Social discrimination is consistently shown to be associated with poor physical and mental health and health-seeking behaviors and outcomes (Banton, 1998; Byrd & Clayton, 2000; D’Anna, Ponce, & Siegel, 2010; Krieger, 2000; Krieger, Smith, Naishadham, Hartman, & Barbeau, 2005; Krieger & Sydney, 1996; Schulz et al., 2006; Williams, Yan, Jackson, & Anderson, 1997; Willie, Kramer, & Brown, 1973). Although the majority of research on social discrimination has been conducted among African Americans (Cain & Kington, 2003), there is evidence suggesting that being a member of a stigmatized social group defined by one’s age, gender, social position and, most relevant to the present study, sexual minority identity (Ayala, Beck, Lauer, Reynolds, & Sundararaj, 2010; Mays & Cochran, 2001; McCabe, Bostwick, Hughes, West, & Boyd, 2010; Mills et al., 2004), can be associated with negative health outcomes and utilization of health services (Karlsen & Nazroo, 2002; Trivedi & Ayanian, 2006).

Social discrimination has been classified into two types: (a) structural discrimination, which refers to social structures and policies that determine access to goods and services for certain groups, and (b) interpersonal discrimination that is directed toward and perceived by individuals (Jones, 2000). Interpersonal discrimination is manifest through behaviors such as mean, unfair, or unequal treatment which may include verbal or physical abuse that is intended to marginalize individuals or communities based on their affiliation with stigmatized opinions, behaviors, or individuals (Ayala, et al., 2010). Interpersonal social discrimination based on sexual orientation or “sexual prejudice” is defined as negative attitudes, or feelings of hostility or dislike directed at individuals because of their sexual “orientation” or identity (Herek, 2000). Individuals who identify as lesbian, gay, or bisexual (LGB) bear the overwhelming burden of this type of discrimination (Ahmed & Hammarstedt, 2009).

Uncovering the prevalence of social discrimination targeted at LGB individuals is an important endeavor for several reasons. First, perceived discrimination has been found to be negatively associated with the use of preventive health services among LGB populations (Ayala, et al., 2010; Bakker, Sandfort, Vanwesenbeeck, van Lindert, & Westert, 2006; O’Hanlan, Dibble, Hagan, & Davids, 2004). LGB individuals may be less likely to obtain preventive health care due to: (1) high rates of being uninsured, particularly among lesbians and bisexual women, and transgender people (Johnson, Mimiaga, & Bradford, 2008; Valanis et al., 2000); (2) feelings that their health provider is not “LGB-friendly” (Beckerman & Fontana, 2009; Bjorkman & Malterud, 2009); (3) lack of trust that their provider is competent or knowledgeable about LGB health-related practices, needs or issues (Ayala, et al., 2010; N. F. Sanchez, Rabatin, Sanchez, Hubbard, & Kalet, 2006); and (4) victimization or negative experiences with health care providers resulting from homophobia, heterosexism, or ignorance (Willging, Salvador, & Kano, 2006). Second, among those who seek health care, an antagonistic environment may cause individuals to hide their sexual orientation, making it extremely difficult to obtain the necessary health services and information specific to their needs (Ayala, et al., 2010; Dean et al., 2000; Meyer & Northridge, 2007; J. P. Sanchez, Hailpern, Lowe, & Calderon, 2007; Willging, et al., 2006). Third, discrimination, prejudice, and verbal harassment are associated with scores of negative health outcomes, including risk of HIV infection among young gay men, other sexually transmitted infections (STIs) in gay, lesbian, and bisexual individuals of both genders (Diaz, Ayala, & Bein, 2004; Krieger, 1999), eating disorders (Coker, Austin, & Schuster, 2010), low self-esteem, depression, suicide and self-harm (Coia et al., 2002; Meyer & Northridge, 2007), alcohol and drug dependence (Cochran & Mays, 2000; Meyer & Northridge, 2007), and homelessness and prostitution among LGB youth (Coker, et al., 2010). Discrimination and prejudice are also related to psychiatric disorders, including major depression, simple phobia, and posttraumatic stress disorder, among lesbians and bisexual women (Cochran & Mays, 2000; Grella, Greenwell, Mays, & Cochran, 2009), and among other LGB populations (Cochran, Mays, Alegria, Ortega, & Takeuchi, 2007; Cochran, Mays, & Sullivan, 2003; Mills, et al., 2004). Discrimination and verbal harassment are also associated with trauma and violence against gay men and lesbians (Herek, 2009; Meyer & Northridge, 2007). Finally, harassment, ridicule, rejection and violence are associated with other health-eroding behaviors among men who have sex with men (MSM), such as concealing relationships, identities, behaviors, and feelings, and detaching from social support or other important resources that may have the potential to improve their general health, opportunities and quality of life (Ayala, et al., 2010; Cole, Kemeny, Taylor, & Visscher, 1996; Diaz, et al., 2004; Koblin et al., 2006).

