Introduction
Racism, including experiencing racial microaggressions, and related stress, have become even more pronounced problems during the COVID-19 pandemic, and they are associated with later inflammation, chronic health conditions, suicidal ideation, externalizing and internalizing symptoms, and alcohol-related problems (Liu and Modir, Reference Liu and Modir2020; Spanierman et al., Reference Spanierman, Clark and Kim2021; Williams, Reference Williams2020; Williams et al., Reference Williams, Osman and Hyon2023). Research needs to further examine the association between racial microaggressions and health/mental health problems to facilitate development of culturally competent strengths-based prevention and intervention efforts (Sosoo et al., Reference Sosoo, Bernard and Neblett2020; Spanierman et al., Reference Spanierman, Clark and Kim2021; Williams, Reference Williams2020).
Racial microaggressions are interactions of BIPOC (Black, Indigenous, and people of colour) with typically White perpetrators in a racialized social system that imply their inferiority or dismiss aspects of BIPOC identity or experience (Spanierman et al., Reference Spanierman, Clark and Kim2021; Sue et al., Reference Sue, Capodilupo, Torino, Bucceri, Holder, Nadal and Esquilin2007). Farahani et al. (Reference Farahani, Nepton, Williams, Martin, Preedy and Patel2022) discuss how often intention to cause harm or perceived lack thereof can lead to difficulties aligning microaggressions as a form of defined aggression/violence, and they state that intention instead is not always part of a definition of violence/aggression, and that microaggressions can be understood as offensive, frustrating to victims, and acts that harm someone not willing to be harmed.
Of note, throughout this paper we used the term ‘minoritized’ to reflect someone who belongs to an under-represented racial or ethnic minority group in the United States (American Psychiatric Association, 2022; Flanagin et al., Reference Flanagin, Frey and Christiansen2021).
Experiencing racial microaggressions may relate to depletion of cognitive resources from related stress, other unhelpful cognitive appraisals about the meaning of these experiences, and subsequent reduced ability to manage intrusive thoughts (Williams et al., Reference Williams, Taylor, Mouzon, Oshin, Himle and Chatters2017b; Willis and Neblett, Reference Willis and Neblett2018). Vigilance or racial battle fatigue may be implicated in this process (Williams et al., Reference Williams, Taylor, Mouzon, Oshin, Himle and Chatters2017b). Vigilance, or monitoring and adjusting one’s behaviour to protect oneself from discrimination has mediated the link between discrimination and stress (Himmelstein et al., Reference Himmelstein, Young, Sanchez and Jackson2015). Racial battle fatigue encompasses social, psychological and physiological responses to experiencing racism and was associated with tension, fatigue, elevated heart beat and blood pressure, feelings of frustration, exhaustion, intrusive thoughts, avoidance and withdrawal, fighting back, or potential acceptance of racist ideas (Smith et al., Reference Smith, Allen and Danley2007). Martinez et al. (Reference Martinez, Suyemoto, Abdullah, Burnett-Zeigler and Roemer2022b) provide a literature review on other models of coping with race-related stress that illustrate how stress from racism may then lead to rumination, negative health behaviours, decreased cognitive flexibility, and unhelpful and rigid use of coping skills.
Obsessive-compulsive disorder
OCD involves obsessions and compulsions (American Psychiatric Association, 2013), and negatively influences quality of life in clinical and non-clinical populations across racial identities (Himle et al., Reference Himle, Muroff, Taylor, Baser, Abelson, Hanna, Abelson and Jackson2008; Macy et al., Reference Macy, Theo, Kaufmann, Ghazzaoui, Pawlowski, Fakhry, Cassmassi and IsHak2013). This paper works from Abramowitz et al.’s (Reference Abramowitz, Deacon, Olatunji, Wheaton, Berman, Losardo, Timpano, McGrath, Riemann, Adams, Björgvinsson, Storch and Hale2010) dimensional model with four symptom categories: responsibility for harm, symmetry, unacceptable thoughts, and contamination. OCD is heterogenous and understanding presentations according to symptom dimensions can help inform treatment (Lochner and Stein, Reference Lochner and Stein2003; Pinciotti et al., Reference Pinciotti, Riemann and Abramowitz2021). Additionally, we use an analogue sample of college students, relevant in research given the prevalence of obsessive-compulsive symptoms (OCS) across the population (Abramowitz et al., Reference Abramowitz, Fabricant, Taylor, Deacon, McKay and Storch2014; Willis and Neblett, Reference Willis and Neblett2018).
Lifetime prevalence of OCD is similar across the whole population (1.6%) (Kessler et al., Reference Kessler, Berglund, Demler, Jin, Merikangas and Walters2005), African American and Caribbean Black people (1.6%), and other racially minoritized groups (Katz et al., Reference Katz, Rufino, Werner, McIngvale and Storch2020). However, experience of OCS across racially minoritized groups and cultures, as well as use of treatments, differs, such as ethnic minority people accessing evidence-based treatments like exposure with response prevention and medication less than people who are not ethnically minoritized, or with potentially higher occurrences of panic disorder amongst Caribbean Black people (Himle et al., Reference Himle, Muroff, Taylor, Baser, Abelson, Hanna, Abelson and Jackson2008; Katz et al., Reference Katz, Rufino, Werner, McIngvale and Storch2020). Furthermore, African Americans and European Americans had similar levels of OCS but different cognitive strategies for addressing them (Nota et al., Reference Nota, Blakey, George-Denn, Jacoby, Schubert, Abramowitz and Coles2014). Similarly, among participants in intensive/residential treatment for OCD, having a higher composite marginalized identity score was associated with greater OCS severity, namely more severe contamination, responsibility for harm, and symmetry symptoms (Wadsworth et al., Reference Wadsworth, Potluri, Schreck and Hernandez-Vallant2020).
