman

Ted

. 2013. “Deficit, Difference, or Both? Autism and Neurodiversity.” Developmental Psychology 49(1):59–71.

52.

Kerns

Connor

Newschaffer

Craig

Berkowitz

Steven

. 2015. “Traumatic Childhood Events and Autism Spectrum Disorder.” Journal of Autism & Developmental Disorders 45(11):3475–86.

53.

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.

Zaslavsky

Alan M.

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

54.

Keyes

Corey L. M.

Wissing

Potgieter

J. P.

Temane

Kruger

van Rooy

2008. “Evaluation of the Mental Health Continuum–Short Form (MHC-SF) in Setswana-speaking South Africans.” Clinical Psychology & Psychotherapy 15(3):181–92.

55.

Legate

Nicole

Ryan

Richard M.

Rogge

Ronald D.

2017. “Daily Autonomy Support and Sexual Identity Disclosure Predicts Daily Mental and Physical Health Outcomes.” Personality and Social Psychology Bulletin 43(6):860–73.

56.

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.

57.

Link

Bruce G.

Cullen

Francis T.

Struening

Elmer

Shrout

Patrick E.

Dohrenwend

Bruce P.

1989. “A Modified Labeling Theory Approach to Mental Disorders: An Empirical Assessment.” American Sociological Review 54(3):400.

58.

Link

Bruce G.

Phelan

Jo C.

1999. “Labeling and Stigma.” Pp.481–94 in Handbook of the Sociology of Mental Health, edited by Aneshensel

C. S.

Phelan

J. C.

Boston: Springer.

59.

Linting

Mariëlle

Meulman

Jacqueline J.

Groenen

Patrick J. F.

van der Kooij

Anita J.

2007. “Stability of Nonlinear Principal Components Analysis: An Empirical Study Using the Balanced Bootstrap.” Psychological Methods 12(3):359–79.

60.

Little

Liza

. 2002. “Middle-class Mothers’ Perceptions of Peer and Sibling Victimisation among Children with Asperger Syndrome and Non-verbal Learning Disorders.” Issues in Comprehensive Pediatric Nursing 25(1):43–57.

61.

Locke

Jill

Ishijima

Eric H.

Kasari

Connie

London

Nancy

. 2010. “Loneliness, Friendship Quality and the Social Networks of Adolescents with High-functioning Autism in an Inclusive School Setting.” Journal of Research in Special Educational Needs 10(2):74–81.

62.

Markowitz

F. E.

1998. “The Effects of Stigma on the Psychological Well-being and Life Satisfaction of Persons with Mental Illness.” Journal of Health and Social Behavior 39(4):335–47.

63.

Marquaridt

Donald W.

1970. “Generalized Inverses, Ridge Regression, Biased Linear Estimation, and Nonlinear Estimation.” Technometrics 12(3):591–612.

64.

McManus

Sally

. 2009. “Adult Psychiatric Morbidity in England, 2007: Results of a Household Survey.” London: National Centre for Social Research.

65.

McNeish

Daniel M.

2015. “Using Lasso for Predictor Selection and to Assuage Overfitting: A Method Long Overlooked in Behavioral Sciences.” Multivariate Behavioral Research 50(5):471–84.

66.

Meyer

Ilan H.

2003. “Prejudice, Social Stress, and Mental Health in Lesbian, Gay, and Bisexual Populations: Conceptual Issues and Research Evidence.” Psychological Bulletin 129(5):674–97.

67.

Meyer

Ilan H.

Dean

Laura

. 1998. “Internalized Homophobia, Intimacy, and Sexual Behavior among Gay and Bisexual Men.” Pp.160–86 in Stigma and Sexual Orientation: Understanding Prejudice against Lesbians, Gay Men, and Bisexuals. Thousand Oaks, CA: Sage.

68.

Meyer

Ilan H.

Rossano

Lindsay

Ellis

James M.

Bradford

Judith

. 2002. “A Brief Telephone Interview to Identify Lesbian and Bisexual Women in Random Digit Dialing Sampling.” Journal of Sex Research 39(2):139–44.

69.

Meyer

Ilan H.

Schwartz

Sharon

Frost

David M.

2008. “Social Patterning of Stress and Coping: Does Disadvantaged Social Status Confer More Stress and Fewer Coping Resources?” Social Science & Medicine 67(3):368–79.

70.