Over the past two decades, initiatives to increase health equity across all groups have been launched nation wide, and attention has turned to uncovering the prevalence of, and factors associated with, discrimination in United States health care settings (Epstein, Taylor, & Seage, 1985; Hooper, Comstock, Goodwin, & Goodwin, 1982; Roter, Hall, & Katz, 1988; Smedley, Stith, & Nelson, 2003; van Ryn, 2002; Waitzkin, 1985). The bulk of the research examining discrimination in health care settings has focused on medical provider perceptions and behaviors including racially and ethnically inequitable medical testing, referrals, treatment, medical procedures and pain management (Smedley, et al., 2003). As such, little is understood about patient perceptions of discriminatory events or interactions that may lead to psychological or physical stress responses including heath care avoidance or other poor health-related behaviors (Klonoff, 2009).

Historically, LGB persons have been underrepresented in the public health literature (Boehmer, 2002; Lee, 2000; Sell & Becker, 2001). In the 1980s and 1990s, the majority of articles published including LGB persons examined STDs, and specifically HIV/AIDS, with an overwhelming focus on gay and bisexual men (Boehmer, 2002). Some attribute the neglect of LGB individuals in research to the estimation that LBG-identified individuals comprise a relatively small proportion of the overall U.S. population and that it is difficult to fund and conduct population-based studies that are large enough to incorporate a representative sample (Hughes & Eliason, 2002). However, others have noted that the fear of stigmatization is a barrier to LGB individuals seeking health care services and participating in health-related research (Lee, 2000), and that the lack of available scientific data on the health of LGB contributes to the lack of prioritization of LGB health issues (Sell & Becker, 2001).

Previous investigations that have included LGB individuals have noted higher utilization of certain types of health services (mental health, substance abuse treatment, and emergency department services) (Grella, et al., 2009; J. P. Sanchez, et al., 2007), but lower use of other services (routine primary healthcare, mammograms, Pap tests) by various groups who identify as LGB (Buchmueller & Carpenter, 2010) (O’Hanlan, et al., 2004), lesbians who identify as “butch” as opposed to “femme” (Hiestand, Horne, & Levitt, 2007), and MSM who identify as heterosexual (Merighi, Chassler, Lundgren, & Inniss, 2010). However, little research has been conducted on the relationships between social discrimination and health, and health care-seeking behaviors within sexual minority groups in the United States (Hughes & Eliason, 2002). Previously published findings based on the Los Angeles Countywide Risk Assessment Survey (CRAS) demonstrated that those who used more health care services were more likely to be HIV positive, to have had a recent HIV test, and to have been the victim of physical partner abuse (Fisher et al., 2010).

In the present study, our research aims were as follows: (a) we examined the prevalence of sexual orientation-based discrimination (in the form of verbal harassment) in a racially and ethnically diverse sample of adult recipients of Los Angeles County Department of Health Services, Office of AIDS Programs and Policy (OAPP)-sponsored health and social services; and (b) we explored possible associations between reported lifetime experiences of sexual orientation-based discrimination (i.e., verbal harassment) and reported utilization of health services across levels of sexual orientation.

Data were collected in May and June of 2004 by 220 HIV prevention service provider staff from fifty-one OAPP-funded agencies located throughout greater Los Angeles County, a 4,000 square mile area composed of urban, suburban and rural areas. This survey utilized a two-tiered sampling method, which consisted of a combination of stratified and systematic sampling. Stratified sampling was employed because it was believed that there were differences in client characteristics among the different agencies. Systematic sampling was used to facilitate a representative sample of participants (Fisher, et al., 2010). Each interviewer was given a number (n) after he/she completed six hours of training on the CRAS and was told to interview each nth client.

The CRAS consisted of four main sections with multiple questions about (a) clients’ demographic information, (b) their alcohol/drug risks (e.g., type of drugs used and needle use/sharing), (c) their sexual history (e.g., both primary and casual partners, methods of HIV protection [condoms], and sexual incentives), and (d) their use of HIV- and STD-prevention services as well as other services made available by a participating OAPP agency and/or within other L.A. County agencies (all within the past 6 months at the time of survey). Client perceptions concerning HIV and AIDS were also assessed. The survey required between 15 and 30 minutes to administer in a face-to-face interview. Participants were offered one of three incentives for completing the CRAS: 1) a non cash food voucher worth $10; 2) two movie tickets with an approximate value of $20; or a non cash gift card in the amount of $15. The data collection methods used for the CRAS have been previously described in detail (Edwards, Fisher, & Reynolds, 2007; Fisher, et al., 2010).

Several items from the 2004 CRAS were used to address the research questions in the present study. The two outcome variables included perceived verbal harassment because of sexual orientation and use of health services. For the first outcome variable, demographic variables (i.e., sex, sexual orientation, race/ethnicity, and education) served as the predicting variables.