The experience of racial microaggressions and obsessive-compulsive symptoms
It is clear that racism has associated mental health difficulties, and it has specific associations with OCS as well that need to be further studied. There is a need for a more thorough understanding of links between anxiety-related diagnoses and diagnoses that have anxiety as a symptom like OCD, and their links with racism (MacIntyre et al., Reference MacIntyre, Zare and Williams2023). They summarize in their review that experiencing racial discrimination is associated with higher OCD symptom severity, with racial identity beliefs sometimes acting as vulnerability factors and sometimes acting as protective factors (MacIntyre et al., Reference MacIntyre, Zare and Williams2023).
Among over 3000 African American participants, everyday racial discrimination was related to presence of contamination and unacceptable thoughts-themed obsessions, multiple types of compulsions, and amount of obsessions and compulsions (Williams et al., Reference Williams, Taylor, Mouzon, Oshin, Himle and Chatters2017b). Symptoms of OCD are associated with distress from experiencing racial microaggressions (Willis and Neblett, Reference Willis and Neblett2018). A large study of Black Caribbean and African American adults found associations with racial discrimination and contamination, harm, washing, and repeating OCS (Williams et al., Reference Williams, Taylor, George, Schlaudt, Ifatunji and Chatters2021). As MacIntyre et al. (Reference MacIntyre, Zare and Williams2023) state, much of the existing literature is on Black Americans, so more work is needed for other ethnic and racial groups of people. Furthermore, research designed to understand causal mechanisms and risk/protective factors in this relationship between racism and OCD is an additional important next step (Willis and Neblett, Reference Willis and Neblett2018).
Psychological flexibility and inflexibility
Psychological flexibility (ability to recognize, adapt, and respond to internal or external experiences while aware of one’s values; Kashdan and Rottenberg, Reference Kashdan and Rottenberg2010) is an important construct. Psychological flexibility and inflexibility are at either end of a spectrum of responding to internal and external experiences in an open, aware, and valued way (Kashdan and Rottenberg, Reference Kashdan and Rottenberg2010). These ideas relate to the literature on the relevance of cognitive appraisals in the link between discrimination and stress (King, Reference King2005; Williams et al., Reference Williams, Taylor, Mouzon, Oshin, Himle and Chatters2017b) and the role of therapists in facilitating protective factors against racial trauma such as cultural identity, social support and sense of belonging (Liu and Modir, Reference Liu and Modir2020).
Similarly, a recent review of how Black people cope with racism found that people flexibly use a variety of strategies, such as active coping and problem-solving for institutional racism, spirituality-based strategies for interpersonal racism, or collective coping, social support and problem-solving for cultural racism (Jacob et al., Reference Jacob, Faber, Faber, Bartlett, Ouimet and Williams2022). Similar to Jacob et al.’s (Reference Jacob, Faber, Faber, Bartlett, Ouimet and Williams2022) emphasis on future research and clinical work that focuses on functional and flexible use of coping strategies, psychological flexibility/inflexibility broadly can also be conceptualized as flexible and functional use of multiple processes including defusion, committed action, experiential avoidance, acceptance, self as context, and awareness of the present moment according to context and someone’s values (Hayes et al., Reference Hayes, Luoma, Bond, Masuda and Lillis2006).
For instance, psychological flexibility measured broadly, or as specific processes such as experiential avoidance, is associated with mental health outcomes after experiencing racism and discrimination, such that more avoidance and inflexible coping leads to worsened mental health and avoidance, and more mindful awareness can buffer these negative impacts (Davis et al., Reference Davis, Xin, Sepeda, Garcia-Romeu and Williams2021; Martinez et al., Reference Martinez, Eustis, Arbid, Graham-LoPresti and Roemer2022a; Masuda et al., Reference Masuda, Barile, Spencer, Juberg, Martin and Vibell2022; Miller and Orsillo, Reference Miller and Orsillo2020; Zvolensky et al., Reference Zvolensky, Rogers, Mayorga, Shepherd, Bakhshaie, Garza and Peraza2022). These studies encompass participants from a variety of racial and ethnic backgrounds. Thus, psychological flexibility and acceptance and commitment therapy (ACT) align with Martinez et al.’s (Reference Martinez, Suyemoto, Abdullah, Burnett-Zeigler and Roemer2022b) review on how coping with racism can lead to rigid and inflexible coping, and how mindfulness and values-based interventions, components of psychological flexibility, can target those struggles. Evidence suggests that these treatments are acceptable to people from marginalized backgrounds (Fuchs et al., Reference Fuchs, West, Graham, Kalill, Morgan, Hayes-Skelton, Orsillo and Roemer2016).