Mikami

Katsunaka

Inomata

Hayakawa

Ohnishi

Enseki

Ohya

Haruki

Kishi

Shinohara

Ichimura

Matsumoto

2009. “Frequency and Clinical Features of Pervasive Developmental Disorder in Adolescent Suicide Attempts.” General Hospital Psychiatry 31(2):163–66.

71.

National Autistic Society. 2012. “NAS Press Release; Survey Results.” London: National Autistic Society.

72.

Pellicano

Elizabeth

Dinsmore

Adam

Charman

Tony

. 2014. “What Should Autism Research Focus Upon? Community Views and Priorities from the United Kingdom.” Autism 18(7):756–70.

73.

Rioux

Marcia H.

Bach

Michael

and Roeher Institute, eds. 1994. Disability Is Not Measles: New Research Paradigms in Disability. North York, Canada: Roeher Institute.

74.

Rosenblatt

Mia

and National Autistic Society. 2008. I Exist: The Message from Adults with Autism in England. London: National Autistic Society.

75.

Rynkiewicz

Agnieszka

Schuller

Marchi

Piana

Camurri

Lassalle

Baron-Cohen

2016. “An Investigation of the ‘Female Camouflage Effect’ in Autism Using a Computerized ADOS-2 and a Test of Sex/Gender Differences.” Molecular Autism 7(1):1–8.

76.

Schwartz

Sharon

Meyer

Ilan H.

2010. “Reflections on the Stress Model: A Response to Turner.” Social Science & Medicine 70(8):1121–22.

77.

Shefcyk

Allison

. 2015. “Count Us In: Addressing Gender Disparities in Autism Research.” Autism 19(2):131–32.

78.

Slopen

Natalie

Williams

David R.

Fitzmaurice

Garrett M.

Gilman

Stephen E.

2011. “Sex, Stressful Life Events, and Adult Onset Depression and Alcohol Dependence: Are Men and Women Equally Vulnerable?” Social Science & Medicine 73(4):615–22.

79.

Smart

Julie

. 2006. “Challenges to the Biomedical Model of Disability.” Advances in Medical Psychotherapy & Psychodiagnosis 12:1–4.

80.

Stucky

Brian D.

Gottfredson

N. C.

Panter

A. T.

Daye

C. E.

Allen

W. R.

Wightman

L. F.

2011. “An Item Factor Analysis and Item Response Theory–Based Revision of the Everyday Discrimination Scale.” Cultural Diversity and Ethnic Minority Psychology 17(2):175–85.

81.

Walters

Stephen J.

2009. Quality of Life Outcomes in Clinical Trials and Health-care Evaluation: A Practical Guide to Analysis and Interpretation. Hoboken, NJ: John Wiley.

82.

Ward

Michael J.

Meyer

Roger N.

1999. “Self-determination for People with Developmental Disabilities and Autism: Two Self-advocates’ Perspectives.” Focus on Autism and Other Developmental Disabilities 14(3):133–39.

83.

Williams

David R.

Williams-Morris

Ruth

. 2000. “Racism and Mental Health: The African American Experience.” Ethnicity & Health 5(3–4):243–68.

84.

Williams

David R.

Yan

Jackson

James S.

Anderson

Norman B.

1997. “Racial Differences in Physical and Mental Health Socio-economic Status, Stress and Discrimination.” Journal of Health Psychology 2(3):335–51.

85.

Wittchen

Hans-Ulrich

. 2010. “Screening for Serious Mental Illness: Methodological Studies of the K6 Screening Scale.” International Journal of Methods in Psychiatric Research 19(S1):1–3.

86.

Wood

Chantelle

Freeth

Megan

. 2016. “Students’ Stereotypes of Autism.” Journal of Educational Issues 2(2):131.

87.

Wright

E. R.

Gronfein

W. P.

Owens

T. J.

2000. “Deinstitutionalization, Social Rejection, and the Self-esteem of Former Mental Patients.” Journal of Health and Social Behavior 41(1):68–90.

88.

Zhang

John

Ibrahim

Mahmud

. 2005. “A Simulation Study on SPSS Ridge Regression and Ordinary Least Squares Regression Procedures for Multicollinearity Data.” Journal of Applied Statistics 32(6):571–88.

89.

Zietsch

Brendan P.

Verweij

Karin J. H.

Bailey

J. Michael

Wright

Margaret J.

Martin

Nicholas G.

2011. “Sexual Orientation and Psychiatric Vulnerability: A Twin Study of Neuroticism and Psychoticism.” Archive of Sexual Behaviour 40(1):133–42.