Analyses were conducted using SAS version 9.2. Sample characteristics using proportions and standard errors are presented first. Second, bivariate analyses were conducted using point-biserial correlation analyses and Pearson chi-square tests of association to determine which predictors and covariates were significantly associated (separately) with being verbally harassed due to one’s sexual orientation.

To assess the relationships between being verbally harassed due to one’s sexual orientation and utilization of health services, mean differences in the utilization of overall health services provided by the OAPP-funded and other L.A. County agencies were examined between those who reported being verbally harassed and those who did not, using a series of independent, two-sided, Satterthwaite t-tests for unequal variances. The mean differences in usage for each of the four sub-dimensions of health services (sexual health treatment, social support, social services, and medical services) were also compared. These analyses were stratified by sexual orientation.

A total of 2,126 individuals with a mean age of 32.7 years participated in the CRAS survey (SD=11.02; range 12-69 years). The survey completion rate was 84.6%. Among those reporting their sex, 66.2% were men (n=1334), and 33.8% were women (n=682). The final analytic sample was 1,995 participants who had responded to the question about being harassed because of their sexual orientation; we excluded those who declared themselves as transgender or transsexual, and 21 additional participants because of missing data on some key measures. Table 2 summarizes the frequency of participants belonging to various sub-groups who experienced verbal harassment versus those who did not. Participants identified themselves as heterosexual (n=1091, 54.7%), gay (n=548, 27.5%), lesbian (n=36, 1.8%), bisexual (n=289, 14.5%) and other (n=31, 1.6%). The majority of participants had less than or equivalent to a high school education (65.8%) (Table 2). Hispanic Americans were the largest ethnic group (44.1%), followed by African Americans (27.0%), Caucasians (17.3%), American Indian/Native Hawaiian and mixed race/ethnicity (7.6%), and Asian Americans (4.0%). Finally, the majority of the participants (n=1553, 77.9%) was born in the United States.

Almost one-third (32.0%) of all participants (n=639) reported that they had experienced verbal harassment attributed to their sexual orientation; the majority of those who experienced this type of discrimination (80.3%) identified as LGB or other sexual orientation, although LGB comprised only 45.3% of the total sample. Additionally, the percentage of participants reporting verbal harassment increased at each level of education, with the exception of those who had achieved a college education or greater (Table 2).

Individuals who reported that their main partner, or any sex partner within the past 6 months, had HIV or AIDS were also more likely to report perceived verbal harassment compared to those who reported that their main partner did not have HIV or AIDS. A total of 324 respondents indicated that they had received a positive result on their last HIV test (16.2% of the total sample); 12.7% of all heterosexuals (n=94), 38.2% of gays (n=169), 14.8% of lesbians (n=4), 22.8% of bisexuals (n=51), and 28.6% of those who identified as being of other sexual orientation (n=6). HIV-positive respondents were more likely to report perceived verbal harassment than those who were HIV-negative. Additionally, having a sexual partner who had injected drugs since 1980 was associated with the experience of verbal harassment (Table 2).

The relationships between licit and illicit substance use and experiences of harassment were also explored (Table 2). Tobacco use in the past six months was associated with the experience of verbal harassment, as was the use of crack, crystal meth, and inhalants, poppers, nitrous oxide, whippets, glues, paint thinners, and aerosol propellants, and steroid use. However, heroin use in the past six months was not significantly associated with the experience of verbal harassment due to one’s sexual orientation.

The examination of the relationships between verbal harassment experience and the utilization of the overall health services at the interviewing agency, as well as the four aforementioned categories of health services, revealed that among all participants those who reported being verbally harassed due to their sexual orientation received significantly more health services overall, more social support, and more social services than those who were not verbally harassed. Table 3 provides details of these analyses.

A similar pattern of findings was found regarding health services received from other L.A. County agencies among all participants. In addition, the significant relationships of interest were also found for the category of medical services received outside of the interviewing agency (Table 4).

As evidenced above, the utilization of health and social services was associated with perceived verbal harassment. Further, t-test results showed that this relationship was modified by levels of sexual orientation. Specifically, among heterosexuals, those who experienced verbal harassment because of their sexual orientation used more overall health services from the OAPP-funded interviewing agency than heterosexuals who did not experience verbal harassment. The same was true for heterosexuals using OAPP-funded social support, social services and medical treatment (Table 3). Among gay men, those experiencing verbal harassment used social support and social services more often, where as participants who identified as bisexual and had experienced verbal harassment used social support more often (Table 3). Heterosexuals, gay men, and bisexuals who experienced verbal harassment used most services available through other L.A. County agencies more often than individuals who did not experience the same type of discrimination (Table 4). Due to small sample size, none of the relationships assessed were significant among lesbians, however, a greater number of all OAPP-funded interviewing agency services, sexual health treatment, social support and medical services, and a greater number of all services and sexual health treatment services within other L.A. County agencies were utilized by lesbians who experienced verbal harassment compared to those who did not.