In one of the first meta-analyses at the time, examining use of acceptance and mindfulness-based treatments for people from diverse backgrounds related to refugee status, race, forensic, and older adults, Fuchs et al. (Reference Fuchs, Lee, Roemer and Orsillo2013) found small to large effect sizes depending on design and comparison condition, with overall small sample sizes, providing initial support but a need for future research. Other research and reviews have echoed that there is theoretical and research support for acceptance and mindfulness-based treatments for Black people and people of colour (POC), but overall there is a need for more studies with these treatments and populations (Dawson et al., Reference Dawson, Jones, Fairbairn and Laurent2022; Dela Cruz et al., Reference Dela Cruz, Johnstone, Kim and Castle2022; Martinez et al., Reference Martinez, Suyemoto, Abdullah, Burnett-Zeigler and Roemer2022b; Trefethen, Reference Trefethen2019). MacIntyre et al. (Reference MacIntyre, Zare and Williams2023) reiterate the general need for new interventions to be validated to help people cope with anxiety disorders related to racism. Thus there is clear literature support for psychological flexibility-informed treatments, with additional research needed exploring their benefits to POC.
Psychological inflexibility and experiential avoidance are also associated with OCS and psychological distress (Angelakis and Pseftogianni, Reference Angelakis and Pseftogianni2021; Bond et al., Reference Bond, Hayes, Baer, Carpenter, Guenole, Orcutt, Waltz and Zettle2011). ACT, of which psychological flexibility/inflexibility is a process of change, is effective for treating OCD, as are mindfulness-based interventions (Bluett et al., Reference Bluett, Homan, Morrison, Levin and Twohig2014; Chien et al., Reference Chien, Tse, Chan, Cheng and Chen2022; Twohig et al., Reference Twohig, Hayes, Plumb, Pruitt, Collins, Hazlett-Stevens and Woidneck2010). Also, psychological flexibility is a process of change in ACT and traditional exposure and response prevention (ERP) for OCD (Schubert et al., Reference Schubert, Hilbert, Favreau, Wolstein and Voderholzer2022). Furthermore, it is important to understand contextual variables that lead to resilience from developing OCD symptoms in response to a stressor such as the COVID-19 pandemic or racism in order to inform clinical and public health strategies (Grant et al., Reference Grant, Drummond, Nicholson, Fagan, Baldwin, Fineberg and Chamberlain2022).
Many recent measures of psychological flexibility do not take into account all six theoretical skill components (acceptance, present moment awareness, defusion, awareness of values, committed action, and self as context) or do not do so in a way that is sensitive to context and time variabilities (i.e. asking participants to answer broadly forcing an answer to speak to trait like differences instead of state differences) (Gloster et al., Reference Gloster, Block, Klotsche, Villanueva, Rinner, Benoy and Bader2021). Additionally, it is important that measures are validated for both clinical and non-clinical samples and are brief for ease of participation, and Gloster et al. (Reference Gloster, Block, Klotsche, Villanueva, Rinner, Benoy and Bader2021) provides one recent example with their Psy-Flex measure.
Aims and hypotheses
The current project is a cross-sectional pilot examination of university undergraduate, graduate and law students who endorse non-White racial identities. It aimed to provide initial research on cognitive mechanisms, specifically psychological flexibility, explaining OCS above and beyond the experience of racial microaggressions using the dimensional model of OCD with an exploration of symptom themes, and a recent conceptualization of psychological flexibility, to pave the way for future longitudinal studies.
Aim 1
Aim 1 was to examine whether or not experiences of racial microaggressions correlated with OCS, and whether or not psychological flexibility correlated with OCS. With baseline correlations we expected significant and moderate to strong relationships such that more experiences of racial microaggressions correlated with increased OCS, and higher psychological flexibility correlated with lower OCS.
Aim 2
Aim 2 was to examine how experiences of racial microaggressions explained OCS, and if psychological flexibility added to that explanation, while co-varying for psychological distress (depression and anxiety). Regression models to test this aim were only built if either total OCD symptoms or a specific symptom theme significantly correlated with experiences of racial microaggressions in Aim 1. Given we expected experiences of racial microaggressions to relate to all total/symptom themes for OCS, we expected this to be the case while co-varying for psychological distress, and we expected that psychological flexibility would significantly add to the models, such that both psychological distress and experiences of racial microaggressions would relate to more OCS, and psychological flexibility would have a negative influence on OCS.
Exploratory aims
Finally, the measure used in this study was a newer measure of psychological flexibility (Gloster et al., Reference Gloster, Block, Klotsche, Villanueva, Rinner, Benoy and Bader2021), with little research examining its use with OCD, let alone in conjunction with experiences of microaggression. Also, the measure has items that relate to specific components of psychological flexibility in addition to its total score. Therefore we examined correlations between components of the psychological flexibility measure and experiences of microaggressions and OCS.
Hypotheses
Generally, we expected (based on past research) that experiences of racial microaggressions would relate positively and strongly to total OCS and all subscales in addition to psychological distress (Angelakis and Pseftogianni, Reference Angelakis and Pseftogianni2021; Williams et al., Reference Williams, Taylor, Mouzon, Oshin, Himle and Chatters2017b). Additionally, we expected that psychological flexibility would also relate strongly, but negatively, to total OCS, all OCS subscales, and experiences of racial microaggressions, with specific contributions to explaining OCS (Angelakis and Pseftogianni, Reference Angelakis and Pseftogianni2021; Wadsworth et al., Reference Wadsworth, Potluri, Schreck and Hernandez-Vallant2020; Williams et al., Reference Williams, Taylor, Mouzon, Oshin, Himle and Chatters2017b).
Method
Procedure
All participants consented to participate in the study and were entered in a gift card raffle if they completed 75% of the survey. The survey took place on the Qualtrics platform (Qualtrics, Provo, UT) (Qualtrics, 2023).
The sample was recruited through dissemination over email to students who are 18 years or older in total at a mid-sized public northeastern university in the United States. The university has degrees/majors organized into colleges of business, engineering, nursing/health sciences, arts and sciences, online and continuing education, law, marine science, and an honours college. Students are accounted for by the university for communication and reporting purposes based on if they are in an undergraduate degree program or graduate degree program in any of the colleges besides law, or if they are a student at the law school. Thus to sample from the entire population of adult students, students were recruited through an email to 7327 students in total, including 5397 undergraduate, 1571 graduate, and 359 law level students.
Data in this manuscript are from participants who completed baseline data during October 2022 as part of a larger longitudinal study where they filled out questionnaires twice, one month a part. They received reminders about the study two weeks apart in October on the email list with encouragement to participate. All emails about the study were sent in messages from the research team that were facilitated by the coordinator for institutional enrolment. Of note, this study had originally planned to also collect data through social media recruitment but ran into many obstacles with data quality and bots; this sample is described in a manuscript in progress by the authors, as well as reflections on data integrity in internet research.
Participants
Participants included 77 undergraduate, graduate and law students from one university. Participants were on average 24.78 years of age (SD=8.26). In terms of education, students were asked to report the highest level of education they had completed, out of high school, some college, associate’s degree, bachelor’s degree, some graduate courses, master’s degree, and doctoral/professional degree. Frequencies in each category are reported in Table 1. Overall, 49 participants (63.6%) of participants reported having completed high school, or some college courses, reflecting that most of the study participants were current undergraduate students spread across progress in their degree. Participants who reported some college or more education were asked if they studied psychology, and most were not a psychology major (n=39, 69.6%).
Table 1. Sociodemographic characteristics of participants at baseline and descriptive statistics
a Scores came from total and subscales of the DOCS.
b Microaggressions measured by the Daily Life Experience–Short Form.
c Psychology major statistics reflect out of those who reported some college and above.
In total, 37 participants (48.1%) identified as male, 36 (46.8%) identified as female, and four (5.2%) identified as trans, gender non-conforming, genderfluid, non-binary, or did not answer. For sexual orientation, 56 (72.7%) identified as heterosexual, 15 (20.8%) as bisexual and two (2.6%) responded that their option was not listed. Participants could endorse more than one racial identity, but all had at least one identity other than White. Endorsed racial identities included: White (n=16, 20.8%), Black (n=31, 40.3%), American Indian/Alaska Native (n=3, 3.9%), Asian (n=31, 40.3%), Middle Eastern/North African (n=6, 7.8%), multi-racial (n=8, 10.4%), and bi-racial (n=9, 11.7%). There was a diverse range of languages with the majority having a second language (n=53, 68.8%). Overall, 17 participants (22.1%) identified as Hispanic/Latino, and roughly 20% felt marginalized in their community as a result of one or more of their identities (n=17, 22.1%). Most participants were not currently in therapy (n=60, 77.9%) but the majority had been diagnosed with at least one psychiatric diagnosis (n=24, 68.8%).
The university student body has about 35% of students who are people of colour, 50% first-generation college students, and 75% of students who receive financial aid (UMass Dartmouth, n.d.).
Measures
Experience of racial microaggressions were measured using the Daily Life Experiences-Frequency Scale (DLEFS), an 18-question scale with a total sum score that assesses past frequency of occurrence of daily discrimination events (Harrell et al., Reference Harrell, Merchant and Young1997; Lee et al., Reference Lee, Gaskin-Wasson, Jones, Harrell, Banks, Kohn-Wood, Sellers and Neblett2021). Cronbach’s α was .951.
OCS were measured using the Dimensional Obsessive-Compulsive Scale (DOCS), a 20-item dimensional measure that offers a total symptom score as well as assessment of four distinct themes: contamination, responsibility for harm and mistakes, incompleteness/asymmetry, and taboo thoughts (Abramowitz et al., Reference Abramowitz, Deacon, Olatunji, Wheaton, Berman, Losardo, Timpano, McGrath, Riemann, Adams, Björgvinsson, Storch and Hale2010). Cronbach’s α for the Overall Obsessive Compulsive Symptoms (OCS) was .939, with individual Cronbach’s α scores of Unacceptable Thoughts OCS being .981, Responsibility for Harm OCS Cronbach’s α being .851, Contamination OCS being .851, and a Cronbach’s α of .872 for Symmetry OCS.
To measure psychological flexibility, the Psy-Flex measure was used, a 6-item measure of the presence of psychological flexibility in the past week according to each of the six components (mindfulness, acceptance, cognitive defusion, self as context, awareness of values and committed action; Gloster et al., Reference Gloster, Block, Klotsche, Villanueva, Rinner, Benoy and Bader2021). We utilized the sum of all items, which is the validated way to score the measure (Gloster et al., Reference Gloster, Block, Klotsche, Villanueva, Rinner, Benoy and Bader2021) and examined item scores in exploratory analyses given their additional usefulness in treatment planning as per Gloster et al. (Reference Gloster, Block, Klotsche, Villanueva, Rinner, Benoy and Bader2021). Cronbach’s α was .735.
For psychological distress, we utilized a combined measure of anxiety and depression, the Patient Health Questionnaire-4, (PHQ-4) (Kroenke et al., Reference Kroenke, Spitzer, Williams and Löwe2009). This measure is a well-validated screener in clinical and non-clinical populations combining two other well-validated questionnaires for depression and anxiety that refers to the last two weeks with two items about depression and two items about anxiety assessing main overall criteria; there is a clinical cut-off of 3 on both anxiety and depression subscales (Kroenke et al., Reference Kroenke, Spitzer, Williams and Löwe2009; Kroenke et al., Reference Kroenke, Spitzer, Williams and Löwe2010; Löwe et al., Reference Löwe, Wahl, Rose, Spitzer, Glaesmer, Wingenfeld and Brähler2010). Cronbach’s α was .893.
Data quality management plan
Steps outlined in this study were designed based on the study team’s experience and consultations with colleagues, but after a retrospective literature review, many strategies did align with those recommended (Griffin et al., Reference Griffin, Martino, LoSchiavo, Comer-Carruthers, Krause, Stults and Halkitis2022; Pozzar et al., Reference Pozzar, Hammer, Underhill-Blazey, Wright, Tulsky, Hong and Berry2020). Age was asked twice, there were three attention-check multiple choice questions phrased in the language of the survey they were embedded in, participants were asked their favourite animal and color, they were asked to provide their emails, and to provide any feedback they had. Additionally, they had to be over 18 and endorse at least one non-White racial identity to be eligible for the study. Participants were told ahead of time that they needed to answer the multiple-choice attention check questions correctly and complete 75% of the questionnaire to be entered in the gift card raffle.
Overall, 178 people opened the survey. Nobody under age 18 opened the survey. In total, one person was screened as spam, two did not consent, and 41 people only endorsed a White identity. This left 134 participants eligible, and of those, 77 completed all three multiple choice data checks correctly (57%). They all entered their age the same at both points it was asked, and although some did not have a colour or animal in the validity check question, given that their answers related to standard English conventions and related to the question, and given that they passed other checks, their answers were kept.
Analytical plan
All analyses were completed with IBM SPSS, version 28 (IBM, 2021).
Aim 1
Aim 1 examined correlational analyses between experiences of racial microaggressions and OCD symptoms and subscales, and between psychological flexibility and total OCS and subscales. Additional correlations between psychological distress (combined depression and anxiety) and all variables, and between experiences of racial microaggressions and psychological flexibility were examined for additional descriptives, and to account for any multicollinearity.
Aim 2
Aim 2 used regression analyses with a combined anxiety and distress score in the first step, and then experience of microaggressions and psychological flexibility entered in steps two and three. Models were only built for that OCD subscale if experience of microaggressions correlated with it significantly.
Exploratory analyses
Exploratory correlations included relationships between psychological flexibility sub-processes and experience of microaggressions and OCD symptoms.
Power analysis
The largest possible cross-sectional model had three predictors (combined depression/anxiety, experiences of microaggressions, and psychological flexibility). We utilized G*Power (Faul et al., Reference Faul, Erdfelder, Lang and Buchner2007) to estimate sample size for desired power with an a priori power analysis for linear regression and F-tests, using a past effect size found by Willis and Neblett (Reference Willis and Neblett2018), a medium effect size relationship between frequency of discrimination and OCD symptoms, and we used .8 for power and a .05 alpha level. These metrics and guidelines for power analysis align with Field (Reference Field2013) and Kang (Reference Kang2021). To detect a medium effect size with three predictors, and .05 alpha level, and a .8/80% power which is often considered to be sufficient for a medium effect size (Cohen, Reference Cohen1988), this necessitated 77 participants. A .7 power necessitated 63 participants.
Results
Descriptive characteristics
Psychological distress scores neared clinical cut-offs for depression and anxiety (Kroenke et al., Reference Kroenke, Spitzer, Williams and Löwe2009). The sample’s mean OCS score (M=18.88, SD=13.90) fell between that of patients with OCD and those with other anxiety disorders, and it was above the clinical cut-off score of 18 (Abramowitz et al., Reference Abramowitz, Deacon, Olatunji, Wheaton, Berman, Losardo, Timpano, McGrath, Riemann, Adams, Björgvinsson, Storch and Hale2010). Mean microaggression scores neared 30 (M=29.63, SD=20.51), suggesting that on average participants experienced each occurrence once to a few times (Harrell et al., Reference Harrell, Merchant and Young1997). See Table 1 for more details on the sample.
Missing data
Within the sample included in analyses, there were no missing data.
Aim 1
According to Cohen’s (Reference Cohen1988) correlation guidelines (.1=weak, .3=moderate and .5=strong), experience of racial microaggressions had significant positive near moderate to moderate correlations with responsibility for harm, contamination, symmetry, total OCS, and psychological distress. Experiences of racial microaggressions did not correlate with psychological flexibility. Psychological flexibility significantly, negatively, and near moderately correlated with total OCS, moderately with unacceptable thoughts symptoms, and significantly, negatively, and near strongly correlated with psychological distress. Psychological distress moderately to strongly positively correlated with total OCS and all symptom dimensions. See Table 2 for the correlations.
Table 2. Correlational analyses between OCD symptoms, experience of racial microaggressions, and psychological flexibility
a Scores came from total and subscales of the DOCS.
b Microaggressions measured by the Daily Life Experience–Short Form.
c Psychological distress measured by the PHQ-4.
d Psychological flexibility measured by the Psy-Flex.
*p<.05; **p<.001; ***p<.001.
Aim 2
Experience of racial microaggressions significantly correlated with all OCD symptoms except for unacceptable thoughts, so four models were built.
For the total OCD model, psychological distress significantly explained 35.6% of the variance in OCS (F 1,75=41.38, p<.001). Experiences of racial microaggressions and psychological flexibility did not significantly add to the model, nor were they significant.
For the responsibility for harm model, while co-varying for psychological distress, experience of racial microaggressions significantly explained 34.3% of the variance in OCS (F 2,74=19.34, p<.001) and experience of racial microaggressions significantly added to the previous model with just psychological distress (F 1,74=4.21, p=.044, ΔR 2=.04).
For the symmetry model, psychological distress explained 17.6% of the variance in OCS, (F 1,75=16.02, p<.001). Adding experiences of racial microaggressions did not significantly add to the model and they were not significant, and the same was true for psychological flexibility.
For the contamination model, psychological distress explained 18.8% of the variance in OCS (F 1,75=17.41, p<.001). Adding experiences of racial microaggressions significantly added to the model (F 1,74=5.99, p=.018, ΔR 2=.06) so that both experiences of racial microaggressions and psychological distress significantly explained 24.8% of the variance in contamination OCS (F 2,74=12.2, p<.001). Psychological flexibility did not significantly add to the model after that, and was not significant.
See Table 3 for regression results.
Table 3. Experience of racial microaggressions, psychological flexibility, and OCD symptoms
a OC symptoms and total measured by the DOCS.
b Experience of racial microaggressions measured by DLEFS scale.
c Psychological flexibility measured by the Psy-Flex.
d Psychological distress measured by the PHQ-4.
*p<.05; **p<0.01; ***p<0.001.
Exploratory analyses
Exploratory correlations demonstrated that experiences of racial microaggressions significantly and near moderately were positively correlated with cognitive defusion. For OCD symptoms, contamination symptoms and responsibility for harm symptoms did not correlate with any psychological flexibility items, but total OCS, unacceptable thoughts and symmetry symptoms had varied patterns of correlations with mindfulness, cognitive defusion, acceptance, and awareness of values. See Table 4 for the exploratory correlations.
Table 4. Exploratory correlations between OCD symptoms, psychological distress, and psychological flexibility components
*p<.05; **p<0.01; ***p<0.001.
Discussion
Aim 1
Our first aim involved baseline correlations to establish in this sample what OCS domains correlated with experiences of racial microaggressions and psychological flexibility. We expected more experiences of racial microaggressions to be correlated with increased OCS, and higher psychological flexibility correlated with lower OCS. We found that more experiences of racial microaggressions were correlated with higher responsibility for harm, contamination, symmetry, and total OCS, and not with unacceptable thoughts OCS. Study results demonstrate the link between OCS and experience of racial microaggressions, which may be due to racial battle fatigue, vigilance, or depleted cognitive resources as Williams et al. (Reference Williams, Taylor, Mouzon, Oshin, Himle and Chatters2017b) suggest. This result generally aligns with research demonstrating variable patterns of symptom levels in various symptom themes within various racial and ethnic identities (Williams et al., Reference Williams, Taylor, George, Schlaudt, Ifatunji and Chatters2021).
Interestingly, while experiences of racial microaggressions did not correlate with unacceptable thoughts/taboo thoughts OCS, psychological flexibility did, so perhaps those symptoms, such as fears of thinking or carrying out harmful or socially taboo actions, might relate more to levels of acceptance and awareness of naturally occurring internal thoughts, versus more potentially external behaviours such as contamination or symmetry rituals that may develop from coping with the trauma of racism.
Aim 2
Aim 2 was to examine specifics of how or if experiences of racial microaggressions and psychological flexibility explain OCS while accounting for depression and anxiety. We expected both things to occur for all OCS symptoms. Results showed that for total OCS and for symmetry symptoms, experiences of racial microaggressions and psychological flexibility did not add to the explanation beyond depression and anxiety. For responsibility for harm symptoms, experiences of racial microaggressions added to the explanation beyond depression and anxiety. These symptoms relate to thoughts and actions surrounding preventing disasters or accidents and other harm (Abramowitz et al., Reference Abramowitz, Deacon, Olatunji, Wheaton, Berman, Losardo, Timpano, McGrath, Riemann, Adams, Björgvinsson, Storch and Hale2010). For contamination OCS, similarly experiences of racial microaggressions significantly added to the explanation of contamination OCS beyond depression and anxiety. These symptoms surround washing, cleaning, checking, and thoughts of illness or uncleanliness (Abramowitz et al., Reference Abramowitz, Deacon, Olatunji, Wheaton, Berman, Losardo, Timpano, McGrath, Riemann, Adams, Björgvinsson, Storch and Hale2010).
Exploratory analyses
Exploratory analyses sought to further examine components of psychological flexibility from the Psy-Flex measure given its novelty, and a need to better understand processes for mental health in racially minoritized young adults as they cope with racism. Correlational results with experiences of racial microaggressions and psychological flexibility sub-components demonstrated a positive relationship with cognitive defusion, being able to look at challenging thoughts from a distance. This may demonstrate that cognitive defusion is adaptive for people experiencing racism, and future work could examine the temporal nature of the relationship, and if added intervention and prevention can bolster the strategy. For total OCS, unacceptable thoughts, and symmetry OCS, there were some correlations with mindfulness, acceptance, and cognitive defusion, but consistent relationships with mindfulness. These were all negative. This suggests that specifically bolstering mindfulness skills (or being able to focus on what is important in the present moment in relation to the wording of the Psy-Flex mindfulness question), may be important for supporting the mental health of racially minoritized students. Martinez et al. (Reference Martinez, Suyemoto, Abdullah, Burnett-Zeigler and Roemer2022b) and their review adds support to this result and other future research directions.
Limitations and future directions
The authors would like to acknowledge our positionality, and that we bring our own experiences, privileges, potential biases, and potential gaps in understanding on the topic to this writing given our own experiences within our own gender and racial identities. One example is when a participant expressed concern that the study was being carried out to find a way to diagnose Black people instead of working on the large challenge of racism. In the future, the study team may adjust the advertisement to acknowledge that social and cultural change needs to occur alongside work to develop strengths-based support for people. Along these lines we want to explicitly state that while it is important to engage in research and clinical work to support individuals and their experiences, it is also essential to acknowledge that societal and structural changes must go hand-in-hand with research on individuals (Steele, Reference Steele2020).
As sampling occurred through email dissemination to all 7327 students of one university, the university had a large audience but generalizability might be limited outside of similar contexts to the university (small, public, northeastern United States).
The sample in the current study was adequately powered to detect a medium effect size in a linear regression with three predictors. It is important to consider in this study that in reality, we were testing two hypotheses, about experience of racial microaggressions, and psychological flexibility, and there were other tests of correlations occurring in the study, so analyses could be under-powered or Type I error could have occurred. We attempted to alleviate some concerns of Type I error by separating main hypotheses and exploratory analyses, and by ensuring that our sample size was powered for our main hypothesis, testing the relationship between experiences of racial microaggressions and OCD symptoms. In the future, replications after this current pilot study should occur with lowered alpha levels according to the amount of tests, larger sample sizes, and further examination of effect sizes of relationships found. Current results should be interpreted with caution.
Upon reflection of racial data in the current study compared with the rest of the United States, it appears that the current study accurately represents Black participants, as well as Hispanic, and Native American/Alaska Indian (US Census Bureau, 2020). There were no Hawaian participants in the current study. The current study had an over-representation of Asian-identifying people, as well as middle eastern/north African, and multi-racial and bi-racial people (US Census Bureau, 2020). This could be due to recent findings that higher percentages of Asian students were enrolled in college compared with White students (NCES, 2019), reports that demographic data on Arab-American people is often undercounted in large surveys (AAI, 2021), and as we provided students an option to self-identify as multi-racial or bi-racial compared with just calculating that afterwards based on selections of other races. In terms of socioeconomic data, we did not collect relevant data for participants, but the university sampled from has high percentages of first-generation students and students who utilize financial aid. This means study results may best represent lower-income adults and young adults with less college education in their family background. Future research could assess participants and their families’ immigrant status to assess the amount of time embedded in the culture of the United States related to experiences of racial microaggressions and their impacts.
The sample in this study included college students, who are a viable option for studying OCD (Abramowitz et al., Reference Abramowitz, Deacon, Olatunji, Wheaton, Berman, Losardo, Timpano, McGrath, Riemann, Adams, Björgvinsson, Storch and Hale2010); however, degree programs/subjects may be a confound, such as for students seeking law degrees, and the current study only assessed the amount of education completed, not specific subjects studied. Future research may recruit participants from a range of academic backgrounds, or focus on clinical samples to provide more specific conclusions about OCD rather than general psychological distress.
This study utilized a measurement of retrospective reported frequency of experiencing microaggressions. For some students, it may be that these events actually happen, and they are accurately reporting. For others, they may have a bias to perceive or notice such slights if they have had them occur in the past, or due to other challenges with being a racially minoritized person in US society or due to other aspects of their identity such as gender and sexual orientation. Therefore, further, more nuanced measurement and conceptualization is needed to best account for peoples’ experiences and their impacts on mental health (Williams et al., Reference Williams, Osman and Hyon2023).
Beyond just better accounting for cognitive and psychological components, other information from qualitative analysis and other ecological levels could support conceptualizing experiences of racism such as through examining other forms of racism, intersectionality, and the societal context such as COVID-19 (Neblett, Reference Neblett2019; Spanierman et al., Reference Spanierman, Clark and Kim2021). Anecdotally, in responses to text feedback questions, participants frequently reported appreciation of the survey topics, desiring to know study results, valuing self-reflection, and sharing other anecdotes of experiences of racism, supporting the usefulness of future qualitative research. Continued opportunities for mixed-methods research that is safe and provides space for peoples’ voices to be heard is vital as this research area continues (Bybee et al., Reference Bybee, Cloyes, Baucom, Supiano, Mooney and Ellington2021; Shelton et al., Reference Shelton, Adsul and Oh2021). Future research must meaningfully engage those with lived experience throughout all parts of the process including through use of qualitative measures (Neblett, Reference Neblett2019; Spanierman et al., Reference Spanierman, Clark and Kim2021). Doing so in a way that provides frameworks, is respectful of time, energy, and labour, but also is open enough to hear novel ideas is important.
The analyses utilized data that controlled for depression and anxiety as well. However, future studies should examine more co-variates such as age, gender, and specific racial identity, and should utilize modelling techniques that better account for the nature of the relationships and possible variation between and within groups across time. This study was composed of people with racially marginalized identities, but did not have enough participants to examine results according to specific identities.
Of note, this study’s data were collected in the autumn of 2022, therefore is in the context of another part of coping with the COVID-19 pandemic, which might have brought internalizing symptoms to the forefront, but greater understanding of vulnerability factors that differentiate increases in OCS can bolster resilience and prevention efforts (Grant et al., Reference Grant, Drummond, Nicholson, Fagan, Baldwin, Fineberg and Chamberlain2022; Wheaton et al., Reference Wheaton, Abramowitz, Berman, Fabricant and Olatunji2011). This study focused on the role of experiencing racial microaggressions, so future work should explore the roles of other aspects of culture (Williams et al., Reference Williams, Chapman, Simms and Tellawi2017a; Willis and Neblett, Reference Willis and Neblett2018). This work should continue in tandem with work exploring the relationship between trauma and aetiology and onset of OCS by examining not just right experiences, and directly compared to indirectly experienced trauma, and with future directions such as those described by Pinciotti and Fisher (Reference Pinciotti and Fisher2022). This trauma and OCD work can be conducted with populations with diverse racial identities, with specific examinations of racism as a trauma, or both.
Future work should examine the role of various conceptualizations of stigma in this OCD and microaggressions relationship given stigma’s relationship to quality of life in people with OCD (Kılıç et al., Reference Kılıç, Görmez, Elbay and Özer2022<6>), the higher endorsements of stigma and shame were associated with family disapproval in ethnic minority people with OCD (Glazier et al., Reference Glazier, Wetterneck, Singh and Williams2015), and the potential relevance of Corrigan et al.’s (Reference Corrigan, Larson and Rüsch2009) ‘Why try?’ effect for stigma implications. Longitudinal studies or experiential sampling may support more accurate and timely recording of experiences.
Implications for clinical practice
This study did have null results for psychological flexibility to explain OCD in the context of experiences of racial microaggressions, but it might be that psychological flexibility acts more clearly on enhancing wellbeing compared with reducing distress/psychopathology (Williams et al., Reference Williams, Botanov, Giovanetti, Perko, Sutherland, Youngren and Sakaluk2022a). Additionally, OCD has a clear previously established impact on quality of life and functioning ( Macy et al., Reference Macy, Theo, Kaufmann, Ghazzaoui, Pawlowski, Fakhry, Cassmassi and IsHak2013) as does psychological flexibility (Kashdan and Rottenberg, Reference Kashdan and Rottenberg2010). There is strong support in the literature for mindfulness, values, and acceptance-based therapies, such as ACT, to support those experiencing microaggressions in a culturally humble way by strengthening psychological flexibility, reflective coping skills, and resilience (Martinez et al., Reference Martinez, Suyemoto, Abdullah, Burnett-Zeigler and Roemer2022b; Neblett, Reference Neblett2019; Spanierman et al., Reference Spanierman, Clark and Kim2021).
Recently, many therapists have developed and adapted clinical frameworks to support people experiencing racism. A framework such as that of ACT that can be incorporated as therapists work to help clients facilitate cultural identity and social support to protect against racial trauma (Liu and Modir, Reference Liu and Modir2020) and emotion regulation strategies can be facilitated to help clients flexibly cope with experiences of racism (Graham et al., Reference Graham, Calloway and Roemer2015). There are numerous recent clinical guidelines and models for supporting people of colour with OCD that could be incorporated, adapted, and studied further (DeLapp and Gallo, Reference DeLapp and Gallo2022; Steele, Reference Steele2020; Williams et al., Reference Williams, Rouleau, La Torre and Sharif2020; William et al., Reference Williams, Holmes, Zare, Haeny and Faber2022b). Future research should examine how to adapt such protocols and treatments for various settings and types of providers given the importance of finding multi-pronged solutions for mental health (Kazdin, Reference Kazdin2022) and brief single session interventions delivered within community specific settings could be an option as well with frameworks such as those described by Schleider et al. (Reference Schleider, Dobias, Sung and Mullarkey2020).
Conclusion
This project advances understanding of the links between experiencing racial microaggressions and OCD, as well as the influence of large-scale social stressors on OCD symptoms. It improves our understanding of racial microaggressions and their contribution to OCD symptomatology with tentative additional support for this relationship, and a need for larger sample sizes and replications. It also suggests a role that psychological flexibility may have in shaping future strengths-oriented interventions that strive to both improve well-being and reduce psychopathology through summarizing the literature and exploratory analyses. However, a lack of results in the current study from main analyses on psychological flexibility, and the potential analyses being under-powered suggest a need for future study on psychological flexibility, OCS, and racism.
Data availability statement
The data that support the findings of this study are available from the corresponding author, E.L.-R., upon reasonable request.
Acknowledgements
We would like to acknowledge Dr Andrew Guzick for his thoughtful feedback on a draft of the manuscript.
Author contributions
Morgan Browning: Conceptualization (lead), Data curation (lead), Formal analysis (lead), Investigation (lead), Methodology (lead), Project administration (lead), Writing – original draft (lead), Writing – review & editing (equal); Elizabeth Lloyd Richardson: Conceptualization (supporting), Formal analysis (supporting), Investigation (supporting), Methodology (supporting), Project administration (supporting), Supervision (lead), Writing – review & editing (lead); Sidney Satterfield: Data curation (supporting), Formal analysis (supporting), Writing – review & editing (supporting); Akshay Trisal: Conceptualization (supporting), Methodology (supporting), Writing – original draft (supporting).
Financial support
This research received no specific grant from any funding agency, commercial or not-for-profit sectors.
Competing interests
The authors declare none.
Ethical standards
The authors have abided by the Ethical Principles of Psychologists and Code of Conduct as set out by the APA (http://www.apa.org/ethics/code/) which align with the Ethical Principles of Psychologists and Code of Conduct as set out by the BABCP and BPS. This study was approved by the university’s Institutional Review Board (approval number 21.056).
Comments
No Comments have been published for this article.